Decentralisation and devolution in Nicaragua ’ s North Atlantic Autonomous Region : natural resources and indigenous peoples ’ rights

A number of governments, particularly in Latin America, have begun to recognise the rights of indigenous peoples and traditional communities to the lands on which they live. Recognition has often taken the form of constitutional provisions or laws that grant use rights in perpetuity or provide land titles. These provisions usually establish rights for multiple communities over a large territory, at a scale that may be ideal for promoting broader, ecosystem management approaches. At the same time, however, indigenous communities often do not have existing territorial governance structures at these scales. Nicaragua’s North Atlantic Autonomous Region provides a rich setting in which to study issues of multilevel natural resource governance. In addition to the devolution policies that have created official indigenous territories, the central government has decentralised important powers over natural resources to the regional autonomous authority, while municipal authorities still exist but it is unclear what their role will be in the future. At the same time, however, the community scale is the one at which local people have traditionally managed resources. This paper examines these issues in light of efforts to establish democratic governance institutions at the territory level and argues that communities continue to lose out under multilevel governance regimes without concerted efforts to level the playing field. The findings are based on several years of research in the region, emerging research on newly titled territories and a six month training and dialogue with territory leaders, organised by a consortium of international and local NGOs. 180 Anne M. Larson and Jadder Lewis-Mendoza


Introduction
Though what it means in practice varies greatly, the collaborative governance of natural resources across institutional levels and scales has been broadly recognised as desirable or even necessary for resource sustainability (Mayers and Vermeulen 2002;Agrawal and Chhatre 2007;Hayes and Persha 2010).It may be even more important for local democracy (Ribot 2004(Ribot , 2008) ) and for human rights (Anaya and Williams 2001;Colchester 2004), particularly the rights of indigenous and traditional peoples.But promoting multiple levels of governance, or multiple centres of power in polycentric schemes (Andersson and Ostrom 2008), is not a simple, technical process (Li 2006) leading to clear improvements in outcome.
Decentralization and devolution efforts, for example, have been obstructed by central governments and captured by elites (Ribot et al. 2006).Efforts at "participation" have sometimes been recognised as a smokescreen for the imposition, but under more amenable circumstances, of decisions "from above" (Poteete 2002;Hickey and Mohan 2004;Meynen and Doornbos 2005).Why would new efforts at multi-level governance be any different?What would it take to assure that multi-level governance gives local people more voice and decisionmaking power in practice?
A number of governments, particularly in Latin America, have begun to recognise the rights of indigenous peoples and traditional communities to the lands on which they live.Recognition has often taken the form of constitutional provisions or laws that grant use rights in perpetuity or provide land titles.These provisions usually establish rights for multiple communities over a large territory, at a scale that may be ideal for promoting broader, ecosystem management approaches.At the same time, however, indigenous communities often do not have existing governance structures at these territorial scales.
Many of these territories coincide with important natural resources, such as forests, fisheries, petroleum or minerals.In many cases indigenous peoples are granted decision-making rights over above-ground resources, though not necessarily over the subsoil, in spite of international conventions calling for Free Prior and Informed Consent (FPIC).The prospect of carbon markets under REDD+ raises similar issues now with regard to carbon rights.
Nicaragua's North Atlantic Autonomous Region (RAAN) provides a rich setting in which to study the unfolding of multi-level resource governance (Figure 1).In addition to the devolution policies that have created official indigenous territories, the central government has decentralised important powers over natural resources to the regional autonomous authority, as well as to municipal governments, which overlap with the jurisdictions of the new territories.The community level is the one at which local people have traditionally managed resources, and the also law formally recognizes community land rights and their elected authorities.This paper will examine these issues in light of efforts to establish democratic governance institutions at the territory scale.
Our central hypothesis is that multi-level governance has led to important new opportunities for indigenous communities to exercise their right to free determination.The overall structure would appear to be ideal: multi-community indigenous territories, with elected leaders, are titled under an autonomous, also Source: Elaborated by Edwin Matamoros based on official and unofficial data elected, regional government regime.Nevertheless, we find that communities are, as yet, unable to take advantage of this opportunity.Rather, legal ambiguity and lack of accountability facilitate the ability of more powerful actors to continue to take advantage of the weaker communities.This article highlights the multiple dimensions of conflict arising from both economic and political interests and argues that theories and practice of multilevel governance must pay greater attention to these issues.
These findings are based on existing literature and several years of experience working in the region, as well as discussions and debates held with the leaders of eight indigenous territories, either recently titled or in process of titling, in the framework of a six month diploma course from January to June 2011 organised by a consortium of international and local NGOs.These discussions were complemented by research on governance and natural resources in three of those territories, representing different ethnic groups (miskitu, mayangna), different stages of titling (titled, in process) and with and without problems of colonist land invasions.This research involved key informant interviews and focus groups in 5-6 communities in each territory aimed at promoting discussion of territorial governance issues with leaders, community members and with women specifically; focus groups were divided, where possible, into two age groups (15-30 and over 30).For a full description of the methods, see Larson and Soto (forthcoming).
This article is organised as follows.The next section provides a brief overview of the struggle for autonomy and land rights in Nicaragua's Caribbean Coast, and hence the background for understanding the current configuration of key governance actors.The third section discusses the literature on multilevel governance.The fourth section presents the particular experience of both decentralisation and devolution of natural resource governance in the RAAN.This is followed by the discussion and conclusions.

The struggle for autonomy and indigenous rights in Nicaragua
Nicaragua's Caribbean coast was a British protectorate beginning in 1740, and the southern portion of the Mosquito Coast, as it was known, was officially passed to the independent Nicaraguan government in the 1860 treaty of Managua, though British influence continued.The Miskitu population, which dominated the smaller sumo-mayangna population, maintained a large degree of autonomy until 1894, when the region was invaded by troops from Managua and was named the department of Zelaya.
Indigenous land rights, mainly to agricultural areas, were formally recognised for the first time with the signing of the Harrison-Altamirano Treaty in 1905.Though the areas recognised were much smaller than those actually used by indigenous communities, these first titles have been an important legitimating factor in land rights claims more recently (CCARC 2000).Local leaders (miskitu and sumo-mayangna) organised in defense of indigenous peoples at different times during the 20th century, but it was not until 1987, through peace negotiations after several years of war against the revolutionary Sandinista government, that indigenous rights were recognised in the Nicaraguan Constitution.Also in 1987, the Autonomy Statute was passed, resulting in the establishment, in 1990, of the first elected autonomous regional councils in the North Atlantic Autonomous Region (RAAN) and South Atlantic Autonomous Region (RAAS).
The creation of these regions created a new layer of government that would, at least in theory, grant greater "local" control to indigenous communities and traditional peoples over these regions and their natural resources.The constitution recognises the right of Caribbean Coast communities to preserve and develop their cultural identity, forms of organization and property, as well as the use and enjoyment of "the waters and forests of their communal lands" (Art.89).It also states that the autonomous regional council should approve all natural resource concessions and contracts.The autonomy statute established the autonomous regional councils as the "maximum authority of the regional autonomous government".The councils are entities created for deliberation, resolution, regional public policy formation and negotiation.Neither this nor other, later legislation addressed the relationship between the new regional councils and the eight municipal governments located within, or overlapping with, the boundaries of the autonomous regions.
It took 16 more years for the implementing regulations of the Autonomy Statute to be passed.During that time, the central government, controlled by neoliberal administrations from 1990 to 2006, resisted granting any significant decision-making power or funds to the regional councils or governments1 , while continuing to exploit the region's natural resources.It also maintained a direct relationship with the municipal governments in the regions, thus undermining the power of the regional entities.As stated in a report by the United National Development Program, the central governments did not support the strengthening of autonomy but rather sought to maintain "the political, economic and cultural subordination" of the Caribbean Coast and its existence as a natural resource reserve at the service of primarily national interests (UNDP 2005).
For their part, the regional councils, with little funding or power, and not without accusations of corruption, failed to maintain broad-based support among the citizens of the coast regions.One opinion survey found that almost 70% of the people interviewed agreed with the statement that "the principle problem with autonomy is that the coast authorities that have been elected have not functioned well to date" (CASC/Ipade 2005, cited in UNDP 2005).(Notably, however, 66% also agreed that "the principle problem is that the central government 'from Managua' has not wanted to support autonomy.")By 2000, ten years after the first regional councils were elected, then, the idea of regional autonomy had lost much of its symbolic power (Gurdián et al. 2002), and a study of land demands by the Caribbean Central American Research Council (CCARC) found that communities had begun to organise in groups in order to claim large blocs of territory (CCARC 2000).The struggle for regional autonomy had thus shifted to a struggle for the recognition of territorial rights, and specifically for the demarcation and titling of indigenous traditional lands.
Key to that struggle was the international court case of the sumo-mayangna community Awas Tingni.Notably, Awas Tingni's case centred on the failure of the central government to obtain the consent of the regional council of the RAAN for a concession, as required by the 1987 constitution.Awas Tingni filed a demand before the Inter-American Court for Human Rights (CIDH) against the Nicaraguan government for granting a forest concession, on their traditional lands and without community consent, to the Korean company SOLCARSA in 1995.The community's legal representatives had fought the concession in the national courts to no avail, in spite of a Supreme Court ruling in 1997 that the concession was unconstitutional for failing to obtain the council's approval (Wiggins 2002).
In 2001, the CIDH ruled in favour of Awas Tingni, holding that "the international human right to enjoy the benefits of property includes the right of indigenous peoples to the protection of their customary land and resource tenure" (Anaya and Grossman 2002, p. 1).It found that the Nicaraguan Government had violated the community's rights to communal property as guaranteed by the Nicaraguan Constitution, as well as the American Convention on Human Rights.The Court ordered the state to adopt the relevant legislative and administrative measures necessary to create an effective mechanism for demarcation and titling for indigenous communities "in accordance with their customary laws, values, customs and mores" (Judgment, cited in Anaya and Grossman 2002, p. 13).The court case was significant for indigenous rights throughout Latin America.
The Communal Lands Law was passed in late 2002.The law formally recognises indigenous land rights and also establishes the institutional framework for demarcation and titling.Specifically, it guarantees indigenous communities "full recognition of rights over communal property, [and] use, administration and management of traditional lands and their natural resources" (Article 2).In 2003, the long-awaited implementing regulations of the Autonomy Statute were passed.Finally, as a result of a change of government in January 2007, and an alliance between the entering Sandinista government (the party that had negotiated the 1987 Constitution and Autonomy Statute to end the war) and the miskitu political party Yatama, the titling of indigenous territories has advanced rapidly, with 15 territories (11 in the RAAN, 2 in the RAAS and 2 in northern Jinotega 2 ) for a total of more than 2 million hectares were titled by mid 2010 (CONADETI 2011, unpublished data; Procuraduría General de la República 2010, unpublished data).The resulting configuration of large, multi-community indigenous territories, titled primarily from 2009 to present, has introduced a new layer of governance in an already complex scenario.

Multilevel governance in theory
The idea of multilevel governance is not new, but its relevance with regard to natural resources has received substantial attention in the past three decades.Decentralisation, devolution, co-management, collaborative management, participatory approaches -all have constituted strategies aimed at increasing the role of local people in natural resource management and decision-making, either directly or through their local representatives.These policies challenged prior centralized and top-down approaches to resource governance.More recently, scholars have begun to promote the term "multilevel governance" specifically, to emphasize the importance of working across scales, and often across disciplines, to solve environmental problems.
Proponents of these policies see them as desirable for a variety of reasons.Berkes (2010, p. 490) emphasizes "basic democratic principles", in reference to the right of citizens to participate in decisions affecting their wellbeing, and "the challenges of governance in a world of complexity and uncertainty".Commons scholars refer mostly to the latter.Environmental problems are rarely only relevant at a single scale, and dynamic, cross-scale influences create "fundamental problems for division of responsibility" among different levels of governance (Pritchard andSanderson 2002, in Armitage 2008).Multilevel approaches are attractive because they permit adaptive and interdisciplinary management approaches to complex problems (Ostrom and Janssen 2004;Paavola 2007;Armitage 2008;Berkes 2010;Leys and Vanclay 2011).
Commons theory has been largely developed on the basis of apparently simple management systems but has evolved to recognize commons governance "as a complex systems problem" (Berkes 2006;Armitage 2008, p. 15).Commons scholars have built on theories of resilience (Olsson et al. 2004;Ostrom and Jenssen 2004;Folke 2006;Armitage 2008) to understand the dynamics of these complex systems.Central to this approach is adaptive learning and a number of other normative features, such as accountability, leadership, knowledge pluralism, trust, and so on, nicely outlined in Armitage (2008).As Armitage (2008) points out, however, this framework has been insufficient to explain outcomes.He proceeds to outline the intersections between resilience theory and political ecology to suggest a way forward, by incorporating the core themes of the latter -"power, power relationships and the mediation of power relationships across scales" (Armitage 2008, p. 21) -into the understanding of multilevel governance.
Commons scholars often fail to take power and politics fully into account, by considering them a problem of external context rather than inherent to the issue of governance.The following statement is telling.In his very useful review of interdisciplinary research examining concepts and experience related to devolution, decentralisation, collaboration, polycentrism, and so on, Berkes (2010, p. 494) writes, "Case studies seem to have produced more negative findings than positive, but that has more to do with the political ecology and unwillingness of some of the actors to make it work, rather than some inherent shortcoming of decentralisation itself."While this statement may be true in theory, it fails to recognize that "decentralisation itself" does not exist in practice: the actors who implement, obstruct or manipulate it are central to the construction of decentralisation (or devolution, or multilevel governance).
Drawing on the work of numerous other scholars (e.g.Mehta et al. 2001;Armitage 2008) and the experience in the RAAN, we argue that there are two central weaknesses to current theoretical conceptions of multi-level environmental governance.First, the issue of power -such as attempts to undermine or capture benefits -in multilevel governance systems is not contextual but is rather central to the nature and practice of these governance systems.Notably, Berkes (2010) concludes that participation and the devolution of management powers are more successful in societies with strong democratic traditions and civil societyrarely the contexts in which we work.Though power issues may be relevant in all societies, authoritarian traditions, clientelist practices and unreliable justice systems clearly make the issue of power and inequity more salient.
Second, environmental governance issues are deeply embedded in other, broader political processes.This is relevant in a number of ways.For example, multilevel governance is seen as desirable in part due to democracy and rights considerations, as mentioned above, yet this aspect is rarely mentioned in the environmental governance literature.Rights considerations, however, may be the primary concern of local people.Interventions may not support democratic principles (Ribot 2004).In the same vein, both theory and interventions tend to emphasize management rather than governance more broadly (see Berkes 2010).Nevertheless, the systems, organizations and institutions of multilevel governance are rarely established for a single purpose (local governments, for example) and often serve multiple social and political purposes as well (Cleaver 2001, in Mehta et al. 2001).These issues must be taken into account in theory and practice.
This article examines multi-level governance in the RAAN.Legal reforms have created an institutional framework that, according to theory, would appear ideal, demonstrating key features of successful devolution and decentralization policies.For example, Agrawal and Ostrom (2001, p. 508) state that successful reforms need to assign local users significant property rights, transforming them into "claimants and proprietors".Devolution reforms in Nicaragua have recognized ancestral property rights for Nicaragua's indigenous communities.For Ribot (2004Ribot ( , 2008)), effective decentralization involves the transfer of important powers to elected local authorities.Again, reforms in the RAAN have transferred powers over natural resources to various levels of elected authoritycommunal, municipal and regional -and recently created a new one at the multi-community territory scale.Hence, the RAAN has polycentric and multilevel governance, with multiple and overlapping centers of authority, empowered communities with titles to multi-community territories and empowered leaders at several spatial levels.

The multilevel governance of territory and natural resources in the RAAN
The decentralisation (or regionalisation, as local political leaders prefer to call it3 ) of power to the regional autonomous governments of Nicaragua's Caribbean coast and the devolution of land through titling of indigenous territories both represent important initiatives aimed at granting greater power and voice to "local people".Each level of government serves a very important political purpose, with direct consequences for both rights and livelihoods in the region.And experience to date demonstrates numerous ways in which the rights and desires of communities continue to be railroaded, manipulated or ignored by other levels of government under the new multi-level governance arrangement.

Regional government
The regional autonomous governments emerged from a demand for indigenous political control over the resources of the region.Prior to this time, the central government claimed the right to sell all natural resources, granted in concession to private interests, without any input or control from the landholders (now owners) or their political representatives in the region.The central government granted logging concessions to forests that were occupied and used by indigenous people who had lived in them historically with customary rights.Mineral concessions were similarly granted to indigenous lands without concern for anyone's approval.Benefits from these concessions accrued to the central government alone.
As mentioned earlier, the Constitution states that the elected regional autonomous councils must approve all such concessions.Later legislation, such as the Forestry Law, mandates the distribution of the income from all resource extraction taxes and fees in the autonomous regions in four equal parts to the central government, the regional government, the municipal government and the local community in which the activity takes place.
In general, particularly since the resolution of the Awas Tingni case, discussed above, the regional authorities have been playing an increasing role in environmental and natural resource management.The Autonomy Law and its implementing regulations permit the regional councils to establish natural resource regulations, within the limits of the norms established by central government ministries.In 2002, the technical office of the Natural Resource Secretary (SERENA) was created to carry out the resource responsibilities of the Regional Councils.SERENA plays a central role in the administration and approval of Environmental Impact Assessments and the evaluation of proposed concessions and forest management plans and participates in activities to control resource management.These activities should always be undertaken in coordination with central ministries, since their legal mandates are overlapping (CABAL 2007, unpublished report).There have been differences of opinion over the interpretation of the law, with some claiming that the autonomous regional governments should have more power than the central government ministries, but central offices vehemently defend their primacy (Larson, A. M., G. Navarro, E. Méndez, M. Sánchez and G. Wallace 2008, unpublished report).Coordination between the regional and central governments increased with the change of government in 2007, due to political alliances and a central government more sympathetic to indigenous interests.
Since that time, there have been few public conflicts between these two levels of government over natural resource decisions.It appears that legal provisions are being respected and agreements have been reached on procedures regarding the vast majority of resource decisions.The only important case that has come to light is one precisely in which the jurisdiction was not as clear: off-shore petroleum concessions.In that case, the central government claimed that the off-shore region was not part of the autonomous authorities' jurisdiction and managed the concession process without regional government4 -or community -participation.Indigenous and environmental activists filed an appeal through regional council authorities in 2007, leading to a renegotiation between the government and petroleum companies.The regional government was apparently satisfied with the results.
Nevertheless, indigenous leaders in the two affected territories, Tawira and Prinsu Auhya Un, are not.Their communities were not included in the negotiations: they were not informed in advance, their opinion was not consulted and they were not informed once the concessions were authorised (interviews with territory leaders, March-April 2011).The central and regional governments argue that the procedures established for environmental consultation through the environmental impact assessment process are sufficient for Free Prior and Informed Consent, as established by international law.Community leaders disagree.There is no established FPIC mechanism in Nicaragua, in spite of several laws requiring community participation.

Municipal governments
Nicaragua's municipal governments were granted autonomy in the 1987 Constitution, a new Municipalities Law was passed in 1988 and the first municipal elections in the autonomous regions took place in 1996.Reforms to the municipal law in 1997 substantially increased municipal powers and responsibilities.With regard to natural resources specifically, municipal authorities obtain a portion of taxes on resources extracted in their jurisdiction and they have the right to present a non-binding opinion on resource contracts and concessions.They also play an important role in land use planning and the administration of permits that are relevant to the indigenous territories, discussed below.
In the RAAN, the municipalities located nearer to the regional government seat in Bilwi, Puerto Cabezas, and those representing the predominant, allied political parties (the FSLN and Yatama), have tended to coordinate more closely with the regional government.With the formation of the territories, however, the role of the municipal governments has become less clear.Also, over half the territories in the RAAN cross more than one municipal boundary.Indigenous leaders have argued that the municipal structure was imposed by the central government and should be eliminated and replaced by the territories, or at least diminished such that this level of government only attends to the urban areas and non-indigenous people (Larson and Mendoza Lewis 2009;Larson 2010).This would require legislative changes that would have to be approved by a two-thirds vote in the National Assembly (Congress). 5or now, the overlapping jurisdictions have led to conflict and confusion.Participants of the diploma course agreed that the relationship between the territories and municipalities was very problematic, with the most contentious issue being "intrigues involving the disposition of natural resource in the territories" (Graham, summary report of course notes, unpublished data).

Indigenous territories and communities
As explained earlier, the Communal Lands law emerged in part in response to the ruling of the international court in the Awas Tingni case.From the perspective of indigenous communities, however, the demand dates back over a century.Recognition of land rights should guarantee access and control over the land and natural resources that these communities have used historically.In the current context, this mainly refers to the right to exclude outsiders from usurping community resources in their own interest and without community benefit.Communities have not objected to all state concessions, but they have received few lasting benefits.Ongoing colonization by non-indigenous campesinos (peasant farmers) into the communities' forests, and the need to resolve the problem of colonists already living in the territories (saneamiento), were two of the three most salient problems in the RAAN identified by the participants of the diploma course (Graham, summary report of course notes, unpublished data).This section will address four paramount issues with regard to territories and rights: colonization, the formation of the territories themselves, the election of territory representatives and authority over natural resources.

Colonization
Over the past century, the central government, primarily through its agrarian institutions, gave out thousands of hectares of lands in the Caribbean region to colonists.From 1963 to 1979, the Nicaraguan Agrarian Institute (IAN) distributed almost 77 thousand hectares in private titles to colonists (PNUD 2005); in the 1990s, the National Agrarian Institute (INRA) gave out titles in collective blocs to former combatants (some indigenous, some non-indigenous), as well as private titles to non-indigenous households, but specific data are not available.In addition, throughout the country, local judges were permitted to authorise titulos supletorios (temporary titles) until recently, meaning, first, that there is no consolidated record of all these titles and, second, there are overlaps among existing claims (Larson and Mendoza Lewis 2009).
The titling of indigenous communal lands is seen as an essential step in curbing the flow of colonists and in removing those who have settled there illegally.The Communal Lands law guarantees the rights of colonists who have titles and occupied land prior to 1987 (the date of the Constitution recognizing indigenous rights); anyone without title should leave or pay rent.It is less clear what should happen in other cases, but there is no doubt that indigenous rights supersede the rights of colonists and that eviction requires indemnification, which could be costly.It is also clear that all new colonization is illegal.
Nevertheless, the flow of new colonists continues with virtually no effort to stop it by state entities.In fact, in the territory of Tasba Pri, which is in an advanced state of demarcation (CONADETI 2011, unpublished data), municipal government officials continue to authorise changes of residence that allow new colonists to obtain local identity cards and to vote in their district (interview with territory leaders, April 2011).The territory of Matungbak hired a lawyer to carry out a census of the colonists and their legal documents for the land in their possession, but the municipal government forced a halt to the process, arguing that the territorial government did not have jurisdiction (focus groups, Matungbak, July 2011).
Finally, though the state presumably no longer authorises individual titles in the areas titled or in process of demarcation as indigenous territories, researchers identified pockets of land that had not been included inside the demarcated areas.In at least one case this land had been assigned to a group of colonists by highlevel political leaders (Larson and Soto forthcoming).

The formation of territories
According to the law, indigenous communities can be titled as individual communities or as groups of communities.Nevertheless, the region's political leaders have not permitted communities to be titled individually, or even to form their territories in groups according to their own design (for a more complete discussion of these issues, see Larson 2010).Rather, political leaders from the miskitu political party Yatama have pushed communities to form territories based on the design that they conceived.In a public meeting in May 2007, Yatama leader Brooklyn Rivera called on RAAN communities to form territories that cover a significant part of the land area, leave little behind as "national land", include all indigenous communities inside territories and move quickly while the political moment is favourable (CRAAN 2007, unpublished meeting transcript).
This design is associated with reshaping the region's electoral districts: eliminating the municipal structure imposed by the central government and replacing it with an "indigenous" structure of territories and territorial authorities (see Larson 2010).For example, the self-designed territory of Tasba Raya, with 7 communities, was forced to become part of a territory with 23 communities (interviews with territory leaders, March 2011); they negotiated the right to retain their name, hence the territory is known as Wangki Twi/Tasba Raya.The selfdesigned territory of Prinsu Rau is actually sliced down the middle by two other territories on official maps (CONADETI 2011, unpublished data).

Territorial representatives
The communal land law formally recognises "traditional communal authorities" as the official representatives of indigenous communities, and the community general assembly (a meeting of all community members) as the maximum authority (Art. 3,4).The communal authorities are elected in each community "according to [its] customs and traditions" (Art.5), and "each community will decide what communal authority will legally represent it" (Art.4).By law, the definition of all procedures regarding election, re-election, removal and periods of office is the decision of the community (Art.6).This community scale is the one at which indigenous communities of the autonomous regions have traditionally managed and governed their affairs.
The formation of territories establishes a new scale of governance.As communities come together to form territories, the law prescribes the formation of territorial authorities as "administrative organs of the territorial units that they legally represent" (Art.5).These authorities should be elected in an assembly of all the traditional communal authorities of the communities that comprise the territory, "according to the procedures they adopt" (Art.3).With regard to the role of the regional government in these elections, the law establishes a few simple procedures.The Regional Council should witness and certify the elections within 8 days, but if not, the territorial authority should be registered and certified based on the official Act of the election by the Regional Council Secretary.And if the Secretary fails to provide the certification within 8 days, the President of the Council should do so (Art.8).
These legal procedures clearly state that the communities should design their own territorial authority structures and that the Council's role is simply to register the people elected.In practice, however, this is not what has happened.The Council has, instead, required all the territories to adopt standardised procedures, such as for the number and responsibilities of representatives.It failed to register territorial authorities that were elected in some territories, such as, for example, from 2006 to 2010 in Tasba Raya.In Li Lamni, a new election was simply convoked without taking into account the results of the election that had already taken place.Some representatives have received their certification immediately while others have had to return repeatedly over a period of months.In one case the authorities registered were not the ones elected.In still others, elections were pre-arranged or manipulated (for a more complete discussion of these issues, see Antonio 2008;Larson and Mendoza Lewis 2009;Larson 2010).Recently, the Regional Council refused to accept the leader elected in the community of Kamla, arguing that the vote should be secret -which is not how voting has occurred traditionally.Community members believe the underlying issue is that the elected leader does not sympathise with the party in power (interviews with territory leaders, March-April 2011).
In our research in three territories, it was clear that there were often tensions between community members and leaders, on the one hand, and territory leaders, on the other.In one territory, community members from most of the participating communities were unable to identify the members of the territory government or explain its functions.In another, communal leaders stated that they respected the territory leaders but did not receive the same in return.In two territories it was evident that few people were informed about the titling process.In all three, two salient issues were identified as problems in territory governance: lack of communication and lack of transparency in the management of funds (Larson and Soto forthcoming).Women in Matungbak were some of the most vocal critics.One group agreed, "The greatest threat is poor management by political leaders at all levels….They are the only ones who benefit from their efforts (gestiones); nothing comes to [community members]."

Powers over land and natural resources
With regard to decisions regarding natural resources, by tradition the síndico (or sometimes the wihta) is the authority in charge of land allocation and the implementation of resource use rules internally; by tradition and now also by law, this is usually the authority in charge of representing the community in negotiation with external actors, such as for signing contracts or receiving funds on behalf of the community.Though some síndicos have played a central role in lobbying for and defending community resource rights (Mairena 2007), in general síndicos have been notoriously unaccountable to the communities that have elected them, "selling" community lands and signing contracts for resource exploitation without providing reports -let alone transparent financial accounting -of these transactions; this may have begun to change, however, as important efforts have been made to increase leaders' accountability in a number of communities (Mairena 2007;Larson and Mendoza-Lewis 2009).
With the formation of territories and territorial authorities, the law now states that territorial authorities will be the ones to authorise exploitation for "natural resources of common use of the communities" comprising the territory.Hence, the implication is that community authorities will maintain the right to represent the community in outside transactions pertaining to community-level resources, while territorial authorities will do so with regard to resources not owned by a particular community but that are shared within the territory.Since communities are not individually demarcated, however, the actual powers and limits of each authority are not clear in practice.
Discussions with territory leaders demonstrate that the process of obtaining the approval for natural resource exploitation is inconsistent.In some cases, regional government officials -the ones usually lobbying for access on behalf of the government or a company -turn to the territorial authority, while in others they turn to the community authority -an example of forum shopping that permits the manipulation of outcomes.In both cases it is the signature of a single person that is legally required to grant access -and stories abound of the way in which such signatures are obtained (often involving alcohol, large sums of money and/ or trips to Managua). 6Similarly, by law communities should receive 25% of tax income from natural resource exploitation, but this is sometimes given to the territorial authority and at others, to the community authority (interviews with territory leaders, January-February 2011).
Notably, the process of obtaining resource access -whereby it is a regional government official who approaches the community or territory leaderdemonstrates a particular vision of development that is reflected in the Communal Lands Law.It is a vision that reflects the prior concession model, rather than any serious turn toward free determination or locally-designed "development".The Communal Lands law never refers to indigenous people, organised in communities or territories, as agents of development or, specifically, natural resource management.The law is written, at least somewhat, in the spirit of free prior and informed consent (FPIC), though it does not use this term.That is, community and territory authorities must authorise extraction by third parties.Even with regard to subsoil resources (for which FPIC is rarely required in national law), procedures are established to gain community consent if the community says no (though it is not clear what happens if it continues to say no).The law is not written, however, in the spirit of free determination, whereby rather than simply accepting (or denying) petitions for resource extraction, the communities themselves would design their own development path, solicit buyers for their resources or organise to manage and sell resources themselves (such as through community forestry).The law does not, however, in any way preclude this.

From imposition to free determination?
One result of the processes of decentralisation and devolution in the RAAN is the formal recognition and empowering of four layers of governance in natural resource decision-making below the central government: regional, municipal, territorial and communal.To what extent has this multilevel governance structure resulted in greater local voice in, and power over, decisions regarding natural resources?What potential does it have to do so?
Legal reforms have resulted in the creation and institutionalization of regional governments and established the right of regional officials to play a greater role in guiding the development of the region.Importantly, they have veto power over central government investment decisions.Most other specific powers and responsibilities regarding natural resources, however, overlap with central ministry jurisdictions and must be negotiated, thus conditioning regional government powers in practice.Even the unambiguous right to approve or veto concessions has not always been respected by the central government, as demonstrated in the Awas Tingni case (under a previous administration) and in the case of off-shore petroleum concessions (under the current administration).Nevertheless the current central administration has negotiated a governance arrangement with the regional government that has clear jurisdictions and that works mostly without conflict.The question for the future is what will happen to this balance under a more hostile central government.
Legal reforms have also granted indigenous peoples the right to land titles and to representation through their community and territory leaders.While the law is unambiguous about these rights, the respective jurisdictions of the two levels of leaders is not entirely clear; the right to free determination is left unstated; the term free prior and informed consent is not used; and clear procedures for free prior and informed consent have not been established.
As above, even the unambiguous right to define their own territories and territorial authorities has not been respected by the regional government, as officials and party leaders have pressured communities to follow other designs, manipulated elections and failed to certify the authorities chosen by communities.Neither the central, regional or municipal governments have enforced the law protecting the boundaries of the newly titled territories, failing to stop ongoing illegal colonization; rather, some public officials have encouraged it.And, if the regional government won the right to participate in the petroleum concessions decisions because these lands form part of the autonomous region, then how can they not also form part of the indigenous territories that comprise the autonomous region?
In summary, in spite of gaining the right to land titles and to participate in natural resource decisions, communities still do not exercise effective decisionmaking power over their territories and natural resources.Where the law is clear, court cases have sometimes led to its enforcement.But filing legal claims is expensive, and even in the case of Awas Tingni, the central government failed to comply until the community went beyond the country's borders. 7Legal ambiguities and overlapping jurisdictions facilitate the ability of more powerful actors to force their own agendas.Though the regional government is in a better position today in relation to central government, thanks largely to political alliances, indigenous communities -and their representatives at the community and territory scale -are still the weakest link in multi-level governance.
As argued by Berkes (2010), the concept of multilevel institutions -and, hence, the practice of multilevel governance -is still poorly understood.The RAAN case shows that more powerful actors maintain control and undermine the ability of communities to make decisions in their own interest.
Is the resolution of this problem just a matter of time, however?After all, the regional government has existed since 1990 but territorial governments are very new.Clear guidelines regarding the extent of each entity's power and responsibilities would likely improve the practice of multi-level governance.If these were largely technical issues, we could expect the problem of overlapping jurisdictions to be resolved through legal arrangements and negotiations, such as has occurred between the central and regional governments.But in that case, arrangements are based solidly on a political alliance that did not exist under previous administrations, and unless they are codified in law, they are at risk under a change of government.The legal jurisdictions of territorial governments do not yet exist in law, and a draft law has been written but, notably, not circulated.The relationship between municipalities and territories will likely be taken up in the National Assembly.It is unclear who will participate in these negotiations, but experience suggests that indigenous communities will probably not be consulted.
Also, as poignantly demonstrated by Fitzpatrick (2006) with regard to contested land tenure, certain more powerful players often benefit from lack of clarity, taking advantage of grey areas or "chaos" to manipulate outcomes to their advantage (see also the extensive literature on forum shopping).That is, not all flaws in the system are unintentional (Fitzpatrick 2006).With regard to leaders' failed accountability, we argue, in line with Chhatre (2008), that accountability is a dynamic process and that it must be constructed and performed.Given the lack of transparency of regional authorities for more than 20 years, accountability relations appear deeply entrenched in the current configuration of power relations.
In practice, then, understanding power relations is fundamental to understanding outcomes of multilevel governance -or, more accurately, the current moment in the evolution of this multilevel governance process.How it evolves -and whether local actors will indeed achieve the exercise of free determination -depends on ongoing negotiations in the context of these power relations and whether or not there are shifts in power.
Multilevel resource governance is also deeply embedded in the politics of relations between the central and regional governments and of indigenous struggles historically; it is deeply embedded in the economic interests that shape that historic relationship as well as the personal economic interests of its most powerful participants.The whole multi-faceted sphere of decision-making is only minimally oriented toward "environmental problems" or even "natural resource management".Relations among actors are not calculated on the basis of social learning or adaptive management.
Multi-level governance may involve a patchwork of institutions and new, local participants who have a degree of voice for the first time, but this does not mean they will have decision-making power.The extent to which they have the right to make decisions over resources depends in part on the law, legal ambiguities and conditions on that right.The extent to which they have the power to make decisions depends on the processes that these changes unleash.
How can multi-level governance avoid simply creating more weak links and the semblance of greater participation and power?How can indigenous communities exercise their rights?How can they begin to break free of the "imposed project" and to formulate their own, freely determined, future project, for development, wellbeing or buen vivir8 ?We argue that the only practical solution is to tip the power balance through the empowerment of local actors -through specific interventions to educate, organize and facilitate their ability to bring about change.
The new rights in law and the formation and titling of territories have unleashed new potential for self-determination in the region.The current dynamic institutional context has fostered discussion and debate and certain challenges to top-down conceptions.The diploma course and community-level research upon which most of the findings here are based constitute small steps toward an attempt to promote dialogue and empower local leaders.

Conclusions
Multilevel governance can be facilitated or mandated by law, but the extent to which it works in practice depends on the relative power of different actors, including the ability to draw on resources, such as the courts, to demand accountability and force one's voice to be heard.The regional government of the RAAN has faced its own battles with the central government, but has tended toward a recentralization of power at the regional level rather than giving voice to the people living in indigenous communities and territories.Legal reforms that would support communities include the development of effective FPIC mechanisms, defining the right to free determination at the community or territory scale and clarifying the relationship between territories and municipalities.
Today the region's indigenous territories are mostly designed and titled, though not always in the way that communities would have desired.Their territorial representatives have been chosen, again not necessarily the ones elected.But initiatives can be taken to help level the playing field: first, to foster a climate of dialogue and debate, and help overcome the fear of retaliation associated with challenging top-down decisions; second, to empower local leaders by increasing their understanding of their own territories and their capacity to negotiate, fostering conditions for the design of endogenous future plans for their territories.
Communities may yet surprise their elected regional officials.

Figure 1 :
Figure 1: Nicaragua's North Atlantic Autonomous Region (RAAN).Source: Elaborated by Edwin Matamoros based on official and unofficial data