Hostname: page-component-745bb68f8f-g4j75 Total loading time: 0 Render date: 2025-02-11T05:54:01.148Z Has data issue: false hasContentIssue false

DECENTRALIZATION AND THE LOCAL DEVELOPMENTAL STATE: PEASANT MOBILIZATION IN OROMIYA, ETHIOPIA

Published online by Cambridge University Press:  06 April 2016

Rights & Permissions [Opens in a new window]

Abstract

This article explores the politics of decentralization and state–peasant encounters in rural Oromiya, Ethiopia. Breaking with a centralized past, the incumbent government of the Ethiopian People's Revolutionary Democratic Front (EPRDF) committed itself to a decentralization policy in the early 1990s and has since then created a number of new sites for state–citizen interactions. In the context of electoral authoritarianism, however, decentralization has been interpreted as a means for the expansion of the party-state at the grass-roots level. Against this backdrop, this article attempts a more nuanced understanding of the complex entanglements between the closure of political space and faith in progress in local arenas. Hence, it follows sub-kebele institutions at the community level in a rural district and analyses their significance for state-led development and peasant mobilization between the 2005 and 2010 elections. Based on ethnographic field research, the empirical case presented discloses that decentralization and state-led development serve the expansion of state power into rural areas, but that state authority is simultaneously constituted and undermined in the course of this process. On that basis, this article aims to contribute to an inherently political understanding of decentralization, development and their entanglement in local and national politics in rural African societies.

Résumé

Cet article explore la politique de décentralisation et les rapports État-paysans dans la région éthiopienne rurale d’Oromia. En rupture avec un passé centralisé, le gouvernement de l’EPRDF (Ethiopian People’s Revolutionary Democratic Front) s’est engagé dans une politique de décentralisation au début des années 1990 et a créé depuis lors de nouveaux sites d’interaction État-citoyen. Dans le contexte d’autoritarisme électoral, cependant, la décentralisation a été interprétée comme un moyen d’expansion de l’État-parti au niveau des citoyens ordinaires. Dans ce contexte, l’article tente une interprétation plus nuancée des entremêlements complexes entre la fermeture de l’espace politique et la foi dans le progrès dans les arènes locales. À cette fin, il suit des institutions sous-kebele au niveau communautaire dans un district rural et analyse leur poids dans le développement dirigé par l’État et la mobilisation paysanne entre les élections de 2005 et de 2010. À partir de travaux de recherche ethnographiques sur le terrain, le cas empirique présenté révèle que la décentralisation et le développement dirigé par l’État servent l’expansion du pouvoir de l’État dans les zones rurales, mais que l’autorité de l’État est simultanément constituée et sapée au cours de ce processus. Fort de ce constat, l’article vise à contribuer à une interprétation intrinsèquement politique de la décentralisation, du développement et de leur entremêlement dans la politique locale et nationale des sociétés africaines rurales.

Type
Ethiopian Landscapes of Power
Copyright
Copyright © International African Institute 2016 

This article explores the politics of decentralization and present-day relations between the peasantry and the state in rural Oromiya, Ethiopia. After the fall of the centralized state of the Derg regime in 1991,Footnote 1 the subsequent Ethiopian People's Revolutionary Democratic Front (EPRDF) government committed itself to a decentralization policy. Since then, a number of policies, processes and reforms have been intended to promote direct citizen engagement and to bring the government closer to the people, who are those most affected by the exercise of power (Keeley and Scoones Reference Keeley and Scoones2000; Harrison Reference Harrison2002; Lefort Reference Lefort2012; Meheret Ayenew Reference Ayenew, Zewde and Pausewang2002; Spielman et al. Reference Spielman, Cohen and Mogues2008).

In Africa and elsewhere, decentralization has been among the highest developmental priorities during the past decades (Boone Reference Boone2003; Diawara Reference Diawara2011; Geiser and Rist Reference Geiser and Rist2009; Wunsch Reference Wunsch2001; Reference Wunsch2008). Despite the theoretical benefits of decentralization policies (Rondinelli Reference Rondinelli1981; Rondinelli et al. Reference Rondinelli, Nellis and Shabbir-Cheema1984), however, the expected results in terms of service delivery, development, democracy and governance have often remained absent. Moving beyond an analysis of the operation and failure of the ideal model of democratic decentralization, a focus on political dynamics increasingly gained scholarly attention (Agrawal and Ostrom Reference Agrawal and Ostrom2001: 487; Boone Reference Boone2003: 355; Olowu Reference Olowu2001: 12). This brought under scrutiny the conditions under which central governments implement or maintain decentralization. Somewhat counterintuitively, it has been illustrated that authoritarian regimes may decentralize further than democratic ones (Riedl and Dickovick Reference Riedl and Dickovick2014). In this vein, many have interpreted decentralization in terms of centre–periphery relations or as an effective tool for retaining central control over rural people and resources (Ribot et al. Reference Ribot, Agrawal and Larson2006). Moreover, the politics of decentralization has brought into consideration the ways in which reforms and state interventions were locally negotiated and how they unfolded within local political arenas and pre-existing power relations (Bierschenk and Olivier de Sardan Reference Bierschenk and Olivier de Sardan1997; Bierschenk et al. Reference Bierschenk, J.-P. and Olivier de Sardan2002; Olowu Reference Olowu2001).

Ethiopia offers useful grounds for such an analysis of the politics of decentralization in the context of authoritarianism. Decentralization in Ethiopia is often interpreted as a means of expanding state power into the rural hinterlands rather than as a genuine attempt to devolve power (Dessalegn Rahmato Reference Rahmato2008b: 321; Harrison Reference Harrison2002: 602; Pankhurst Reference Pankhurst2002: 12). In particular, scholars have criticized top-down decision-making processes, undemocratic practices and serious limitations to direct citizen engagement (Abbink Reference Abbink2011: 515; Dessalegn Rahmato Reference Rahmato1993: 42; Reference Rahmato2008b: 253; Keeley and Scoones Reference Keeley and Scoones2000: 94). A number of studies have further identified the overriding authority of the ruling party and a less visible party structure that accompanies the state structure as the main reasons for the failure of local decentralization (Dessalegn Rahmato Reference Rahmato2008b: 260; Tegegne Gebre-Egziabher and Kassahun Berhanu Reference Gebre-Egziabher, Berhanu, Assefa and Gebre-Egziabher2007: 37; Vaughan and Tronvoll Reference Vaughan and Tronvoll2003: 40).

In contrast, the recent literature on the developmental state in Ethiopia has pointed to the significant achievements resulting from the EPRDF's development policies and reforms. On the one hand, Lefort (Reference Lefort2013: 463) has argued that the ruling EPRDF, in its attempt to retain central control, has nevertheless failed to translate these ‘achievements into true legitimacy’. Moreover, Planel (Reference Planel2014: 420) has demonstrated, in the case of agricultural extension, that technical development approaches can even operate as a powerful instrument ‘that reinforces the local disempowerment of the most vulnerable peasants'. On the other hand, Mains (Reference Mains2012: 9) looked at the perception of infrastructure projects and found a strong ‘faith in progress’, and concluded that state-led development also provides ‘a means of legitimizing political rule’.

Against this backdrop, this article draws attention to the complex entanglements between faith in progress and the closure of political space in rural Ethiopia. In the name of development, the peasant household has long been a site and testing ground for local-level state interventions, a phenomenon that continues in current rounds of decentralization (Chinigò Reference Chinigò2014a; Reference Chinigò2014b; Dessalegn Rahmato Reference Rahmato and Assefa2008a: 130–7; Lefort Reference Lefort2012). In Ethiopia's largest regional stateFootnote 2 of Oromiya, a proliferating series of ever-smaller units amplified the existing administrative structure of federal, regional, district (woreda) and sub-district (kebele) levels – in particular the creation of the gott (hamlet) and garee (team) in 2004 (HRW 2005). These two latter units, glossed as sub-kebele, were said to provide a twofold advantage: linking households to the lower tiers of the decentralized system and enabling people to engage in development-related activities as so-called development teams (garee misoma). The creation of these teams officially responded to a popular demand for development and to the EPRDF's rural development targets,Footnote 3 but has been accompanied by considerable controversy. Critical voices have depicted these sub-kebele in the context of electoral authoritarianism as highly effective mechanisms of control and repression at the grass-roots level (Aalen and Tronvoll Reference Aalen and Tronvoll2009; HRW 2005; ICG 2009; Lefort Reference Lefort2007).

This article considers the sub-kebele as a new site for state–peasant interactions in rural Oromiya that has been under-researched so far. As an ethnographic account, it pays attention to encounters between local government officials and the rural citizenry in a local political arena in Meta Robi district.Footnote 4 In these encounters, the garee's involvement in rural road construction provides a valid empirical entry point into state-led development more generally. The district's road infrastructure mirrors the decentralized hierarchies in which it is constructed, administered or maintained, and qualifies Meta Robi as an appropriate study site. The data upon which this analysis is based were gathered through field research in the district during the rainy season between June and September 2009. The main data were derived from fifty-eight unstructured and semi-structured interviews, twenty-three group discussions and participant observation in five of the district's kebele. On that basis, this article provides an appraisal of recurrent state interventions opening and closing electoral cycles in Ethiopia between the contested 2005 federal and regional elections and the elections in 2010 that resulted in an overwhelming victory for the ruling party and its regional affiliates (cf. Aalen and Tronvoll Reference Aalen and Tronvoll2009; Tronvoll Reference Tronvoll2011; Lefort Reference Lefort2007; Reference Lefort2010).

Each of the empirical sections that follow illuminates the ambiguous interplay between the manifestation of decentralization and state-led development in a local political arena. State power expanded into the rural district with the creation of the garee, its embedding in local government and its instrumentalization by the local party-state. Its establishment was also accompanied by the proliferation of the idea of a developmental state among rural dwellers, defining narratives of progress and the terms in which development had to be conceived and achieved – while being constantly negotiated and contested. Thus, the following sections disclose how state authority is simultaneously constituted and undermined in the course of decentralization and state-led development. Overall, this paper aims to unravel ‘the making of the EPRDF's state and authority’ on the lower rungs of rural society and complements Di Nunzio's (Reference Di Nunzio2014: 460) account of community policing in urban Ethiopia.

DECENTRALIZED ROAD CONSTRUCTION

State-led road construction and the growing incorporation of Meta Robi during the past decades have been met with ambivalence in the rural district. On the one hand, the district's inhabitants largely depend on subsistence farming, while off-farm employment opportunities are highly limited. The agricultural productivity of the district is low due to ‘small land holdings, traditional farming practices, crop and livestock diseases, limited use of agricultural input, and erratic rainfall’ (Meserete Kristos Reference Kristos2009). Access roads to Shino are therefore pivotal for rural dwellers as the district's capital is the main economic centre and hosts a weekly market. On the other hand, such roads also enable the ruling government to attain central control, and this has significant consequences for the district's predominantly ethnic Oromo population.

Located about 100 kilometres north-west of Addis Ababa, Shino is connected to the national road network today, which makes it possible to reach the Ethiopian capital by daily buses throughout the year. This permanent link between the national and the district capital was established during the Derg regime by the extension of the national road network and the construction of an all-weather linking road to Shino. The Derg had identified the lack of access to rural areas as a serious hindrance for agricultural development and therefore promoted the construction of low-cost and low-standard roads (ETCA 1986: 4; Emmenegger Reference Emmenegger2012). As a consequence, the Ethiopian Transport Construction Authority (ETCA) started its involvement in rural road construction at a national level and in Meta Robi.

Unlike Shino, however, access to most of the district's forty-two kebele has been difficult due to the poor condition of the road infrastructure, or even its complete absence. Only a few kebele are connected to the road network and predominantly by dry-weather roads that are regularly destroyed during the two annual rainy seasons – the small belg rain (March and April) and the big meher rain (June to September). According to the Woreda Rural Road Office (WRRO), the district's network amounted to a total length of 183.8 kilometres in 2009, of which 85.8 kilometres were all-weather roads and 98 kilometres were dry-weather roads (see Figure 1).Footnote 5 In technical terms, however, only 52 kilometres of the total network are gravelled roads and classified according to design standards DS-6.Footnote 6

Figure 1 Meta Robi road map. The sub-kebele does not appear on this map found in the WRRO office at district level in 2009. As this indicates, state territorialization through the sub-kebele is not primarily a cartographic exercise, but rather advances through the establishment of administrative hierarchies at the local level.

In 2009, a 15 kilometre section of the gravel surfaced network was under construction, further extending the all-weather road system and newly connecting a settlement called Ketiketto along the range of hills with Shino and Addis Ababa. As a construction project of the regional Oromiya Road Authority (ORA) – implemented without local government involvement – this network expansion reflected the acknowledgement that rural road infrastructure was a decisive factor in the EPRDF's development policy.Footnote 7 Asked about the meaning of roads, most rural dwellers appropriated this policy discourse and emphasized the significance of roads for their district. Some more critical voices, however, also underlined a political dimension of road infrastructure and the district's significance in the resistance against the former Derg and the EPRDF today. According to some, the Ketiketto area used to be a ‘difficult place for the government’, as its location at a topographic elevation allowed the liberation movement to oversee the surrounding valleys and to hide in the inaccessible northern lowlands of the district.Footnote 8 In the height of the resistance against the Derg, Ketiketto arguably served as a node ‘where different movements came together’ and as a corridor for their joint movement towards Addis Ababa.Footnote 9

The Oromo Liberation Front (OLFFootnote 10 ), the main ethnic resistance movement concerned, became part of the EPRDF-led transitional coalition government in the early 1990s but withdrew in 1992. In its place, the Oromo People's Democratic Organization (OPDO), as a constituent regional party of the ruling EPRDF, claimed to legitimately represent Oromo people and established a near-total monopoly of political power in the region (Chanie Reference Chanie2007: 362; HRW 2005: 10; Pausewang Reference Pausewang2009: 5). During field research, critical voices argued that the OLF had returned to the area behind Ketiketto, and has continued to ‘struggle against the regime even today’.Footnote 11 From that point of view, the Ketiketto construction not only reflects the government's development policy, but also its counter-insurgency strategy.

The bulk of the district's network is administered by local government institutions, however. It concerns primarily dry-weather roads that are vital for the rural population's market access as well as wood extraction from the district. The WRRO was established to administer these roads at the district level in 2002–03 (Tegegne Gebre-Egziabher and Kassahun Berhanu Reference Gebre-Egziabher, Berhanu, Assefa and Gebre-Egziabher2007: 34). Since then, this sector office has been responsible for planning and extending the community road network according to the government's overall development plan. In an increasingly professionalized manner, the WRRO works from an action plan that ‘defines the amount to be done in one or the other kebele . . . based on survey, observation and . . . experience’.Footnote 12 WRRO documents identified 35.5 kilometres of dry-season roads planned for construction or maintenance in the previous year, of which 25.5 kilometres were successfully realized.

DECENTRALIZED LOCAL GOVERNMENT

The Ethiopian state is intrinsically tied to its peasantry since most of its citizens live in rural areas. In the modern period, imperial, socialist and federal governments ‘have been intent on extending their reach and authority over the peasantry . . . with different styles, approaches and justifications’ (Dessalegn Rahmato Reference Rahmato2008b: 244). Most significant has been the expansion of the state into rural areas through the establishment of a complex administrative structure at village level (Vaughan and Tronvoll Reference Vaughan and Tronvoll2003: 40). Since the creation of the sub-kebele, the administrative hierarchy has further expanded and has enabled the idea of the developmental state to proliferate.

In Meta Robi, a decentralized structure is built by the woreda administration in the district's capital – Shino, located in the southern part of the district – linking kebele administrations with higher government institutions. This structure constitutes a ‘physical and political manifestation of the state’ and embeds a wide range of actors who act in its name (Chinigò Reference Chinigò2014a: 48). While the state is increasingly visible in the peasants' everyday lives, the kebele in particular is its most important reification and a site for everyday encounters between the citizenry and the decentralized government. Under the EPRDF, the kebele has been a constant object of administrative and political reforms and of continuous rounds of decentralization (Dessalegn Rahmato Reference Rahmato2008b: 244; Emmenegger et al. Reference Emmenegger, Keno and Hagmann2011: 733; Lefort Reference Lefort2012: 684; Tegegne Gebre-Egziabher and Kassahun Berhanu Reference Gebre-Egziabher, Berhanu, Assefa and Gebre-Egziabher2007: 10). Thus, the kebele has become increasingly professionalized and reinvented as a differentiated administrative body up until the present day.

Historically, the socialist Derg established kebele institutions as so-called peasant associations with the nationalization of land as the focal point of its radical reform after the revolution in the mid-1970s (Dessalegn Rahmato Reference Rahmato2008b). Since then, the Ethiopian Constitution has stipulated that land belongs jointly to the state and the people, a commitment that was later maintained by the EPRDF (Crewett and Korf Reference Crewett and Korf2008: 203; Devereux et al. Reference Devereux, Teshome and Sabates-Wheeler2005: 121). Kebele councils consisted of household heads, to whom authority was given concerning land distribution, conflict resolution and other community-related affairs (Dessalegn Rahmato Reference Rahmato1993: 39). With the adoption of ‘Soviet-style socialism’, however, the peasantry became a ‘passive recipient of socialist directives channelled from above’ (Dessalegn Rahmato Reference Rahmato1993: 38; cf. Aspen Reference Aspen, Zewde and Pausewang2002: 63). Moreover, the peasant associations were entrusted with a variety of administrative, defence and political tasks and were instrumentalized by the Derg to ‘capture’ the rural population (Clapham Reference Clapham1989: 7; Dessalegn Rahmato Reference Rahmato1993: 39). Vaughan and Tronvoll (Reference Vaughan and Tronvoll2003: 41) pointed to the ‘twofold capacity’ that this (subsequent) kebele system had developed, functioning as an extended arm of central government and as a grass-roots intelligence-gathering tool.

The Derg significantly altered state–society relations in Ethiopia and made the state increasingly present in peasants' everyday lives. In rural areas such as Meta Robi, the state became embodied by the ‘peasant[s] made by the revolution’ (Dessalegn Rahmato Reference Rahmato1993: 47, emphasis in original), forming an elite who stood out by their active roles in various rural mass organizations, their close relationships to state or party officials, and their resulting access to state-controlled resources (ibid.: 40–8). The chairmen of the kebele in particular derived their power directly from the state and were consequently incorporated in, and dependent on, the established administrative hierarchy (Clapham Reference Clapham1989: 8). Although the EPRDF later redistributed official positions in the local administration after coming to power in 1991, this incorporation continued with the successive younger generation of peasant society, which gained fresh control over leading local positions (Aspen Reference Aspen, Zewde and Pausewang2002: 66).

State intervention continued with the creation of the sub-kebele Footnote 13 in Meta Robi and other parts of the region around 2004 (see Figure 2).Footnote 14 In territorial terms, the gott and garee divided up the kebele into groups of households, usually grouping between sixty and ninety households for a gott and about twenty to thirty for a garee (cf. HRW 2005: 30). Although no official documents could be identified during field research, informants in Meta Robi had a common feeling that government standards legally define the number of households organized. In contrast to this belief, examples revealed deviations as the actual clustering of households was adjusted to the specific patterns of settlement and topography of each kebele. However, the creation of these sub-kebele incorporates the ‘household’ as a state category into an administrative hierarchy and signifies the expansion of the state into the district.

Figure 2 Administrative hierarchy of the existing decentralized structure including the sub-kebele.

Like existing institutions in other regions,Footnote 15 the gott and garee came to organize a given number of households, which are represented by a committee. Their clustering should enable people to engage as development teams (garee misoma) in ‘what the authorities describe as community projects’ (Dessalegn Rahmato Reference Rahmato2008b: 253). In the eyes of a woreda official, garee activities reflect the government's pragmatic stance towards development: ‘“I cannot do everything by myself” . . . “the people have to participate”.’Footnote 16 Although community projects seem to have decreased in Ethiopia, government directives nevertheless stipulate that ‘peasants are to provide 20 percent of their working time to public work schemes (without payment) and 80 percent to their own livelihoods’ (Ministry of Capacity Building 2007 quoted in Dessalegn Rahmato Reference Rahmato2008b: 253).

COMMUNITY DEVELOPMENT

In Meta Robi, the gott and garee were introduced in the course of a regional pre-2005 election campaign. As part of a governmental delegation, selected members of the Meta Robi district cabinet travelled to Hararghe, eastern Oromiya region, to share experiences of an irrigation project and to learn ‘how to dig and what to do by the garee and the gott’.Footnote 17 Because these eastern parts of the region had gained experience in constructing water wells in harsh ecological conditions, the visiting delegation now learned about the technical side and the organization of people for such community projects. After their return to Meta Robi, the gott and garee were established within each kebele and resumed work with an initial water well construction project. At that time, the head of the Woreda Security Office, who was among the delegates, went into different kebele and ‘mobilized people for digging’.Footnote 18 In 2009, some rural dwellers remembered their initial engagement in these ‘digging’ activities and referred to the remaining holes that still attested to their engagement. Initially intended as a measure to improve water access for rural dwellers, ‘the project failed . . . because there is enough rain in Meta Robi’, as a local government official confessed.Footnote 19

This initial project of the sub-kebele is representative of the often limited capacity of development blueprints to address local conditions and peculiarities. Despite its failure, however, a structure and the corresponding ‘working culture of the people’ was established and remained operational from then onwards. Since then, the garee has engaged in various development-related projects in the fields of irrigation, education, health and mobility, and has mainly carried out the construction of the public infrastructure required.

Roads have also been constructed and maintained by the garee misoma in most of the visited kebele, usually after the rainy season. The WRRO office head argued that community participation is needed due to the budget constraints of his office. But he denied any direct connection to the sub-kebele: ‘Our link is not with the garee, but we link to the kebele. How it [the road] is divided is not our, but their [the kebele's] issue.’Footnote 20 The empirical focus during interviews on the subject of community roads, in particular, was a dry-weather road connecting a kebele called Kuyu Gicci and the district capital. Leading 6 kilometres north-west from Shino, the road was constructed in the initial years of the sub-kebele. The initiative originated from the woreda when access to the kebele had become urgent because of an irrigation project that was planned, and led district officials to order the kebele chairman to mobilize people.

In Kuyu Gicci, road building reportedly started with a meeting where local government officials and rural people came together and the kebele administration, gott and garee leaders informed people about the upcoming construction. Informants also reported that the kebele administrator and other kebele officials accompanied sub-kebele leaders to the households in order to advertise the road's significance. As a former garee leader remembered: ‘The kebele chairman encouraged the people because the road was really bad before.’Footnote 21 The former kebele chairman himself remembered that he succeeded in convincing people by explaining: ‘If you are sick or your wife is pregnant . . . you can go to Shino.’Footnote 22 In their encounters, government officials similarly appropriated much of the national policy discourse by articulating the connection between road construction and its development impact in regard to health, communication, education, transport and agriculture. Finally, the road's potentially positive impacts on service delivery and market access convinced officials and citizens alike. Many garee members stated that they were initially persuaded by the prospect of fertilizer being stored in their kebele; for Kuyu Gicci, this had been distributed by the woreda office in the district capital until that point.

For the actual construction of the road, the kebele administration then divided it geographically, assigned a particular section to each garee, and advised gott and garee leaders to organize garee members at a specific place and date. Although the Kuyu Gicci road partly crossed the territory of a neighbouring kebele, only Kuyu Gicci's nine garee were mobilized because ‘the road serves [only] us directly’, as the deputy kebele chairman clarified.Footnote 23 Participants stated that the length of each garee's section varied according to the topography along the way. The total working hours were the result of the performance of each team and therefore differed from three to five working days; this is more or less in line with the reported workload in other garee activities. While the garee leaders participated in and coordinated the activities, overseeing the presence of garee members and regularly reporting to the kebele administration were also part of their responsibilities for the worksite. The garee members contributed labour and tools as equipment for the manual tasks, which comprised clearing and levelling the surface.

In principle, members of the garee are the household heads who are referred to as abba warra in Oromiffa and who qualify by paying taxes to local government on the land they ‘own’. Household heads are usually adult or elderly males, but they can be replaced by their widows in case of their death; the widows then inherit the land, the garee membership and the duty to pay taxes. Although it is said that ‘land owners have a strong obligation to participate’, the actual construction of the Kuyu Gicci road revealed slight adjustments in the seemingly permanent development teams.Footnote 24 As a garee leader explained: ‘We selected the able people who could accomplish their own [farm] work and additionally our work.’Footnote 25 As garee activities are usually labour intensive, elderly people and widows are often exempted from the work, and are either replaced by the younger generation or compensate for their absence by contributing money or construction materials. In this particular project, rumours circulated of individuals staying absent from physical work due to their personal relationships with the kebele administrators. ‘Community projects’, in sum, are thus less popular than the notion suggests: they have limited appeal among the community members whom they are intended to embody and mobilize.

SECURITIZED DEVELOPMENT

The use of planned intervention has a history in Ethiopia. The Derg used so-called government teams (mengistawi budin) for mobilizing peasants and for carrying out development-related activities (Bevan and Pankhurst Reference Bevan and Pankhurst2007: 70; Dessalegn Rahmato Reference Rahmato2008b: 254; Pankhurst Reference Pankhurst2008: 12; Segers et al. Reference Segers, Dessein, Hagberg, Develtere, Haile and Deckers2008: 13). In Meta Robi, some elderly informants still remembered their engagement in construction work through the budin during the Derg's ‘villagization’ programme.Footnote 26 Comparing the present and the past, some informants equated the mengistawi budin and garee misoma, despite significant differences in institutional and territorial terms between the two.Footnote 27 This section illustrates the increasing dependence on the state of rural inhabitants and unravels their involvement in state-led development as a necessity given the authoritarian context into which these sub-kebele have been embedded.

With the arrival of the EPRDF, the newly created armed militia took over responsibility for organizing people, since, at that time, ‘almost everything was done by the militia’.Footnote 28 In Meta Robi, the militia has constituted a tight, but often invisible, network through which ‘peace and security’ in the district is maintained.Footnote 29 The armed militia consists of people living in the kebele, mainly ordinary farmers, who have received basic military training and ‘were given [a] Kalashnikov’.Footnote 30 Since then, there have been numerous militia members in each kebele under the orders of the kebele's chairman and head of security. While the latter, in principle, is accountable to the kebele chairman and the head of the Woreda Security Office, this institutional structure is often bypassed because the kebele chairman orders the militia directly. Thus, the militia has reportedly ‘ordered people to come to the meeting place’, where they were informed about upcoming community projects within the kebele, among other issues.Footnote 31

With the creation of the sub-kebele, peasant mobilization in Meta Robi continued through the gott and garee. For people's involvement in road work, various informants mentioned the importance of letters in which the garee leader receives orders from the kebele chairman. The advantage of having garee leaders in place was underlined by a young kebele manager as follows: ‘We directly order the leaders, or if the district orders us . . . we write a letter to the garee leaders.’Footnote 32 Such a letter – either in written form or in a rhetorical sense – makes the garee leader's orders formal and mandatory for participants. Their orders are further enforced by the militia that accompanies and empowers the garee leader. A garee leader involved in the Kuyu Gicci road construction confirmed: ‘The militia assisted us in ordering the people who refused to participate, and solved conflicts between people during the construction.’Footnote 33 Participants also mentioned the authoritative function of the militia and particularly its guns as a coercive means of mobilization.

The involvement of security personnel in road construction raises a number of questions in regard to the relation between state violence and development. When asked about the role the militia plays in development-related activities, the head of the Woreda Security Office denied any involvement and declared that ‘the militia's [only] function is to maintain peace and security for people’.Footnote 34 Corroborated reports of collaboration between militia and garee leaders, however, make it difficult to distinguish between the role of the two and highlight the overlap between the security and development arenas at sub-kebele level. Moreover, several individuals held the position of a gott or garee leader and were members of the militia at the same time. This further underlines the difficulties of drawing a strict line between the two, in terms of both their practices and people's everyday experiences.

Given the interplay of bureaucratic and authoritarian methods, involvement in state-led development activities is inevitable in Meta Robi as people's refusal can have significant consequences. Informants explained that any failure to take part is punished with a relatively small fine from the garee and a bigger one from the kebele, which does not exempt the shirker from manual work.Footnote 35 Although only a few cases were reported during my fieldwork, falling into disfavour with the ruling government comes with more profound and threatening consequences. Informants reported that those who fail to take part risk accusations of being members of the regional opposition movement – the OLF. In Meta Robi, refusing to participate in state-led development is obviously seen as an anti-government stance, which further underlines the overlapping of development and security. Beyond Meta Robi, similar accusations have been reported and documented in other studies conducted in the region (HRW 2005).

The basis for peasant mobilization appears, however, to be the combined result of the local government's authoritarian traits and its control over the means of agricultural production (land, fertilizer, seeds, etc.). Thus, the farmers' ability to maintain their livelihoods crucially depends on their relationship with the government – and their willingness to obey government directives. Two young farmers argued: ‘Those who have a relation with the garee and the kebele administration were given land.’Footnote 36 They concluded: ‘[It is] through the participation in garee activities that we hope that the government will give us money.’Footnote 37 In contrast, the risks of not participating become obvious in a garee leader's statement: ‘If someone violates our order and keeps absent, then we just tell him that he will not get the government's benefit and that he will not get fertilizer. Additionally, his application will not be considered even in case of problems.’Footnote 38 Thus, the threatening consequences for those who refuse provide a compelling reason for rural people to engage in so-called community projects (cf. Harrison Reference Harrison2002: 600; Lefort Reference Lefort2007: 254; Pausewang Reference Pausewang, Zewde and Pausewang2002: 98).

EMBEDDING THE SUB-KEBELE

As an extension of the decentralized structure, the garee has democratic potential. Initially, each garee was represented by a five-member committee. A gott five-member committee further represented three garee and was established as a link of accountability between garee committees and kebele officials. According to some woreda officials and kebele managers, the purpose of the garee is to work efficiently for development and to enable people to express their concerns through the government. Following this ideal, the garee is supposed to organize regular meetings where its members discuss and make decisions about development issues. On that basis, the garee misoma is supposed to implement specific activities within its own territory. However, the depiction of the garee in terms of a constitutional ‘right to participate’ or as a democratic opportunity was undermined in the eyes of many when it appeared as a kebele instrument for mobilization and control.Footnote 39

During field research, several informants indeed reported the vitality of the garee during its initial years, both in terms of meetings held and activities undertaken. Captured in an elder's terms: ‘In the beginning, they told us that the garee enables cooperation between people and there was even the slogan “If someone cries, we look after him.”’Footnote 40 Many of those who had initially considered the garee a promising platform for direct citizen engagement and progress expressed their unease with its failure shortly afterwards. Various informants remembered their involvement in a variety of garee activities, but now complained about its inactivity. In contrast, several kebele officials continued to emphasize the significance of the garee in a variety of development-related activities. Such activities could be documented in different kebele in 2009, and particularly concerned construction work relating to irrigation, health and education. Moreover, various informants expected their repeated engagement in road maintenance work after the end of the rainy season, only two months ahead. Nevertheless, it was clear that the role and level of activity of the garee varied widely across the district.

In fact, the involvement of the garee in development-related community projects was ephemeral and had faded in various kebele after 2005. As in many other parts of Oromiya, the gott and garee had fallen into neglect because they were too labour intensive as a set of structures to be properly maintained once the elections passed.Footnote 41 In particular, the Oromiya regional government had dissolved the gott committees in 2008 and placed the garee committees under the supervision of the kebele. In line with this restructuring, the size of the garee committee was reduced from five to three members. The woreda also neglect the garee, as there ‘would be too many’ to deal with.Footnote 42 The district administrator of Meta Robi underlined that his institution deals only with the kebele administrations and no longer focuses on the garee.Footnote 43 He interpreted the decline of garee activities in many kebele as a consequence of people's attitude towards work: ‘People don't work, they don't like to work, people like to rest. Consequently, people don't participate.’Footnote 44

Nevertheless, the garee exists within pre-existing power relations at kebele level. In 2009, the role of the garee was highly dependent on the strength of the administrators in the kebele concerned as they ‘are widely perceived in terms of their potentially repressive and punitive powers’, in Meta Robi as elsewhere (Vaughan and Tronvoll Reference Vaughan and Tronvoll2003: 58). Being dependent on a privileged relationship with the kebele administrator and armed militia, the social standing of garee leaders remains weak. In their ‘community’, garee leaders run the gauntlet of being criticized or even mocked by their neighbours if they try to implement orders from above. They also risk being accused ‘of being an OLF member’Footnote 45 if they do not obey or if they refuse to implement such orders. During field research, several garee leaders complained about their role as powerless intermediaries between the kebele and the people, and mentioned the problems they face in their attempt to convince rural people to follow kebele directives.

Other garee leaders, in contrast, could be regularly observed joining their kebele chairmen at public gatherings in the kebele or at meetings at woreda level. Although garee leaders did not receive any salary in 2009 (apart from small per diems during the 2005 election campaign), holding this office came with potential advantages for exemplary leaders. ‘My garee was ranked as the first of all in the kebele,’ a garee leader boasted, and, in his case, good performance was rewarded by his later promotion as the kebele chairman's deputy.Footnote 46 The garee leader's performance was evaluated based on the garee's work and on his personal loyalty to the government and its political agenda.

FUSING THE PARTY-STATE

Despite the increasing weakness of the garee after 2005, the perpetuation of a leadership structure at sub-kebele level clearly indicates the ability of the party-state to retain political control at the local level. A number of informants ascribed the strength of the structure to its social mechanism and the fact that ‘people [who are organized in groups] can more easily control each other’.Footnote 47 Beyond Meta Robi, Di Nunzio (Reference Di Nunzio2014: 455) has documented similar dynamics of state surveillance in his analysis of community policing in Addis Ababa in the mid-2000s, finding that the ‘boundary between spying and gossiping was blurred’. Informants also stressed the significance of having garee leaders who know and live in their neighbourhood and who efficiently channel the kebele's order downwards. In the context of electoral authoritarianism, however, we learned that repeated state interventions are discredited not just among the peasants who are subjected to them, but also among those embodying the state in these interventions.

The garee's ability to control and monitor the rural population has enhanced the kebele's administrative capacity and has opened up a new channel for government propaganda at the household level. A number of informants reported the garee's involvement in mobilizing voters in the run-up to the contested 2005 elections. These informants mentioned in particular the use of the garee by the ruling OPDO for channelling propaganda, surveying the rural population and intimidating any form of opposition. In the words of a former garee leader: ‘At the beginning, the garee was us, but they simply passed down information from above.’Footnote 48 The resulting control at the local level is nicely articulated in an elder's statement: ‘We say, “That's good!” to anything they [the government] do, because we won't get anywhere by opposing.’Footnote 49

However, the decrease in garee-led development activities after 2005 and the increasing uncertainty in the pre-2010 election period allowed room for people to ‘resist’ government directives. As described by a local informant: ‘People are listening for the coming elections, and because of that the people started disregarding and suspecting each other . . . if another government will come, why should we suffer with this one?’Footnote 50 In a similar vein, other informants also presented the garee's inactivity as a sign of resistance and as the result of people's refusal to accept orders from the garee leader and the kebele. In 2009, several local farmers declared that they had ignored the kebele's call for meetings, although it was clear that the kebele could enforce its will at any time – and that people ‘may participate in kebele activities if they are ordered’.Footnote 51

In Meta Robi, the government's democratic legitimacy was crucially undermined in the eyes of many by the way in which the garee had been embedded in the local party-state. Also, several garee leaders privately expressed their scepticism about the garee's role, for example as an effective means of achieving development or as an instrument of the OPDO during the 2005 elections. ‘They [the woreda] ordered us to gather the people for an assembly but the people refused. Then they ordered us to punish each with a fine of 50 BirrFootnote 52 but we refused to do so. Then we were said to be OLF members.’Footnote 53 These former garee leaders reported that ‘they [the ruling party] put force on people if they did not elect the OPDO’.Footnote 54 They went on: ‘We are at risk, because the OPDO people came and preached, and if you don't elect them you don't get fertilizer.’Footnote 55 These informants were strongly convinced that the vast majority of people in the district, including government officials, did not vote for the OPDO in the 2005 elections, but the ruling party nevertheless maintained power. Regardless of such private criticism, most of these leaders kept their position within the government structure, which can be interpreted as a strategy to avoid powerlessness and to maintain their position as a local ‘relay point of power relations’ (Vaughan and Tronvoll Reference Vaughan and Tronvoll2003: 34).

Given the fact that the garee was in decline, a new set of structures was in the process of being formed through the creation of local party cells, locally called celli; this illustrated the ruling government's mobilization strategy for the coming 2010 elections. A newly nominated cell leader explained the ongoing transformation as follows: ‘The government collapsed it [the gott] and it does not any longer focus on the garee. According to the government's plan, focus is given to the celli.’Footnote 56 In contrast to the sub-kebele, which encompasses ‘all members’, the cell was now established as a local extension of the party structure, selectively including ‘the party members’.Footnote 57 Four nominated cell leaders who were already in position in 2009 represented these party members. The role of these leaders was to meet regularly with the members of the cell in order to discuss and promote the ruling party's developmental vision and achievements.

This pre-2010 mobilization was characterized by the expansion of the OPDO's outreach and the promotion of state-led development at local level. As the district administrator framed it: ‘In the celli, party members talk together.’Footnote 58 Thus, ‘it advises people for development. Info from the newspaper are given to celli members two times a month and the members read this newspaper and tell the people about the successful development activities. In regard to the enemy [the OLF], the celli advises people not to follow them.’Footnote 59 After contestations in the 2005 elections, the ruling government had learned that ‘forcing people for the OPDO is not good’, as a kebele chairman underscored.Footnote 60 Instead, he argued that ‘membership must be based on interest’.Footnote 61 Although the cell had only just started its operation in 2009, and only a few activities had taken place, it was already perceived as a tool of grass-roots intelligence by local government representatives and citizens alike. A kebele manager explained, for example, that ‘the celli collects information in regard to who is acting against the regime’.Footnote 62

The establishment of the cell involved a re-classification of households and a re-selection of leaders. Several cells had been formed within the territory of a single garee in some kebele, while selected party members from different garee territories were organized in a single cell in others. In some of the visited kebele, the names of the cell leaders were listed for every cell in a registry book stored in the kebele office. This pre-2010 mobilization involved a re-selection process through which leaders were either dismissed or confirmed. At that time, many garee leaders and former gott leaders assumed positions as cell leaders and confirmed their loyalty to the ruling government and its developmental vision. As their reward, they gained influence through their ability to draw on the ruling party and its development agenda in the local political arena.

The obvious overlap between state and party structure, however, caused a great deal of confusion among ordinary citizens and government leaders alike. During field research, most informants were unable to explain the difference between the garee and the celli, since the latter had been created recently and was often embodied by the same leaders. Some cell leaders explained that ‘the kebele orders the celli and the celli orders the gareeFootnote 63 or that ‘the celli orders the garee and the garee orders the people’.Footnote 64 A former gott leader understood his new position as follows: ‘In place of the gott, they say celli, but there is no change at all.’Footnote 65 In a kebele visited, the kebele chairman had similar difficulties explaining the institutional structure of the cell, saying: ‘Let me see the documents; these things change frequently.’Footnote 66 In another kebele, the chairman forcefully convinced his fellow that ‘the celli is about the party and the garee about the state’,Footnote 67 despite his companion having just claimed the opposite – that ‘the celli and the garee are the same’.Footnote 68

THE LOCAL DEVELOPMENTAL STATE

The garee constitutes a new site for state–peasant interactions, where state authority is produced, maintained and contested. Poluha (Reference Poluha, Poluha and Rosendahl2002: 101) has characterized the relationship between the peasantry and ‘the power at large’ (Lefort Reference Lefort2007: 256) as vertical and highly hierarchical. In Meta Robi, government officials are able to place their actions and practices in a meaningful social framework of state-led development in order to subject the peasantry to the workings of a system of institutionalization and power (cf. Raeymaekers et al. Reference Raeymaekers, Menkhaus and Vlassenroot2008: 13). The persisting gap between the peasantry and the state, however, continues to undermine the legitimacy of a developmental state among the rural citizenry.

In Meta Robi, local government officials derived the importance of the state for development from an assumption of the peasant's backwardness. In the eyes of these officials, the single peasant is someone who is lazy, illiterate and individualistic, and who is exclusively interested in his own livelihood rather than in community development. They further point out that the peasantry lacks the awareness as well as the working culture required for development work. As the head of the Woreda Women's Affairs Office explained: ‘People dislike taking part in the meeting because they do not perceive it as development. Development for them is more about farming and keeping cattle.’Footnote 69 Dessalegn Rahmato (Reference Rahmato2008b) also observes the conception of the backward peasant in the field of rural development in Ethiopia, where it is ‘a common refrain today as well as in the past among government agents’.

The depiction of the district's peasantry as backward logically calls out for government intervention in order to achieve development. During field research, a number of local officials portrayed the creation of the garee and the nomination of garee leaders in themselves as an intervention to overcome backwardness and achieve development. For them, the garee helps to organize and control the people, because ‘organization is necessary for development’.Footnote 70 They assume that the garee encourages the required working culture by organizing its members in a structured way, for example for more efficient road construction. In addition, local officials presented the garee leader as someone who is qualified to represent people because of his ability to lead the community. In particular, garee leaders themselves underlined their commitment to the community as a crucial condition for achieving development in the district. As Harrison (Reference Harrison2002: 600) has documented elsewhere in Ethiopia, participation is perceived as people ‘working together to help their community’, an ideal that also lies at the very basis of the perceived superiority of local officials in Meta Robi (cf. Dessalegn Rahmato Reference Rahmato2008b: 253). This belief is nicely illustrated by the following statement by a kebele official: ‘Sometimes when people struggle against us, then we tell the people that we will stop working and that we could also do the work on the farm as they do.’Footnote 71

As a result, state-led development provides the logical and only solution for overcoming the backwardness of the peasantry. As a kebele manager confessed: ‘There [has been a] change since the establishment of the garee, a positive change. The government is also helping us and if it continues, there will be good progress.’Footnote 72 Hope for a better future, aspiration and desire have been of key relevance in Meta Robi and have powerfully convinced rural dwellers to engage as participants or leaders and to work for the people or the government. ‘I was convinced by the advantages of the road for us,’Footnote 73 a farmer confessed. Recalling Mains (Reference Mains2012: 5), this illustrates that ‘faith in progressive narratives and a developmentalist state continues to be quite powerful’ in urban as well as rural Ethiopia.

The local political arena in Meta Robi is divided between those who evaluate the garee as contributing to development and those who contest that view. In private conversations (i.e. in the absence of the kebele administrator), ordinary people often questioned the government and its development initiatives. As a farmer complained: ‘Meles [former Prime Minister and chairman of the EPRDF] fails to understand the problems of the poor. We travel on the road but Meles fails to understand us.’Footnote 74 While scepticism abounds in private, state-led development is barely challenged in public. Vaughan and Tronvoll (Reference Vaughan and Tronvoll2003: 34) similarly observed that people at the local and village level seldom question state authority in public and that local administrators traditionally control the political arena. What this means in Meta Robi became clear during an interview with a farmer, who initially denied that the newly built Kuyu Gicci road had been maintained regularly, but later remained silent when the newly arrived kebele chairman started to claim the opposite.

In Meta Robi, the government has been repeatedly disparaged by rural dwellers when state-led development interventions have not materialized as expected. ‘They say that the garee will bring change in the future, [but] nothing has changed so far. Everything is as it was,’ as a participant objected.Footnote 75 Similarly, many informants criticized the government for its inability to achieve its policy goals in a variety of development-related fields. In the case of the Kuyu Gicci road, for example, a former gott leader expressed his disappointment as follows: ‘At the beginning, the government had promised to bring fertilizer [to Kuyu Gicci], but this year it is not even available.’Footnote 76 In particular, this informant complained about the promise made at initial kebele meetings that fertilizer would be brought to Kuyu Gicci after the road construction. The deputy kebele chairman confirmed that: ‘At the beginning, we also planned an office where fertilizer could be distributed but it was never constructed.’Footnote 77 For this government representative, the absence of a fertilizer storehouse in Kuyu Gicci was not an expression of the government's failure. He countered: ‘We have plans for the future.’Footnote 78

Others, however, expressed their confusion about the government's plan: ‘We don't know why, but they say something today and they change it tomorrow.’Footnote 79 In this sense, state-led development morphed from being a shared desire into an object of contestation. The statement of a local informant nicely illustrates this: ‘The road only benefits some officials who can come easily from the woreda in order to give us information.’Footnote 80 The project of building the road, which should have attested to the government's commitment to community development, therefore turned here into an expression of its failure. Thus, state-led development as road construction does not automatically legitimize political rule or the ruling government, but it can provide a reason for its contestation.

CONCLUSION

An analysis of state–peasant encounters illustrates the complex entanglements between faith in progress and the closure of political space in rural Oromiya, Ethiopia. The empirical case presented reveals that state authority is simultaneously constituted and undermined in the course of decentralization and state-led development in an authoritarian context. Thereby, it discloses how state-led development is propagated, politicized and contested locally, and can turn from being a shared desire into an object of contestation.

The creation of the gott and garee has proved very significant in state-led development and peasant mobilization in rural Meta Robi. Initially introduced for state-led community projects in the rural district, the sub-kebele turned into an instrument of the local party-state for political mobilization and control in the run-up to the 2005 elections. Although it fell into neglect soon afterwards, the structure continued to unfold within a tradition of authoritarian rule at the kebele level, where local government officials and garee leaders continued to mobilize the peasantry through bureaucratic and authoritarian means. Moreover, the garee, its decentralized constitution and development activities became increasingly politicized in the overlapping arenas of development and security. In preparation for the 2010 elections, however, the ruling party again intervened with the creation of a new set of structures as its new mobilization strategy in the rural district.

In Meta Robi, the instrumentalization of the sub-kebele and its fusion with the local party-state in the context of electoral authoritarianism have undermined its democratic legitimacy among the rural citizenry. Nevertheless, the garee has also carried notions of state-led development down to the grass roots, where the peasantry has been subject to the operation of a meaningful system of institutionalization and power. On the one hand, this includes rendering the rural household into a state category and incorporating it into an administrative hierarchy. On the other hand, it contains the specification of a local developmental state and the terms in which faith in progress and aspiration for change are conceived and desired. In this sense, decentralization has not only extended the reach of the state into rural society, but has also diversified the ways in which the state is represented in people's everyday lives.

ACKNOWLEDGEMENTS

I am indebted to Sibilo Keno and Tobias Hagmann for their support in the process of writing this article. I also gratefully acknowledge the help of my research assistant Girma Hundessa during fieldwork. I am furthermore beholden to Jennifer Bartmess, Ilona Imoberdorf, Siegfried Pausewang, Ephraim Pörtner, Timothy Raeymaekers, Anna-Katharina Thürer and three anonymous reviewers for helpful comments on earlier versions of this article. Sadly, Sibilo Keno and Siegfried Pausewang died before they could see the article published. I would also like to acknowledge the support of the University of Zurich (Department of Geography), the Swiss National Centre of Competence in Research North-South (NCCR North-South) and the Addis Ababa University (Department of Anthropology).

Footnotes

1 Derg means ‘committee’ or ‘council’ in Amharic and refers to the 1974–91 Leninist–Marxist military government (Donham Reference Donham1999: 19).

2 In 1994, an ethnic-based federation of nine regional states was defined in the constitution. Article 39/3 of the Ethiopian Constitution (FDRE 1994) defines the power balance between the federal and regional level, and guarantees ‘every Nation, Nationality and People in Ethiopia . . . [the] unconditional right to self-determination’.

3 Getachew Bedane, former head of Oromiya Regional Government, in a telephone interview with Human Rights Watch in 2005 (HRW 2005: 31).

4 Conceptually, this article draws on the notion of the actor (Long Reference Long, Long and Long1992; Reference Long2001) and the metaphor of the local political arena (Bierschenk et al. Reference Bierschenk, J.-P. and Olivier de Sardan2002; Bierschenk and Olivier de Sardan Reference Bierschenk and Olivier de Sardan1997; Hagmann and Péclard Reference Hagmann and Péclard2010; Olivier de Sardan Reference Olivier de Sardan2005).

5 In a national comparison, Meta Robi's road density appears relatively high, with a network length of about 190 km per 1,000 km2 compared with the national average of 104 km per 1,000 km2 (cf. ERA 2008b: 8).

6 DS-6 is equivalent to a RR-50 standard according to the previous classification system of rural roads (cf. Emmenegger Reference Emmenegger2012: 13).

7 The significance of roads in the EPRDF's development policy is outlined in Road Sector Development Programmes (RSDP I to III) and the Ethiopian Rural Travel and Transportation Sub-Programme (ERTTP) (ERA 2007; 2008a; 2009; ORSG 2009).

8 Interview with Oromiya Road Construction Enterprise professional, Meta Robi, 27 July 2009.

10 The OLF emerged from the Macha and Tulama Association in the 1970s (Mohammed Hassen Reference Hassen and Pausewang2009: 32).

11 Interview with former garee leader, Meta Robi, 9 August 2009.

12 Interview with head of WRRO, Meta Robi, 24 August 2009.

13 In the broader literature, the term ‘sub-kebele’ is used in different ways. In some cases, it is used as an umbrella term for various institutions and layers below the kebele. In other cases, it refers to a specific institution or layer structurally located below the kebele. In the latter sense, ‘each kebelle is divided into three sub-kebelles and ten gotts for purposes of administration and service delivery. Each gott is further divided into five ye-limat bouden [development teams]’ (Dessalegn Rahmato Reference Rahmato2008b: 254).

14 HRW (2005) dates the creation of the gott and garee to 2004. During field research in 2009, however, many informants found it difficult to indicate the exact date of their establishment, with dates ranging between 2002 and 2004.

15 Similar institutions exist in other regions and have been documented in Amhara, Tigray and Southern Nations regional states. The gott in Oromiya equates to the gott in Amhara or the qushet in Tigray, which are also structurally located between the kebele and the garee. Also, the garee or garee misoma (development team) find their equivalent in the limat budin in Amhara and in the gudjle limat in Tigray (Bevan and Pankhurst Reference Bevan and Pankhurst2007: 70; Pankhurst Reference Pankhurst2008: 12; Segers et al. Reference Segers, Dessein, Hagberg, Develtere, Haile and Deckers2008: 13).

16 Interview with head of Woreda Security Office, Meta Robi, 27 August 2009.

17 Interview with former head of WRRO, Meta Robi, 14 August 2009.

18 Interview with head of Woreda Security Office, Meta Robi, 27 August 2009.

19 Interview with former head of WRRO, Meta Robi, 14 August 2009.

20 Interview with head of WRRO, Meta Robi, 24 August 2009.

21 Interview with former garee leader, Meta Robi, 30 August 2009.

22 Interview with former kebele administrator, Meta Robi, 27 August 2009.

23 Interview with kebele deputy administrator, Shino, 17 August 2009.

24 Interview with a member of the militia, Meta Robi, 25 August 2009.

25 Interview with garee leader, Meta Robi, 16 August 2009.

26 Interview with a farmer, Meta Robi, 30 August 2009.

27 As an example of the institutional reconfiguration, the size of the committee was changed from three team leaders (representing women, youth and administration) to initially five and later three leaders (chairman, secretary and cashier) with the creation of the garee. The territorial reconfiguration included the re-drawing of boundaries within the kebele. Moreover, the budin reportedly engaged in activities outside the kebele's territory and was not constrained by its borders as the garee are today. Nevertheless, the government teams of the Derg provide a conceptual basis for the later establishment of development teams (cf. Vaughan and Tronvoll Reference Vaughan and Tronvoll2003: 40).

28 Interview with head of Woreda Security Office, Shino, 27 August 2009.

29 According to an informant, there are seventy-four members of the armed militia in one of the kebele around Ketiketto, which has about 900 inhabitants. In a neighbouring kebele, the same informants even reported that there were 250 militia members selected from among its 1,100 inhabitants.

30 Interview with member of the militia, Meta Robi, 25 August 2009.

31 Interview with kebele deputy administrator, Shino, 17 August 2009.

32 Interview with kebele administrator, Meta Robi, 18 August 2009.

33 Interview with garee leader, Meta Robi, 30 August 2009.

34 Interview with head of Woreda Security Office, Shino, 27 August 2009.

35 Such a fine amounts to about US$2 in total in the cases documented.

36 Interview with young farmers, Shino, 17 August 2009.

38 Interview with garee leader, Meta Robi, 26 August 2009.

39 With the 1994 Constitution, all people gained ‘the right to participate in national [and regional] development and, in particular, to be consulted with respect to policies and projects affecting their community’ (FDRE 1994: Article 43, 2; cf. ORSG 2001: 22, Article 46).

40 Interview with elder, Meta Robi, 31 August 2009.

41 Email exchange with the author of the HRW report, Chris Albin-Lackey, 22 June 2009.

42 Interview with woreda administrator, Meta Robi, 31 August 2009.

45 Interview with former garee leader, Meta Robi, 9 August 2009.

46 Interview with kebele deputy administrator and former garee leader, Shino, 17 August 2009.

47 Interview with former kebele administrator, Meta Robi, 30 August 2009.

48 Interview with former garee leader, Meta Robi, 9 August 2009.

49 Interview with elder, Meta Robi, 22 August 2009.

50 Interview with farmer, Meta Robi, 30 August 2009.

51 Interview with former gott leader, Meta Robi, 30 August 2009.

52 The Ethiopian Birr (ETB) is the currency used in Ethiopia.

53 Interview with former garee leaders, Meta Robi, 9 August 2009.

56 Interview with celli and garee leader, Meta Robi, 31 August 2009.

57 Interview with kebele administrator, Meta Robi, 31 August 2009.

58 Interview with woreda administrator, Meta Robi, 31 August 2009.

59 Interview with kebele manager, Shino, 26 August 2009.

60 Interview with kebele administrator, Meta Robi, 31 August 2009.

62 Interview with kebele manager, Shino, 26 August 2009.

63 Interview with celli and garee leader, Meta Robi, 31 August 2009.

64 Interview with celli and gott leader, Meta Robi, 24 August 2009.

66 Interview with kebele administrator, Meta Robi, 18 August 2009.

67 Interview with kebele administrator, Meta Robi, 31 August 2009.

68 Interview with kebele official, Meta Robi, 31 August 2009.

69 Interview with head of Women's Affairs Office, Meta Robi, 29 August 2009.

70 Interview with garee leader, Meta Robi, 27 August 2009.

71 Interview with kebele deputy administrator, Meta Robi, 16 August 2009.

72 Interview with kebele manager, Shino, 26 August 2009.

73 Interview with farmer, Shino, 27 August 2009.

74 Interview with farmer, Meta Robi, 18 August 2009.

75 Interview with labourer, Shino, 27 August 2009.

76 Interview with former gott leader, Meta Robi, 31 August 2009.

77 Interview with kebele deputy administrator, Shino, 17 August 2009.

79 Interview with former garee leaders, Meta Robi, 9 August 2009.

80 Interview with farmer, Meta Robi, 25 August 2009.

References

REFERENCES

Aalen, L. and Tronvoll, K. (2009) ‘The end of democracy? Curtailing political and civil rights in Ethiopia’, Review of African Political Economy 36 (120): 193207.Google Scholar
Abbink, J. (2011) ‘“Land to the foreigners”: economic, legal, and socio-cultural aspects of new land acquisition schemes in Ethiopia’, Journal of Contemporary African Studies 29 (4): 513–35.Google Scholar
Agrawal, A. and Ostrom, E. (2001) ‘Collective action, property rights and decentralization in resource use in India and Nepal’, Politics and Society 29 (4): 485514.Google Scholar
Aspen, H. (2002) ‘Models of democracy: perceptions of power, government and peasantry in Ethiopia’ in Zewde, Bahru and Pausewang, S. (eds), The Challenge of Democracy from Below. Uppsala and Addis Ababa: Nordiska Afrikainstitutet and Forum for Social Studies.Google Scholar
Bevan, P. and Pankhurst, A. (2007) ‘Power structures and agency in rural Ethiopia: development lessons from four community case studies’. Paper prepared for the Empowerment Team in the World Bank Poverty Reduction Group. Bath: University of Bath.Google Scholar
Bierschenk, T. and Olivier de Sardan, J.-P. (1997) ‘Local powers and a distant state in rural Central African Republic’, Journal of Modern African Studies 35 (3): 441–68.Google Scholar
Bierschenk, T., J.-P., Chauveau and Olivier de Sardan, J.-P. (2002) Local Development Brokers in Africa: the rise of a new social category. Working papers 13. Mainz: Johannes Gutenberg University.Google Scholar
Boone, C. (2003) ‘Decentralization as political strategy in West Africa’, Comparative Political Studies 36: 355–80.Google Scholar
Chanie, P. (2007) ‘Clientelism and Ethiopia's post-1991 decentralization’, Journal of Modern African Studies 45 (3): 355–84.Google Scholar
Chinigò, D. (2014a) ‘Decentralization and agrarian transformation in Ethiopia: extending the power of the federal state’, Critical African Studies 6 (1): 4056.Google Scholar
Chinigò, D. (2014b) ‘The politics of land registration in Ethiopia: territorialising state power in the rural milieu’, Review of African Political Economy 42 (144): 174–89.Google Scholar
Clapham, C. (1989) ‘The state and revolution in Ethiopia’, Review of African Political Economy 16 (44): 417.Google Scholar
Crewett, W. and Korf, B. (2008) ‘Ethiopia: reforming land tenure’, Review of African Political Economy 35 (2): 203–20.Google Scholar
Rahmato, Dessalegn (1993) ‘Agrarian change and agrarian crisis: state and peasantry in post-revolution Ethiopia’, Africa 63 (1): 3655.Google Scholar
Rahmato, Dessalegn (2008a) ‘Ethiopia: agriculture policy review’ in Assefa, Taye (ed.), Digest of Ethiopia's National Policies, Strategies and Programme. Addis Ababa: Forum for Social Studies.Google Scholar
Rahmato, Dessalegn (2008b) The Peasant and the State: studies in agrarian change in Ethiopia, 1950s–2000s. Addis Ababa: Custom Book Publishing.Google Scholar
Devereux, S., Teshome, Amdissa and Sabates-Wheeler, R. (2005) ‘Too much inequality or too little? Inequality and stagnation in Ethiopian agriculture’, IDS Bulletin 36 (2): 121–6.Google Scholar
Di Nunzio, M. (2014) ‘Thugs, spies and vigilantes: community policing and street politics in inner city Addis Ababa’, Africa 84 (3): 444–65.CrossRefGoogle Scholar
Diawara, M. (2011) ‘Development and administrative norms: the Office du Niger and decentralization in French Sudan and Mali’, Africa 81 (3): 434–54.Google Scholar
Donham, D. (1999) Marxist Modern: an ethnographic history of the Ethiopian revolution. Berkeley CA and Oxford: University of California Press and James Currey.Google Scholar
Emmenegger, R. (2012) The Roads of Decentralization: the history of rural road construction in Ethiopia. NCCR North-South Dialogue 39. Bern and Zurich: National Centre of Competence in Research (NCCR) North-South.Google Scholar
Emmenegger, R., Keno, Sibilo and Hagmann, T. (2011) ‘Decentralization to the household: expansion and limits of state power in rural Oromiya’, Journal of Eastern African Studies 5 (4): 733–54.Google Scholar
ERA (2007) Road Sector Development Programme III (2007–2010). Addis Ababa: Ethiopian Roads Authority (ERA), Ministry of Works and Urban Development, Federal Democratic Republic of Ethiopia.Google Scholar
ERA (2008a) An Account of the Ethiopian Rural Travel and Transport Programme and Prospects: assessment of the pilot implementation of the ERTTP. Addis Ababa: Ethiopian Roads Authority (ERA), Ministry of Works and Urban Development, Federal Democratic Republic of Ethiopia.Google Scholar
ERA (2008b) RSDP Performance: eleven years later. Addis Ababa: Ethiopian Roads Authority (ERA), Ministry of Works and Urban Development, Federal Democratic Republic of Ethiopia.Google Scholar
ERA (2009) The Ethiopian Rural Travel and Transport Program (ERTTP): implementation arrangement of the ERTTP. Addis Ababa: Ethiopian Roads Authority (ERA), Ministry of Works and Urban Development, Federal Democratic Republic of Ethiopia.Google Scholar
ETCA (1986) Labor-based, Equipment-supported Rural Road Construction and Maintenance in Ethiopia: a programme proposal. Addis Ababa: Rural Roads Department, Ethiopian Transport Construction Authority (ETCA), Provisional Military Government of Socialist Ethiopia.Google Scholar
FDRE (1994) Constitution of the Federal Democratic Republic of Ethiopia. Addis Ababa: Federal Democratic Republic of Ethiopia (FDRE).Google Scholar
Geiser, U. and Rist, S. (2009) Decentralization Meets Local Complexity: local struggles, state decentralization and access to natural resources in South Asia and Latin America. Perspectives of the Swiss National Centre of Competence in Research (NCCR) North-South, University of Bern, volume 4. Bern: Geographica Bernensia.Google Scholar
Hagmann, T. and Péclard, D. (2010) ‘Negotiating statehood: dynamics of power and domination in Africa’, Development and Change 41 (4): 539–62.CrossRefGoogle Scholar
Harrison, E. (2002) ‘The problem with the locals: partnership and participation in Ethiopia’, Development and Change 33 (4): 587610.Google Scholar
HRW (2005) Suppressing Dissent: human rights abuses and political repression in Ethiopia's Oromia region. New York NY: Human Rights Watch (HRW).Google Scholar
ICG (2009) Ethiopia: ethnic federalism and its discontents. Africa Report 153. Nairobi and Brussels: International Crisis Group (ICG).Google Scholar
Keeley, J. and Scoones, I. (2000) ‘Knowledge, power and politics: the environmental policy-making process in Ethiopia’, Journal of Modern African Studies 38 (1): 89120.CrossRefGoogle Scholar
Lefort, R. (2007) ‘Powers – mengist – and peasants in rural Ethiopia: the May 2005 elections’, Journal of Modern African Studies 45 (2): 253–73.Google Scholar
Lefort, R. (2010) ‘Powers – mengist – and peasants in rural Ethiopia: the post-2005 interlude’, Journal of Modern African Studies 48 (3): 435–60.Google Scholar
Lefort, R. (2012) ‘Free market economy, “developmental state” and party-state hegemony in Ethiopia: the case of the “model farmers”’, Journal of Modern African Studies 50 (4): 681706.Google Scholar
Lefort, R. (2013) ‘The theory and practices of Meles Zenawi: a response to Alex de Waal’, African Affairs 448 (112): 460–70.Google Scholar
Long, N. (1992) ‘Part I: an actor-oriented paradigm’ in Long, N. and Long, A. (eds), Battlefields of Knowledge: the interlocking of theory and practice in social research and development. London: Routledge.Google Scholar
Long, N. (2001) Development Sociology: actor perspectives. London: Routledge.Google Scholar
Mains, D. (2012) ‘Blackouts and progress: privatization, infrastructure and a developmentalist state in Jimma, Ethiopia’, Cultural Anthropology 27 (1): 327.Google Scholar
Ayenew, Meheret (2002) ‘Decentralization in Ethiopia: two case studies on devolution of power and responsibilities to local authorities’ in Zewde, Bahru and Pausewang, S. (eds), The Challenge of Democracy from Below. Uppsala and Addis Ababa: Nordiska Afrikainstitutet and Forum for Social Studies.Google Scholar
Kristos, Meserete (2009) Meta Robi Alternative Basic Education School Project: terminal evaluation. Addis Ababa: Meserete Kristos Church Relief and Development Association.Google Scholar
Hassen, Mohammed (2009) ‘Oromo nationalism, and the continuous multifaceted attack on the Oromo cultural civic and political organisations’ in Pausewang, S. (ed.), Exploring New Political Alternatives for the Oromo in Ethiopia: report from Oromo workshop and its after-effects. Bergen: Chr. Michelsen Institute.Google Scholar
Olivier de Sardan, J.-P. (2005) Anthropology and Development: understanding contemporary social change. London: Zed Books.Google Scholar
Olowu, D. (2001) Decentralization Policies and Practices under Structural Adjustment and Democratization in Africa: democracy, governance and human rights. Paper no. 4. Geneva: United Nations Research Institute for Social Development.Google Scholar
ORSG (2001) Revised Constitution: megeleta Oromiya. Proclamation 46. Addis Ababa: Oromiya Regional State Government (ORSG).Google Scholar
ORSG (2009) ‘Sectoral development: infrastructure – road’. Addis Ababa: Oromiya Regional State Government (ORSG) <http://www.oromiagov.org/Communication.asp>, accessed 15 April 2009.,+accessed+15+April+2009.>Google Scholar
Pankhurst, A. (2002) The Influence of the State and Market on Local Level Management of Natural Resources: case studies of forest, irrigation and pasture sites in South Wello, Ethiopia. Madison WI: Broadening Access and Strengthening Input Market Systems Collaborative Research Support Program (BASIS CRSP) Management Entity, University of Wisconsin-Madison.Google Scholar
Pankhurst, A. (2008) Enhancing Understanding of Local Accountability Mechanisms in Ethiopia: Protection Basic Service project, PBS II preparation studies. Revised summary report. Addis Ababa: no publisher.Google Scholar
Pausewang, S. (2002) ‘No environmental protection without local democracy? Why peasants distrust their agricultural advisers’ in Zewde, Bahru and Pausewang, S. (eds), The Challenge of Democracy from Below. Uppsala and Addis Ababa: Nordiska Afrikainstitutet and Forum for Social Studies.Google Scholar
Pausewang, S. (2009) Exploring New Political Alternatives for the Oromo in Ethiopia: report from Oromo workshop and its after-effects. Bergen: Chr. Michelsen Institute. Google Scholar
Planel, S. (2014) ‘A view of a bureaucratic developmental state: local governance and agricultural extension in rural Ethiopia’, Journal of Eastern African Studies 8 (3): 420–37.Google Scholar
Poluha, E. (2002) ‘Learning political behaviour: peasant–state relations in Ethiopia’ in Poluha, E. and Rosendahl, M. (eds), Contesting ‘Good’ Governance: crosscultural perspectives on representation, accountability and public space. London: Routledge Curzon.Google Scholar
Raeymaekers, T., Menkhaus, K. and Vlassenroot, K. (2008) ‘State and non-state regulation in African protracted crises: governance without government?’, Africa Focus 21 (2): 721.Google Scholar
Ribot, J. C., Agrawal, A. and Larson, A. (2006) ‘Recentralizing while decentralizing: how national government reappropriate forest resources’, World Development 34 (11): 1864–86.CrossRefGoogle Scholar
Riedl, R. B. and Dickovick, J. T. (2014) ‘Party systems and decentralization in Africa’, Studies in Comparative International Development 49 (3): 321–42.CrossRefGoogle Scholar
Rondinelli, D. (1981) ‘Government decentralization in comparative perspective’, International Review of Administrative Sciences 47 (2): 133–45.Google Scholar
Rondinelli, D., Nellis, J. and Shabbir-Cheema, G. (1984) Decentralization in Developing Countries: a review of recent experience. Washington DC: World Bank.Google Scholar
Segers, K., Dessein, J., Hagberg, S., Develtere, P., Haile, Mitiku and Deckers, J. (2008) ‘Be like bees: the politics of mobilizing farmers for development in Tigray, Ethiopia’, African Affairs 108 (430): 119.CrossRefGoogle Scholar
Spielman, D. J., Cohen, M. J. and Mogues, Tewodaj (2008) Mobilizing Rural Institutions for Sustainable Livelihoods and Equitable Development: a case study of local governance and smallholder cooperatives in Ethiopia. Washington DC: International Food Policy Research Institute (IFPRI).Google Scholar
Gebre-Egziabher, Tegegne and Berhanu, Kassahun (2007) ‘A literature review of decentralization in Ethiopia’ in Assefa, Taye and Gebre-Egziabher, Tegegne (eds), Decentralization in Ethiopia. Addis Ababa: Forum for Social Studies.Google Scholar
Tronvoll, K. (2011) ‘The Ethiopian 2010 federal and regional elections: re-establishing the one-party state’, African Affairs 110 (438): 121–36.Google Scholar
Vaughan, S. and Tronvoll, K. (2003) The Culture of Power in Contemporary Ethiopian Political Life. Sida Studies 10. Stockholm: Swedish International Development Cooperation Agency (Sida).Google Scholar
Wunsch, J. (2001) ‘Decentralization, local governance and “recentralization” in Africa’, Public Administration and Development 21: 277–88.Google Scholar
Wunsch, J. (2008) ‘Why has decentralization failed in Africa? Assessing the lessons of self-organized, local governance initiatives’. Workshop on Political Theory and Policy Analysis. Bloomington IN: Indiana University.Google Scholar
Figure 0

Figure 1 Meta Robi road map. The sub-kebele does not appear on this map found in the WRRO office at district level in 2009. As this indicates, state territorialization through the sub-kebele is not primarily a cartographic exercise, but rather advances through the establishment of administrative hierarchies at the local level.

Figure 1

Figure 2 Administrative hierarchy of the existing decentralized structure including the sub-kebele.