Responding to land-based conflict in Ethiopia: The land rights of ethnic minorities under federalism

Responding to land-based conflict in Ethiopia: The land rights of ethnic minorities under federalism Abstract There is a common perception that Ethiopia is unusual in Africa in having a relatively uniform system of state land ownership. While highly influential, state ownership is not the only body of law with implications for land administration. This article argues that the institutionalization of ethnic federalism and the persistence of neo-customary tenure result in considerable ambiguity, particularly regarding the land rights of non-indigenous minorities. The analysis highlights tensions between these three sets of land tenure institutions—state ownership, ethnic federalism and neo-customary tenure—and their implications for minority land rights. A case study of land-based conflict in Oromiya region, based on fieldwork conducted in 2009 and 2010, demonstrates the continuing relevance of these land tenure institutions and associated ideas in land debates in Ethiopia, both in terms of the use of these ideas by protagonists as means of justifying land claims, and the ambiguous state response to the conflict, which goes well beyond the provisions of the land policy. As such, while there are certainly particular characteristics of the Ethiopian case, many of the key issues regarding ethnicity and land mirror debates taking place across the continent. Research has repeatedly highlighted the ‘multiple and overlapping’ nature of land tenure in sub-Saharan Africa. This institutional plurality results from the claims of colonial and post-colonial authorities to ultimate ownership of all land, while states manipulate or reinforce neo-customary tenure as a mechanism of social control.1 In contrast, the land question in Ethiopia is frequently considered to be ‘rather unique’ in sub-Saharan Africa, with a uniform policy of state land ownership with usufruct rights distributed to farmers.2 Not only was Ethiopia’s brush with European colonialism relatively brief, but the 1975 land reform that nationalized all land was ‘particularly extensive’3 and ‘effectively abolished’4 or ‘wiped out’5 previous tenure systems. While the significance of the 1975 land reform is indisputable, there nonetheless remains ambiguity in land tenure regarding the authority of state and non-state actors and the implications of different land tenure institutions for citizenship. This ambiguity is partly due to the continuing influence of neo-customary tenure regimes, which in many places were not actually wiped out by state ownership, and, arguably, have become increasingly influential in recent years.6 However, ambiguity also results from the territorial implications of ethnic federalism introduced in the 1990s. While ethnic federalist principles resonate in certain respects with neo-customary tenure by prioritizing the land rights of ‘indigenous’ inhabitants, they are in direct tension with state ownership regarding the rights of ethnic minorities outside their ‘home’ region. This article focuses specifically on this question of the land rights of non-indigenous minorities. The analysis uses a case of land conflict in Oromiya to illustrate the continuing relevance of these competing perspectives on land tenure and to highlight how the state’s attempts to resolve the dispute reflect this ambiguity regarding minority land rights. In doing so, the article links to important recent debates on land tenure in Africa. Land tenure regimes structure intra-community relations and relations between state and society.7 Notably, land tenure regimes have direct implications for the political salience of ethnicity, the form that resource conflict takes, and citizenship rights.8 Ethiopian land tenure is certainly an extreme case on the African continent. In the majority of sub-Saharan Africa, the dominant form of land tenure remains neo-customary, whereas in Ethiopia state ownership predominates. Furthermore, Ethiopia’s ethnic federal system is unusual and provides an alternate rationale in land debates absent in most other countries. That said, while the relative importance of different land tenure systems varies between countries, the tensions between these systems raise closely related issues. Consequently, in Ethiopia, as in much of Africa, an important thread of contemporary land debates is the question of whether land should be made available to all citizens—who have a right to live anywhere within the country—or whether the land rights of ‘indigenous’ populations should be prioritized. The contemporary relevance of these issues in Ethiopia has been vividly highlighted by the wave of protests that have swept across Oromiya and then Amhara since 2014. These protests have resulted from a confluence of causes, including youth unemployment, inadequate representation of individual and group interests, government corruption, fragmentation within the ruling party since 2012 and growing ethno-nationalism. However, an important flashpoint for the protests and a significant grievance concerned questions of land and territory under ethnic federalism, both in terms of the delineation of ethno-regional borders and the relative prioritization of land access for the indigenous population and outside investors. There is already a large literature on Ethiopia’s ethnic federal project.9 Research specifically on the links between land and federalism has tended to focus on the challenge of demarcating borders between ethnic regions, leading to territorial competition along ethnic lines.10 In contrast, there has been little attempt to examine the implications of federalism for non-indigenous ethnic minority land rights. This is, perhaps, surprising given the clear implications of federalism for territory and citizenship. Both Endrias Eshete and Alemante Selassie note the potential weakness of minority land rights outside their home regions in their broad discussions of federalism, while Christophe Van der Beken highlights the limited rights of non-indigenous minorities in general.11 None of these valuable contributions, however, explores the inconsistencies regarding land rights between federalism and state ownership of land in policy and practice. The case study presented in the article is based on fieldwork conducted in 2009 and 2010 involving 53 key informant interviews with residents of the village in question, former residents who had been displaced, and government officials. Respondents within the community were purposively selected to provide variation in ethnicity and gender, as well as to identify people who were old enough to have directly experienced the events. Key informant interviews were also conducted with the leaders of the association formed to represent the displaced population and who were therefore in a position to provide insights into the subsequent attempts at dispute resolution. Representatives of the displaced ethnic minority also provided extensive documentation relating to the appeals that they made to government officials and the government’s attempts to resolve the dispute. This written documentation constitutes a vital source of data, providing information from the time about these events, and enabling triangulation of interview testimony. The article proceeds as follows. The first section conceptualizes the linkages between land tenure institutions and ideas before outlining the contrasting implications of three sets of ideas and laws for the land rights of ethnic minorities, notions of citizenship and territorial authority. Section two then examines the case study of a land conflict in Oromiya region and government attempts at dispute resolution over the subsequent twenty years. This case study demonstrates the continuing relevance of these three ideas and laws in land debates in Ethiopia, the use of these ideas by protagonists as means of justifying land claims, and the ambiguous state response to the conflict, which appears to go well beyond the provisions of the land policy. Section three reflects on the implications of this case for broader debates about land, ethnicity, and conflict in Ethiopia, particularly in relation to recent protests and unrest. The fourth and final section summarizes the arguments and concludes with a discussion of the parallels with debates on land tenure in other parts of sub-Saharan Africa. Contrasting ideas and laws on land and territory in Ethiopia Land tenure regimes are not just means of distributing resources but key mechanisms by which states relate to their societies and project authority into rural areas. In addition to a system of property rights, therefore, as Catherine Boone argues, land tenure regimes comprise three main elements.12 First, a locus of political authority—usually the state or neo-customary authorities—with responsibility for the adjudication and enforcement of land rights. Second, a territorial jurisdiction that is subject to the land tenure regime. This jurisdiction may be restricted to a particular locality or span the entire national territory. Third, land tenure regimes imply different forms of citizenship, defined as ‘the boundaries of membership in a political community’.13 Neo-customary tenure regimes frequently imply local citizenship based on membership of a particular descent group and confer economic and political rights, including entitlements to land. In contrast, state ownership establishes a direct relationship between land users and the central state and implies national citizenship. For discursive institutionalists, institutions, including those related to land tenure, ‘are built on ideational foundations’.14 Institutions are not the simple product of the preferences of competing interest groups but draw on ideas for inspiration. However, ideas are rarely adopted tout court as institutions. Instead, ideas are likely to be adapted through processes of bricolage—combining one idea with others—or translation—the adaptation of an idea when taken from one context to another.15 As such, ideas that help shape land tenure institutions at one point in time may continue to have a life of their own within public and political debate, independent from the institutions that they helped to shape. As such, these ideas may continue to be influential in policy debates or in the process of implementation. This section examines three sets of influential ideas and associated institutions with direct implications for land tenure regimes, highlighting their similarities and contradictions. The first two have been promoted directly by the current government: state land ownership and ethnic federalism. The third is the neo-customary land tenure regime in Arssi, Oromiya, which, though originating in Arssi Oromo society, has been promoted and adapted by the state in recent years. The first, and most influential, set of land tenure institutions and associated ideas relate to the system of state land ownership. The Derg regime (1974–91) nationalized all rural land in 1975, distributing usufruct rights to smallholders. The Derg undoubtedly had political motivations, namely the destruction of the landholding class that was the base of Imperial power.16 However, land reform was justified in egalitarian and social justice terms encapsulated in the rallying cry of ‘land to the tiller’. The new tenure system was universalistic—providing land access for all Ethiopian citizens—and class-based stating that, ‘any person who is willing to personally cultivate land shall be allotted rural land sufficient for his maintenance and that of his family’.17 The Tigrayan People’s Liberation Front (TPLF) began its insurrection against the Derg in the mid-1970s. When the TPLF’s ambitions grew from liberating Tigray to seizing national power, the TPLF formed the Ethiopian Peoples’ Revolutionary Democratic Front (EPRDF), a coalition of ethno-nationalist parties to represent Ethiopia’s different ethnic groups, of which the TPLF was the leading member. On coming to power in 1991, the EPRDF largely maintained the Derg’s land policy. Again, there were important political motivations for doing so, namely to consolidate the support of the peasantry, to prevent the emergence of a landholding class, and to limit urban migration that might threaten social and political instability.18 However, there was a subtle shift in the discourse used to justify state land ownership. In addition to a shared emphasis on social equity and the prevention of class differentiation, former Prime Minister Meles Zenawi argued that state land ownership protected smallholders against displacement by market forces, providing a form of ‘social security’.19 Notwithstanding, Ethiopian citizenship and class remained key principles underpinning the land policy. As stated in the 1994 constitution: ‘Ethiopian peasants have right to obtain land without payment’.20 A similar sentiment is re-iterated in the 2005 federal land proclamation, ‘Any citizen of the country who is 18 years of age or above and wants to engage in agriculture for a living shall have the right to use rural land’.21 This is not to say that EPRDF land policy has remained a straightforward statement of Marxist principles. Rapid population growth has meant diminishing landholdings, while critics argued that agricultural productivity was impeded by tenure insecurity.22 The government’s response was to conduct land registration to enhance tenure security and investment incentives and prevent the sub-division of landholdings into unviable plots.23 In doing so, land policy has moved a step away from past universalistic principles; there are now very large numbers of landless people in rural areas who reached adulthood after the last land redistribution and have little hope of accessing land.24 Nonetheless, the idea of universal land access remains a key part of the debate. A second, distinct set of institutions and ideas with direct implications for land tenure relates to the system of ethnic federalism. The rights of ethnic groups within Ethiopia’s political system have been a polarizing feature of political debate since the late 1960s, with opposing camps drawing on very different interpretations of Ethiopian history.25 On one hand, centrists emphasize Ethiopia’s historical unity and fear its break up. On the other hand, regionalists use narratives of internal colonization and cultural hegemony by the Amhara and Tigrayan ethnic groups as the basis of claims for greater ethnic autonomy, even secession. While the Derg dismissed ethnicity as a distraction from class consciousness, for ethno-nationalist movements including the TPLF and the Oromo Liberation Front (OLF), ethno-nationalist struggle was seen as paving the way for class struggle. On taking national power in 1991, the EPRDF established an ethnic federal system that, in Christopher Clapham’s words, constituted a ‘wholesale takeover of Stalin’s theory of the national question as an approach to the problem of Ethiopia’s ethnic diversity’.26 Ethiopia’s ethnic federalism is based on the primordialist idea that ethnic groups—or nations, nationalities and peoples—are objectively identifiable, independent of the views of their members.27 As such, the EPRDF primarily relied on an ethno-linguistic criterion to ‘draw round’ ethnic groups—aiming to delineate regional boundaries that provided a perfect fit between ethno-linguistic groups and territorial borders.28 In Ethiopia’s new constitution these ethnic groups were given the ‘unconditional right to self-determination, including the right to secession’.29 Ethnic federalism therefore has important implications for land administration. Not only does federalism associate ethnic groups with defined territories, but, by providing the right to secession, it also explicitly prioritizes the land rights of ethnic groups over those of individuals. As argued by Endrias Eshete, To confer the right to secession on national communities is to grant that a regional state’s collective property rights take priority over the property rights of outsiders—nonmembers and federal government—in the region. What is now held by nonmembers can be legitimately taken by a seceding state.30 Federalism implies, therefore, that each ethnic group has its own home region and, consequently, that ethnic outsiders have a weaker claim to land than indigenous inhabitants. In effect, federalism implies that individuals are first and foremost citizens of ethnic regions, rather than of Ethiopia. There is a direct contradiction between the universalistic principles of the land policy, which accords all Ethiopian farmers equal land rights, and the territorial implications of ethnic federalism that suggest lesser rights for non-indigenous ethnic minorities. Under federalism, land administration has been devolved to regional governments, which are required to formulate land proclamations within the framework of the federal land policy.31 Despite the implications of ethnic federalism for land administration, however, regional land proclamations—in line with federal land policy—do not mention ethnicity. For example, the Oromiya land proclamation states that ‘any resident of the region’ has the right to land.32 Furthermore, when asked about the potential for ethnicity to influence land administration, the head of the federal Land Use Administration denied that there could be any ethnic differentiation in land rights: ‘ethnicity should not influence policy, farmers are farmers’.33 The only instance in which land administration explicitly distinguished between ethnic groups was the government’s resettlement programme that relocated food insecure farmers from densely populated highlands to more sparsely populated lowlands, primarily in the west and south. In an attempt to limit ethnic conflict between resettled households and existing inhabitants of resettlement sites, resettlement was conducted within, rather than between, regions.34 Although the full territorial implications of ethnic federalism are not articulated in land laws, the ideas underpinning ethnic federalism continue to be widely debated in Ethiopia. On numerous occasions during fieldwork people—mostly outside state structures, but occasionally within them—argued for a more extreme interpretation of federalism: that it is necessary to differentiate between ethnic groups depending on whether they are in their ‘home’ region or not, even justifying the removal of ethnic minorities to their ‘home’ regions to preserve resources for the dominant ethnic group—effectively ethnic cleansing.35 As such, there remains the possibility that these more extreme ideas on ethnic federalism could influence implementation, beyond the explicit provisions of land laws. The third and final set of land tenure institutions and related ideas concerns the neo-customary tenure system in Arssi, Oromiya. From a legal perspective, the 1975 nationalization eradicated any other form of tenure. However, in practice neo-customary tenure systems remain important influences on land administration in many parts of the country. I use the term neo-customary rather than customary in acknowledgement that ‘customary’ tenure systems are the product of inherited traditions adapted by state and societal actors. As Boone argues, it is invariably the case that ‘[p]ostcolonial governments have been deeply and actively implicated in upholding and reproducing the institutions and political relationships embedded in the (neo)customary land tenure regimes. It is incorrect to see customary land regimes as informal in the sense of existing beyond the purview of the state’.36 This rationale applies equally to neo-customary tenure in Ethiopia. Neo-customary tenure remains a major influence on land administration in areas where pastoralism and shifting cultivation are the dominant forms of production and where the territorial reach of the state continues to be much more limited than in the highlands, namely in Afar, Benishangul-Gumuz, Gambella, the south of Oromiya, Southern Nations, Nationalities and Peoples’ Region (SNNPR), and Somali.37 However, even in areas of smallholder cultivation in Oromiya, including the part of Arssi covered by the case discussed below, and SNNPR, neo-customary tenure continues to be an influential alternative to state land administration.38 According to the ideals of Arssi Oromo neo-customary tenure, land is the property of the patrilineal descent group or clan, and a principal objective of neo-customary tenure is to retain land within the descent group.39 As such, women—who are required to marry out of the clan—lose rights to land in their birth clan upon marriage, while women who marry into a clan only have rights to land through their husband.40 As such, Arssi Oromo neo-customary tenure is in direct tension with the universalist principles of state land ownership. Rather than land being a right of national citizenship, according to neo-customary ideals land access is limited to clan members, in effect implying a form of local citizenship. To a degree, therefore, neo-customary tenure resonates with the territorial implications of ethnic federalism. Like federalism, neo-customary law associates an ethnic group with a particular territory and, in doing so, curtails land rights for ethnic outsiders, prioritizing local citizenship. However, the locus of authority and the territorial jurisdiction differ. Under the ideals of federalism, the regional government has ultimate authority, whereas clan elders are responsible under neo-customary tenure. Furthermore, local citizenship under neo-customary tenure is based on the clan, excluding not just non-Oromo, but also other Oromo clans, in contrast to federalism, which provides equal rights for all those from the dominant ethnic group of the region. Table 1 summarizes the differences between these ideal-type land tenure regimes. While the distinctions between these ideals are clear in principle, the reality of actual land tenure institutions is rather more ambiguous. Table 1 Contrasting ideals on land in Ethiopia State ownership Ethnic federalism Arssi Oromo neo-customary Locus of authority over land Federal government Regional government Neo-customary leaders Territorial jurisdiction National Regional Local Citizenship principle National Ethnicity Clan membership State ownership Ethnic federalism Arssi Oromo neo-customary Locus of authority over land Federal government Regional government Neo-customary leaders Territorial jurisdiction National Regional Local Citizenship principle National Ethnicity Clan membership Source: author, based on categories developed in Boone, Property and political order in Africa. Table 1 Contrasting ideals on land in Ethiopia State ownership Ethnic federalism Arssi Oromo neo-customary Locus of authority over land Federal government Regional government Neo-customary leaders Territorial jurisdiction National Regional Local Citizenship principle National Ethnicity Clan membership State ownership Ethnic federalism Arssi Oromo neo-customary Locus of authority over land Federal government Regional government Neo-customary leaders Territorial jurisdiction National Regional Local Citizenship principle National Ethnicity Clan membership Source: author, based on categories developed in Boone, Property and political order in Africa. The picture is further complicated by Oromiya regional government attempts to re-establish the Gadaa institution. Gadaa is the customary system of social and political organization in Oromo society. During Imperial and Derg rule Gadaa was effectively wiped out in most of Oromiya and only really persisted among the Boran Oromo pastoralists. Since the early 2000s, the Oromiya People’s Democratic Organization (OPDO)—the regional EPDRF party—has sought to re-establish Gadaa as a means of compensating for its limited legitimacy. The OPDO has been regarded by many as an artificial creation of the TPLF/EPRDF that has struggled to establish legitimate authority in Oromiya. Indeed, the regional government recognizes that when it ‘works through the Gadaa it is more successful than through the local government’.41 As such, the position of Aba Gadaa—the political leader of the Gadaa and the ultimate authority over customary law—has been institutionalized in line with the government structures of zone and wereda42 and reports to the regional Bureau of Culture and Tourism.43 The regional land policy now also requires community elders to make the first attempt at dispute resolution, including with respect to land disputes. These elders are given training on government policy and issued with identity cards.44 Only if one party to a dispute is dissatisfied with the outcome does a case proceed to the courts or higher levels of government administration for adjudication. In the process, the regional government has sought to transform neo-customary tenure, jointly announcing changes to customary law with the Aba Gadaa and codifying neo-customary law in a book held in the wereda offices.45 Amongst other things, these changes seek to enhance women’s rights to hold land and, most importantly for this article, allow ethnic minorities to hold land and even to nominate elders from ethnic minorities. As such, there is now a distinct separation between popular ideas about the neo-customary tenure system as described above—frequently idealized as unchanging since the sixteenth century46—and what currently passes as customary tenure in parts of the region, which the regional government has sought to co-opt as a means of pursuing its policy objectives and enhancing its authority. Conflict and land rights in practice The case study presented in this section is used to illustrate the relevance of these ideas and institutions for land administration and to highlight the ambiguous state responses to one particular land conflict. The analysis builds on insights from Joel Migdal’s State-in-society approach and Tobias Hagmann and Didier Péclard’s Negotiating statehood framework.47 Rather than a unitary actor, the state is conceptualized as a network of distinct organizations, each with their own interests, in competition for domination with societal organizations. State officials, meanwhile, are embedded in society through their membership of families, clans, businesses or clubs. As such, policy implementation is not simply the impartial application of formal rules and laws by bureaucrats, but rather a negotiated process that may vary from the intentions of higher levels of government. In such circumstances, formal and informal institutions, and ideas all have the potential to influence decision-making. Turufe Kechema occupies a fertile agricultural area suited to cereal production in the Rift Valley with good road links to nearby Shashemene town and Addis Ababa. Agricultural production is dominated by smallholder farmers, most of whom hold approximately 0.5–3 hectares of land.48 Turufe is part of the territory claimed by the Weyrera clan from the Arssi Oromo ethnic group. The Weyrera continued to apply neo-customary tenure on land that was not expropriated by the Imperial regime and in secret alongside state ownership during the Derg’s rule.49 Despite neo-customary tenure, however, successive governments have ignored clan and ethnicity in land administration and there has been a steady flow of migrants into Turufe. In addition to the Weyrera, there are members of several other Arssi clans in Turufe who, under ideal-type neo-customary law would have no rights to land. Meanwhile, some of the first non-Oromo arrived during the Imperial conquest in the late nineteenth century. Northern soldiers (neftegna), especially Amhara and some Eritreans, were allocated rights to extract tribute from Arssi Oromo tenants (gebbar) as reward for military service.50 From the 1950s, migrants came for treatment at the nearby clinic, now the Shashemene Referral Hospital, a centre for leprosy treatment. At that time, Swedish Protestant missionaries, who established the clinic, held 12 gasha (480 hectares) around Turufe. These missionaries rehabilitated people treated for leprosy, in particular Kembata, by dividing two gasha between them.51 The area also attracted migrant labour during busy agricultural periods and some migrants have subsequently settled. Under the Derg, land was redistributed several times between 1977 and 1990 to adjust for changing household size, taking no account of ethnicity, with the result that all kebele residents received land. Due to immigration and population growth, land shortages became a growing problem during the Derg era and an important driver of tensions over land. Since the EPRDF took power in 1991, land has not been redistributed with the result that landlessness has become a major issue, especially for young adults. At present there are significant minorities of Amhara, Tigrayan, Kembata, and Wolayita in Turufe, with the Oromo making up 54 per cent of the population according to one survey.52 Turufe Kechema is part of Shashemene Zuria wereda and one of three villages in Turufe-Wetera-Elemo kebele. Ever since 1991, the OPDO has ruled Oromiya and Turufe-Wetera-Elemo. Under the Derg the peasant association leadership comprised several ethnic groups and the working language of government was Amharic. Since 1991 under ethnic federalism, the working language of government throughout Oromiya is Afaan Oromo and in Turufe all local government officials are Oromo. The original land conflict considered in this case study dates to 1991, when the Derg lost power and there was a period of instability around Turufe, exacerbated by local clashes between the EPRDF and the OLF.53 At this point of limited state authority, the Weyrera called on other Arssi Oromo to join them in attacking the Kembata, both in Turufe and neighbouring kebele, burning their houses.54 In total, 3,601 households across 10 kebele were expelled from their villages.55 In Turufe, 117 of 413 households at the time were Kembata,56 and although a few Kembata who had married Oromo were allowed to stay, most were evicted.57 The vast majority of those expelled were Kembata; however, some Wolayita were also forced to leave.58 The local Oromo population distributed the Kembata land amongst themselves. However, a few Tigrayan farmers also obtained land by paying Oromo who had seized land.59 A few evicted Kembata moved to Kembata zone, in SNNPR, though most settled on the outskirts of Shashemene. Respondents offered a number of explanations for these events. According to most Oromo respondents, the conflict originated in an incident in the 1980s in which Oromo thieves (shifta) were stealing crops from local people, both Kembata and Oromo, or in which Oromo farmers allowed their cattle to graze on the harvest of Kembata farmers.60 The Kembata reported the crime to the authorities, and the Oromo defendants were imprisoned and, subsequently, executed. According to some Oromo, the Kembata peasant association chairman at the time signed the execution papers.61 The former chairman himself denied any involvement, blaming the executions on the vicious nature of the Derg.62 According to the Weyrera, an entire clan is held responsible for a murder carried out by one individual from that clan. As such, they argued that the entire Kembata ‘clan’ was responsible for the execution.63 Furthermore, the Kembata reported the crop loss directly to the peasant association, bypassing customary dispute resolution since they believed that the elders were biased towards the Weyrera.64 Customary dispute resolution is a highly valued social institution in Oromiya, while the pursuit of justice through the courts is considered disrespectful of customary law. Some Kembata acknowledged that these actions probably deepened Oromo resentment.65 Kembata respondents, however, dismiss the executions as a pretext. They argue that the eviction was actually a response to growing land shortages as a result of population growth and the Weyrera belief that the land belonged to their clan: The Oromo made a pretext but they were saying that they needed the land for their children. They burned the houses of the Kembata and stole from them.66 They were saying that the land belongs to us, that you are alien to the area and that there is a shortage of land … The local [Oromo] families had land but the Oromo children had none. So they devised a plan to get the land for their children.67 Indeed, a significant minority of Oromo respondents supported these claims, highlighting the effect of migration on population growth and justifying the evictions in terms of indigenous land rights. These arguments clearly resonate with both neo-customary and ethnic federalist rationales, Migration has caused me a problem because the other people [other ethnic groups] have claimed the land as their own. The local community told them [the Kembata] that they had come for treatment [at the hospital] not to take the land of the local people.68 The Kembata were living on Oromo land. When the Derg left, the people were waiting for the opportunity to throw out the Kembata.69 The origins of the dispute pre-date federalism, and the eviction occurred during a period of instability and state weakness. Clearly then, the conflict cannot be solely attributed to ethnic federalism. Nevertheless, several respondents from different ethnic groups argued that ethno-nationalist ideas promoted by both the EPRDF and the OLF polarized the community, contributing to an expectation that ethnic groups should return to their own ethnic regions: Federalism exacerbated the problem. It divided the community. Those who did not speak the local language were expected to return to their own homeland.70 This government came up with ethnic federalism based on language and the government here started to work in Afaan Oromo. All the people were expected to go to their original land.71 When the government changed many people were expected to go back to their homeland. Some left voluntarily, some forcefully.72 Although the Kembata constituted the majority of those evicted in 1991, other minorities were also threatened, but were armed and able to defend themselves.73 During the hotly contested 2005 national elections, in which the EPRDF suffered significant electoral losses, Tigrayan residents were concerned that victory for the Oromo National Congress (ONC)—an Oromo opposition party—would have led to retribution from the local Oromo population against the Tigrayans who remain strongly associated with the TPLF/EPRDF.74 EPRDF victory ensured that this did not occur but they remain concerned about the future: The local people have a great sense of belonging to the land. They divided the Kembata land amongst themselves. I fear that if this government fails, their eyes are on our land and the same thing may happen again. When there is a transfer of power, if there is instability there may be a window of opportunity. Currently the government is in control so I do not think that this could happen.75 While many Oromo were reluctant to discuss the problems, a few were more candid. For example, one respondent acknowledged that the government prevents the Oromo from taking further action but the fall of the government could bring change: If we get the chance to remove them [the Tigrayan and other ethnic groups] we would not give them even one day in Turufe. They [the current ‘Tigrayan’ government] are killing our children… [Oromo] Children here mix Amharic and Afaan Oromo, they are not like the Wellega who speak pure Afaan Oromo. So we want to remove the other groups—it is beyond just land but it is to realize our right to language and culture. It is to avert such problems that we want to remove the others, to avert this cultural distortion.76 While the full territorial implications of ethnic federalism are not reflected in the land policy, therefore, this more extreme interpretation of ethnic federalism—the idea that individuals are first and foremost members of ethnic groups with limited rights outside their home region—is clearly present in land debates in Turufe and strongly resonates with the local citizenship implied by the ideal form of neo-customary tenure. Over the years since this original dispute, the Kembata have appealed for support to a variety of state agencies, several of which have made some attempt to resolve the conflict. In 1992 the Kembata registered the dispute with the wereda Agricultural Office, responsible for land administration.77 They have subsequently lodged appeals with the zone and regional governments, the Federal Ministry of Agriculture and Rural Development (MoARD), the Federal House of Representatives, the Prime Minister’s Office and the Federal Ombudsman’s Office. The Kembata made their appeals to government agencies, they say, because the courts under federalism are biased in favour of the Oromo majority.78 In their letters, they repeatedly called upon the government to intervene on their behalf as ‘citizens of Ethiopia’, framing their appeals in the universalistic language of the land policy.79 This is in stark contrast to the Oromo justification of the eviction in terms of indigeneity, which finds support in the more extreme interpretation of ethnic federalism and the ideal form of neo-customary tenure. Based on federal and regional land laws, which make no mention of ethnicity, the case is straightforward. The last Derg land redistribution is accepted as final and any ethnic minorities that received land at that time have the same rights as indigenous Oromo.80 Since the Kembata were allocated land under the Derg, their land should be returned to them. Indeed, according to the letters provided by the Kembata, no state agency has ever denied the legitimacy of their claim. However, various agencies have proposed solutions to the dispute that have not only failed to achieve a resolution but are also highly ambiguous regarding the land rights of non-indigenous minorities. The Prime Minister’s Office made the first concrete proposal in 1995/96, asking the wereda to find common land in the vicinity of Shashemene to allocate to the displaced households.81 It was, however, stipulated that this should not include the return of their original land. While this proposal failed to respect the displaced population’s rights to their previous land, it was broadly in line with the principles of the land policy, namely that the Kembata—an ethnic minority residing in Oromiya—had the right to replacement land in that region. Given the challenge of finding sufficient land in Shashemene wereda, the MoARD subsequently approached the regional governments of Oromiya and SNNPR to find a solution. The regional governments identified 3,958 hectares of forestland straddling the border between the two regions.82 While the Kembata accepted the proposal, in 1999/2000 the federal government ordered a stay of execution since the land in question had itself already been the focus of ‘recurrent ethnic conflict’ among the existing inhabitants.83 At this stage East Shewa zone asked Shashemene wereda to investigate the original dispute and to consider returning the farmers to their original land in Turufe.84 From the documentation available, this was the only occasion that a return to Turufe was even considered by the state as a possibility. Nevertheless, two years later a letter from Shashemene wereda to East Shewa zone merely reported that the amount of land previously held by the Kembata was unknown, and that all the disputed land was currently in use.85 The wereda’s failure to restore the Kembata’s land is perhaps not surprising. Ethnic federalism ‘tends to encourage, even require, political leaders to view themselves primarily, if not exclusively, as agents of their own ethnic communities’.86 The vast majority of government employees in Oromiya are Oromo, while many in the kebele administration in Turufe are from Weyrera clan. Indeed, it is likely that some current incumbents were directly involved in the eviction of the Kembata or have close ties to those that were. Furthermore, in light of the precarious nature of the OPDO’s relationship with the ethnic group that it is supposed to represent, it would be very difficult for the local administration to side with the Kembata over the Oromo majority. After several years of impasse, in 2005, the Minister for Agriculture and Rural Development wrote to the Kembata noting that following consultations with the evicted population and the regional governments of Oromiya and SNNPR, his final decision was that the Kembata would be given priority in either the Oromiya or SNNPR resettlement programmes, depending on their preference.87 The Kembata, unhappy with this offer, wrote to the Federal Ombudsman to appeal. Although the Ombudsman did investigate, at the time of fieldwork, he had not reached any conclusion. In the meantime, Oromiya began land registration in 2005, including in Turufe. Registration is organized by the wereda and the measurement of the plots is conducted by a land administration committee (LAC) elected from and by members of the kebele. When land is measured, all those with claims to land in a particular locality are called and, if there is any dispute over a plot, the LAC refers the matter to the courts. In Turufe, all five members of the LAC were Oromo and there were no disputes recorded regarding land previously held by the Kembata. The wereda administration—despite knowledge of ongoing investigations—issued land certificates to all occupants of the disputed land. In doing so, the wereda significantly strengthened the occupiers’ position, providing legal documents to support their claims and, in all likelihood, extinguishing any hopes of return for the Kembata. The proposals to resettle the Kembata in SNNPR and/or Oromiya and the decision to bring the SNNPR government into discussions represent something of a shift from the initial proposal to find land within Shashemene wereda. While that first proposal is broadly in line with the land policy, the inclusion of the SNNPR administration in dispute resolution and the possibility of relocating the Kembata to SNNPR are not. The SNNPR has no jurisdiction over the Kembata living in Oromiya and inter-regional resettlement is explicitly prohibited. While the rationale for relocating the Kembata to SNNPR is not stated, the most likely explanation would appear to be the ideas underpinning ethnic federalism, namely that the Kembata belong first and foremost to that region and having been displaced from Turufe, the logical step is for them to return to their ‘home’ region. While the land policy might be clear about the rights of non-indigenous ethnic minorities, the state’s response to the eviction of the Kembata is ambiguous, to say the least, and appears to draw on a more extreme interpretation of ethnic federalism, in addition to the land policy. The proposals to relocate the Kembata to SNNPR also serve to illustrate some of the inherent tensions within the federal project. SNNPR comprises about 45 ethnic groups that were combined into one region not based on common ethnic bonds, but because these groups were considered too small to be administratively viable.88 Consequently, relocating the Kembata to the region in which Kembata zone is located does not necessarily mean that they will join people with whom they share any bonds. Kembata zone has one of the highest population densities in the country, with the result that it would be impossible to find sufficient land there to resettle the displaced population, and so the proposals involved relocation to other parts of the region. The first proposal regarding forestland on the border of SNNPR and Oromiya involved land in Sidama zone, while the SNNPR resettlement programme focuses on Bench-Maji, faraway in the southwest of the region. Though part of the same linguistic family, the Kembata nevertheless have a distinct ethnic identity to the Sidama, while they have no linguistic or other ties with the Bench, the dominant ethnic group in Bench-Maji. Land, ethnicity and conflict in Ethiopia The case of Turufe illustrates some of the tensions between competing laws and ideas on land administration and the ambiguous state response to dispute resolution. However, Turufe is only one case and—it might be argued—a rather exceptional one. While land conflict is common in rural Ethiopia and there are numerous cases in which conflicts take on an ethnic dimension,89 widespread ethnic cleansing—as might be implied by ethnic federalism—has not taken place. That said, however, Turufe is far from exceptional. Indeed, there are several examples of comparable situations in which non-indigenous minorities have been relocated to their home region or where they have been threatened with such action. As such, the tensions between these competing laws and ideas on land rights have broader relevance, while state responses to infringements of minority land rights are far from uniform. A comparable case is provided by Tesfaye Tafesse’s study of the displacement of 12,000 Amhara from East Wellega, Oromiya in 2000.90 Amhara had migrated to the area over several decades and, while relations between Amhara settlers and the indigenous Oromo were good for many years, growing land shortages and ethnic federalism resulted in a deterioration. Under federalism, Afaan Oromo was made the language of local government and previously ethnically-diverse local administrations were increasingly dominated by Oromo. Tafesse argues that Oromo local government officials incited violence against the migrants, in line with the territorial implications of federalism, in stating that ‘[t]his land belongs to Oromos and only Oromos and not to you, Amharas’.91 Accusations that the Amhara had been involved in cattle theft, illegal deforestation and a murder led in June 2000 to orders from wereda and kebele officials for the Amhara settlers to be expelled and to return to Amhara. While many did leave, the violent intervention of local government militias at the end of the year is thought to have led to the deaths of several hundred Amhara and further large-scale displacement. The majority fled to Amhara and were subsequently included in the Amhara resettlement programme that focused on the western Amhara lowlands. Both the displaced population and the government rejected any possibility of returning to Wellega. As in Turufe, the territorial implications of ethnic federalism provided a rationale for the displacement of an ethnic minority in response to growing resource shortages. Meanwhile, the state failed to enforce the universalist principles of state ownership. In line with the rationale of ethnic federalism, state agencies promoted the eviction of non-indigenous minorities, facilitating their return to their ‘home’ region. Similar questions regarding the land rights of ethnic minorities are raised by the case of the Guji Oromo pastoralists living in the Nech Sar national park in Gamo Gofa zone, SNNPR, as described by Tadesse Berisso.92 Despite the park being located within SNNPR, Guji Oromo have lived in the area for generations. Nonetheless, when park authorities decided to remove all inhabitants to ‘protect’ the park and promote tourism, repeated attempts were made to relocate the Guji Oromo to Oromiya. Once again this decision is in line with the more extreme interpretation of ethnic federalism and in direct contradiction of the universalist principles of state ownership. A common perception remains that members of particular ethnic groups ultimately belong in the home region of that ethnic group and, therefore, when there are disputes over land, one of the main options available is to return the ethnic minority to their home region. The legal and ideational tension between federalism and state ownership exists throughout the country, while varying forms of neo-customary tenure provide an alternate rationale in most regions. However, the ways in which these institutions and ideas combine to shape policy implementation in different parts of the country are contingent on the government’s political economic priorities and ever-evolving state-society and intra-state relations. In the case of Turufe and that of East Wellega, the status of the Oromo as the largest ethnic group in the country and the limited legitimacy of the OPDO within the region appears to have influenced decision making, with local government favouring or at least turning a blind eye to the dispossession of ethnic minorities for fear of alienating the local population. In contrast, recent government promotion of large-scale agricultural investments provides a clear counter-example in which land access for outsiders has been prioritized over that of politically marginal indigenous populations. In particular, the federal government has re-centralized authority from regional administrations to lease land to agricultural investors in lowland areas of Benishangul-Gumuz, Gambella and South Omo.93 In the process, pastoralists and shifting cultivators have found their land access curtailed in direct contradiction of the principles of ethnic federalism. While the EPRDF promoted and designed the federal system, its support for federalism has long been ambiguous and the government has resisted its full implementation. Instead, the EPRDF has sought out the political middle ground between the centrists and regionalists.94 In particular, the EPRDF has employed federalism in an attempt to neutralize the ethno-national sentiments of the latter, while limiting the implementation of the more extreme and highly problematic implications of ethnic federalism, including with respect to land administration. Certainly the EPRDF has no interest in further stoking ethno-nationalism or permitting ethnic cleansing. Far from resolving the ‘national question’, however, federalism has, arguably, led to growing ethno-nationalist sentiments in many parts of the country. For Abbink, instead of solving conflicts, federalism has merely ‘decentralized’ them, pushing them down to lower levels of government administration.95 While this may be true, by-and-large, the importance of ethnicity to national-level politics has been re-asserted during unprecedented anti-government protests, first in 2014 and then from late 2015 to the present, and large-scale conflict along ethnic lines in the Oromiya–Somali border region in 2017. This unrest culminated in a state of emergency from October 2016 to August 2017, and once again beginning in February 2018. The original flashpoint was the Addis Ababa Masterplan that sought to integrate Addis Ababa with the surrounding area in Oromiya as part of an integrated urban development plan. Many Oromo claim Addis Ababa (or Finfinnee in Afaan Oromo) as the capital city of Oromiya. Consequently, its status as a separate city administration is a source of considerable resentment, while the proposed city expansion is seen as a further attempt to expropriate Oromo territory. Subsequently protests spread to Amhara, partly based on solidarity with Oromo protestors but also concerning the contested incorporation of Wolkayit wereda, previously part of the Imperial Province of Gondar, into Tigray region. The government has tended to attribute the protests to high rates of unemployment and rural landlessness. While there is some truth in this, the protests generated such unprecedented support as a result of a confluence of a wider range of issues, including fragmentation within the ruling party, frustration at the absence of democracy, the perceived domination of the TPLF within the EPRDF and the lack of popular legitimacy and corruption in the EPRDF.96 Clearly, given the focus of protests on the Addis Ababa Masterplan and the status of Wolkayit wereda, the territorial implications of ethnic federalism are also among the central grievances. Indeed, as was the case in Turufe in the early 1990s, there is some evidence that in the temporary absence of state authority created by the protests, some Oromo protestors have attacked non-indigenous Amhara and Tigrayan residents in parts of Oromiya, seizing their land and other property in the process.97 While the relative strength of the EPRDF has to date limited any possibility of the widespread pursuit of the full territorial implications of ethnic federalism, these ideas are widely debated and supported by at least some in Ethiopia. Rather than resolving the national question, ethnic federalism has merely changed its form. Finding a way to balance growing ethno-nationalism with the protection of minority rights remains one of the central challenges facing the government, notably with respect to what are, at the time of writing, fast moving events in Ethiopian politics. Conclusion There is a common perception in much of the literature that Ethiopia’s 1975 land reform eradicated previous diverse tenure systems, resulting in a relatively uniform and unambiguous system of state land ownership. This article has shown that the reality is not quite so clear-cut. While state land ownership remains the dominant influence on land tenure, the creation of ethnic federalism has direct implications for land administration, in particular with respect to the rights of non-indigenous ethnic minorities. Furthermore, neo-customary tenure systems have retained some influence in many parts of the country. Indeed, the government has sought to reinvigorate and co-opt customary tenure as a means of pursuing its own policy objectives in recent years. As such, land tenure in Ethiopia is more ambiguous than is often assumed. This article has explored the overlap and divergence between these three laws and the broader ideas that have been used to justify them, specifically with respect to the land rights of non-indigenous minorities. The case study of land conflict in Turufe demonstrates that these three bodies of law and associated ideas remain of the utmost relevance to contemporary land debates, and that parties to land disputes draw on these ideas to justify their particular resource claims. Meanwhile, the state response to conflict has been highly ambiguous, failing to defend the land rights of non-indigenous minorities in line with state ownership, and instead presenting proposals that align, in part, with the principles of ethnic federalism. While competing institutions and ideas on land tenure result in considerable ambiguity, the outcome of tensions between these competing rationales is not pre-determined but varies with evolving political dynamics—including both local state-society and intra-state relations. Although this article has provided examples in which non-indigenous inhabitants have been displaced, in line with more extreme interpretations of ethnic federalism, such events are, fortunately, relatively rare. Equally, however, there are other examples in which indigenous rights have been undermined by the state’s prioritization of outsiders, most clearly through the state promotion of large-scale agricultural investments in the west of the country. The broad contours of the Ethiopian debate regarding the land rights of non-indigenous minorities and the political implications of these issues mirror those taking place across much of the rest of Africa.98 As elsewhere, ambiguity in land tenure in Ethiopia results in considerable uncertainty regarding the land rights of non-indigenous minorities. This, in turn, raises major questions regarding citizenship and belonging—whether individuals are first and foremost national citizens or members of ethnic groups. While the broad contours of this debate mirror those taking place elsewhere, the form that this takes is specific to the Ethiopian case. Notably, the predominance of state land ownership sets Ethiopia apart from much of the rest of the continent, where neo-customary tenure is prevalent. Furthermore, Ethiopia is rather unusual regarding the system of ethnic federalism, which is an additional source of ambiguity regarding minority rights, introducing further uncertainty regarding the jurisdiction of federal and regional governments. Footnotes 1. Liz Alden Wily, ‘‘The law is to blame’: The vulnerable status of common property rights in Sub-Saharan Africa’, Development and Change 42, 3 (2011), pp. 733–57; Pauline Peters, ‘Conflicts over land and threats to customary tenure in Africa’, African Affairs 112, 449 (2013), pp. 543–62. 2. Gebru Mersha and Mwangi Wa Githinji, ‘Untying the Gordian Knot: The question of land reform in Ethiopia’ (Policy Paper 9, International Institute for Social Studies/United Nations Development Programme, The Hague, 2005), p. 8. 3. Mersha and Githinji, ‘Untying the Gordian Knot’, p. 8. 4. Wibke Crewett, Ayalneh Bogale, and Benedikt Korf, ‘Land tenure in Ethiopia: Continuity and change, shifting rulers, and the quest for state control’ (Working Paper 91, CGIAR Systemwide Program on Collective Action and Property Rights (CAPRi), Washington DC, 2008), p. 13. 5. Berhanu Adenew and Fayera Abdi, ‘Research report 3: Land registration in Amhara region, Ethiopia’ (IIED, London, 2005), p. 6. 6. Allan Hoben, ‘Ethiopian land tenure revisited: Continuity, change and contradictions’ (Working Paper 236, African Studies Center, Boston University, 2001); Crewett et al., ‘Land tenure in Ethiopia’; Bereket Kebede, ‘Land reform, distribution of land and institutions in rural Ethiopia: Analysis of inequality with dirty data’, Journal of African Economies 17, 4 (2008), pp. 550–77; Tom Lavers, ‘Land registration and gender equality in Ethiopia: How state-society relations influence the enforcement of institutional change’, Journal of Agrarian Change 17, 1 (2017), pp. 188–207. 7. Catherine Boone, ‘Property and constitutional order: Land tenure reform and the future of the African state’, African Affairs 106, 425 (2007), pp. 557–86; Catherine Boone, Property and political order in Africa: Land rights and the structure of conflict (Cambridge University Press, Cambridge, 2014). 8. Mahmood Mamdani, Citizen and subject: Contemporary Africa and the legacy of late colonialism (Princeton University Press, Princeton, NJ, 1996); Helène Maria Kyed and Lars Buur, ‘New sites of citizenship: Recognition of traditional authority and group-based citizenship in Mozambique’, Journal of Southern African Studies 32, 3 (2006), pp. 563–81; Boone, Property and political order in Africa. 9. Notable contributions include Alemante Selassie, ‘Ethnic federalism: Its promise and pitfalls for Africa’, Yale Journal of International Law 28 (2003), pp. 51–107; Sarah Vaughan, ‘Ethnicity and power in Ethiopia’ (University of Edinburgh, unpublished PhD thesis, 2003); David Turton, Ethnic federalism: The Ethiopian experience in comparative perspective (James Currey, London, 2006). 10. Asnake Kefale, ‘Federal restructuring in Ethiopia: Renegotiating identity and borders along the Oromo–Somali ethnic frontiers’, Development and Change 41, 4 (2010), pp. 615–35; Fekadu Adugna, ‘Overlapping nationalist projects and contested spaces: The Oromo–Somali borderlands in southern Ethiopia’, Journal of Eastern African Studies 5, 4 (2011), pp. 773–87. 11. Endrias Eshete, ‘Ethnic federalism: New frontiers in Ethiopian politics’ (First National Conference on Federalism, Conflict and Peace Building, Addis Ababa, 2003); Selassie, ‘Ethnic federalism’; Christophe Van der Beken ‘Federalism, local government and minority protection in Ethiopia: Opportunities and challenges’, Journal of African Law 59, 1 (2015), pp. 150–77. 12. Boone, Property and political order in Africa. 13. Ibid., p. 68. 14. Colin Hay, ‘Ideas and the construction of interests’, in Daniel Béland and Robert Henry Cox (eds), Ideas and politics in social science research (Oxford University Press, Oxford, 2011), pp. 65–82, 69. 15. Daniel Béland, ‘Mapping changing social policy ideas: A global, actor-centred approach’ (UNRISD workshop on New Direction in Social Policy, 7–8 April 2014, Geneva, 2014). 16. Christopher Clapham, Transformation and continuity in revolutionary Ethiopia (Cambridge University Press, Cambridge, 1988). 17. PDRE, ‘A proclamation to provide for the public ownership of rural land’, Federal Negarit Gazeta Proclamation 31/1975 (1975), para. 4.1. Emphasis added. 18. Dessalegn Rahmato, The peasant and the state: Studies in agrarian change in Ethiopia 1950s—2000s (Addis Ababa University Press, Addis Ababa, 2009). 19. Meles Zenawi, cited in Harold Marcus, A breakfast meeting with Prime Minister Meles, 1995, <http://www.hartford-hwp.com/archives/33/008.html> (15 July 2009). 20. FDRE, ‘Constitution of the Federal Democratic Republic of Ethiopia (FDRE)’, Federal Negarit Gazeta, (1994), para. 40. Emphasis added. 21. FDRE, ‘Federal Democratic Republic of Ethiopia (FDRE) – Rural land administration and land use proclamation’, Federal Negarit Gazeta, Proclamation 456/2005 (2005), para. 5.1.b. Emphasis added. 22. Klaus Deininger and Songqin Jin, ‘Tenure security and land-related investment: Evidence from Ethiopia’, Proceedings of the first international conference on the Ethiopian economy II (Ethiopian Economics Association, Addis Ababa, 2004), pp. 19–50. 23. Interview, respondent FG4, head, Federal Land Use Administration, MoARD, Addis Ababa, 17 November 2009. 24. In most parts of the country, during the latter days of the Derg. 25. Eshete, ‘Ethnic federalism’; Merera Gudina, ‘Contradictory interpretations of Ethiopian history: The need for a new consensus’, in Turton, Ethnic federalism, pp. 119–31. 26. Christopher Clapham ‘Ethiopian development: The politics of emulation’, Commonwealth and Comparative Politics 44 (2006), pp. 137–50, p. 148. 27. Vaughan, ‘Ethnicity and power in Ethiopia’. 28. Van der Beken, ‘Federalism, local government and minority protection in Ethiopia’. 29. FDRE, ‘Constitution of the Federal Democratic Republic of Ethiopia (FDRE)’, para. 39.1. 30. Eshete, ‘Ethnic federalism’, p. 21. 31. Crewett et al., ‘Land tenure in Ethiopia’, highlight some limited variation between regional policies. 32. ONRG, ‘Oromia National Regional Government (ONRG) – Proclamation to amend the proclamation nos. 56/2002, 70/2003, 103/2005 of Oromia Land Use and Administration’, Megeleta Oromia Proclamation 130/2007 (2007), para. 5.1. 33. Interview, respondent FG4. 34. MoFED, ‘Rural development policy and strategies’ (Ministry of Finance and Economic Development [MoFED], Addis Ababa, 2003), p. 41. 35. Such views were raised in a number of informal discussions in cafes and on long bus journeys in Oromiya and Tigray, and by a number of Oromo university students. In a few cases, this included people involved in extended state structures, for example, those holding unpaid positions in the many local committees that play an important role in implementation, albeit not paid officials. 36. See Boone, Property and political order in Africa, p. 38. 37. Crewett et al., ‘Land tenure in Ethiopia’; Tom Lavers, ‘Agricultural investment in Ethiopia: Undermining national sovereignty or tool for state building?’ Development and Change 47, 5 (2016), pp. 1078–101. 38. Interview, respondents SR1, land administration experts, SNNPR land administration, Awassa, 23 February 2010. 39. Mamo Hebo, Land, local custom and state policies: Land tenure, land disputes and disputes settlement among the Arsii Oromo of southern Ethiopia (Shoukadoh, Kyoto, 2006). 40. Evidently, neo-customary tenure also has important gender implications see Lavers, ‘Land registration and gender equality in Ethiopia’. 41. Interview, respondent OR2, deputy head, Oromiya Bureau of Culture and Tourism, Addis Ababa, 16 February 2010. This co-optation of customary institutions as a means of enhancing authority is a common government strategy, see Tobias Hagmann, ‘Beyond clannishness and colonialism: Understanding political disorder in Ethiopia’s Somali region, 1991–2004’, Journal of Modern African Studies 43, 4 (2005), pp. 509–36; Marco Bassi, ‘The politics of space in Borana Oromo, Ethiopia: Demographics, elections, identity and customary institutions’, Journal of Eastern African Studies 4, 2 (2010), pp. 221–46. 42. The structure of government administration (from highest to lowest level) is: federal, regional, zone, wereda, kebele. 43. To the point where the Aba Gadaa’s office is situated in the local government compound and the Aba Gadaa required a letter of permission from the Bureau of Culture and Tourism to participate in the research. 44. Interview, respondent TM8, male Oromo elder, Turufe Kechema, 26 January 2010. 45. Interview, respondents OR2; TK2, male Oromo elder and kebele vice-chairman, Turufe Kechema, 13 January 2010; TF5, female Oromo resident and head of kebele women’s affairs, Turufe Kechema, 27 January 2010. 46. Hebo, Land, local custom and state policies; Abbas Gnamo, Conquest and resistance in the Ethiopian Empire, 1880−1974: The case of the Arsi Oromo (Brill, Leiden, 2014). 47. Joel Migdal, State in society: Studying how states and societies transform and constitute one another (Cambridge University Press, Cambridge, 2001); Tobias Hagmann and Didier Péclard (eds), Negotiating statehood: Dynamics of power and domination in Africa (Wiley-Blackwell, Chichester, 2011). 48. Since the mid-2000s, the government has promoted capital-intensive agricultural investments, particularly to the north of Turufe, but also in the area around Turufe itself. While these investments and urban expansion constitute important additional pressures on land in the area, these have only had a more recent impact and were not influential in the events discussed below. 49. Getachew Fule and Mesfin Tadesse, Ethiopian village studies: Turufe Kechema (Addis Ababa University & University of Oxford, Addis Ababa & Oxford, 1996). 50. Temesgen Gebeyehu, ‘A history of land measurement in Shashemene (Ethiopia), 1941–1974’, African Journal of History and Culture 1, 4 (2009): pp. 67–75. 51. Interview, respondent TM26, male Kembata displaced from Turufe, Shashemene, 29 January 2010. 52. CSAE, Ethiopian rural household survey 2004 (Centre for the Study of African Economies [CSAE], Oxford University, Oxford, 2004). 53. Interview, respondent TM1, male Oromo resident, Turufe Kechema, 31 January 2010. The OLF was briefly a member of the post-Derg transitional government, but left before the 1992 elections and was banned as a terrorist organization. 54. Interview, respondent TM17, male Kembata displaced from Turufe, Shashemene, 29 January 2010. 55. Record 3a, Letter from the Oromiya Bureau of Agriculture and Rural Development to the Oromiya Rural Land Administration Authority, date illegible. 56. Fule and Tadesse, Ethiopian village studies. 57. Interview, respondents TM3, male Oromo resident, Turufe Kechema, 28 January 2010; TM12, male Wolayita resident, Turufe Kechema, 14 January 2010; TM18, male Kembata resident, Turufe Kechema, 16 January 2010. 58. Interview, respondents TM23, male Tigrayan resident, Turufe Kechema, 17 January 2010; TM26. 59. Interview, respondents TM21; TM21, male Tigrayan resident, Turufe Kechema, 17 January 2010. 60. Interview, respondents TM12; TK5, male member of kebele Land Administration Committee, Turufe Kechema, 14 January 2010. 61. Interview, respondent TK5. 62. Interview, respondent TM26. 63. Interview, respondents TK5; TM4, male Oromo resident, Turufe Kechema, 28 January 2010. 64. Interview, respondent TM18. 65. Interview, respondents TM12, TM18. 66. Interview, respondent TM18, a Kembata who remained in Turufe. 67. Interview, respondent TM17, a Kembata displaced to Shashemene. 68. Interview, respondent TM4, an Oromo. 69. Interview, respondent TK5, an Oromo. 70. Interview, respondent TM26, a Kembata displaced to Shashemene. 71. Interview, respondent TM1, an Oromo. 72. Interview, respondent TM23, a Tigrayan. 73. Interview, respondents TM17; TM23. 74. Yohannes Gezahegn, Bizuayehu Ayele, Getachew Fule, and Mesfin Tadesse, Ethiopian village studies II: Turufe Kechema (WeD-Ethiopia, Addis Ababa, 2006). 75. Interview, respondent TM20, male Tigrayan resident, Turufe Kechema, 17 January 2010. 76. Interview, respondent TM1, an Oromo. 77. Interview, respondent TM17. 78. Interview, respondents TM17; TM18; TM24, male Kembata resident, Turufe Kechema, 19 January 2010. 79. Record 1e, Letter from the evicted farmers to Walta Information Centre, 14 October 2002; Record 1 f, Letter from the evicted farmers to the House of Peoples’ Representatives, 10 October 2002. 80. Interview, respondent OR1, senior expert, Oromiya Land Use and Environmental Protection Bureau, Addis Ababa, 17 November 2009; ONRG, ‘Proclamation to amend the proclamation nos. 56/2002, 70/2003, 103/2005’. 81. Record 1e. 82. Record 2a, Letter from the federal Ministry of Agriculture and Rural Development to the association of evicted farmers, 5 November 2005; Record 4c, Letter from the federal Ombudsman to the Oromiya and SNNP regional governments, 23 August 2007. 83. Record 2a. 84. Record 7a, Letter from the East Shewa zone administration to the Shashemene wereda administration, 9 March 2000. 85. Record 6a, Letter from Shashemene wereda administration to the East Shewa zone administration, 29 January 2002. 86. Selassie, ‘Ethnic federalism’, p. 92. 87. Record 2a. 88. Vaughan, ‘Ethnicity and power in Ethiopia’. 89. Abbink, ‘Ethnicity and conflict generation in Ethiopia’. 90. Tesfaye Tafesse, The migration, environment and conflict nexus in Ethiopia: A case study of Amhara migrant-settlers in East Wollega Zone (OSSREA, Addis Ababa, 2007). 91. Ibid., p. 86. 92. Tadesse Berisso, ‘Planning resettlement in Ethiopia: The experience of the Guji Oromo and the Nech Sar National Park’, in Alula Pankhurst and François Piguet (eds), Moving people in Ethiopia: Development, displacement and the state (James Currey, London, 2009), pp. 93–101. 93. Lavers, ‘Agricultural investment in Ethiopia’. 94. Vaughan, ‘Ethnicity and power in Ethiopia’. 95. Jon Abbink, ‘Ethnicity and conflict generation in Ethiopia: Some problems and prospects of ethno-regional federalism’, Journal of Contemporary African Studies 24, 3 (2006), pp. 389–413, p. 390. This is in line with Boone’s hypothesis that the scale of land conflicts is shaped by institutional settings, with land conflicts ‘bottled up’ at local levels under more localized land tenure regimes. See Boone, Property and political order in Africa. 96. Africa Confidential, ‘Oromia erupts’, Africa Confidential, 18 March 2016, <https://www.africa-confidential.com/article-preview/id/11594/Oromia_erupts> (4 March 2018); René Lefort ‘Ethiopia’s crisis’, openDemocracy, 19 November 2016, <https://www.opendemocracy.net/ren-lefort/ethiopia-s-crisis> (4 March 2018). 97. EHRCO, ‘140th Special report: Executive summary’ (Ethiopian Human Rights Council, Addis Ababa, 2016), p. 5; René Lefort, ‘The ‘Ethiopian Spring’: ‘Killing is not an answer to our grievances’’, openDemocracy, 9 September 2016, https://www.opendemocracy.net/ren-lefort/ethiopian-spring-killing-is-not-answer-to-our-grievances (12 September 2016). Given that the area around Turufe Kechema was among the main flashpoints of the protests in 2016 leading to a temporary breakdown in state control, this raises serious concerns about the situation in Turufe and the fate of the ethnic minorities there. Unfortunately, it has not yet been possible to conduct additional fieldwork. 98. Boone, ‘Property and constitutional order’; Boone, Property and political order in Africa; Kyed and Buur, ‘New sites of citizenship’. © The Author(s) 2018. Published by Oxford University Press on behalf of Royal African Society This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted reuse, distribution, and reproduction in any medium, provided the original work is properly cited http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png African Affairs Oxford University Press

Responding to land-based conflict in Ethiopia: The land rights of ethnic minorities under federalism

African Affairs , Volume Advance Article (468) – Mar 29, 2018

Loading next page...
 
/lp/ou_press/responding-to-land-based-conflict-in-ethiopia-the-land-rights-of-qU0m6bXDV2
Publisher
Oxford University Press
Copyright
© The Author(s) 2018. Published by Oxford University Press on behalf of Royal African Society
ISSN
0001-9909
eISSN
1468-2621
D.O.I.
10.1093/afraf/ady010
Publisher site
See Article on Publisher Site

Abstract

Abstract There is a common perception that Ethiopia is unusual in Africa in having a relatively uniform system of state land ownership. While highly influential, state ownership is not the only body of law with implications for land administration. This article argues that the institutionalization of ethnic federalism and the persistence of neo-customary tenure result in considerable ambiguity, particularly regarding the land rights of non-indigenous minorities. The analysis highlights tensions between these three sets of land tenure institutions—state ownership, ethnic federalism and neo-customary tenure—and their implications for minority land rights. A case study of land-based conflict in Oromiya region, based on fieldwork conducted in 2009 and 2010, demonstrates the continuing relevance of these land tenure institutions and associated ideas in land debates in Ethiopia, both in terms of the use of these ideas by protagonists as means of justifying land claims, and the ambiguous state response to the conflict, which goes well beyond the provisions of the land policy. As such, while there are certainly particular characteristics of the Ethiopian case, many of the key issues regarding ethnicity and land mirror debates taking place across the continent. Research has repeatedly highlighted the ‘multiple and overlapping’ nature of land tenure in sub-Saharan Africa. This institutional plurality results from the claims of colonial and post-colonial authorities to ultimate ownership of all land, while states manipulate or reinforce neo-customary tenure as a mechanism of social control.1 In contrast, the land question in Ethiopia is frequently considered to be ‘rather unique’ in sub-Saharan Africa, with a uniform policy of state land ownership with usufruct rights distributed to farmers.2 Not only was Ethiopia’s brush with European colonialism relatively brief, but the 1975 land reform that nationalized all land was ‘particularly extensive’3 and ‘effectively abolished’4 or ‘wiped out’5 previous tenure systems. While the significance of the 1975 land reform is indisputable, there nonetheless remains ambiguity in land tenure regarding the authority of state and non-state actors and the implications of different land tenure institutions for citizenship. This ambiguity is partly due to the continuing influence of neo-customary tenure regimes, which in many places were not actually wiped out by state ownership, and, arguably, have become increasingly influential in recent years.6 However, ambiguity also results from the territorial implications of ethnic federalism introduced in the 1990s. While ethnic federalist principles resonate in certain respects with neo-customary tenure by prioritizing the land rights of ‘indigenous’ inhabitants, they are in direct tension with state ownership regarding the rights of ethnic minorities outside their ‘home’ region. This article focuses specifically on this question of the land rights of non-indigenous minorities. The analysis uses a case of land conflict in Oromiya to illustrate the continuing relevance of these competing perspectives on land tenure and to highlight how the state’s attempts to resolve the dispute reflect this ambiguity regarding minority land rights. In doing so, the article links to important recent debates on land tenure in Africa. Land tenure regimes structure intra-community relations and relations between state and society.7 Notably, land tenure regimes have direct implications for the political salience of ethnicity, the form that resource conflict takes, and citizenship rights.8 Ethiopian land tenure is certainly an extreme case on the African continent. In the majority of sub-Saharan Africa, the dominant form of land tenure remains neo-customary, whereas in Ethiopia state ownership predominates. Furthermore, Ethiopia’s ethnic federal system is unusual and provides an alternate rationale in land debates absent in most other countries. That said, while the relative importance of different land tenure systems varies between countries, the tensions between these systems raise closely related issues. Consequently, in Ethiopia, as in much of Africa, an important thread of contemporary land debates is the question of whether land should be made available to all citizens—who have a right to live anywhere within the country—or whether the land rights of ‘indigenous’ populations should be prioritized. The contemporary relevance of these issues in Ethiopia has been vividly highlighted by the wave of protests that have swept across Oromiya and then Amhara since 2014. These protests have resulted from a confluence of causes, including youth unemployment, inadequate representation of individual and group interests, government corruption, fragmentation within the ruling party since 2012 and growing ethno-nationalism. However, an important flashpoint for the protests and a significant grievance concerned questions of land and territory under ethnic federalism, both in terms of the delineation of ethno-regional borders and the relative prioritization of land access for the indigenous population and outside investors. There is already a large literature on Ethiopia’s ethnic federal project.9 Research specifically on the links between land and federalism has tended to focus on the challenge of demarcating borders between ethnic regions, leading to territorial competition along ethnic lines.10 In contrast, there has been little attempt to examine the implications of federalism for non-indigenous ethnic minority land rights. This is, perhaps, surprising given the clear implications of federalism for territory and citizenship. Both Endrias Eshete and Alemante Selassie note the potential weakness of minority land rights outside their home regions in their broad discussions of federalism, while Christophe Van der Beken highlights the limited rights of non-indigenous minorities in general.11 None of these valuable contributions, however, explores the inconsistencies regarding land rights between federalism and state ownership of land in policy and practice. The case study presented in the article is based on fieldwork conducted in 2009 and 2010 involving 53 key informant interviews with residents of the village in question, former residents who had been displaced, and government officials. Respondents within the community were purposively selected to provide variation in ethnicity and gender, as well as to identify people who were old enough to have directly experienced the events. Key informant interviews were also conducted with the leaders of the association formed to represent the displaced population and who were therefore in a position to provide insights into the subsequent attempts at dispute resolution. Representatives of the displaced ethnic minority also provided extensive documentation relating to the appeals that they made to government officials and the government’s attempts to resolve the dispute. This written documentation constitutes a vital source of data, providing information from the time about these events, and enabling triangulation of interview testimony. The article proceeds as follows. The first section conceptualizes the linkages between land tenure institutions and ideas before outlining the contrasting implications of three sets of ideas and laws for the land rights of ethnic minorities, notions of citizenship and territorial authority. Section two then examines the case study of a land conflict in Oromiya region and government attempts at dispute resolution over the subsequent twenty years. This case study demonstrates the continuing relevance of these three ideas and laws in land debates in Ethiopia, the use of these ideas by protagonists as means of justifying land claims, and the ambiguous state response to the conflict, which appears to go well beyond the provisions of the land policy. Section three reflects on the implications of this case for broader debates about land, ethnicity, and conflict in Ethiopia, particularly in relation to recent protests and unrest. The fourth and final section summarizes the arguments and concludes with a discussion of the parallels with debates on land tenure in other parts of sub-Saharan Africa. Contrasting ideas and laws on land and territory in Ethiopia Land tenure regimes are not just means of distributing resources but key mechanisms by which states relate to their societies and project authority into rural areas. In addition to a system of property rights, therefore, as Catherine Boone argues, land tenure regimes comprise three main elements.12 First, a locus of political authority—usually the state or neo-customary authorities—with responsibility for the adjudication and enforcement of land rights. Second, a territorial jurisdiction that is subject to the land tenure regime. This jurisdiction may be restricted to a particular locality or span the entire national territory. Third, land tenure regimes imply different forms of citizenship, defined as ‘the boundaries of membership in a political community’.13 Neo-customary tenure regimes frequently imply local citizenship based on membership of a particular descent group and confer economic and political rights, including entitlements to land. In contrast, state ownership establishes a direct relationship between land users and the central state and implies national citizenship. For discursive institutionalists, institutions, including those related to land tenure, ‘are built on ideational foundations’.14 Institutions are not the simple product of the preferences of competing interest groups but draw on ideas for inspiration. However, ideas are rarely adopted tout court as institutions. Instead, ideas are likely to be adapted through processes of bricolage—combining one idea with others—or translation—the adaptation of an idea when taken from one context to another.15 As such, ideas that help shape land tenure institutions at one point in time may continue to have a life of their own within public and political debate, independent from the institutions that they helped to shape. As such, these ideas may continue to be influential in policy debates or in the process of implementation. This section examines three sets of influential ideas and associated institutions with direct implications for land tenure regimes, highlighting their similarities and contradictions. The first two have been promoted directly by the current government: state land ownership and ethnic federalism. The third is the neo-customary land tenure regime in Arssi, Oromiya, which, though originating in Arssi Oromo society, has been promoted and adapted by the state in recent years. The first, and most influential, set of land tenure institutions and associated ideas relate to the system of state land ownership. The Derg regime (1974–91) nationalized all rural land in 1975, distributing usufruct rights to smallholders. The Derg undoubtedly had political motivations, namely the destruction of the landholding class that was the base of Imperial power.16 However, land reform was justified in egalitarian and social justice terms encapsulated in the rallying cry of ‘land to the tiller’. The new tenure system was universalistic—providing land access for all Ethiopian citizens—and class-based stating that, ‘any person who is willing to personally cultivate land shall be allotted rural land sufficient for his maintenance and that of his family’.17 The Tigrayan People’s Liberation Front (TPLF) began its insurrection against the Derg in the mid-1970s. When the TPLF’s ambitions grew from liberating Tigray to seizing national power, the TPLF formed the Ethiopian Peoples’ Revolutionary Democratic Front (EPRDF), a coalition of ethno-nationalist parties to represent Ethiopia’s different ethnic groups, of which the TPLF was the leading member. On coming to power in 1991, the EPRDF largely maintained the Derg’s land policy. Again, there were important political motivations for doing so, namely to consolidate the support of the peasantry, to prevent the emergence of a landholding class, and to limit urban migration that might threaten social and political instability.18 However, there was a subtle shift in the discourse used to justify state land ownership. In addition to a shared emphasis on social equity and the prevention of class differentiation, former Prime Minister Meles Zenawi argued that state land ownership protected smallholders against displacement by market forces, providing a form of ‘social security’.19 Notwithstanding, Ethiopian citizenship and class remained key principles underpinning the land policy. As stated in the 1994 constitution: ‘Ethiopian peasants have right to obtain land without payment’.20 A similar sentiment is re-iterated in the 2005 federal land proclamation, ‘Any citizen of the country who is 18 years of age or above and wants to engage in agriculture for a living shall have the right to use rural land’.21 This is not to say that EPRDF land policy has remained a straightforward statement of Marxist principles. Rapid population growth has meant diminishing landholdings, while critics argued that agricultural productivity was impeded by tenure insecurity.22 The government’s response was to conduct land registration to enhance tenure security and investment incentives and prevent the sub-division of landholdings into unviable plots.23 In doing so, land policy has moved a step away from past universalistic principles; there are now very large numbers of landless people in rural areas who reached adulthood after the last land redistribution and have little hope of accessing land.24 Nonetheless, the idea of universal land access remains a key part of the debate. A second, distinct set of institutions and ideas with direct implications for land tenure relates to the system of ethnic federalism. The rights of ethnic groups within Ethiopia’s political system have been a polarizing feature of political debate since the late 1960s, with opposing camps drawing on very different interpretations of Ethiopian history.25 On one hand, centrists emphasize Ethiopia’s historical unity and fear its break up. On the other hand, regionalists use narratives of internal colonization and cultural hegemony by the Amhara and Tigrayan ethnic groups as the basis of claims for greater ethnic autonomy, even secession. While the Derg dismissed ethnicity as a distraction from class consciousness, for ethno-nationalist movements including the TPLF and the Oromo Liberation Front (OLF), ethno-nationalist struggle was seen as paving the way for class struggle. On taking national power in 1991, the EPRDF established an ethnic federal system that, in Christopher Clapham’s words, constituted a ‘wholesale takeover of Stalin’s theory of the national question as an approach to the problem of Ethiopia’s ethnic diversity’.26 Ethiopia’s ethnic federalism is based on the primordialist idea that ethnic groups—or nations, nationalities and peoples—are objectively identifiable, independent of the views of their members.27 As such, the EPRDF primarily relied on an ethno-linguistic criterion to ‘draw round’ ethnic groups—aiming to delineate regional boundaries that provided a perfect fit between ethno-linguistic groups and territorial borders.28 In Ethiopia’s new constitution these ethnic groups were given the ‘unconditional right to self-determination, including the right to secession’.29 Ethnic federalism therefore has important implications for land administration. Not only does federalism associate ethnic groups with defined territories, but, by providing the right to secession, it also explicitly prioritizes the land rights of ethnic groups over those of individuals. As argued by Endrias Eshete, To confer the right to secession on national communities is to grant that a regional state’s collective property rights take priority over the property rights of outsiders—nonmembers and federal government—in the region. What is now held by nonmembers can be legitimately taken by a seceding state.30 Federalism implies, therefore, that each ethnic group has its own home region and, consequently, that ethnic outsiders have a weaker claim to land than indigenous inhabitants. In effect, federalism implies that individuals are first and foremost citizens of ethnic regions, rather than of Ethiopia. There is a direct contradiction between the universalistic principles of the land policy, which accords all Ethiopian farmers equal land rights, and the territorial implications of ethnic federalism that suggest lesser rights for non-indigenous ethnic minorities. Under federalism, land administration has been devolved to regional governments, which are required to formulate land proclamations within the framework of the federal land policy.31 Despite the implications of ethnic federalism for land administration, however, regional land proclamations—in line with federal land policy—do not mention ethnicity. For example, the Oromiya land proclamation states that ‘any resident of the region’ has the right to land.32 Furthermore, when asked about the potential for ethnicity to influence land administration, the head of the federal Land Use Administration denied that there could be any ethnic differentiation in land rights: ‘ethnicity should not influence policy, farmers are farmers’.33 The only instance in which land administration explicitly distinguished between ethnic groups was the government’s resettlement programme that relocated food insecure farmers from densely populated highlands to more sparsely populated lowlands, primarily in the west and south. In an attempt to limit ethnic conflict between resettled households and existing inhabitants of resettlement sites, resettlement was conducted within, rather than between, regions.34 Although the full territorial implications of ethnic federalism are not articulated in land laws, the ideas underpinning ethnic federalism continue to be widely debated in Ethiopia. On numerous occasions during fieldwork people—mostly outside state structures, but occasionally within them—argued for a more extreme interpretation of federalism: that it is necessary to differentiate between ethnic groups depending on whether they are in their ‘home’ region or not, even justifying the removal of ethnic minorities to their ‘home’ regions to preserve resources for the dominant ethnic group—effectively ethnic cleansing.35 As such, there remains the possibility that these more extreme ideas on ethnic federalism could influence implementation, beyond the explicit provisions of land laws. The third and final set of land tenure institutions and related ideas concerns the neo-customary tenure system in Arssi, Oromiya. From a legal perspective, the 1975 nationalization eradicated any other form of tenure. However, in practice neo-customary tenure systems remain important influences on land administration in many parts of the country. I use the term neo-customary rather than customary in acknowledgement that ‘customary’ tenure systems are the product of inherited traditions adapted by state and societal actors. As Boone argues, it is invariably the case that ‘[p]ostcolonial governments have been deeply and actively implicated in upholding and reproducing the institutions and political relationships embedded in the (neo)customary land tenure regimes. It is incorrect to see customary land regimes as informal in the sense of existing beyond the purview of the state’.36 This rationale applies equally to neo-customary tenure in Ethiopia. Neo-customary tenure remains a major influence on land administration in areas where pastoralism and shifting cultivation are the dominant forms of production and where the territorial reach of the state continues to be much more limited than in the highlands, namely in Afar, Benishangul-Gumuz, Gambella, the south of Oromiya, Southern Nations, Nationalities and Peoples’ Region (SNNPR), and Somali.37 However, even in areas of smallholder cultivation in Oromiya, including the part of Arssi covered by the case discussed below, and SNNPR, neo-customary tenure continues to be an influential alternative to state land administration.38 According to the ideals of Arssi Oromo neo-customary tenure, land is the property of the patrilineal descent group or clan, and a principal objective of neo-customary tenure is to retain land within the descent group.39 As such, women—who are required to marry out of the clan—lose rights to land in their birth clan upon marriage, while women who marry into a clan only have rights to land through their husband.40 As such, Arssi Oromo neo-customary tenure is in direct tension with the universalist principles of state land ownership. Rather than land being a right of national citizenship, according to neo-customary ideals land access is limited to clan members, in effect implying a form of local citizenship. To a degree, therefore, neo-customary tenure resonates with the territorial implications of ethnic federalism. Like federalism, neo-customary law associates an ethnic group with a particular territory and, in doing so, curtails land rights for ethnic outsiders, prioritizing local citizenship. However, the locus of authority and the territorial jurisdiction differ. Under the ideals of federalism, the regional government has ultimate authority, whereas clan elders are responsible under neo-customary tenure. Furthermore, local citizenship under neo-customary tenure is based on the clan, excluding not just non-Oromo, but also other Oromo clans, in contrast to federalism, which provides equal rights for all those from the dominant ethnic group of the region. Table 1 summarizes the differences between these ideal-type land tenure regimes. While the distinctions between these ideals are clear in principle, the reality of actual land tenure institutions is rather more ambiguous. Table 1 Contrasting ideals on land in Ethiopia State ownership Ethnic federalism Arssi Oromo neo-customary Locus of authority over land Federal government Regional government Neo-customary leaders Territorial jurisdiction National Regional Local Citizenship principle National Ethnicity Clan membership State ownership Ethnic federalism Arssi Oromo neo-customary Locus of authority over land Federal government Regional government Neo-customary leaders Territorial jurisdiction National Regional Local Citizenship principle National Ethnicity Clan membership Source: author, based on categories developed in Boone, Property and political order in Africa. Table 1 Contrasting ideals on land in Ethiopia State ownership Ethnic federalism Arssi Oromo neo-customary Locus of authority over land Federal government Regional government Neo-customary leaders Territorial jurisdiction National Regional Local Citizenship principle National Ethnicity Clan membership State ownership Ethnic federalism Arssi Oromo neo-customary Locus of authority over land Federal government Regional government Neo-customary leaders Territorial jurisdiction National Regional Local Citizenship principle National Ethnicity Clan membership Source: author, based on categories developed in Boone, Property and political order in Africa. The picture is further complicated by Oromiya regional government attempts to re-establish the Gadaa institution. Gadaa is the customary system of social and political organization in Oromo society. During Imperial and Derg rule Gadaa was effectively wiped out in most of Oromiya and only really persisted among the Boran Oromo pastoralists. Since the early 2000s, the Oromiya People’s Democratic Organization (OPDO)—the regional EPDRF party—has sought to re-establish Gadaa as a means of compensating for its limited legitimacy. The OPDO has been regarded by many as an artificial creation of the TPLF/EPRDF that has struggled to establish legitimate authority in Oromiya. Indeed, the regional government recognizes that when it ‘works through the Gadaa it is more successful than through the local government’.41 As such, the position of Aba Gadaa—the political leader of the Gadaa and the ultimate authority over customary law—has been institutionalized in line with the government structures of zone and wereda42 and reports to the regional Bureau of Culture and Tourism.43 The regional land policy now also requires community elders to make the first attempt at dispute resolution, including with respect to land disputes. These elders are given training on government policy and issued with identity cards.44 Only if one party to a dispute is dissatisfied with the outcome does a case proceed to the courts or higher levels of government administration for adjudication. In the process, the regional government has sought to transform neo-customary tenure, jointly announcing changes to customary law with the Aba Gadaa and codifying neo-customary law in a book held in the wereda offices.45 Amongst other things, these changes seek to enhance women’s rights to hold land and, most importantly for this article, allow ethnic minorities to hold land and even to nominate elders from ethnic minorities. As such, there is now a distinct separation between popular ideas about the neo-customary tenure system as described above—frequently idealized as unchanging since the sixteenth century46—and what currently passes as customary tenure in parts of the region, which the regional government has sought to co-opt as a means of pursuing its policy objectives and enhancing its authority. Conflict and land rights in practice The case study presented in this section is used to illustrate the relevance of these ideas and institutions for land administration and to highlight the ambiguous state responses to one particular land conflict. The analysis builds on insights from Joel Migdal’s State-in-society approach and Tobias Hagmann and Didier Péclard’s Negotiating statehood framework.47 Rather than a unitary actor, the state is conceptualized as a network of distinct organizations, each with their own interests, in competition for domination with societal organizations. State officials, meanwhile, are embedded in society through their membership of families, clans, businesses or clubs. As such, policy implementation is not simply the impartial application of formal rules and laws by bureaucrats, but rather a negotiated process that may vary from the intentions of higher levels of government. In such circumstances, formal and informal institutions, and ideas all have the potential to influence decision-making. Turufe Kechema occupies a fertile agricultural area suited to cereal production in the Rift Valley with good road links to nearby Shashemene town and Addis Ababa. Agricultural production is dominated by smallholder farmers, most of whom hold approximately 0.5–3 hectares of land.48 Turufe is part of the territory claimed by the Weyrera clan from the Arssi Oromo ethnic group. The Weyrera continued to apply neo-customary tenure on land that was not expropriated by the Imperial regime and in secret alongside state ownership during the Derg’s rule.49 Despite neo-customary tenure, however, successive governments have ignored clan and ethnicity in land administration and there has been a steady flow of migrants into Turufe. In addition to the Weyrera, there are members of several other Arssi clans in Turufe who, under ideal-type neo-customary law would have no rights to land. Meanwhile, some of the first non-Oromo arrived during the Imperial conquest in the late nineteenth century. Northern soldiers (neftegna), especially Amhara and some Eritreans, were allocated rights to extract tribute from Arssi Oromo tenants (gebbar) as reward for military service.50 From the 1950s, migrants came for treatment at the nearby clinic, now the Shashemene Referral Hospital, a centre for leprosy treatment. At that time, Swedish Protestant missionaries, who established the clinic, held 12 gasha (480 hectares) around Turufe. These missionaries rehabilitated people treated for leprosy, in particular Kembata, by dividing two gasha between them.51 The area also attracted migrant labour during busy agricultural periods and some migrants have subsequently settled. Under the Derg, land was redistributed several times between 1977 and 1990 to adjust for changing household size, taking no account of ethnicity, with the result that all kebele residents received land. Due to immigration and population growth, land shortages became a growing problem during the Derg era and an important driver of tensions over land. Since the EPRDF took power in 1991, land has not been redistributed with the result that landlessness has become a major issue, especially for young adults. At present there are significant minorities of Amhara, Tigrayan, Kembata, and Wolayita in Turufe, with the Oromo making up 54 per cent of the population according to one survey.52 Turufe Kechema is part of Shashemene Zuria wereda and one of three villages in Turufe-Wetera-Elemo kebele. Ever since 1991, the OPDO has ruled Oromiya and Turufe-Wetera-Elemo. Under the Derg the peasant association leadership comprised several ethnic groups and the working language of government was Amharic. Since 1991 under ethnic federalism, the working language of government throughout Oromiya is Afaan Oromo and in Turufe all local government officials are Oromo. The original land conflict considered in this case study dates to 1991, when the Derg lost power and there was a period of instability around Turufe, exacerbated by local clashes between the EPRDF and the OLF.53 At this point of limited state authority, the Weyrera called on other Arssi Oromo to join them in attacking the Kembata, both in Turufe and neighbouring kebele, burning their houses.54 In total, 3,601 households across 10 kebele were expelled from their villages.55 In Turufe, 117 of 413 households at the time were Kembata,56 and although a few Kembata who had married Oromo were allowed to stay, most were evicted.57 The vast majority of those expelled were Kembata; however, some Wolayita were also forced to leave.58 The local Oromo population distributed the Kembata land amongst themselves. However, a few Tigrayan farmers also obtained land by paying Oromo who had seized land.59 A few evicted Kembata moved to Kembata zone, in SNNPR, though most settled on the outskirts of Shashemene. Respondents offered a number of explanations for these events. According to most Oromo respondents, the conflict originated in an incident in the 1980s in which Oromo thieves (shifta) were stealing crops from local people, both Kembata and Oromo, or in which Oromo farmers allowed their cattle to graze on the harvest of Kembata farmers.60 The Kembata reported the crime to the authorities, and the Oromo defendants were imprisoned and, subsequently, executed. According to some Oromo, the Kembata peasant association chairman at the time signed the execution papers.61 The former chairman himself denied any involvement, blaming the executions on the vicious nature of the Derg.62 According to the Weyrera, an entire clan is held responsible for a murder carried out by one individual from that clan. As such, they argued that the entire Kembata ‘clan’ was responsible for the execution.63 Furthermore, the Kembata reported the crop loss directly to the peasant association, bypassing customary dispute resolution since they believed that the elders were biased towards the Weyrera.64 Customary dispute resolution is a highly valued social institution in Oromiya, while the pursuit of justice through the courts is considered disrespectful of customary law. Some Kembata acknowledged that these actions probably deepened Oromo resentment.65 Kembata respondents, however, dismiss the executions as a pretext. They argue that the eviction was actually a response to growing land shortages as a result of population growth and the Weyrera belief that the land belonged to their clan: The Oromo made a pretext but they were saying that they needed the land for their children. They burned the houses of the Kembata and stole from them.66 They were saying that the land belongs to us, that you are alien to the area and that there is a shortage of land … The local [Oromo] families had land but the Oromo children had none. So they devised a plan to get the land for their children.67 Indeed, a significant minority of Oromo respondents supported these claims, highlighting the effect of migration on population growth and justifying the evictions in terms of indigenous land rights. These arguments clearly resonate with both neo-customary and ethnic federalist rationales, Migration has caused me a problem because the other people [other ethnic groups] have claimed the land as their own. The local community told them [the Kembata] that they had come for treatment [at the hospital] not to take the land of the local people.68 The Kembata were living on Oromo land. When the Derg left, the people were waiting for the opportunity to throw out the Kembata.69 The origins of the dispute pre-date federalism, and the eviction occurred during a period of instability and state weakness. Clearly then, the conflict cannot be solely attributed to ethnic federalism. Nevertheless, several respondents from different ethnic groups argued that ethno-nationalist ideas promoted by both the EPRDF and the OLF polarized the community, contributing to an expectation that ethnic groups should return to their own ethnic regions: Federalism exacerbated the problem. It divided the community. Those who did not speak the local language were expected to return to their own homeland.70 This government came up with ethnic federalism based on language and the government here started to work in Afaan Oromo. All the people were expected to go to their original land.71 When the government changed many people were expected to go back to their homeland. Some left voluntarily, some forcefully.72 Although the Kembata constituted the majority of those evicted in 1991, other minorities were also threatened, but were armed and able to defend themselves.73 During the hotly contested 2005 national elections, in which the EPRDF suffered significant electoral losses, Tigrayan residents were concerned that victory for the Oromo National Congress (ONC)—an Oromo opposition party—would have led to retribution from the local Oromo population against the Tigrayans who remain strongly associated with the TPLF/EPRDF.74 EPRDF victory ensured that this did not occur but they remain concerned about the future: The local people have a great sense of belonging to the land. They divided the Kembata land amongst themselves. I fear that if this government fails, their eyes are on our land and the same thing may happen again. When there is a transfer of power, if there is instability there may be a window of opportunity. Currently the government is in control so I do not think that this could happen.75 While many Oromo were reluctant to discuss the problems, a few were more candid. For example, one respondent acknowledged that the government prevents the Oromo from taking further action but the fall of the government could bring change: If we get the chance to remove them [the Tigrayan and other ethnic groups] we would not give them even one day in Turufe. They [the current ‘Tigrayan’ government] are killing our children… [Oromo] Children here mix Amharic and Afaan Oromo, they are not like the Wellega who speak pure Afaan Oromo. So we want to remove the other groups—it is beyond just land but it is to realize our right to language and culture. It is to avert such problems that we want to remove the others, to avert this cultural distortion.76 While the full territorial implications of ethnic federalism are not reflected in the land policy, therefore, this more extreme interpretation of ethnic federalism—the idea that individuals are first and foremost members of ethnic groups with limited rights outside their home region—is clearly present in land debates in Turufe and strongly resonates with the local citizenship implied by the ideal form of neo-customary tenure. Over the years since this original dispute, the Kembata have appealed for support to a variety of state agencies, several of which have made some attempt to resolve the conflict. In 1992 the Kembata registered the dispute with the wereda Agricultural Office, responsible for land administration.77 They have subsequently lodged appeals with the zone and regional governments, the Federal Ministry of Agriculture and Rural Development (MoARD), the Federal House of Representatives, the Prime Minister’s Office and the Federal Ombudsman’s Office. The Kembata made their appeals to government agencies, they say, because the courts under federalism are biased in favour of the Oromo majority.78 In their letters, they repeatedly called upon the government to intervene on their behalf as ‘citizens of Ethiopia’, framing their appeals in the universalistic language of the land policy.79 This is in stark contrast to the Oromo justification of the eviction in terms of indigeneity, which finds support in the more extreme interpretation of ethnic federalism and the ideal form of neo-customary tenure. Based on federal and regional land laws, which make no mention of ethnicity, the case is straightforward. The last Derg land redistribution is accepted as final and any ethnic minorities that received land at that time have the same rights as indigenous Oromo.80 Since the Kembata were allocated land under the Derg, their land should be returned to them. Indeed, according to the letters provided by the Kembata, no state agency has ever denied the legitimacy of their claim. However, various agencies have proposed solutions to the dispute that have not only failed to achieve a resolution but are also highly ambiguous regarding the land rights of non-indigenous minorities. The Prime Minister’s Office made the first concrete proposal in 1995/96, asking the wereda to find common land in the vicinity of Shashemene to allocate to the displaced households.81 It was, however, stipulated that this should not include the return of their original land. While this proposal failed to respect the displaced population’s rights to their previous land, it was broadly in line with the principles of the land policy, namely that the Kembata—an ethnic minority residing in Oromiya—had the right to replacement land in that region. Given the challenge of finding sufficient land in Shashemene wereda, the MoARD subsequently approached the regional governments of Oromiya and SNNPR to find a solution. The regional governments identified 3,958 hectares of forestland straddling the border between the two regions.82 While the Kembata accepted the proposal, in 1999/2000 the federal government ordered a stay of execution since the land in question had itself already been the focus of ‘recurrent ethnic conflict’ among the existing inhabitants.83 At this stage East Shewa zone asked Shashemene wereda to investigate the original dispute and to consider returning the farmers to their original land in Turufe.84 From the documentation available, this was the only occasion that a return to Turufe was even considered by the state as a possibility. Nevertheless, two years later a letter from Shashemene wereda to East Shewa zone merely reported that the amount of land previously held by the Kembata was unknown, and that all the disputed land was currently in use.85 The wereda’s failure to restore the Kembata’s land is perhaps not surprising. Ethnic federalism ‘tends to encourage, even require, political leaders to view themselves primarily, if not exclusively, as agents of their own ethnic communities’.86 The vast majority of government employees in Oromiya are Oromo, while many in the kebele administration in Turufe are from Weyrera clan. Indeed, it is likely that some current incumbents were directly involved in the eviction of the Kembata or have close ties to those that were. Furthermore, in light of the precarious nature of the OPDO’s relationship with the ethnic group that it is supposed to represent, it would be very difficult for the local administration to side with the Kembata over the Oromo majority. After several years of impasse, in 2005, the Minister for Agriculture and Rural Development wrote to the Kembata noting that following consultations with the evicted population and the regional governments of Oromiya and SNNPR, his final decision was that the Kembata would be given priority in either the Oromiya or SNNPR resettlement programmes, depending on their preference.87 The Kembata, unhappy with this offer, wrote to the Federal Ombudsman to appeal. Although the Ombudsman did investigate, at the time of fieldwork, he had not reached any conclusion. In the meantime, Oromiya began land registration in 2005, including in Turufe. Registration is organized by the wereda and the measurement of the plots is conducted by a land administration committee (LAC) elected from and by members of the kebele. When land is measured, all those with claims to land in a particular locality are called and, if there is any dispute over a plot, the LAC refers the matter to the courts. In Turufe, all five members of the LAC were Oromo and there were no disputes recorded regarding land previously held by the Kembata. The wereda administration—despite knowledge of ongoing investigations—issued land certificates to all occupants of the disputed land. In doing so, the wereda significantly strengthened the occupiers’ position, providing legal documents to support their claims and, in all likelihood, extinguishing any hopes of return for the Kembata. The proposals to resettle the Kembata in SNNPR and/or Oromiya and the decision to bring the SNNPR government into discussions represent something of a shift from the initial proposal to find land within Shashemene wereda. While that first proposal is broadly in line with the land policy, the inclusion of the SNNPR administration in dispute resolution and the possibility of relocating the Kembata to SNNPR are not. The SNNPR has no jurisdiction over the Kembata living in Oromiya and inter-regional resettlement is explicitly prohibited. While the rationale for relocating the Kembata to SNNPR is not stated, the most likely explanation would appear to be the ideas underpinning ethnic federalism, namely that the Kembata belong first and foremost to that region and having been displaced from Turufe, the logical step is for them to return to their ‘home’ region. While the land policy might be clear about the rights of non-indigenous ethnic minorities, the state’s response to the eviction of the Kembata is ambiguous, to say the least, and appears to draw on a more extreme interpretation of ethnic federalism, in addition to the land policy. The proposals to relocate the Kembata to SNNPR also serve to illustrate some of the inherent tensions within the federal project. SNNPR comprises about 45 ethnic groups that were combined into one region not based on common ethnic bonds, but because these groups were considered too small to be administratively viable.88 Consequently, relocating the Kembata to the region in which Kembata zone is located does not necessarily mean that they will join people with whom they share any bonds. Kembata zone has one of the highest population densities in the country, with the result that it would be impossible to find sufficient land there to resettle the displaced population, and so the proposals involved relocation to other parts of the region. The first proposal regarding forestland on the border of SNNPR and Oromiya involved land in Sidama zone, while the SNNPR resettlement programme focuses on Bench-Maji, faraway in the southwest of the region. Though part of the same linguistic family, the Kembata nevertheless have a distinct ethnic identity to the Sidama, while they have no linguistic or other ties with the Bench, the dominant ethnic group in Bench-Maji. Land, ethnicity and conflict in Ethiopia The case of Turufe illustrates some of the tensions between competing laws and ideas on land administration and the ambiguous state response to dispute resolution. However, Turufe is only one case and—it might be argued—a rather exceptional one. While land conflict is common in rural Ethiopia and there are numerous cases in which conflicts take on an ethnic dimension,89 widespread ethnic cleansing—as might be implied by ethnic federalism—has not taken place. That said, however, Turufe is far from exceptional. Indeed, there are several examples of comparable situations in which non-indigenous minorities have been relocated to their home region or where they have been threatened with such action. As such, the tensions between these competing laws and ideas on land rights have broader relevance, while state responses to infringements of minority land rights are far from uniform. A comparable case is provided by Tesfaye Tafesse’s study of the displacement of 12,000 Amhara from East Wellega, Oromiya in 2000.90 Amhara had migrated to the area over several decades and, while relations between Amhara settlers and the indigenous Oromo were good for many years, growing land shortages and ethnic federalism resulted in a deterioration. Under federalism, Afaan Oromo was made the language of local government and previously ethnically-diverse local administrations were increasingly dominated by Oromo. Tafesse argues that Oromo local government officials incited violence against the migrants, in line with the territorial implications of federalism, in stating that ‘[t]his land belongs to Oromos and only Oromos and not to you, Amharas’.91 Accusations that the Amhara had been involved in cattle theft, illegal deforestation and a murder led in June 2000 to orders from wereda and kebele officials for the Amhara settlers to be expelled and to return to Amhara. While many did leave, the violent intervention of local government militias at the end of the year is thought to have led to the deaths of several hundred Amhara and further large-scale displacement. The majority fled to Amhara and were subsequently included in the Amhara resettlement programme that focused on the western Amhara lowlands. Both the displaced population and the government rejected any possibility of returning to Wellega. As in Turufe, the territorial implications of ethnic federalism provided a rationale for the displacement of an ethnic minority in response to growing resource shortages. Meanwhile, the state failed to enforce the universalist principles of state ownership. In line with the rationale of ethnic federalism, state agencies promoted the eviction of non-indigenous minorities, facilitating their return to their ‘home’ region. Similar questions regarding the land rights of ethnic minorities are raised by the case of the Guji Oromo pastoralists living in the Nech Sar national park in Gamo Gofa zone, SNNPR, as described by Tadesse Berisso.92 Despite the park being located within SNNPR, Guji Oromo have lived in the area for generations. Nonetheless, when park authorities decided to remove all inhabitants to ‘protect’ the park and promote tourism, repeated attempts were made to relocate the Guji Oromo to Oromiya. Once again this decision is in line with the more extreme interpretation of ethnic federalism and in direct contradiction of the universalist principles of state ownership. A common perception remains that members of particular ethnic groups ultimately belong in the home region of that ethnic group and, therefore, when there are disputes over land, one of the main options available is to return the ethnic minority to their home region. The legal and ideational tension between federalism and state ownership exists throughout the country, while varying forms of neo-customary tenure provide an alternate rationale in most regions. However, the ways in which these institutions and ideas combine to shape policy implementation in different parts of the country are contingent on the government’s political economic priorities and ever-evolving state-society and intra-state relations. In the case of Turufe and that of East Wellega, the status of the Oromo as the largest ethnic group in the country and the limited legitimacy of the OPDO within the region appears to have influenced decision making, with local government favouring or at least turning a blind eye to the dispossession of ethnic minorities for fear of alienating the local population. In contrast, recent government promotion of large-scale agricultural investments provides a clear counter-example in which land access for outsiders has been prioritized over that of politically marginal indigenous populations. In particular, the federal government has re-centralized authority from regional administrations to lease land to agricultural investors in lowland areas of Benishangul-Gumuz, Gambella and South Omo.93 In the process, pastoralists and shifting cultivators have found their land access curtailed in direct contradiction of the principles of ethnic federalism. While the EPRDF promoted and designed the federal system, its support for federalism has long been ambiguous and the government has resisted its full implementation. Instead, the EPRDF has sought out the political middle ground between the centrists and regionalists.94 In particular, the EPRDF has employed federalism in an attempt to neutralize the ethno-national sentiments of the latter, while limiting the implementation of the more extreme and highly problematic implications of ethnic federalism, including with respect to land administration. Certainly the EPRDF has no interest in further stoking ethno-nationalism or permitting ethnic cleansing. Far from resolving the ‘national question’, however, federalism has, arguably, led to growing ethno-nationalist sentiments in many parts of the country. For Abbink, instead of solving conflicts, federalism has merely ‘decentralized’ them, pushing them down to lower levels of government administration.95 While this may be true, by-and-large, the importance of ethnicity to national-level politics has been re-asserted during unprecedented anti-government protests, first in 2014 and then from late 2015 to the present, and large-scale conflict along ethnic lines in the Oromiya–Somali border region in 2017. This unrest culminated in a state of emergency from October 2016 to August 2017, and once again beginning in February 2018. The original flashpoint was the Addis Ababa Masterplan that sought to integrate Addis Ababa with the surrounding area in Oromiya as part of an integrated urban development plan. Many Oromo claim Addis Ababa (or Finfinnee in Afaan Oromo) as the capital city of Oromiya. Consequently, its status as a separate city administration is a source of considerable resentment, while the proposed city expansion is seen as a further attempt to expropriate Oromo territory. Subsequently protests spread to Amhara, partly based on solidarity with Oromo protestors but also concerning the contested incorporation of Wolkayit wereda, previously part of the Imperial Province of Gondar, into Tigray region. The government has tended to attribute the protests to high rates of unemployment and rural landlessness. While there is some truth in this, the protests generated such unprecedented support as a result of a confluence of a wider range of issues, including fragmentation within the ruling party, frustration at the absence of democracy, the perceived domination of the TPLF within the EPRDF and the lack of popular legitimacy and corruption in the EPRDF.96 Clearly, given the focus of protests on the Addis Ababa Masterplan and the status of Wolkayit wereda, the territorial implications of ethnic federalism are also among the central grievances. Indeed, as was the case in Turufe in the early 1990s, there is some evidence that in the temporary absence of state authority created by the protests, some Oromo protestors have attacked non-indigenous Amhara and Tigrayan residents in parts of Oromiya, seizing their land and other property in the process.97 While the relative strength of the EPRDF has to date limited any possibility of the widespread pursuit of the full territorial implications of ethnic federalism, these ideas are widely debated and supported by at least some in Ethiopia. Rather than resolving the national question, ethnic federalism has merely changed its form. Finding a way to balance growing ethno-nationalism with the protection of minority rights remains one of the central challenges facing the government, notably with respect to what are, at the time of writing, fast moving events in Ethiopian politics. Conclusion There is a common perception in much of the literature that Ethiopia’s 1975 land reform eradicated previous diverse tenure systems, resulting in a relatively uniform and unambiguous system of state land ownership. This article has shown that the reality is not quite so clear-cut. While state land ownership remains the dominant influence on land tenure, the creation of ethnic federalism has direct implications for land administration, in particular with respect to the rights of non-indigenous ethnic minorities. Furthermore, neo-customary tenure systems have retained some influence in many parts of the country. Indeed, the government has sought to reinvigorate and co-opt customary tenure as a means of pursuing its own policy objectives in recent years. As such, land tenure in Ethiopia is more ambiguous than is often assumed. This article has explored the overlap and divergence between these three laws and the broader ideas that have been used to justify them, specifically with respect to the land rights of non-indigenous minorities. The case study of land conflict in Turufe demonstrates that these three bodies of law and associated ideas remain of the utmost relevance to contemporary land debates, and that parties to land disputes draw on these ideas to justify their particular resource claims. Meanwhile, the state response to conflict has been highly ambiguous, failing to defend the land rights of non-indigenous minorities in line with state ownership, and instead presenting proposals that align, in part, with the principles of ethnic federalism. While competing institutions and ideas on land tenure result in considerable ambiguity, the outcome of tensions between these competing rationales is not pre-determined but varies with evolving political dynamics—including both local state-society and intra-state relations. Although this article has provided examples in which non-indigenous inhabitants have been displaced, in line with more extreme interpretations of ethnic federalism, such events are, fortunately, relatively rare. Equally, however, there are other examples in which indigenous rights have been undermined by the state’s prioritization of outsiders, most clearly through the state promotion of large-scale agricultural investments in the west of the country. The broad contours of the Ethiopian debate regarding the land rights of non-indigenous minorities and the political implications of these issues mirror those taking place across much of the rest of Africa.98 As elsewhere, ambiguity in land tenure in Ethiopia results in considerable uncertainty regarding the land rights of non-indigenous minorities. This, in turn, raises major questions regarding citizenship and belonging—whether individuals are first and foremost national citizens or members of ethnic groups. While the broad contours of this debate mirror those taking place elsewhere, the form that this takes is specific to the Ethiopian case. Notably, the predominance of state land ownership sets Ethiopia apart from much of the rest of the continent, where neo-customary tenure is prevalent. Furthermore, Ethiopia is rather unusual regarding the system of ethnic federalism, which is an additional source of ambiguity regarding minority rights, introducing further uncertainty regarding the jurisdiction of federal and regional governments. Footnotes 1. Liz Alden Wily, ‘‘The law is to blame’: The vulnerable status of common property rights in Sub-Saharan Africa’, Development and Change 42, 3 (2011), pp. 733–57; Pauline Peters, ‘Conflicts over land and threats to customary tenure in Africa’, African Affairs 112, 449 (2013), pp. 543–62. 2. Gebru Mersha and Mwangi Wa Githinji, ‘Untying the Gordian Knot: The question of land reform in Ethiopia’ (Policy Paper 9, International Institute for Social Studies/United Nations Development Programme, The Hague, 2005), p. 8. 3. Mersha and Githinji, ‘Untying the Gordian Knot’, p. 8. 4. Wibke Crewett, Ayalneh Bogale, and Benedikt Korf, ‘Land tenure in Ethiopia: Continuity and change, shifting rulers, and the quest for state control’ (Working Paper 91, CGIAR Systemwide Program on Collective Action and Property Rights (CAPRi), Washington DC, 2008), p. 13. 5. Berhanu Adenew and Fayera Abdi, ‘Research report 3: Land registration in Amhara region, Ethiopia’ (IIED, London, 2005), p. 6. 6. Allan Hoben, ‘Ethiopian land tenure revisited: Continuity, change and contradictions’ (Working Paper 236, African Studies Center, Boston University, 2001); Crewett et al., ‘Land tenure in Ethiopia’; Bereket Kebede, ‘Land reform, distribution of land and institutions in rural Ethiopia: Analysis of inequality with dirty data’, Journal of African Economies 17, 4 (2008), pp. 550–77; Tom Lavers, ‘Land registration and gender equality in Ethiopia: How state-society relations influence the enforcement of institutional change’, Journal of Agrarian Change 17, 1 (2017), pp. 188–207. 7. Catherine Boone, ‘Property and constitutional order: Land tenure reform and the future of the African state’, African Affairs 106, 425 (2007), pp. 557–86; Catherine Boone, Property and political order in Africa: Land rights and the structure of conflict (Cambridge University Press, Cambridge, 2014). 8. Mahmood Mamdani, Citizen and subject: Contemporary Africa and the legacy of late colonialism (Princeton University Press, Princeton, NJ, 1996); Helène Maria Kyed and Lars Buur, ‘New sites of citizenship: Recognition of traditional authority and group-based citizenship in Mozambique’, Journal of Southern African Studies 32, 3 (2006), pp. 563–81; Boone, Property and political order in Africa. 9. Notable contributions include Alemante Selassie, ‘Ethnic federalism: Its promise and pitfalls for Africa’, Yale Journal of International Law 28 (2003), pp. 51–107; Sarah Vaughan, ‘Ethnicity and power in Ethiopia’ (University of Edinburgh, unpublished PhD thesis, 2003); David Turton, Ethnic federalism: The Ethiopian experience in comparative perspective (James Currey, London, 2006). 10. Asnake Kefale, ‘Federal restructuring in Ethiopia: Renegotiating identity and borders along the Oromo–Somali ethnic frontiers’, Development and Change 41, 4 (2010), pp. 615–35; Fekadu Adugna, ‘Overlapping nationalist projects and contested spaces: The Oromo–Somali borderlands in southern Ethiopia’, Journal of Eastern African Studies 5, 4 (2011), pp. 773–87. 11. Endrias Eshete, ‘Ethnic federalism: New frontiers in Ethiopian politics’ (First National Conference on Federalism, Conflict and Peace Building, Addis Ababa, 2003); Selassie, ‘Ethnic federalism’; Christophe Van der Beken ‘Federalism, local government and minority protection in Ethiopia: Opportunities and challenges’, Journal of African Law 59, 1 (2015), pp. 150–77. 12. Boone, Property and political order in Africa. 13. Ibid., p. 68. 14. Colin Hay, ‘Ideas and the construction of interests’, in Daniel Béland and Robert Henry Cox (eds), Ideas and politics in social science research (Oxford University Press, Oxford, 2011), pp. 65–82, 69. 15. Daniel Béland, ‘Mapping changing social policy ideas: A global, actor-centred approach’ (UNRISD workshop on New Direction in Social Policy, 7–8 April 2014, Geneva, 2014). 16. Christopher Clapham, Transformation and continuity in revolutionary Ethiopia (Cambridge University Press, Cambridge, 1988). 17. PDRE, ‘A proclamation to provide for the public ownership of rural land’, Federal Negarit Gazeta Proclamation 31/1975 (1975), para. 4.1. Emphasis added. 18. Dessalegn Rahmato, The peasant and the state: Studies in agrarian change in Ethiopia 1950s—2000s (Addis Ababa University Press, Addis Ababa, 2009). 19. Meles Zenawi, cited in Harold Marcus, A breakfast meeting with Prime Minister Meles, 1995, <http://www.hartford-hwp.com/archives/33/008.html> (15 July 2009). 20. FDRE, ‘Constitution of the Federal Democratic Republic of Ethiopia (FDRE)’, Federal Negarit Gazeta, (1994), para. 40. Emphasis added. 21. FDRE, ‘Federal Democratic Republic of Ethiopia (FDRE) – Rural land administration and land use proclamation’, Federal Negarit Gazeta, Proclamation 456/2005 (2005), para. 5.1.b. Emphasis added. 22. Klaus Deininger and Songqin Jin, ‘Tenure security and land-related investment: Evidence from Ethiopia’, Proceedings of the first international conference on the Ethiopian economy II (Ethiopian Economics Association, Addis Ababa, 2004), pp. 19–50. 23. Interview, respondent FG4, head, Federal Land Use Administration, MoARD, Addis Ababa, 17 November 2009. 24. In most parts of the country, during the latter days of the Derg. 25. Eshete, ‘Ethnic federalism’; Merera Gudina, ‘Contradictory interpretations of Ethiopian history: The need for a new consensus’, in Turton, Ethnic federalism, pp. 119–31. 26. Christopher Clapham ‘Ethiopian development: The politics of emulation’, Commonwealth and Comparative Politics 44 (2006), pp. 137–50, p. 148. 27. Vaughan, ‘Ethnicity and power in Ethiopia’. 28. Van der Beken, ‘Federalism, local government and minority protection in Ethiopia’. 29. FDRE, ‘Constitution of the Federal Democratic Republic of Ethiopia (FDRE)’, para. 39.1. 30. Eshete, ‘Ethnic federalism’, p. 21. 31. Crewett et al., ‘Land tenure in Ethiopia’, highlight some limited variation between regional policies. 32. ONRG, ‘Oromia National Regional Government (ONRG) – Proclamation to amend the proclamation nos. 56/2002, 70/2003, 103/2005 of Oromia Land Use and Administration’, Megeleta Oromia Proclamation 130/2007 (2007), para. 5.1. 33. Interview, respondent FG4. 34. MoFED, ‘Rural development policy and strategies’ (Ministry of Finance and Economic Development [MoFED], Addis Ababa, 2003), p. 41. 35. Such views were raised in a number of informal discussions in cafes and on long bus journeys in Oromiya and Tigray, and by a number of Oromo university students. In a few cases, this included people involved in extended state structures, for example, those holding unpaid positions in the many local committees that play an important role in implementation, albeit not paid officials. 36. See Boone, Property and political order in Africa, p. 38. 37. Crewett et al., ‘Land tenure in Ethiopia’; Tom Lavers, ‘Agricultural investment in Ethiopia: Undermining national sovereignty or tool for state building?’ Development and Change 47, 5 (2016), pp. 1078–101. 38. Interview, respondents SR1, land administration experts, SNNPR land administration, Awassa, 23 February 2010. 39. Mamo Hebo, Land, local custom and state policies: Land tenure, land disputes and disputes settlement among the Arsii Oromo of southern Ethiopia (Shoukadoh, Kyoto, 2006). 40. Evidently, neo-customary tenure also has important gender implications see Lavers, ‘Land registration and gender equality in Ethiopia’. 41. Interview, respondent OR2, deputy head, Oromiya Bureau of Culture and Tourism, Addis Ababa, 16 February 2010. This co-optation of customary institutions as a means of enhancing authority is a common government strategy, see Tobias Hagmann, ‘Beyond clannishness and colonialism: Understanding political disorder in Ethiopia’s Somali region, 1991–2004’, Journal of Modern African Studies 43, 4 (2005), pp. 509–36; Marco Bassi, ‘The politics of space in Borana Oromo, Ethiopia: Demographics, elections, identity and customary institutions’, Journal of Eastern African Studies 4, 2 (2010), pp. 221–46. 42. The structure of government administration (from highest to lowest level) is: federal, regional, zone, wereda, kebele. 43. To the point where the Aba Gadaa’s office is situated in the local government compound and the Aba Gadaa required a letter of permission from the Bureau of Culture and Tourism to participate in the research. 44. Interview, respondent TM8, male Oromo elder, Turufe Kechema, 26 January 2010. 45. Interview, respondents OR2; TK2, male Oromo elder and kebele vice-chairman, Turufe Kechema, 13 January 2010; TF5, female Oromo resident and head of kebele women’s affairs, Turufe Kechema, 27 January 2010. 46. Hebo, Land, local custom and state policies; Abbas Gnamo, Conquest and resistance in the Ethiopian Empire, 1880−1974: The case of the Arsi Oromo (Brill, Leiden, 2014). 47. Joel Migdal, State in society: Studying how states and societies transform and constitute one another (Cambridge University Press, Cambridge, 2001); Tobias Hagmann and Didier Péclard (eds), Negotiating statehood: Dynamics of power and domination in Africa (Wiley-Blackwell, Chichester, 2011). 48. Since the mid-2000s, the government has promoted capital-intensive agricultural investments, particularly to the north of Turufe, but also in the area around Turufe itself. While these investments and urban expansion constitute important additional pressures on land in the area, these have only had a more recent impact and were not influential in the events discussed below. 49. Getachew Fule and Mesfin Tadesse, Ethiopian village studies: Turufe Kechema (Addis Ababa University & University of Oxford, Addis Ababa & Oxford, 1996). 50. Temesgen Gebeyehu, ‘A history of land measurement in Shashemene (Ethiopia), 1941–1974’, African Journal of History and Culture 1, 4 (2009): pp. 67–75. 51. Interview, respondent TM26, male Kembata displaced from Turufe, Shashemene, 29 January 2010. 52. CSAE, Ethiopian rural household survey 2004 (Centre for the Study of African Economies [CSAE], Oxford University, Oxford, 2004). 53. Interview, respondent TM1, male Oromo resident, Turufe Kechema, 31 January 2010. The OLF was briefly a member of the post-Derg transitional government, but left before the 1992 elections and was banned as a terrorist organization. 54. Interview, respondent TM17, male Kembata displaced from Turufe, Shashemene, 29 January 2010. 55. Record 3a, Letter from the Oromiya Bureau of Agriculture and Rural Development to the Oromiya Rural Land Administration Authority, date illegible. 56. Fule and Tadesse, Ethiopian village studies. 57. Interview, respondents TM3, male Oromo resident, Turufe Kechema, 28 January 2010; TM12, male Wolayita resident, Turufe Kechema, 14 January 2010; TM18, male Kembata resident, Turufe Kechema, 16 January 2010. 58. Interview, respondents TM23, male Tigrayan resident, Turufe Kechema, 17 January 2010; TM26. 59. Interview, respondents TM21; TM21, male Tigrayan resident, Turufe Kechema, 17 January 2010. 60. Interview, respondents TM12; TK5, male member of kebele Land Administration Committee, Turufe Kechema, 14 January 2010. 61. Interview, respondent TK5. 62. Interview, respondent TM26. 63. Interview, respondents TK5; TM4, male Oromo resident, Turufe Kechema, 28 January 2010. 64. Interview, respondent TM18. 65. Interview, respondents TM12, TM18. 66. Interview, respondent TM18, a Kembata who remained in Turufe. 67. Interview, respondent TM17, a Kembata displaced to Shashemene. 68. Interview, respondent TM4, an Oromo. 69. Interview, respondent TK5, an Oromo. 70. Interview, respondent TM26, a Kembata displaced to Shashemene. 71. Interview, respondent TM1, an Oromo. 72. Interview, respondent TM23, a Tigrayan. 73. Interview, respondents TM17; TM23. 74. Yohannes Gezahegn, Bizuayehu Ayele, Getachew Fule, and Mesfin Tadesse, Ethiopian village studies II: Turufe Kechema (WeD-Ethiopia, Addis Ababa, 2006). 75. Interview, respondent TM20, male Tigrayan resident, Turufe Kechema, 17 January 2010. 76. Interview, respondent TM1, an Oromo. 77. Interview, respondent TM17. 78. Interview, respondents TM17; TM18; TM24, male Kembata resident, Turufe Kechema, 19 January 2010. 79. Record 1e, Letter from the evicted farmers to Walta Information Centre, 14 October 2002; Record 1 f, Letter from the evicted farmers to the House of Peoples’ Representatives, 10 October 2002. 80. Interview, respondent OR1, senior expert, Oromiya Land Use and Environmental Protection Bureau, Addis Ababa, 17 November 2009; ONRG, ‘Proclamation to amend the proclamation nos. 56/2002, 70/2003, 103/2005’. 81. Record 1e. 82. Record 2a, Letter from the federal Ministry of Agriculture and Rural Development to the association of evicted farmers, 5 November 2005; Record 4c, Letter from the federal Ombudsman to the Oromiya and SNNP regional governments, 23 August 2007. 83. Record 2a. 84. Record 7a, Letter from the East Shewa zone administration to the Shashemene wereda administration, 9 March 2000. 85. Record 6a, Letter from Shashemene wereda administration to the East Shewa zone administration, 29 January 2002. 86. Selassie, ‘Ethnic federalism’, p. 92. 87. Record 2a. 88. Vaughan, ‘Ethnicity and power in Ethiopia’. 89. Abbink, ‘Ethnicity and conflict generation in Ethiopia’. 90. Tesfaye Tafesse, The migration, environment and conflict nexus in Ethiopia: A case study of Amhara migrant-settlers in East Wollega Zone (OSSREA, Addis Ababa, 2007). 91. Ibid., p. 86. 92. Tadesse Berisso, ‘Planning resettlement in Ethiopia: The experience of the Guji Oromo and the Nech Sar National Park’, in Alula Pankhurst and François Piguet (eds), Moving people in Ethiopia: Development, displacement and the state (James Currey, London, 2009), pp. 93–101. 93. Lavers, ‘Agricultural investment in Ethiopia’. 94. Vaughan, ‘Ethnicity and power in Ethiopia’. 95. Jon Abbink, ‘Ethnicity and conflict generation in Ethiopia: Some problems and prospects of ethno-regional federalism’, Journal of Contemporary African Studies 24, 3 (2006), pp. 389–413, p. 390. This is in line with Boone’s hypothesis that the scale of land conflicts is shaped by institutional settings, with land conflicts ‘bottled up’ at local levels under more localized land tenure regimes. See Boone, Property and political order in Africa. 96. Africa Confidential, ‘Oromia erupts’, Africa Confidential, 18 March 2016, <https://www.africa-confidential.com/article-preview/id/11594/Oromia_erupts> (4 March 2018); René Lefort ‘Ethiopia’s crisis’, openDemocracy, 19 November 2016, <https://www.opendemocracy.net/ren-lefort/ethiopia-s-crisis> (4 March 2018). 97. EHRCO, ‘140th Special report: Executive summary’ (Ethiopian Human Rights Council, Addis Ababa, 2016), p. 5; René Lefort, ‘The ‘Ethiopian Spring’: ‘Killing is not an answer to our grievances’’, openDemocracy, 9 September 2016, https://www.opendemocracy.net/ren-lefort/ethiopian-spring-killing-is-not-answer-to-our-grievances (12 September 2016). Given that the area around Turufe Kechema was among the main flashpoints of the protests in 2016 leading to a temporary breakdown in state control, this raises serious concerns about the situation in Turufe and the fate of the ethnic minorities there. Unfortunately, it has not yet been possible to conduct additional fieldwork. 98. Boone, ‘Property and constitutional order’; Boone, Property and political order in Africa; Kyed and Buur, ‘New sites of citizenship’. © The Author(s) 2018. Published by Oxford University Press on behalf of Royal African Society This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted reuse, distribution, and reproduction in any medium, provided the original work is properly cited

Journal

African AffairsOxford University Press

Published: Mar 29, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off