1. Introduction

Reduced Emissions from Deforestation and Forest Degradation (REDD+)1 is a global program for disbursing funds, primarily to pay national government in developing countries, to reduce forest carbon emission (UN-REDD 2009, 4). The title sounds good. Nobody can disagree with reducing deforestation and degradation. But is a market-based approach the right mechanism, and what are the associated risks? REDD’s framers acknowledge that REDD risks ‘decoupling conservation from development’, enabling ‘powerful REDD consortia to deprive communities of their legitimate land-development aspirations’, undermining ‘hard-fought gains in forest management practices’, and eroding ‘culturally rooted not-for-profit conservation values’ (FAO et al. 2008, pp. 4–5).2

The framers view these risks as justified by ecological sustainability, ‘the potential to achieve significant sustainable development benefits for millions of people worldwide’ and to ‘help sustain or improve livelihoods and food security for local communities.’ In addition, they foresee that ‘a premium may be negotiable for emission reductions that generate additional benefits’ for local people. They even acknowledge ‘that REDD benefits in some circumstances may have to be traded off against other social, economic or environmental benefits’ and call for care in taking local place-based complexity into account when designing REDD interventions. (FAO et al. 2008, pp. 4–5).

What will prevent the promised ‘premium’ from being competed down to nothing, as is the tendency in any competitive market (economics 101)? Why won’t it be captured by intermediaries (Munden Project 2011)? Who will do the trading-off of REDD benefits? Isn’t the converse, local needs being traded off for REDD carbon benefits, more likely? These tradeoffs involve people’s lives and histories at the edge of the legal world. How will REDD proponents ensure that tradeoffs are just? How will REDD strategies take local peoples’ needs and aspirations into account? How will rights be established and enforced?

Social safeguards are, of course, being developed. In the United Nations Framework Convention on Climate Change (UNFCCC) 16th Conference of Parties (COP) meeting in Cancun in 2010, a series of principles were approved, including ‘transparent … national forest governance structures,’ ‘respect for the knowledge and rights of indigenous peoples and members of local communities’, and ‘the full and effective participation of relevant stakeholders, in particular indigenous peoples and local communities’ (FCCC/CP/2010/7/Add.1). Free Prior and Informed Consent (FPIC) is broadly assumed to be an integral part of such safeguards. (For a recent review of REDD safeguards see Rutt 2012.)

Again, these provisions raise new questions. When is ‘transparent’ accountable? We see in many places transparent abuse and corruption (visible to everyone who cares to look). These are not corrected by ‘transparency’. Transparency only leads to accountability when there are sanctions – shame is not enough for the shameless (Fox 2007). Who will determine who holds stakes and which stakes are relevant? We often see stakeholder processes where thousands of villagers have one representative, NGOs have one or more (since there may be more than one NGO, even in only one village), forestry merchants have a representative, donors are represented, the government is represented and then they all vote. When citizen voices are represented as one vote among multiple ‘stakeholders’, this is not democracy. Indeed the term ‘stakeholder’ gives license to whoever organizes a stakeholder process to decide who participates. The term ‘stakeholder’ should be banished.

Further, what is ‘full and effective participation’? Is it the kind of participation we typically see in which local people voice their needs and someone else decides what to do, or whether to do anything (Cooke and Kothari 2001; Baviskar 2005)? And shouldn’t the use of FPIC be mandatory? By international law it is only required for indigenous communities, but what about all other communities, or those that fail to fit the formal definition of ‘indigenous’? Finally, who represents local communities in FPIC? Who provides consent ‘for’ the community? Is it a leader who is accountable to the people? Is it a dictator who inherited a ‘traditional’ post? Who determines whose consent is valid? Guidelines and protections are needed to ensure that these processes are not a representation charade – a performance of democracy and justice in their total absence.

The proposed safeguard principles include some excellent provisions – like FPIC. But these protections still need to be mandatory, universal and defined well enough to be consistently and meaningfully implemented. In practice, local representation has been tried many times, and the results have been less than stellar (Lemos and Agrawal 2006; Tacconi et al. 2006; Lund et al. 2009; Larson et al. 2010; Ribot et al. 2010). The forestry and conservation institutions that are asked to apply them resist being subjected to such principles (Ribot and Oyono 2005; Poteete and Ribot 2011). The complexity of an illegible (a laScott 1998) context also makes implementation very difficult. Most programs and associated protections have not addressed the needs and aspirations nor established and protected rights of resource dependent rural populations. Also, representation is only meaningful if there are rights to defend. In many instances local rights are weak or ambiguous. It is notable, for example, that the UN safeguard agreements do not mention tenure rights – or even use rights and prior access claims that might have historical and ethical but no legal basis.

Critics also argue that the COP17 meeting in Durban, South Africa, marked a step backward on safeguards, because it failed to make them mandatory or to develop performance indicators for compliance.3 Instead, it calls for voluntary reporting on ‘how these safeguards are being addressed and respected…and included in, where appropriate, all phases of implementation’ (FCCC/SBSTA/2011/L.25/Add.1), suggesting, among other things, that some countries may decide they are not always appropriate. Some analysts are optimistic, arguing that fostering safeguard principles as universal norms will provide ample pressure on national governments ‘not to deviate too far’ (Jagger et al. 2012). Norms are good, but since when do we ask foxes to guard the henhouse? Like transparency, we should not let policies depend on principles of shame that have little influence on the shameless (Fox 2007).

This article examines evidence from legal reforms of statutory rights in Senegal to argue that safeguards are unlikely to work for local people under REDD unless they increase the power of forest communities to participate meaningfully – by shifting binding decision-making powers to accountable local representatives in ways that challenge the power of elites and transform ‘business as usual’. The case demonstrates how the creation, application, effectiveness, and ultimate meanings of legal reforms are shaped by entrenched rural inequalities embedded in disabling social, political-economic, and legal hierarchies. Lack of empowered representation along with policy-backed marginalisation is deepened even by so-called ‘neutral’ or seemingly ‘fair’ policies, because of unequal access to capital, labor and credit, rooted in class, identity and social relations (Baviskar 2001; Ribot and Peluso 2003; Larson et al. 2006; Bandiaky 2007). Together these factors slant the access playing field, pitting marginal people against the more powerful, reshaping the intention and effects of legal instruments.

How can safeguards be effective in light of such entrenched inequalities? They can be effective by being mandatory. By being clear about what specific changes they require – in laws, access to resources, means of enforcement. By creating guidelines for their design, implementation and monitoring that make their existence or absence knowable – and transparent. And, by having sanctions so that transparency might result in accountability (Agrawal and Ribot 2012). But even this is not enough. Social protections, like any legal reforms, are easily fettered, stymied, manipulated and circumvented. Local people are often given strong rights to valueless resources, rights to forests rather than markets, rights to implement rather than decide, rights to participate rather than control. Protecting these is not enough. Both the content of rights and the procedures of justice must be upheld.

Will FPIC be selectively limited to questions involving the inadequate ‘rights’ held by the rural poor? Will something called ‘FPIC’ be extracted, coerced, cajoled, persuaded or hoodwinked out of communities? Will it remain selectively targeted to indigenous people, leaving out the many non-indigenous long-standing forest communities who deserve equal protection? Real emancipation requires the establishment of universal representation (the procedural side) – via empowered and locally accountable authorities (the content side). Establishing such conditions – even if impossible – remains the central challenge to fair and just REDD+. Such goals will likely never be ‘achieved’. But if not well defined and aimed for they will not even be approximated.

This articles takes an ‘access’ approach to policy analysis, described below, by analyzing the political economy that shapes the distribution of benefits from forests under a particular policy regime. It focuses on the real-world problem that forest policies and/or policy implementation systematically exclude various groups from forest benefits. In doing so, forestry policies and practices, sometimes inadvertently, impoverish and maintain the poverty of these groups. Poverty is not just about being left out of economic growth. It is produced by the very policies that enable some to profit – today from timber, firewood and charcoal, tomorrow from carbon.

The next section of this article frames our access approach. The following presents a case study of charcoal production in Senegal – little to do with REDD yet, but it is everything to do with the uneven fields on which REDD is already beginning to play out.4

2. From disabling to enabling policies: rights with access

Governments have long mediated forest access (Thompson 1977; Guha 1989; Scott 1998). Sunderlin et al. (2005, pp. 1390) describe how ‘forestry laws and regulations in many countries were written to assure privileged access to timber wealth and to prevent counter-appropriation by the poor.’ In Africa the colonial antecedents of many of today’s forestry policies were unapologetic in favoring profit by Europeans over Africans (Ribot 1999a). Writing on Gabon, for example, the colonial historian R.L. Buell reported that

“…before 1924, natives held [forest] concessions and sold wood upon the same basis as Europeans. But the competition became so keen … that in a 1924 administrative order, the government declared that a native could not cut and sell wood except for his own use without making a deposit with the government of twenty-five hundred francs – a prohibitive sum.”(Buell 1928, Vol. II, 256.)

Over 80% of the world’s forests are on public lands, and the state is often the first gateway to forest access (FAO 2006).5 Forestry authorities are still using many exclusionary strategies directly descendent from these earlier techniques, keeping forest peoples poor.

The World Bank (2002) estimates that 1.6 billion people depend on forests for livelihoods (see also Kaimowitz 2003). At least in some countries, there is an important correlation between forests and poverty (Blaikie 1985; Peluso 1992; Dasgupta 1993; Taylor et al. 2006). Communities living in and near forests suffer from outsiders’ commercial exploitation of forest resources (see Colchester 2006 for a list of studies and consequences; Ribot 2004; Oyono et al. 2006), and it is clear from commodity chain and forest-village studies that vast profits are extracted through many commercial forest activities, yet little remains local (Blaikie 1985; Peluso 1992; Dasgupta 1993; Ribot 1998, 2006). Retaining forest benefits locally may offer options for improved well-being in these areas. Indeed, the great commercial and subsistence value of forests is drawing increased attention to their potential role in poverty alleviation (Kaimowitz and Ribot 2002; Oksanen et al. 2003; Sunderlin et al. 2005), though there may also be tradeoffs between forest conservation and poverty alleviation (Wunder 2001; Tacconi et al. 2006; Lund et al. 2009; Ribot et al. 2010).

Over the past two decades there has been a wave of reforms designed to increase local participation and benefits for forest dwellers. Studies of community forestry in Mexican ejidos (Bray 2005) and Guatemala’s Petén (Taylor 2006) have demonstrated substantial economic and other livelihood benefits, such as increased income, greater human and social capital, natural resource conservation, decreased vulnerability, greater equity, democratisation of power and empowerment. Community forestry in Cameroon, Nepal and Senegal has also significantly increased income to forest villages (Agrawal 2001, 2005; Oyono 2004, 2006; Ribot 2009a). But few such studies are available precisely because communities rarely have policy-supported access to forests, the resources that are valuable in them, or policy-supported access to the capital and markets that would make increased income possible (Ribot 1998, 2004). These experiments in inclusion are important trail blazers toward more progressive and pro-poor forestry, but they still represent only small enclaves of change in the vast wilderness of forestry practice.6

Colchester (2006) point out that many governments have signed numerous ‘soft laws’ such as international agreements that, among other things, recognise indigenous land rights and customary resource management practices, but that these have rarely been incorporated into forestry legislation. In cases where land rights have been granted, this does not necessarily include rights over trees or forest management.7 Where laws have passed granting communities greater access to land and/or forests, these have often been adopted through processes outside the realm of forest policy specifically, such as in Nicaragua’s autonomous regions or Panama’s indigenous comarcas, though there are exceptions such as Bolivia (Larson et al. 2006). For their part, forest policy frameworks tend to be developed with significant influence of timber interests, as well as the state and multilateral financial institutions, but less often, despite the widespread discourse, with the effective participation of community or indigenous groups (Silva et al. 2002). It is no surprise that forest policy usually reflects the multiple interests of ‘stakeholders’ – at the expense of these under-represented forest-dependent populations.

In light of strong discourse yet generally weak implementation of decentralization and devolution policies, there is concern that REDD may lead to recentralization, particularly if large sums of money are involved (Phelps et al. 2010). REDD is primarily conceived of as a national project through central governments, though there is increasing discussion of the importance of multiple scales of governance for REDD implementation (e.g. Hayes and Persha 2010). Karsenty and Ongolo (2011) argue that the assumptions behind REDD’s financial incentives – that central governments have the ability to change the development path and implement the policies and measures to do so – are inappropriate for ‘fragile states’ and that alternative designs working through subnational scales will be more effective.

The risks to local people of REDD implementation are numerous, with the main concerns related to rights to forest land and livelihoods. Larson et al. (2012) argue:

“If forest tenure is currently insecure, unclear or in conflict, more powerful actors could gain rights to the land in the interest of obtaining REDD+ benefits. On state-owned lands, customary land users without formal rights could be subject to new rules and regulations, including restrictions on land use that lead to new hardships. If forest tenure is currently secure, unknowing or unscrupulous leaders could sign away rights and/or commit to obligations without fully understanding the consequences or obtaining the consent of those who live on the land.”

It is no surprise, then, that the UNFCCC safeguard principles, mentioned earlier, emphasize respect for rights and the need for participation and that these two aspects are at the center of most debates in the literature on safeguards (e.g. Lawlor et al. 2010; Agrawal et al. 2011; Lemaitre 2011).

Nevertheless, if policy and legislation establish rights, often inadequate or weak rights, and rarely translates these rights into greater benefits for average rural citizens, how can safeguards, presumably to protect rights and participation, be effective? How do we explain the paradox of increasing recognition of rights at a broad scale alongside the failure to establish adequate rights and to guarantee basic access in practice, and what lessons can be drawn regarding the design and implementation of safeguards under REDD? We propose that access can be supported by 1) social safeguards aimed at broadening local access through significant rights to the resource and to markets, and 2) social safeguards reinforcing accountable local representation in the shaping of access and access rules. In short, we propose that forestry must aim to be emancipatory. That is, forestry and REDD must include provisions that empower local people to shape the political economy that shapes their access (cf. Bohle et al. 1994). This is otherwise called local ‘democracy’ – local leaders with sufficient and meaningful discretionary powers who are accountable to local citizens (Agrawal and Ribot 1999; Ribot et al. 2008).

In their ‘theory of access’, Ribot (1998) and Ribot and Peluso (2003) contrast the common formulation of property as a ‘bundle of rights’ with their broader conception of access as a ‘bundle of powers’. Property is the right to benefit from things (McPherson 198:78) while access is the ability to benefit. Ability is broader than rights – one may have rights but still not be able to benefit. To procure entry and use of forest resources through property rights – temporarily, such as short- or long-term contracts for concessions, or permanently, such as land titles or constitutional guarantees8 – is one element of access, but the power to act on those rights depends on the negotiation of a number of complementary access mechanisms. For example, in the Senegal case we present below, even when charcoal producers hold rights to the forest resource itself they do not have rights to enter markets, access to which is mediated by the forestry service. In this case they cannot derive the full benefits from their forests. Further, the access approach highlights the role of power, emphasizing that many people gain and maintain access throughothers who control it. Hence, access to markets and even to forests themselves are mediated by higher authorities – such as the Forest Service and legislators – who are influenced by wealthy urban-based charcoal merchants. Under these conditions authorities who implement and enforce forestry laws systematically favor charcoal merchants and create multi-layered access barriers for local producers.

The theory of access shows that to benefit from forests, property rights are not enough. Producers must also have access, in this case, to markets – which in turn requires an ability to influence those who control market access – that is, forest service agents and powerful merchants.

The access approach provides a way to explain empirically ‘who benefits’ from things, and shows that rights are only one aspect of this ability. The access approach complements the rights-based approach. Rights-based approaches, if practiced according to their original conception, aim to alter power dynamics in development (Nyamu-Musembi and Cornwall 2004). In this framework, gaining rights, such as those established through the signing of international treaties and inscribed in national laws, is only a first step. Rights, however, only take effect when implemented in practice – also a political process that will likely challenge vested interests at every step. On the ground, then, a rights-based approach is successful when the power dynamics of access are altered and access to livelihood assets and markets are improved for formerly excluded and marginalised groups.

The case below shows how current forestry policies in Senegal – even when called community based or participatory – and the ways they are selectively implemented continue to reproduce the double standards and conditions that disadvantage, create and maintain the rural poor.

3. Charcoal in Senegal

“There is a certain complicity of the Forest Service – it is not against us, it is for the interest of the patrons.” (Elected Rural Council President in discussion 14 February 2006 at Tamba Atelier with four Rural Council Presidents)

Until 1998 the system of forest management in Senegal was organised around a system of licenses, permits and quotas allocated by the national forest service. A national quota for charcoal production was fixed by the Forest Service each year. Forest service officials and agents claimed this quota was based on estimates of the total national demand for charcoal and the potential for the forests to meet this demand. But these estimates were neither based on surveys of consumption nor forest inventories. Indeed, there was (and still is) a persistent gap between the quantity set for the quota and the much-higher figures from consumption surveys. In practice, the quota is based on the previous year’s quota which is lowered or raised depending on various political considerations. Over the past decade, the quota was lowered almost every year – regardless of demand – thus increasing illegal production (since demand was always met) (Ribot 2006).

Prior to the new decentralised forestry laws, the nationally set quota was divided among some 120–170 forestry patrons, or merchants, at the head of forestry enterprises – cooperatives, economic interest groups (GIE) and corporations – who hold professional forest producer licenses delivered by the forest service. Allocation of quotas among these entities was based on their previous year’s quota with adjustments based on whether or not the enterprise had fully exploited its quota and had engaged in positive forest management activities, such as reforestation. Some forestry patrons did plant trees by the side of the road to demonstrate such efforts – they called these plantations their ‘chogo goro’ or bribes – since these helped them get larger quota allocations from the forest service. During this period, new professional licenses were also allocated most years (enabling new cooperatives to enter the market).

Each year after the allocation of quotas, the forest service and ministry of environment held a national meeting to ‘announce’ the opening of the new season. They passed a decree listing the quotas for each enterprise and indicating in which of the two production regions, Tambacounda or Kolda, these quotas were to be exploited. Soon after, the Regional Forest Services then called a meeting in each regional capital to inform the recipients of the location they would be given to exploit their quotas. Sites were chosen by foresters based on ‘eyeballing’ of standing wood. The forest agents organised the zone into very loose rotations and chose sites by eye, such that some areas that were considered exhausted would be closed, while others that had not been official production sites for a time would be re-opened. There was no local say in the matter.

Progressive legal changes gave the rural populations new rights in the late 1990s. Senegal’s 1996 decentralisation law gave Rural Communities (the most-local level of local government) jurisdiction over forests in their territorial boundaries. The Rural Council (the elected body governing the Rural Community) was given jurisdiction over ‘management of forests on the basis of a management plan approved by the competent state authority’ (RdS 1996a, art.30), and the 1998 forestry code (RdS 1998) gave the council the right to determine who can produce in these forests (art.L8, R21). Further, even the more general decentralisation framing law gave the council jurisdiction over ‘the organisation of exploitation of all gathered plant products and the cutting of wood’ (RdS 1996b, art.195). Finally, the forestry code states that ‘Community Forests are those forests situated outside of the forested domain of the State and included within the administrative boundaries of the Rural Community who is the manager’ (RdS 1998, art.R9). The forested domain of the state consists of areas reserved for special uses and protection (RdS 1998, R2), and most of Senegal’s forests are not reserved. In short, under the new laws most Rural Communities control large portions of the forests – if not all of the forests – within their territorial boundaries.

To protect the rights over these forests, the forestry code requires the forest service to obtain the signature of the rural council president (PCR, elected from among the rural councilors) before any commercial production can take place (art.L4). For their part, PCR presidents play an executive role and cannot take action prior to the deliberation of the council whose decisions are taken by a majority vote (RdS 1996b, arts.200, 212). In short, the new laws require a majority vote of the rural council approving production before anyone can produce in Rural Community forests.

The radical new 1998 forestry code changed everything – at least on paper. The amount of production would be based on the biological potential of each Rural Community’s forests rather than by decree in Dakar and the regional capital. The enterprises that could work in a given forest would be chosen by the rural council rather than the National Forest Service in Dakar. If implemented, the new system would empower rural councilors to manage their forests for the benefit of the Rural Community. The law stated that quota system was to be entirely eliminated in 2001 (RdS 1998, art.R66). But despite all the new Rural Community rights, as of 2009 little had changed. The forest service continued to manage and to allocate access to the forests via centrally allocated licenses, quotas, and permits.

In implementation, the Rural Council’s new rights to decide over forest use are being attenuated by double standards concerning forest access and market access. The new laws give the PCR rights over forests, but the forest service refuses to transfer the powers. Rural populations in Senegal lose out mainly due to two double standards: access to forests and access to commercial opportunities are both skewed against them. These are discussed below.

3.1. Double standards in forest access

The rural council president legally controls the rights to access forests, but foresters do not allow him to exercise his prerogative. Foresters argue that villagers and councilors are ignorant of forest management and that national priorities trump local ones. They treat the PCR’s signature as a requirement rather than as a transfer of powers or change in practice. They and the merchants coerce – threaten and pressure – the council presidents to sign away forest rights (Ribot 2008).9 Rural council presidents say no but are ultimately pressured to sign.

The Regional Forest Service deputy director was asked, ‘Given that the majority of Rural Council Presidents do not want production in the forests of their Rural Communities, how do you choose their Rural Community as a production site?’ He replied with a non-comprehending expression, ‘If the PCRs have acceptable reasons? [pause] If the local population would not like? [pause]’ He then asserted, ‘the resource is for the entire country. To not use it, there must be technical reasons. The populations are there to manage. There is a national imperative. There are preoccupations of the state. This can’t work if the populations pose problems for development.’ Nevertheless, the deputy director knew the letter of the law that he was breaking every day. When asked to explain the function of the PCR’s signature, he replied, ‘the PCR signature must come before the quota is allocated, before the regional council determines which zones are open to exploitation.’ (Interview, Deputy Director of the Regional Forest Service, Tambacounda, 3 December 2005.) In short, Rural Councils are asked for their signature, with the expectation they will sign – the Forest Service does not seem them as having a right to say ‘no’ – despite that the population they represent opposes production.

In four Rural Communities where donors have set up model forest management projects, the new forestry laws are being applied – albeit selectively. In project areas rural people have the opportunity to participate in forest exploitation, but only if they engage in forest management activities required by the forest service. The ecological evidence indicates that few measures are necessary since natural regeneration in the zone is robust (Ribot 1999b). Further, Wurster (2010) has shown using transects and satellite imagery that managed forests regenerate at the exact same rate as non-managed forests – there is no detectable ecological effect of management. Forest villagers know this and do not see the need for most management activities. Nevertheless, to be allowed to manage their own forests, rural communities must use management plans created by the forest service. That is, whereas urban-based merchants install migrant labourers in non-project areas without management plans, villagers wishing to engage in charcoal production must do so under strictly supervised and highly managed circumstances. (Ironically, even in these areas most of the PCRs and councilors did not want production, but were forced to sign off under pressure from the forest service – similarly to PCRs in non-project areas.)

By creating a spatially limited implementation zone for existing policies, the projects serve as an excuse not to implement the laws more generally. Foresters argue that the projects represent cutting edge practices that are being tested before expanding to other sites, but this argument does not justify the Forest Service prohibiting forest villagers outside of the production areas from producing charcoal while allocating forests to the migrant woodcutters working for urban-based merchants. In fact, the project areas serve as a decoy. When donors come to visit the forests, they are shown project areas where management – rather, the labour to implement management obligations imposed by the forest service – is decentralised. They do not see the rest of the forests where forest service activities have barely changed since colonial times (including those areas where production is closed without consultation of the rural councils). The project in this case reduces the progressive 1998 forestry laws to a territorially limited experiment.

3.2. Double standards in market access

The forest service requires all those wishing to trade in the charcoal market (called charcoal patrons) to be members of a registered cooperative, economic interest group (GIE) or a private enterprise in order to request from the Forest Service a license (Cart Professionnelle d’Exploitant Forestière) in the name of their organisation (see Bâ 2006). Despite the elimination of the quota in 2001, production and marketing remain impossible without quotas, since at least until 2011 permits are still only allocated to those with quotas.10

Upon receipt of a professional card, the member’s organisation can be allocated a portion of the national quota in the annual process of quota allocation. In 2004 the national quota of 50,000 tonnes was divided into 46,265 tonnes initial quotas and 3735 tonnes of encouragement quotas (7.5%) (RdS 2004d, 11–12). The initial quotas are allocated at the beginning of the season and the encouragement quotas are allocated at the discretion of the forest service and minister later in the season (Bâ 2006).

Each year new cooperatives and GIE (economic interest groups – a kind of for-profit collective business) have been added to the market. In 2005 there were 164 organisations (RdS 2005), up by 18 new organisations from 147 organisations in 2004 (RdS 2004d, 12). Unfortunately, all of the rural-based cooperatives we have spoken with who have requested professional cards have been refused. The quota per patron, however, is shrinking, and many patrons believe that new licenses are being allocated to relatives of powerful merchants and political allies. ‘The registration of new entities is due to the officials: the president of the national union and the state. Most of the entities are family businesses – brothers and sisters.’ In particular, they are the brothers and sisters of other already registered patrons. According to older patrons, some of the new organisations do nothing but resell their quotas to others (Patron2 25 Dec 05). As one patron told us in disgust, ‘most of the large quota people are new entrants into the market’ (Interview AMD, Cooperative president, Patron Charbonnier, Tamba 26 Dec 05).

In recent years, the Forest Service, upon recommendation by the director of the national union, has been allocating licenses and quotas to women (Interview, union leader, 22 Feb 2006). This is a new phenomenon. In an interview with one such woman, we learned that she was the wife of an established patron. Forming her own cooperative appears to be a strategy to increase her husband’s quota (interview by Salieu Core Diallo Feb 2006). Other patrons are not happy with this. One told us that the national union president ‘was given a supplementary quota [officially called an ‘encouragement’ quota]. They give quotas and supplementary quotas to women. These women are behind [the national union president].’ (Interview, PCR at 14 Feb 06 workshop.)

Over the past several years, rural councilors and other rural community members have requested licenses so that they could get quotas.11 In one case a rural GIE president went to the Director of the Forestry Service in Dakar to request the card. He explained: ‘We put together a GIE in 1998 with its own forest production unit. We filed our registration papers at Tamba [the regional capital] – it went all the way to Dakar. I saw the dossier at Hann [National Forestry Office]…. We asked for cooperative member cards and for a quota. We were discouraged. We went to Hann and to Tamba. In Dakar, they wanted to give us quotas as individuals. I said “no” in solidarity with the rest of my colleagues with whom I was putting together the GIE’ (Interview, elected rural council member, Tambacounda Region, 22 Dec 05). A similar story was recounted by a GIE president in Missirah (Interview December 05).

The forestry service explains their refusal to give professional cards to local GIE by saying ‘they need to be trained’ and explaining that ‘if we let them produce, they will learn the bad techniques of the surga [migrant woodcutters]’ who work for the current patrons (Interviews, 2 IREF officials in Tamba Dec 05 and three ATEF). First the community has to be organised into village committees and trained to manage and survey forest rotations and to use the Casamance kiln [these are all requirements within project areas but not requirements under the law]. Meanwhile, however, the forest service continues to admit new cooperatives that have no knowledge of production whatsoever and to hand out quotas to patrons who are producing without any training or management within managed and non-managed zones.

After the initial and encouragement quotas are allocated, illegal production and transport fill in the gap between legal supply and actual consumption. But these illegal activities can only be done by those who hold licenses and quotas – since license and quota holders can use their licenses to obtain supplementary permits and can hide extra charcoal with their legal loads. This is how the gap between the quota and consumption is filled. The market – legal and illegal – is tied up in the hands of a small privileged group of well-connected patrons (Ribot 2006).

Despite that Senegal’s progressive forestry policies have given away little of the state’s control, they are at this moment being reformed and replaced by less-progressive new forestry laws (Ribot 2009b). Senegal’s current forestry bill takes back many of the rights hard won over the decades of decentralisation. Further, Senegal’s forest service went from being a civilian service to a military service in 2008. No donors in Senegal’s forestry sector made any protests. This militarisation is quite opposite the movement in most countries. The forestry bill, still in discussion in mid 2011, is likely to pass. It promises to consolidate control over commercial access to forests with the forest service – something the current laws had threatened but never achieved. If Senegal’s new forestry bill passes, then the quota will be officially renamed ‘the contract’. It will have the same function as before but under a new name (Faye and Ribot 2010). Although nothing will change, Senegal’s foresters and their supporters will all celebrate the great progressive reform.12

4. Conclusion

Senegal’s 1998 forestry laws are beautifully written. They place key decisions over forest exploitation in the hands of democratic local authorities and open the markets for communities to sell their products. But these laws are not respected in implementation. Old forestry laws favouring the urban elite were eliminated by progressive new law, but practice only changed slightly. Through long-abrogated but still-practiced policies, Senegal’s Forest Service allocates licenses and quotas (now called ‘contracts’) so as to retain market access in elite hands. Senegal’s forest access and management standards are singular and fair in the law, but double in practice. Urban elites are systematically favored while rural forest dwelling populations are excluded with total disregard for their rights, wishes and needs.

Inequalities favoring outside commercial interests over those of local communities are maintained in Senegal and elsewhere by a large repertoire of access means (see Ribot and Oyono 2005; Smith 2006; Toni 2006; Larson and Ribot 2008; Nayak and Berkes 2008; He 2010; Neimark 2010; Saito-Jensen et al. 2010; Poteete and Ribot 2011). Though the specific dynamics vary from country to country, poor communities and smallholders remain at a disadvantage in comparison to more powerful outside interests. Laws may create uniform standards or access asymmetries; they may even transfer decision-making powers and lucrative opportunities to poor rural populations. But even when laws would create fair access, they are not fair when unevenly implemented or selectively enforced, and they are not sufficient to overcome existing inequities unless they are designed and implemented with bias toward the underprivileged (Ribot 2004; Bandiaky 2007; Baviskar 2007).

Despite a new language concerning decentralisation and the recognition of indigenous or rural peoples’ rights, forest services around the world still treat local people as subjects and continue to colonise forested territories. The policies they apply today are almost all – even when given a participatory or decentralised patina – relics of colonial management based on earlier European practice (as in Africa) or of post-colonial entrenched bureaucracies (as in Latin America). REDD will build on this tradition of domination if it does not actively and vigorously seek to transform the structure and cultures of forestry and forest services. Weak checks, balances and safeguards are not enough. Targeting the poor is not enough. New progressive policies will have to target the rich to shoot down some of their inordinate privilege. It will not be enough to tweak, or enforce existing (inadequate) ‘rights’. New policies that favor benefits for local people over outsiders are needed. New politics that regulate through minimum standards (see Ribot 2004) rather than maximum central control may have transformative power. The poor must be represented in the making and implementation of these processes – proportionally to their inordinate numbers.

The outcomes of forest policy and implementation processes worldwide demonstrate the multiple and competing interests and goals of different stakeholders and the weaker power of those who consistently lose out. The existence of apparently fair laws, however, also demonstrates that advocacy by and for forest-based populations has its successes and that further progress is possible. Senegal’s forestry policies are much better for rural people today than twenty years ago. New policies should include deepening forestry decentralisations through effective representation and participation, seeking common ground across myriad local goals and interests, and identifying opportunities to challenge unjust privilege. Representation will mean that when local people say ‘no’ to exploitation of local forests, then there will be no exploitation. It may not mean that when they say yes, that exploitation should necessarily take place. Such a ‘yes’ could have negative ecological externalities for higher scales of social, economic and political organisation. Environmental standards are needed (Ribot 2004). The right to say no to exploitation, however, gives them the ability to negotiate – the cost of this negotiation, the costs of real ‘participation’ and ‘representation’, is less privilege to outside interests.

REDD is entering this slanted world with the primary objective of carbon emissions reduction – not justice or equity. If community rights are already limited, as in Senegal, will they be limited in the future under REDD in the name of carbon sequestration? Who will control forests? What rules for resource use will be developed to meet carbon targets under REDD, who will create and enforce these rules and how might they limit community access to forests for livelihoods? If communities carry new burdens – such as limitations on activities permitted in forests (‘no’ imposed from above) – will they be fairly compensated? Will the rights to forest benefits – this time to carbon funds – once again be captured by outsiders? (Larson 2011).

Rights are only real when they are enforced – rights are ‘enforceable claims’ (MacPherson 1978) – so rules not enforced are not rights. Likewise, safeguards that are not enforced, that are voluntary, or that protect rights that are not enforced, are meaningless. If REDD is to challenge business as usual and to benefit local populations, safeguard policies must not just protect rights, but must also establish, strengthen and secure rights (including the right to say no through effective FPIC policies) and other forms of local access control. In this context, voluntary principles are insufficient since those who can choose to apply or circumvent them are the very same people who stand to benefit most from circumventing them: safeguards must be enforced with clear mandatory compliance criteria and multi-dimensional monitoring.

The weak must have the means of enforcement, whether through representation, resistance or withdrawal, to fight for favorable policies and fair implementation. The support of rigorous analysis and of sympathetic allies (a role of scholars) can back progressive claims and help exert pressure on those who resist change.

Policy objectives are damped out in the transition from discourse to law and transformed in implementation. Hildyard et al. (2001) observe that participatory projects and policies ‘however carefully prepared, generally flounder the moment they leave the drawing board. By the time they are implemented, they are frequently unrecognizable even to their authors.’ Lele (2000, in Nayak and Berkes 2008, pp. 707) postulates ‘that (a) participatory management involves the devolution of power, (b) but the state is by nature interested in maintaining and accumulating power, and therefore (c) joint forest management must be a “sleight of hand” carried out by state to coopt activists and placate donors while retaining control and even expanding it in new ways.’

These are fair observations, but policy is not something that is made and implemented once and for all. It is an iterative process that requires constant vigilance and struggle. Stratification is a constant process. Inequity always comes back. Governments (like the businesses they support) perform (enact, portray, pretend) change while maintaining business as usual (Poteete and Ribot 2010). Still, progressive policies are better than regressive ones. There are many politicians, foresters, donors, NGOs, and administrators fighting for greater justice in forestry. Their efforts can make things better even if they do not make all well. REDD will have to be hyper-progressive, indeed, emancipatory if it is to benefit the rural poor. Given its global scale and potential power, REDD has the opportunity – and moral responsibility – to be an emancipatory program.