1In the wake of democratisation and the demand for national conferences in African countries in the late 1980s, African governments and international agencies manifested a renewed interest in decentralisation. Decentralisation was seen as a means by which the over-centralisation of political decisions at the top of the state could be altered. Civil society actors and political opponents supported decentralisation for its potentials to favour the devolution of central power to local institutions and structures. Decentralisation became one central element of democratisation processes, making some refer to current policies as ‘democratic decentralisation’ (Crook & Manor 1998). Others see decentralisation as a legal technique of territorial administration, and as a political mode of sharing powers between central and local authorities in a country (Mback 2003 :31-2). The last decade has seen a growing literature on decentralisation policies and their political and economic implications (Bierschenk & Olivier de Sardan 2003 ; Blundo & Olivier de Sardan 2006 ; Crook & Manor 1998 ; Dahou 2003 ; Fay et al. 2006 ; Kassibo 1997 ; Laurent et al. 2004 ; Mback 2003 ; Sawadogo 2001 ; Seely 2001 ; Totté et al. 2003). With years of experience of decentralising West African States, it is time to focus on everyday practices. Instead of describing how decentralisation should be, ought to be, or, once problems arise, should have been, I propose that we look at the wide range of activities that are carried out ‘in the name’ of decentralisation. Hence, I address decentralisation as a series of political, economic and cultural practices currently at work in local arenas across West Africa. Although these practices are neither coherent nor uniform, they are commonly invoked to justify various claims made by political actors.
- 1 In addition to the five articles constituting the thematic issue of this APAD-Bulletin I am very h (...)
2The collection of essays assembled in this APAD-Bulletin revisits decentralisation and citizen participation in local arenas. All articles concern decentralisation as part of political and cultural processes at work everywhere in West Africa and, yet, differently articulated depending inter alia on country, context and sector. In some cases we see a ‘decentralisation by default’ (Andrae) in the sense that the withdrawal or disengagement (Körling) – or absence – of the state opens up for non-state actors. In others we listen to the voices of voters who have been abandoned by political parties and still continuously vote for the majority party (Rudebeck). We see how pastoralists losing out in decentralisation adopt a strategy of territorialisation of land (Mohamadou). We also look at decentralisation in various societal sectors, such as water provision (Andrae), education (Körling), party politics (Rudebeck) and state administration (Mohamadou). In addition, all articles deal, in one way or another, with citizen participation through different local actors’ struggles over political, economic and cultural resources. Decentralisation practices concern the ways in which citizens are appropriating politics and making sense of state policies, as well as the means to solve daily problems for making ends meet. Meanwhile, these practices also articulate processes of exclusion and inclusion of different social categories.1
- 2 The ideas developed in this article were first formulated in a key note to the international works (...)
3In this article I attempt to develop some general remarks on decentralisation practices, leading to a conceptual reflection of how the local is currently invented and mobilised.2 The purpose is to understand the mundane everyday practices of decentralisation in West Africa by reflecting upon the different ways in which the local is transformed. In addition to conclusions from the in-depth case-studies presented in the four articles concerning Guinea-Bissau, Niger and Nigeria, I also develop my argument by drawing upon long-term fieldwork conducted in Burkina Faso and a recent short-term fieldwork in Mali.
4Across West Africa we may identify three transformative processes favoured by current decentralisation practices. First, decentralisation has changed the politico-administrative stakes in local arenas across West Africa. Whereas decentralisation has historically been associated with the creation of urban communes, current practices seek to create rural communes far away from urban centres (Bierschenk & Olivier de Sardan 2003 ; Fay et al. 2006 ; Hahonou 2006; Koné 1997; Laurent et al. 2004; Sawadogo 2001 ; Olivier de Sardan 2003). Today rural districts are involved in, and affected by, municipal elections. Second, the primacy of urbanity and modern education are recast in present-day decentralisation. The fact that citizens now have the possibility to elect mayors with little or no formal education is hotly debated in many West African countries. It is frequently held that mayors without skills of budgeting and bookkeeping are exposed to all forms of manipulation and dependency. And, still, most locally legitimate political actors are not necessarily those who have been to school (Camara 2006 ; Hagberg 2002). Third, decentralisation attempts to marry legality and legitimacy, and has often been interpreted as the return of traditional chieftaincy. While such interpretations are not unambiguously accepted and supported by all political actors, it is a political fact that discourses of autochthony, whereby family decent and ethnic belonging are made criteria for defining legitimate political actors, are increasingly ventured in local political arenas (Buur & Kyed 2007 ; Geschiere & Gugler 1998 ; Geschiere & Nyamnjoh 2000 ; Kassibo 1997). The reigns of the city must, so it is often argued, be held by a son of the locality (cf. Hagberg 2006a). Common to these three transformative processes is that they demonstrate how seemingly unproblematic categories – ‘rural’, ‘local’ and ‘home’ – are filled with new meanings.
5Beyond these transformative processes, the very definition of the local and its boundaries is highly problematic, and subject to a wide range of interpretations and meanings. Discourses of the local are a central feature of political culture in West Africa (and certainly elsewhere) in the era of decentralisation and, yet, ambiguously referred to by political actors. On the one hand, the local is positively referred to in elections. The necessity to bring ‘local development’ and work for ‘local people’ is a recurrent theme. Politicians often seek to mobilise ‘the local electoral constituency’ by invoking their status as sons or daughters of the village or the district. It is not rare that politicians represent a village where they trace ancestry but in which they have never lived. The local also represents tradition and belonging. In village rituals and cultural manifestations, the local is associated with origin and authenticity. Only persons who have undergone initiation may attend many village rituals, such as agrarian rites in the sacred grove. On the other hand, the local is simultaneously attributed negative connotations by political actors when they represent the village as backward, static and inward-looking. Being ‘the local representative’ of a political party or a state service is fraught with ambiguities as it may stand for dynamic brokerage and conservative backwardness at the same time. The negative connotations of the local have for a long time been reproduced within state administration. In Burkina Faso local government officials are often labelled fonctionnaires de brousse by their peers higher up in the hierarchy (Hagberg 2005). In addition to these ambiguities it is important to remember that not all social actors fit into the category of the local. Fulbe pastoralists are often excluded as they are perceived as strangers, foreigners and those coming from far away (Hagberg 2000). Other social categories often marginalised in official decision-making processes are ‘migrants’ and ‘women’. Therefore it is necessary to explore the ambiguous notions of the local that prevail in the era of decentralisation.
- 3 For the sake of clarity let us accept, in Mark Hobart’s phrasing, that ignorance “is not a simple (...)
6For political actors, there is a need to invent the local in order to be able to mobilise it. In such processes of inventing and mobilising the local, decentralisation practices have reshaped the fundamental relations between citizens and the State. From a top-down and highly authoritarian State with its colonial legacy we see a process that at least holds promises of a possible change because, despite manipulations, frauds and deviations, democracy provides the framework for present-day local politics in West Africa. Even when national politics experiences back-lash, e.g. the personalisation of presidential power, local democratic processes tend to prevail. Inventing and mobilising the local furthermore opens up for the possibility of recasting the distinction between educated and urban-based salaried actors, and ‘the illiterate’ and ‘the ignorant’3 ones in rural areas, or perhaps better, the divide of citizens and subjects in the bifurcated legacy of the colonial State (Mamdani 1996). In this vein, practices of decentralisation present new ways in which the postcolonial state is re-appropriated, while strengthening the agenda of certain categories of political actors.
7This introductory article in which I develop an argument on how the local is invented and mobilised by political actors is organised as follows. Firstly, I make an overview of decentralisation policies so as to situate current practice in the colonial and postcolonial history. Secondly, I discuss the invention and mobilisation to explore actors’ ambiguous stance towards the local. Thirdly, I propose that we need to simultaneously study political, economic and cultural dimensions if we are to understand current decentralisation practices. Fourthly, I develop some examples of how specific social categories (here exemplified with ‘women’) may be discursively included while being practically excluded. Fifthly, I introduce the four case-studies contributing to this thematic issue of the APAD-Bulletin as to highlight everyday decentralisation and development across West Africa. In conclusion, I argue that we need to seriously integrate a cultural analysis of decentralisation and citizen participation in order to understand the heat and intensity of public debate and the strong emotions related to what at first glance may seem to be marginal political and economic issues.
8Much hope has been invested in decentralisation policies since the early 1990s when decentralisation was presented as something new, a radical breach with past authoritarian regimes. Yet various attempts to decentralise, that is, the deliberate and planned transfer of authority and resources away from the central state institutions to peripheral institutions, have for a long time been the subject of experimentation. And, locally elected government bodies in urban areas are not a new phenomenon in West Africa. The Quatre Communes were erected in Senegal by the French colonial administration in the 19th century: St. Louis and Gorée in 1872, and Rufisque in 1880, and Dakar in 1887. Residents of these communes became French citizens whereas the rest of the population of the territory remained African subjects (cf. Mamdai 1996). In other words, the creation of communes required the distinction between citizens and subjects. In the first half of the 20th century urban communes were created elsewhere in Africa. The communes of Ouagadougou and Bobo-Dioulasso were, for example, erected to mixed communes of 1st degree in 1926 (Sawadogo 2001). An administrative mayor governed these communes with the assistance of a municipal commission. The commission was composed of three different categories of members : first, there were nominated members ; second, there were elected members with a limited electoral base ; and, third, there were elected members by universal suffrage (Sawadogo 2001 :207 ; Fourchard 2001). The salient feature of these early attempts to decentralise was that they involved urban communes with a high degree of French residents. Hence, decentralisation was partial, and concerned only a small share of the population.
9A word of caution is needed here. Decentralisation as a mode of governing tends to focus on the colonial and postcolonial State and, consequently, disregarding more or less traditional modes of rule. Yet it is important to keep in mind that decentralisation articulates state-society relations not only related to the postcolony, but also to traditional socio-political structures. For instance, although a hierarchical kingdom is more likely to find locally legitimate political structures in the context of decentralisation, what have been called stateless or segmentary societies are likely to appropriate decentralisation quite differently (Amselle & M’Bokolo 1985 ; Hagberg 1998, 2004b ; Laurent 1995 ; Laurent et al. 2004 ; Lentz 2000 ; Savonnet-Guyot 1986).
10Bamidele Olowu locates the first phase of decentralisation, at that time labelled as a system of local government, from the World War II until the early 1960s. During this phase decolonisation was on the agenda due to a variety of reasons : reward for the colonised peoples’ participation in the war, agitation by the growing elite from the colonies, and the rise of social liberal parties in Britain and France (Olowu 2001 :6). The development of an efficient and democratic local government was seen as a key to success in African administration. Yet, the decentralised focus was abandoned with national independence. In the 1960s the newly independent African states subscribed to central planning and consolidation of the nation-state via the single party mechanism. Local administrations were essentially instruments of control, and the relationship between the centre and the periphery was strongly hierarchical. When the economic crisis of the 1970s emerged most African states responded by adopting structural adjustment programmes. Decentralisation then presented itself as a mechanism for cutting back central government expenditures (Olowu 2001 :8), and was more economically than politically motivated. Yet despite these policy responses, the structural adjustment policies incited what has been termed ‘decentralisation by default’, that is, “a situation in which a variety of non-state organisations filled the void left by the absence of state institutions in the production of goods and services” (Olowu 2001 :11).
11Since the early 1990s, however, several West African countries have undertaken reforms to devolve political and administrative power to elected local government bodies in rural areas, such as communes and regions. In contrast to previous decentralisation efforts these attempts were conceived of in the context of political liberalisation and democratisation, and the policies have sometimes been referred to as ‘democratic decentralisation’ (Crook & Manor 1998). The models adopted in the 1990s differ considerably with respect to the composition of local government institutions and to the powers and resources that they should control. But, decentralisation policies have been differently implemented in various countries. Whereas in Senegal rural communes have been in place for more than thirty years, they are relatively recent in Mali (Fay et al. 2006 ; Kassibo 1997). Niger held its first local elections in July 2004 (Hahonou 2006 ; Olivier de Sardan 2006). In Burkina Faso decentralisation has been under way for a decade (Hagberg 1998 ; Laurent et al. 2004 ; Sawadogo 2001), and the communalisation was completed with local elections in April 2006. And, yet, Ghana has been involved in a decentralisation process since the early 1980s (Amanor & Annan 1999 ; Ayee 1996 ; Crook & Manor 1998). The federation of Nigeria is a different case, even though the relationship between the federal government and the different states is much debated with respect to the adoption of the shari’a as civil law by several states (Olowu 2001).
12This overview demonstrates the necessity to reflect upon the concept of decentralisation itself. It is indeed relevant to question whether one can use this concept to cover all these different policy propositions mentioned so far. Is it useful to see the creation of the Quatre Communes in Senegal in the 19th century as decentralization ? By doing this we may run the risk of being trapped into a purely legalistic perspective intimately associated with the colonial and postcolonial State, overlooking segmentary socio-political structures outside the modern state in, for instance, northern Ghana and southern Burkina Faso (Hagberg & Tengan 2000 ; Savonnet-Guyot 1986).
13At a very general level decentralisation is a distribution of power that is horizontal rather than hierarchical (Kasfir 1990, cited by Seely 2001), and depicts the transfer of power away from a central authority to lower levels in a territorial hierarchy. Yet, the transfer of power from a centre to a periphery involves the construction of such a periphery in the first place ! In order to make the transfer of competences to local governments (collectivités locales) the latter must already exist as an administrative entity (Mback 2001). So, while decentralisation is the transfer of power a critical question is which are the institutions, structures and actors that should be granted this power. When local decision-making is advocated, the local is as much a social construction as other political categories and units.
14In the mainstream debate on decentralisation a distinction is often made between deconcentration and devolution. The concept of deconcentration refers to the delegation of responsibility and authority to field units of the same department or level of government. It extends the reach of the central government to strengthen its authority “by moving executive agencies controlled by the centre down to lower levels in the political system” (Crook & Manor 1998 :6). Yet what is called devolution has the opposite effect, “since it cedes control of such agencies and resources to political actors and institutions at lower levels” (Crook & Manor 1998 :7). The concept of devolution indicates the transfer of authority to locally constituted units of government or special purpose authorities.
- 4 “Vous déciderez vous-mêmes de vos affaires”.
15The distinction between deconcentration and devolution is reasonable for understanding the general concept of decentralisation. According to a legal-administrative perspective it makes a difference whether the central government rules through local representatives or the local government is invested with autonomous decision-making powers. Yet while at the policy level the distinction between deconcentration and devolution makes sense, once we move to actual practice the distinction is less clear-cut. Decentralisation is, like any other centrally implemented policy, subject to a multitude of local interpretations. While some political actors see decentralisation as part of a patrimonial agenda aimed at preserving the monopoly of power and ensuring control over resources, others pursue reformist objectives aimed at creating more transparency, accountability and efficiency in the management of local affairs (De Jong et al. 1999 :5). For donors, decentralisation is supposed to open many doors : “decentralisation is not only supposed to lead to the introduction of local democracy and political accountability, it is also supposed to activate the dynamics of development at local level” (Bierschenk & Olivier de Sardan 2003 :145). An unofficial interpretation of decentralisation says that “you decide on your affairs yourself” (Mback 2003 :36).4 In Mali an impressive work was invested in the 1990s in order to find appropriate local notions for the concept of decentralisation. Two main terms in Bambara express power. While the term mara refers to the exercise of authority at all levels of society, the term fanga implies force, strength. The concept of decentralisation was translated into mara ka ségi so, that is, ‘the power returns home’ (Béridogo 1997 :30). But the precise location and meaning of ‘the home’ was sometimes questioned, and gave rise to intense debates on political legitimacy. Once again we see that the construction of ‘the local’ (or ‘the home’) is central to decentralisation practices.
16In this section I discuss how the ambiguous concept of the local is invented and mobilised by social actors. Localism is a common term used to denote a wide range of ideas and practices that support ‘local production and consumption of goods’, ‘local control of government’, and ‘local culture and identity’. However, the local is not an unproblematic category, but rather a social and political construction that is filled with different meanings by various actors.
17In classical anthropology ideas of locality and belonging were for a long time regarded as juxtaposed kindred concepts. Yet, as pointed out by David Parkin, “anthropologists can no longer assume that the people they study see themselves as attached to a particular, bounded locality” (Parkin 1998 :ix). Rather than being a erritorial reference point, locality could be a starting point for understanding concepts of origin, home, village and, ultimately, belonging. In the introduction to a volume on locality and belonging, Nadia Lovell develops how belonging to a particular locality evokes the loyalty to a place. Belonging is multifarious and related not only to social relations and rights, but also to emotions. “The exploration of how notions of belonging, localities and identities are constructed seems particularly relevant in current political contexts of ‘globalisation’, where the interface between localised understandings of belonging, locality and identity often seem to conflict with wider national and international political, economic and social interests” (Lovell 1998 :1). Gupta and Ferguson argue for an anthropology that is less an evocation of the local and more an exploration of the location. Anthropological fieldwork should be less of “a time-honoured commitment to the local but with an attentiveness to social, cultural, and political location” (Gupta & Ferguson 1997 :5). Shifting our own location means that one may also explore links to other political or epistemological locations. That is why we need to represent the local in decentralisation practices as an emic concept. So, while all people are located, notions of the local are socially, politically, economically, and culturally constructed.
- 5 A study on market women and taxi-drivers in urban Ghana describes that democracy is popularly defi (...)
18In colonial time the definition of administrative units was a critical task for state administration, mostly for reasons of governance and efficiency. The impact of colonial administration on ethnicity and belonging can hardly be contested today, but the relative importance of ethnicity in colonial statecraft is debatable. To my understanding ethnic identities were not only the result of colonialism, but, in Philip Burnham’s words, the colonial period was “a particularly productive moment for ethnic political ‘projects’” (Burnham 1996 :158). In situations of dramatic change – the prime example is the colonial situation – one could expect that ethnic discourses react to change and rupture as to assert distinctions between ‘Us’ and ‘Them’. Today’s decentralisation and political pluralism could, I would argue, well be understood as another ‘particularly productive moment for ethnic political projects’. It is in this context of socio-political and economic change that we may understand the construction – or, as I would prefer, the invention – of the local, because in ‘democratic decentralisation’ the interconnection between group and locality is given new political, economic and cultural meaning. When prominent members of an ethnic group see that they should occupy their ‘own’ municipality, issues of belonging and autochthony are likely to be ventured in party politics, associational life and economic infrastructure investment (Geschiere & Nyamnjoh 2000 ; Hagberg 2004a, 2006b ; Lentz 2000; Zougouri 2008). In this process references to the local that result from the specific historical context also resonate with localism, because it is hard to argue against attempts to decentralise West African countries that have historically experienced such a high degree of centralism (cf. Young 2003). Yet as shown in many recent writings on local arenas in West Africa (e.g. Bierschenk & Olivier de Sardan 2003 ; Hagberg 2004b, 2006b ; Laurent et al. 2004 ; Zougouri 2008), decentralisation practices have often led to sharpened identity politics and community conflict. Today many actors see local democracy as synonymous with dispute and conflict. For instance, a local politician in Burkina Faso told me recently: “Since the municipal elections in 2006 my village is divided”. Hence divide and rivalry rather than unity and understanding are associated with democracy. Freedom of speech is of particular importance ; democracy has popularly come to imply that nowadays people may say whatever they want without consideration for other people.5
19The struggle over definitions and boundaries of the local urge us to be cautious with the concept itself. With respect to decentralisation, Olivier de Sardan (2006 :408) conceptualises the local as both an arena where heterogeneous actors intervene with local and external resources, and an emergent (and state-centred) public space where state representatives and representatives of the population (or citizens) interact around multiple norms with respect to access to public goods and services. To my mind, Olivier de Sardan’s approach of the local as ‘an arena’ and ‘a public space’ has the methodological advantage that the local becomes tangible and possible to study. However, the local is more than that. It also pertains to political imagination, cultural representation and symbolic meaning. Therefore, I would to argue that the actors’ perceptions of the local – the emic point of view – need to be integrated into our analysis.
- 6 In Francophone Africa the notion of pays (‘country’) as to indicate the territory associated with (...)
- 7 Congrès pour la Démocratie et le Progrès.
- 8 Rassemblement pour le Développement du Burkina.
20The theme of this issue of the APAD-Bulletin is Inventing and mobilising the local. By inventing I would like to draw attention to the fact that the local does not exist ‘out there’ as a category itself, but must be constructed or created in order to make sense. Entities referred to as traditional or ancestral by social actors may be recent creations. The colonial entity of canton was most often a new creation albeit drawing bits and pieces of historical entities. A very telling example is that of Koné Gnagwan, a former interpreter who, thanks to his political skilfulness, became the canton chief of Karaborola and Komono in the Banfora region of present-day Burkina Faso (Soma 2000 ; see also Hagberg 1998). From 1921, Koné Gnagwan, a Gouin from Bounouna, ruled over Karaboro ‘country’ and in 1938 his reign was extended to cover Komono and southern parts of Tiefo ‘countries’.6 Koné Gnagwan was feared and ruled over life and death because this was the time of forced labour. But, in 1946 he was removed due to power abuse (Hagberg 1998 :93-94; Soma 2000). Still, Koné Gnagwan is today a legend of the Banfora region. Over the last decade a main political actor in Banfora was Koné’s grandson, the late Mamadou Koné. Seen as a young dynamic private entrepreneur in the construction business, he also based his legitimacy on the ancestry. Mamadou Koné was an MP for the ruling party CDP7 and then turned opposition politician for RDB8. In April 2006 he was elected mayor of Banfora against all odds as he contested the ruling CDP. With strong popular support and mobilisation Koné won the elections. However, unfortunately, he passed away later the same year suffering from an incurable disease. Some said Mamadou Koné died from cancer, whereas others held that political adversaries killed him by poison and/or witchcraft. In any case, the late Koné Mamadou based his power on his political dynamism and capacity to mobilise in conjunction with the historical legacy of his grand-father. In this example we see how invention is a process where political actors use the bits and pieces of political imagination, collective memory, and the pragmatics of power. While it is not ‘pure invention’, it results from the conscious attempts to create meaningful and workable political entities. Yet it is not enough to invent a historical legitimacy, one equally needs to have the capacity to mobilise. The example of the late Mamadou Koné therefore points towards the necessary combination of legality, legitimacy and dynamism of political actors in the process of inventing and mobilising the local.
21The capacity to mobilise the local is a central challenge for any political actor. For instance, the visibility of political action is an important element for mobilising voters, state support and international donors. I have often heard how people refer to political actors in Burkina Faso along the following lines : “X est devenu une pièce incontournable pour toute intervention dans le village”. The politician has great interest to become a gatekeeper or broker. A political actor who succeeds in becoming the broker able to mobilise actors and resources in ‘the local’, as well as actors and resources from ‘the global’ (most often in the form of state intervention and development aid) is likely to become the gatekeeper for any political action, development intervention and other collective action to be undertaken (Bierschenk et al. 2000 ; Lewis & Mosse 2006). The invention of the local therefore goes hand in hand with its mobilisation.
22Decentralisation processes carried out so far in Africa remain utterly contradictory. Despite the important steps taken, the strong policy focus obscures much of actual decentralisation practices. Firstly, the ‘irreversibility’ of decentralisation so often invoked is more of wishful thinking than grounded in empirical observation. The first decade of full-scale territorial communalisation in Burkina Faso, Mali and Niger currently shows sign of fatigue and increased questioning of decentralisation. One sign of this fatigue could be detected in the idealisation of the time of military presidents like Séyni Kountché 1974-87 (Niger) and Thomas Sankara 1983-87 (Burkina Faso) as it articulates local actors’ longing for a central and powerful state. Secondly, while it is easy to be impressed by the social engineering in, for instance, Mali’s decentralisation, a political analysis more than a policy analysis of decentralisation is needed to understand why ex-President Konaré pushed for its implementation (Seely 2001), whereas his successor Amadou Toumani Touré – popularly called ATT – has shown much less interest in decentralisation. Thirdly, attention must also be paid to less positive developments that prevail in the context of decentralisation. Despite Senegal’s early decentralisation attempts, corruption is, in the words of Giorgio Blundo, a fact of decentralisation, not as a lack of mechanisms in its implementation, but as a mode of local governance (Blundo 2001). Fourthly, while decentralisation cannot be attributed all forms of transformations and diversions at work in municipalities across West Africa, the neo-liberal strive to ‘de-scale’ (Andrae), to ‘disengage’ (Körling), or to ‘decentralise’ (Mohamadou, Rudebeck) all involve the conscious politico-administrative transfer of power away from a centre to a periphery. In this section I discuss decentralisation processes in relation to how social actors in the local aim to acquire power, livelihoods and cultural meaning.
23The first dimension of decentralisation is political as it goes without saying that decentralisation affects the distribution and the exercise of powerwithin society. Therefore, the materialisation of decentralisation depends not only on policy and legislative reform, but also on the political will of the central government to transfer real, discretionary decision-making powers to local government bodies. An elected mayor in Mali told me in 2003 that “we cannot wait until the power and the resources are given to us by the central government, we have ‘to pull out’ the power”. Another political aspect is related to the legitimacy of the local government. Local governments may well be legal in that they have acquired decision-making powers without being legitimate vis-à-vis other local power holders (particularly traditional chiefs). One notes, for instance, that land tenure is likely to be a contested issue ; local governments may have the legal right to exercise power whereas traditional chiefs uphold legitimacy (Benjaminsen & Lund 2003 ; Juul & Lund 2002 ; Kuba & Lentz 2006 ; Lund 2006). But power is also central for understanding the relations between local governments, traditional institutions and the so-called civil society in terms of citizen participation, representation and accountability (Hagberg 2004a, 2006b). To some extent decentralisation complicates local political arenas, because rather than legislating away former power structures decentralisation represents a new layer of political institutions. Bierschenk and Olivier de Sardan argue that decentralisation is not a radical rift opening the way to good governance, but that “the current processes of decentralisation in Africa merely represent another moment in a long series of regime changes imposed from above by the state and experienced in Africa since the end of the Second World War” (Bierschenk & Olivier de Sardan 2003 :167).
24The second dimension of decentralisation concerns economics, or perhaps better, livelihoods and local economic development. Even though decentralisation introduces new political relations between the electors and the elected, its raison d’être largely relies on newly elected government bodies’ abilities to support the livelihoods of local people and to address everyday problems of poor health and education facilities, inadequate water supplies, and lack of employment, marketing and investment opportunities. This requires ensuring effective management of natural resources, the provision of appropriate, efficient and affordable services, the seizing of new economic opportunities, and the reconciling of competing interests of different social groups. In economic terms decentralisation has, positively or negatively, an impact on people’s livelihoods. The paying of taxes and the increased control over people’s livelihoods through land tenure and cattle markets are likely to have implications for rural producers’ capacities to subsist. However, the logics of micro-projects seem to be more oriented towards ‘the management of poverty’ (Mback 2001 :110-111). Donors that have been supporting decentralisation processes often undermine local governments by channelling public aid through NGOs, leaving district assemblies with scarce resources. Another economic aspect is the decentralisation of patrimonial networks. Giorgio Blundo identifies the decentralisation of corrupt practices, what he terms décentralisation détournée, as emanating from the logics of bureaucratic predation coupled with the logics of factional sharing of the resources that are administered by members of local governments (Blundo 2001 :124). And in local discourses on corruption, the corrupted is not the one who has eaten per se, but the one who has eaten in an egoistic manner (see also Blundo & Olivier de Sardan 2006).
25The third dimension of decentralisation is cultural in the sense that it involves people’s perceptions of, and meanings attributed to, decentralisation. Beyond political and economic practices of decentralisation, there is a need to better understand the cultural meaning of decentralisation (Hagberg 2004b ; Laurent at al. 2004). Even though it is generally admitted that cultural practices of decentralisation are important, these practices are rarely addressed in any systematic manner. Decentralisation may enable a re-appropriation not only of local resources but also of local cultural identities and values. It could allow for local arrangements building on local rules and institutions (Djiré & Dicko 2007). By devolving more power to local communities decentralisation may help make sense of the postcolonial state administration and, at best, favour the emergence of local democratic culture. In Mali the elected mayor rather than the former sous-préfet, who is appointed by the state, is nowadays the strong political actor. This represents a radical shift in terms of power, but it has cultural bearings as well. The tricky issue of how to draw politico-administrative boundaries prevails in most countries and articulates with belonging and exclusion. Niger has opted for preserving the former canton boundaries whereas Mali has had serious problems in getting economically viable communes and has instead seen the creation of micro-communes (Koné 1997a ; Koné 1997b). The reshuffling of provincial boundaries has often been interpreted culturally in the sense that those on the other side of the border are different. A hunter leader explained to me when a hunters’ association had been divided into two as a consequence of decentralisation’s new provincial boundaries : “It is good because of languages. We do not understand them. We the Karaboro, Gouin and other peoples here [in the Comoé Province] we understand each other but not them [that is those in the new Léraba Province][…] But the government has regrouped us. Now we are divided and it is not bad” (Hagberg 2004b). In other words, the politico-administrative reshuffling was justified culturally by this hunter leader.
26In addition, ‘culture’ may be manipulated by local elites to back their claims over local power and resources in the process of the invention of tradition (Hobsbawm & Ranger 1983 ; Ranger 1983 ; Buur & Kyed 2007 ; Perrot & Fauvelle-Aymar 2003; Hagberg 2004a, 2006a, 2006b, 2007). This has been articulated as ‘the return of the kings’ (Perrot & Fauvelle-Aymar 2003) or as ‘power returns home’ (Béridogo 2006 ; Kassibo 2006 ; Koné 1997b). The Nigerien case is most obvious where the canton chiefs are still on the payroll of the state. But it is a common feature that descendants of the former canton chiefs are among today’s power-holders in the capacity of mayors and municipal advisors. In Burkina Faso, the Emir of Liptaako was mayor of Dori 1995-2000 and the traditional chief Pô Pê was the mayor of Pô in the early 2000s. And when the chief himself remains outside the local government, a younger brother, a son or another close relative is likely to be among the local power-holders. But in this process traditional chiefs are not exempt from the public gaze. In 2000-2001, Burkina Faso experienced three cases of political violence in which chiefs were deeply involved (Hagberg 2007). In these cases the fuzzy boundary between traditional power and modern politics was one key element. Another was that these chiefs became publicly criticised for acting far beyond the limits of the rule of law.
27The cultural dimension of decentralisation practices is thus ambiguous. Decentralisation provides the opportunity for chiefs, legitimised by tradition and culture, to access new political and economic resources, but decentralisation also exposes tradition and culture to public scrutiny, debate and contest. Cultural practices of decentralisation fundamentally concern exclusion. Some political actors are defined as legitimate and apt for political offices, while others are excluded. Social categories, such as women, nomads, latecomers and various groups of casted peoples are particularly vulnerable when ‘culture’ is instrumentally fed into local governments.
28In current decentralisation practices women are ambiguously portrayed and represented in local mobilisation. Initially women tend to loose out when power is transferred to the local, but given the strong advocacy for gender and development they simultaneously seem to gain weight with decentralisation. In this section I elaborate on two examples of women in the local so as to reflect upon the constraints and opportunities of moving beyond what I would like to label ‘the female space’.
29The first example illustrates that decentralisation not only implies the devolution of power but also a re-centralisation at local level. When local administration is strengthened, citizens get closer to public authorities. But the public and administrative sphere of people’s life is simultaneously enlarged, formalised and recentred by decentralisation. In committees and councils one or two seats are generally reserved for women, but as a matter of fact they do not compete for political power with the male actors. Similarly, in municipal elections women tend to be less represented locally than in national assembly. And, still, political competition between women tends to be seen as ‘a women’s affair’.
30In recent years there has been a fierce struggle among women’s political leadership in Burkina Faso’s second city Bobo-Dioulasso. With legislative elections in 2007 there was competition about which one among leading women politicians of the city who was to be on an eligible position of the majority party CDP’s list. In 2008 the competition was once again launched when the coordinating structure for women’s associations under the Ministry of Women’s Promotion was to be elected. One faction was led by a female MP, who has for long time been a leading personality of the city and who is the president of a collective of HIV/AIDS associations. The other faction was led by the deputy mayor of Bobo-Dioulasso and who is also the president of a collective of environmental associations. Both women leaders represented the ruling party CDP, because at least officially the struggle concerned a civil society organisation (la coordination provinciale des associations féminines) under the umbrella of the ministry. In fact this structure is positioned at the interface of party politics, state administration and civil society.
31While the battle included many intricate details that I will describe elsewhere, suffice to note that the battle was about who should be the women’s leader in the city rather than how to increase the number of women in politics. The two factions were competing for the space available for women in the local, while leaving the remaining male-dominated largely space untouched. This observation would perhaps lead to a pessimistic conclusion of the possibilities to enhance women’s political participation. But the case of women’s factional battle could also represent a shift in which women leave the public political category of caring mothers and sisters, and bring the fierce battle and factional fight to the fore.
32The second example demonstrates that when women in the local are able to influence politics and development, they are often legitimised in terms of their husband, father or other male relatives. A dispute around a cereal mill offered to women in Gongasso in Southern Mali is particularly revelatory. Different women’s associations and groups entered into conflict about the management of the mill. The mill was a donation to ‘the women of the village’. While according to one version the mill came from a local organisation called Centre de Santé Communautaire (CSCOM), another version held that the mill was a gift from the ruling political party Alliance pour la démocratie au Mali (ADEMA, party in power since the early 1990s). The prominent women leaders in Gongasso were ‘the wife of the village chief’, ‘the wife of the mayor’, ‘the wife of the Imam’ and ‘the wife of the schoolteacher’. Although such labelling does not necessarily exclude these women’s agency outside their husbands’ influence, women do gain legitimacy through their matrimonial position (Hagberg et al. 2009). In addition, many actors take on the task to speak ‘in the name of women’, especially given the fact that international development agencies give priority to fund women’s activities. However, the mill that was donated to the village women was later captured by political actors and outside women’s control. It ultimately ended at the village chief’s courtyard. This example shows that while many actors speak in the name of women, local women are often unable to influence key management decisions when it comes to decentralisation and development.
33These two cases clearly exemplify the problems related to women in the local and the importance of looking at processes of exclusion in decentralisation practices. Yet it should also be stated that the politics of exclusion has also meant that public debate and discussion increasingly highlight those who are loosing out. In Burkina Faso women’s representation on electoral lists has become an issue and in the 2006 elections the ruling party CDP decided a quota on electoral lists. In 2009 a law on quotas was enacted by parliament meaning that at least 30% of eligible seats on the electoral list should be reserved for women. These steps are important in the sense that they prepare the ground for increased political participation of women, but resistance is continuously strong in local arenas.
34The articles of this APAD-Bulletin represent various scholarly, geographical and thematic locations, but they are all held together by a concern for how the local is mobilised in daily political and development actions. Human geographer Gunilla Andrae draws upon fieldwork from Kano in Nigeria to raise the issue of water provision and decentralisation by default. She points to how the liberalisation of water supplies has dramatically changed the provisioning patterns in African cities. The dominating form for liberalisation of urban water supplies has been the take-over by transnational companies of previously publicly owned water utilities. With support of the World Bank, State authorities have let go supply functions as well as regulating functions concerning water supplies, especially targeting poor populations in urban peripheries. As a consequence, the field has opened up for a wide range of other actors outside state control, meaning that water provision is “de-scaled without democratic substance”. Andrae investigates different modes of regulations of water supplies in present-day Kano. In addition to state actors we observe several commercial and neo-traditional actors involved in water provision. Three main forms of regulation are identified: the secular state, the leaders of community organisations, and the neo-traditional rulers. Andrae concludes that the degree of popular influence around water supplies is not particularly democratic, neither by standards of electoral parliamentary democracy nor of direct ‘substantial’ democracy. This is not the intention of decentralisation policies, but these practices are the consequence of liberalisation by default. In this process the local is mobilised so as to counteract the withdrawal of the State.
- 9 Former liberation movement and present-day ruling party is Partido Africano da Independência de Gu (...)
35The democratic substance is also the concern of political scientist Lars Rudebeck who looks into electoral processes and politics in the village of Kandjadja in Guinea-Bissau. The article deals with ‘substantial democracy’ as to integrate constitutional, minimalistically defined democracy and political equality in practice. Substantial democracy is about political power being more equally distributed between citizens in a process where constitutional rules and actual practice are mutually supportive. Rudebeck, who has followed the village of Kandjadja in Guinea-Bissau since 1976, provides a diachronic account of local politics. The central paradox explored is that while the whole area is practically untouched by development interventions and literally no benefits of national policy have reached the village since the last school teacher left in 1989, the ruling party PAIGC9 and state power under PAIGC have always received political support in the Kandjadja area. The most common explanation is that “this is an old liberated area”, possibly indicating that the only political contacts between the village and the capital go through PAIGC. So, although people hold that they have been left “in a hole” and instead demand for change, in elections more votes were cast for PAIGC than any of the other parties. Rudebeck suggests that the support also originates in the 1960s during the liberation war, when the PAIGC political structure was grafted onto local ‘traditional’ socio-political structure. Politics in Kandjadja is furthermore linked to ethnicities, because Rudebeck observes how in the elections in 1999 ‘Mandinga’ of Kandjadja lined up against a presidential candidate of ‘Balanta’ origin. Thus, the case of Kandjadja is instructive both by demonstrating how little local people get in return for their political support, and by suggesting the limits of their patience. It also exemplifies how a specific form of the local was invented in the 1960s when PAIGC used local political structures to mobilise the area for the liberation war, and the power of that invention despite the fact that socio-economic betterment has yet to come.
36Marginalisation and exclusion are central aspects described in the article by anthropologist Abdoulaye Mohamadou. He analyses how pastoral peoples are marginalised in decentralisation and, in consequence, how they seek to territorialise their claims. Decentralisation has meant that various forms of chieftaincy have provided primacy to sedentary groups in contrast to mobile ones. In Dakoro, Niger, pastoralists have engaged in settlement logics as to obtain chieftaincies and territories on the municipal level. These logics include the creation of socio-economic infrastructures (wells, schools, cereal and fodder banks etc.) and changing transhumance patterns according to which family members settle permanently in the home area to avoid that farmers settle there. Traditional chieftaincy is administratively recognised in Niger, and chiefs are major actors in local arenas. But Mohamadou demonstrates that today chiefs do not “speak with one voice” any longer, due to the growing impact of development projects and programmes. Managers of projects have become a local noblesse, building legitimacy through NGOs and development agencies. The case of Dakoro is illuminating when it comes to what I call the invention of the local. The district was delimited in 1947 to form part of the Maradi Cercle. Dakoro was then divided into an agricultural zone on which chefs de canton exercised power and authority, and a pastoral zone inhabited by nomads. While farmers and nomads had different chieftaincies, the nomads did not have legally recognised territories. The difference with regard to territorial claims prevailed through colonial and postcolonial eras. With decentralisation the issue of territorial claims and land rights has become a central concern that articulates political stakes around the local. The overlapping of administrative and customary territories awakened old rivalries between local notabilities as well as between agriculturalists and pastoralists, and pastoralists used decentralisation as a pretext to delimit the boundaries of their territories in a conflict-ridden agro-pastoral zone. Beyond the invention of the local decentralisation articulates the problem of State building: the disengagement of the State has given way to traditional chieftaincy with a patrimonial and extractive mode of governance.
37In another article on Niger anthropologist Gabriella Körling looks into the education sector, and follows the rise and fall of a school in the periphery of Niamey. The article highlights how the Millennium Development Goals, notably to achieve universal primary education, interact with social, political and economic stakes around schools in local arenas. Schooling is a key strategy for donors, whereas the neo-liberal paradigm of a disengaged State seems to be in conflict with education for all. Körling details how various local stakeholders (e.g. parents’ associations, management committees and chiefs) got involved in the creation of a new school in the village of Saga, and how the classrooms, due to the lack of basic infrastructure, were built of straw. The ‘action plan’ of the committee was to ensure that pupils would at least get such classrooms. The principal of the school, employed by the State, had to rely on the mobilisation of local resources and hope for ‘development projects’ (in this case Canadian and Chinese organisations) to finance this basic infrastructure. With a thick ethnographic description Körling shows how by the end of the school year the classrooms were taken down and stored until next school year. For people in Saga the disengagement of the State concretely meant increasing personal costs of education for school books, school material, uniforms etc. Or, as Körling’s informants put it : “we have done everything and the State has done nothing.” In the context of a disengaged State we see how the local is mobilised to raise funding among local socio-political actors. Körling’s example of the Chinese development project that was to come, but never arrived, furthermore illustrates how local actors are squeezed in-between donors and rumours.
38So far I have discussed different ways in which the local is invented and mobilised. I have argued that the local must not only be conceptualised as an arena and a public space but also as an emic category that is invented and mobilised by social and political actors. When mobilising the local certain social categories tend to be included and portrayed as ‘local people’, ‘the population’ or ‘stakeholders’, while excluding others that are defined as ‘strangers’ or ‘outsiders’. In line with this I would like to emphasise that we see not only political and economic transformations in decentralisation practices but also cultural ones. It is therefore necessary to rethink cultural dimensions of decentralisation, not as a culturalist explanatory model (Olivier de Sardan 2005) but as how people live culturally and invest in what they see as culturally meaningful projects in situations where the economic and political benefits remain uncertain. A cultural analysis is needed to understand why successful individuals on a national and sometimes international arena opt to return to the local and compete to become mayor of a rural municipality. To do something concrete and meaningful for the home area and to search of recognition from people back home seem to be driving forces. But such a cultural analysis also depicts the manipulation of, and fight over, cultural categories. In this vein, I propose that the local is also an emic category and a locus of political imagination, cultural representation and symbolic meaning.
39Although historically associated with the creation of urban communes and its consequence local government, current decentralisation practices seek to create rural communes far away from urban centres. There is a strong historical record of reproducing distinctions between urban citizens and rural populations, that is, the bifurcated State of citizens and subjects (Mamdani 1996). Mahmood Mamdani argues that such a divide of citizens and subjects was a central feature of the colonial state. Citizenship would be a privilege of the civilised, whereas the uncivilised would be subject to an all-round tutelage : “The rights of free association and free publicity, and eventually of political representation, were the rights of citizens under direct rule, not of subjects indirectly ruled by a customarily organized tribal authority” (Mamdani 1996 :19). So, in a way it could be argued that decentralisation constitutes an attempt to overcome the historical divide between urbanites and rural dwellers, and between citizens and subjects. But, citizenship is far more than a legal category. It also includes rights and access to basic goods and services, and the case-studies clearly show that this is far from certain in contemporary West Africa.
40Decentralisation practices concern the fundamental relations between citizens and the State. Key relations include the ways in which the local are invented and mobilised, presenting new ways in which the postcolonial state is re-appropriated. The return to ‘the local’ or ‘the home’ could therefore also be expressed as the celebration of rurality in an urban context, representing emerging West African urbanities. But decentralisation should first and foremost be approached as the most recent layers of political and administrative reform to which local actors need to adjust, or to resist. Far from irreversible, it is my contention that decentralisation practices may well give way to the return of the strong, central State with financial and political resources to implement policies. That is why historically grounded and ethnographically dense analyses, aptly illustrated by the four case-studies of this APAD-Bulletin, are needed for understanding how policy is translated into the daily practices of citizens of West African municipalities.