The role of ideology in Ethiopia’s war and peacetime

The study of the EPRDF’s use of revolutionary democracy as an ideological guiding star is a contribution to the understanding of how rebel parties practise politics after war, and how they may transform wartime ideas to serve peacetime political needs. Peacetime policies are not only influenced by the ‘ethos of the liberation ideology’ (Salih Reference Salih2003). As demonstrated by several articles in this collection on former armed movements, ideology and visions of state transformation (Curtis and Sindre Reference Curtis and Sindre2019), political identity and wartime ideas of rebel groups also matter for how the successor parties govern and mobilize, and which political issues become salient and possible to mobilize on in peacetime. This shows that ideology is not simply a part of the rhetorical repertoire or merely a pragmatic instrument, but that it matters in practical politics and in political behaviour and strategies.

The EPRDF’s use of wartime ideology to consolidate power in peacetime was facilitated by the peculiar history of the Ethiopian state and the way the war ended. Being a victorious rebel party in a society with a well-established and strong state has made the EPRDF able to navigate the changing political field in a particular way. This combination has created exceptionally favourable conditions for the party to monopolize power and use the incumbent position to legitimize its rule to secure continued hold on power (Lyons Reference Lyons2016), demonstrating the importance of the structure of the indigenous society for the success of insurgencies (Clapham Reference Clapham1998). Despite this peculiar setting, this article argues that the ideological choices the party has made remain significant.

As argued by Devon Curtis and Gyda M. Sindre (2019), ideology is not necessarily a strictly defined or all-encompassing system of thought but should at the least be a belief system that provides a guide to action, and a way in which actors or systems rationalize themselves. Or, as Francisco Gutiérrez Sanín and Elisabeth Wood (2014: 215) define it, ideology can be ‘a more or less systematic set of ideas’ that includes the identification of a referent group (ethnic, class or other social group), an expression of the grievances or challenges that the group faces, its goals (for example, political change), a programme of action and, possibly, a set of institutions with which to work for these goals. Revolutionary democracy has for the EPRDF fulfilled all the defining purposes of an ideology. It has served as a guiding and unifying vision, helped mobilize supporters and recruits, and prescribed a certain way of organizing in order to reach a defined goal.

The Tigray People’s Liberation Front (TPLF), the winner of the military struggle against the Derg regime and the creator of the EPRDF coalition, has – more than any of the other three coalition partners – provided the ideological direction of the EPRDF and the post-transition government (Tadesse and Young Reference Tadesse and Young2003).Footnote ¹ The TPLF was formed by a group of Tigrayan students in the north of the country in 1975. In the mid-1980s it was transformed into a Marxist–Leninist vanguard organization, and in 1989 it established the EPRDF. The aim of the TPLF’s ideology from the start was to transform society from a pre-capitalist to a socialist order, along Marxist–Leninist lines (Vaughan Reference Vaughan2011; Young 1997).

Marxist–Leninism and Maoism were the TPLF’s first ideological inspiration during the struggle. But its approach differed from the traditional proletarian understanding of class struggle. Instead of defining the working class as the referent group, the TPLF identified oppressed ethnic groups, termed ‘nationalities’, where the majority were peasants and not proletariat, as the referent group. It started off as a nationalist movement, fighting the cause of the Tigrayan people and for greater autonomy or independence for the Tigray region (Markakis Reference Markakis2011; Milkias Reference Milkias2003). The call for an independent Tigray and the exclusive focus on the Tigrayan cause were, however, controversial within the movement, and were later abandoned in favour of a solution which included all oppressed ethnic groups in the country. This implied that the state had to be transformed into a polity where all ethnic groups would be represented, and where the hegemony of the dominant Amhara ethnic group was defeated, in order to reach the ultimate goal of a socialist revolution.

The revolutionary democratic ideology’s main programme of action was therefore to introduce ‘self-determination’ for the nationalities, through autonomy for the various ethnic groups, implemented as ethnic federalism from 1991. Similar to other Marxist–Leninist organizations, this transformation was to be led by an enlightened elite, the vanguard, who would mobilize and organize the masses. As stated in the later EPRDF Statute, ‘the EPRDF is expected to play the role of a vanguard by bracing up its organizational capacity to lead the people in their efforts to raise their consciousness and organize themselves’ (EPRDF 2006). Democratic centralism was determining the internal processes within the movement. This meant that issues would be thoroughly discussed and evaluated, but once the leadership had decided, no one could question or object to the party line, leaving no room for dissent. By the use of gim gimma, an elaborate process of evaluation, criticism and self-criticism, defects and mistakes in members were reprimanded. In some of the party documents, revolutionary democracy is also termed ‘popular democracy’, based on communal collective participation and representation based on consensus (EPRDF 1993). The emphasis on the collective implied that members of ethnic groups and the peasantry were seen as a homogenous mass, speaking with one voice. This made divergent opinions irrelevant, and dissent was therefore not possible (Vaughan and Tronvoll Reference Vaughan and Tronvoll2003).

The ideas of a vanguard mobilization of the masses and communal collective participation led to a concrete set of institutions to organize the peasantry in the liberated areas of Tigray during the struggle and throughout Ethiopia after the war: the baito system, and later the kebele system.Footnote ² These were vanguard-led village assemblies taking care of local administration, land distribution and social reforms. They served the dual aim of mobilization and control (Aalen and Muriaas Reference Aalen and Muriaas2018a) and became key in the party and state apparatuses’ surveillance system in the post-war period (Pausewang et al. Reference Pausewang, Tronvoll and Aalen2002). In addition to these local political-administrative structures, from the early 1990s the EPRDF and affiliated parties gained control and established regional and local party structures all over Ethiopia, which have controlled the governments in all nine regional states of the federation to date (Aalen and Muriaas Reference Aalen and Muriaas2018b; Abbink 1995). The central principle of federalism, regional self-rule, has therefore been severely compromised by centralized party power (Aalen Reference Aalen2002), an issue further elaborated on in the discussion on federalism in the next section.

Despite its clear organizational direction, the otherwise vague character of revolutionary democracy as an ideology has created internal tension, contributing to major political crises in the EPRDF’s post-war governance. These crises, although disruptive, have, however, opened up the potential for the party to redirect its policies and adapt to new circumstances. The crack in the TPLF in 2001, which led to the dismissal of a number of TPLF members from the party organization and the government, was about how to interpret revolutionary democracy – and what consequences this ideology should have for practical policy (Milkias Reference Milkias2003). The split paved the way for Chairman and Prime Minister Meles Zenawi to take more of the party’s power in his own hands, allowing him to redirect revolutionary democracy towards the developmental state. This reorientation was driven by policy needs, which naturally changed from the transition period to the consolidation period. The EPRDF’s key priority during the first decade was to establish its political base – essentially through ethnic federalism. But, as pointed out by Christopher Clapham (Reference Clapham2017: 4), ‘this was ultimately counter-productive to the development enterprise’. A new policy was therefore needed for the consolidation period, a policy which could facilitate and not obstruct the consolidation of a cohesive national economy.

Ethnic federalism: revolutionary democracy in the transition period

The outline of the EPRDF’s ideology has demonstrated that the perpetuation of ideology after war does not imply static policies, but can instead facilitate policy change. Equally, ideological commitment does not prevent pragmatic policy implementation. Instead, ideology and ideological battles can be used to justify policies which are directed by political needs in a changing post-conflict situation. The EPRDF’s implementation of ethnic federalism since taking power in 1991 is an illustration of how an ideologically grounded policy has served the interests of new rulers to consolidate national power. When the EPRDF ousted the Derg regime, it had – like any post-insurgency movement taking power – to get rid of the legacies of the ancient regime, to create a clear break with the past and to demarcate a new role for itself. Ideology was important in this process. By framing the change of power in an ideological way, it could add legitimacy, a source of solidarity and a tool for discrediting possible dissent. Revolutionary democracy served all these purposes. Drawing on the ethnic-based opposition to the old regimes of Emperor Haile Selassie (1931–74) and the Derg (1974–91), the party emphasized that the best way of organizing was through ethnic groups – in their concept, ‘nations and nationalities’.

Though clearly ideologically motivated, ethnic federalism served pragmatic needs by legitimizing the replacement of members of the old elite who were often of Amhara origin. The adoption of a decentralized form of government is not surprising in a post-conflict setting, where a rebel movement takes power. As shown in other similar contexts, wartime social networks put successful insurgent groups in an excellent position to mobilize support in subnational elections and to reward those loyal to the organization. In the African context, governments that are in a secure position due to a one-sided victory after war and have great confidence in winning elections – exemplified by Ethiopia and Uganda – are the most eager to introduce and implement decentralizing reforms (Aalen and Muriaas Reference Aalen and Muriaas2018b). The reorganization into the nine ethnic-based regional administrations, all with elected legislatures and autonomous executive and judicial powers, was performed together with the mobilization of new ethnic elites within every community in the Ethiopian state. These joined local branches or affiliates of the ruling EPRDF coalition. Thus, the establishment of the federal system and the locally based party organizations were efficient avenues for the regime to penetrate new areas and create new elites loyal to the party in every corner of the vast state territory (Aalen and Muriaas Reference Aalen and Muriaas2018b). In many communities, young and inexperienced but loyal local leaders took the place of Derg officials and traditional leaders (Aalen Reference Aalen2011).

The system of ethnic federalism was implemented alongside liberalization of the economy and introduction of a multiparty system, which satisfied the interests of the international community of donors and global powers after the end of the Cold War. This ensured Ethiopia’s position as an important ally of the West in the fragile and unstable Horn of Africa region and facilitated a steady flow of aid to reconstruct the country and mitigate food insecurity. As pointed out by Jean-Nicolas Bach (Reference Bach2011), the political and economic opening during the 1990s did not imply that revolutionary democracy was abandoned. Instead, the formally democratic institutions became both revolutionary and liberal, as liberal institutions were ‘appropriated by the EPRDF to paradoxically support revolutionary democracy’ (Bach Reference Bach2011: 642). Similar to other electoral authoritarian regimes (Schedler Reference Schedler2006), elections became an important tool for maintaining the EPRDF’s hegemony and dominant position. Through the intimidation of opposition and pressure on voters, the party managed to win every election from the first local polls in 1992 (Pausewang et al. Reference Pausewang, Tronvoll and Aalen2002; Tronvoll and Hagmann Reference Tronvoll and Hagmann2012), ensuring the hegemony of the revolutionary democratic project. Privatization was also given a revolutionary democratic twist. Key economic sectors were dominated by companies connected to the ruling party, and banking, energy, telecoms and transport remained predominantly state-owned. Most notably, land remained state-owned under the rationale that the people own the land, the government represents the people and thus the government administers the land on behalf of the people. This was a clear expression of ideological conviction, prompting international pressure from the World Bank and the IMF to privatize land ownership (Lie and Mesfin Reference Lie and Mesfin2018).

Although reinforcing ruling party dominance, the establishment of regional and local government structures and the holding of what was formally presented as multiparty elections facilitated the participation of large numbers of the population in local affairs and led to the inclusion of previously marginalized ethnic groups into the body politic of Ethiopia. This blend of ethnic federalism and mass participation has added a dimension of inclusiveness in an otherwise repressive autocracy (Aalen and Muriaas Reference Aalen and Muriaas2018b). A side effect of ethnic politics was the thriving of ethnic entrepreneurs, aiming to take advantage of the new system. The political mobilization of ethnic cleavages led to increased interethnic tension and claims for new separate ethnic-based administrations and parties (Aalen Reference Aalen2011; Donham et al. Reference Donham, James, Kurimoto and Triulzi2002; Kefale Reference Kefale2013).

From the end of the 1990s, the EPRDF actively used references to the revolutionary democratic ideology to try to contain ethnic political mobilization. The promotion of the particular interests of ethnic groups was increasingly framed as a threat to national development and, ultimately, to revolutionary democracy. Local leaders challenging the central or regional governments were called ‘rent seekers’ or in even more ideologically laden terms, ‘narrow nationalists’. To restrain what the EPRDF gradually came to see as problematic ethnic entrepreneurship, it introduced administrative reforms, geared towards strengthening of the non-ethnic local administration. This implied devolving power to the local level, through the so-called District Level Decentralisation Programme. From 2001, the local woreda (county) administrations received both larger budgets and a greater number of qualified personnel than before, largely bypassing the ethnic regional administrations (Dickovick and Gebre-Egziaber Reference Dickovick and Gebre-Egziaber2014).

Similar to the unintended and destabilizing consequences of ethnic federalism, the calculated opening for more multiparty elections unexpectedly challenged the EPRDF hegemony during the national elections of 2005. While the ruling party never intended opening up for real competition, a limited liberalization before the 2005 election led to unprecedented support for a pan-Ethiopian opposition party, Coalition for Unity and Democracy (CUD). The opposition took a third of the parliament’s seats. The EPRDF’s reaction was uncompromising. The government crackdown on post-election protesters led to the deaths of more than 200 people and the imprisonment of thousands of youth in addition to the whole opposition leadership as well as a number of civil society and independent media actors (Lefort Reference Lefort2010). Since the 2005 elections, the space for political life and discussion has been severely restricted. Following the enactment of a range of laws regulating civil society, independent media and opposition parties, as well as a new and excessive anti-terrorism law, voices outside the ruling party’s immediate control have hardly been heard (Tronvoll and Hagmann 2012).

The developmental state: revolutionary democracy in the consolidation period

The evolution of revolutionary democracy to emphasize the overarching aim of economic development has happened gradually since the beginning of the 2000s. Since then, the emphasis on democratic legitimacy and the liberation of the ‘nations and nationalities’ have been replaced by a focus on national economic delivery or performance legitimacy. In this way, the initial programme of action of the rebel group’s ideology has been fundamentally changed. In hindsight, when re-reading the works on the original ideological thinking of the TPLF/EPRDF, the gradual move from ethnic federalism to the developmental state model is not as surprising and contradictory as at first sight. We can see that ideologically, self-determination and ethnic federalism were not end goals in themselves for the TPLF (Aalen Reference Aalen2011; Eyob Reference Eyob2018; Vaughan and Tronvoll Reference Vaughan and Tronvoll2003). They were instead seen as tools to solve the economic iniquity created by Amhara dominance and were steps towards greater national integration and cohesion once economic access was democratized and expanded. So the logic was: first solve the ethnic question, then move on to the real issue – the economy.

In this perspective, the turn towards the developmental state model from the early 2000s can be seen as a logical, and ideologically grounded, next step in the path towards economic and political integration. Since the early 2000s, the EPRDF started exploring the developmental state model, drawing on the experiences of the East Asian ‘Tigers’, South Korea and Taiwan, the classical developmental state of Japan and, not least, the Chinese growth model (Fourie Reference Fourie2015). Common to all these experiences is that the state has had an active role in the economy, defining the priorities of both public and private actors. In the EPRDF’s view, this fitted well with the revolutionary democratic ideology. Just as the party plays a vanguard role in mobilizing the masses, the state should play the vanguard role in the economy, intervening and deciding the priorities for the private sector. The private sector should be supported only when it follows the principles decided by the party-state.

Even though it is possible to discern clear ideological reasoning for the EPRDF’s reorientation from ethnic federalism to the developmental state, it is equally sure that this policy change was beneficial for the changing political needs of the ruling party. In the move from a transition to a consolidation period, political demands change. Ethnic federalism acted as a door-opener for the EPRDF in all corners of the country in the transition period. Instability caused by ethnic entrepreneurship and local ethnic conflicts, in combination with threats from a nationally oriented opposition, urged the party to consolidate central power further. In addition, the war with Eritrea from 1998 to 2000 had revealed strong nationalistic sentiments in the Ethiopian population and had rehabilitated many national symbols that had been dead since the end of the Derg period (Clapham Reference Clapham2017; Tronvoll Reference Tronvoll2009). This made the ruling party look more actively for other policy options and to time its move to the next step in the revolutionary democratic programme of action – that of developing a cohesive and strong national economy. The split in the TPLF in 2001, partly because of disagreements about the war, allowed Meles Zenawi to clear out his challengers and come out as the sole definer of EPRDF policies, including the economic policy. Furthermore, external events had an impact. Though primarily defined by domestic and internal events, certain changes in the international context aided the EPRDF’s change of policy. The emergence of China as a global power, in combination with aid donors increasingly occupied with promoting social and economic progress before liberal democratic values, made the international community highly receptive to the Ethiopian government’s apparently effective, albeit autocratic, growth model (Lie and Mesfin Reference Lie and Mesfin2018). The EPRDF’s promotion of the developmental state was therefore very much in line with the global zeitgeist.

While the overall aim of revolutionary democracy was maintained, the reorientation from ethnic federalism to the developmental state implied that the EPRDF changed its ideological programme of action from representation of oppressed nationalities to that of fighting for economic development. Its slogan turned from ‘liberation of the nationalities’ to ‘poverty is our enemy’ (Bach Reference Bach2011). This also had implications for the party’s ideological definition of the referent group. The original and relatively clearly defined referent group of ‘oppressed nationalities’ was replaced by a vaguer and much wider group among the population, whose interest was in fighting against the ‘rent seekers’ and other unproductive elements that were obstructing the building of the developmental state. A drawback with a vaguer definition of the referent group was that it became unclear to the constituencies who the EPRDF was now fighting for. Was it the successful investors and farmers who had managed to contribute to the national economy, or was it the small-scale farmers who were still struggling to make a living and fighting poverty? As argued by Lefort (Reference Lefort2012), the EPRDF’s new policy implied a reorientation towards elite ‘model farmers’ who were using new agricultural inputs and technologies, leaving behind the focus on subsistence peasants, who had been the main constituency since the days of struggle. The move from transition to consolidation, from ethnic federalism towards the developmental state, also had implications for the way the EPRDF mobilized and organized its constituencies. After 2005, the ruling coalition made concerted efforts to enlarge and strengthen the vanguard by the mass recruitment of members to the party. While the overall frame and origin of directions continued to be the revolutionary democratic ideology and the vanguard mode of organization, the mass base of the party was expanded and diversified. The new vanguard were model peasants and women, mass organizations (women’s and youth leagues) and party cells at the local level.

The new revolutionary democratic programme of action for development also led to changes in the organizational structures of the party and the state. A more fine-meshed political-administrative structure was established at the grassroots level, below the original kebele system. This has gradually been formalized and presented as Ethiopia’s ‘development army’, linked to the EPRDF vision of creating a developmental state. Its official purpose was to reach all the way down to the family level to organize communities into small groups charged with implementing local elements of national development policies in a harmonized manner (Khan et al. Reference Khan, Faguet and Ambel2015). Within each kebele, rural households were grouped into development armies of 25 to 30 families. These were further divided into groups of four to five families, the so-called one to five networks, which were led by a ‘model family’ that in turn led the rest ‘in the acquisition of skills and attitudes conducive for the implementation of official policy’ (Khan et al. Reference Khan, Faguet and Ambel2015). The establishment of these structures has entailed a dramatic increase in the presence of the party and the state at the local level. They are predominantly led by ruling party cadres and are actively used to mobilize in pre-election periods, to survey the population and to intimidate any kind of opposition, as demonstrated by Lovise Aalen and Kjetil Tronvoll (2009). Although membership is presented as voluntary, refusal to join the groups is seen as an expression of an anti-government stance (Emmenegger Reference Emmenegger2016: 274).

The ideologically founded organizational models are also reflected in the major strategic plans of the government. In the efforts to firm up national development policies, the first National Growth and Transformation Plan was launched in 2010. In this five-year plan, economic and developmental goals were defined, with the aim of making Ethiopia a middle-income economy within the next two decades. In order to achieve this, vanguards across all kinds of institutions, within every economic sector and societal subgroup, have to play an active role. Women and youth are seen as particularly important vanguards in a range of sectors and activities (Federal Democratic Republic of Ethiopia 2010: 209). The plan emphasizes the need to coordinate between the vanguards of the political leadership, the civil servants and the public ‘to be able to address the rent-seeking political economy and ensure the hegemony of a developmental political economy’ (Federal Democratic Republic of Ethiopia 2010: 47). This implies that the technocratic elite within the civil service, normally seen as independent of vested interests in the classical developmental state (Johnson Reference Johnson1982), ought not be autonomous, but integrated in the ruling party.

Another consequence of the EPRDF’s revolutionary democratic interpretation of the developmental state is that the call for ‘developmentalism’ has been actively used in curbing political dissent, including TPLF dissidents expelled after 2001, but also after the controversial elections in 2005, when the ruling party was for the first time seriously challenged by the opposition (Aalen and Tronvoll Reference Aalen and Tronvoll2009). The nationally oriented opposition parties have since the early 1990s been labelled as supporters of Amhara dominance, a dominance which the EPRDF has fought since the struggle started in 1975. In this way, the EPRDF portrayed the opposition as an existential threat to the original goal of revolutionary democracy – to defeat the oppression of the non-Amhara. After the 2005 elections, however, political opponents, including independent civil society actors, were more frequently described as ‘neoliberals’ and ‘rent-seekers’, meaning that they represented the unproductive elements of society which immorally and illegitimately make use of public resources, thereby inhibiting Ethiopia’s escape from poverty. The Societies and Charities Proclamation of 2009, which makes it illegal for civil society organizations working in the area of human rights to receive more than 10% of their funding from abroad, is also justified in these terms.Footnote ³

The 2015/16 revision of the EPRDF programme expresses the same disregard and suspicion of any opposition. The programme describes those who are dissatisfied with the aspirations and goals of revolutionary democracy as parasitic elements who make no real contributions to development: ‘Although these forces profess to stand for democracy, they are unable to practice their avowed slogans except as façade covering for their deeds. To carry out the objectives of revolutionary democracy embraced by the EPRDF, it is mandatory to fight against these forces’ (EPRDF 2016). In this way, the ruling party also makes it clear that the only way for Ethiopia to reach more advanced development is through the continued power of the EPRDF. This shows not only how the EPRDF has attempted to make itself irreplaceable, but also how the party has not made any attempt to make its new developmental policies take root among those outside the circles of the ruling party, in the opposition parties, private businesses and civil society. The EPRDF has therefore failed to incorporate significant national elites in its transformative project and therefore lacks another crucial element of the classical developmental state models – a national consensus on central developmental goals (Johnson Reference Johnson1982).

The protests and their aftermath: revolutionary democracy reformed or undermined?

Throughout the post-war period, revolutionary democracy has effectively been used to justify but also conceal policy changes, and has been called upon in times of crisis to manage dissent and reinforce cohesion. The protests of 2015 and 2016 in the two largest regions of Ethiopia and the following rupture within the EPRDF coalition are, however, signs of revolutionary democracy losing some of its power. The protests were first spurred by resistance to the national government’s acquisition of land under regional jurisdiction and the way the national authorities had privileged national and international investors’ access to land. This shows that the ruling party had underestimated the effects of the inherent contradictions between a decentralized federal system, which means regional control of land, and a centralized developmental state, implying national planning and control of crucial developmental assets, including land. The EPRDF may also have overestimated the ability of the party to impose consensus across the member parties representing the largest regions of the federation.

The Oromo protests that broke out in November 2015 were triggered by a new plan from the government of extending the capital into the areas surrounding Addis Ababa. It was perceived as an annexation of Oromo territories by the federal government, and a way of undermining their right to govern their own affairs. Oromia has been one of the regions hardest hit by expropriation of land from farmers to domestic and foreign investors, in the name of national development. Factories and commercial farms were therefore targeted by the protesters. Protests in Amhara region were also triggered by the issue of control of land. They erupted in August 2016 in support of a committee requesting a referendum on a decision made in 1991 to include an area of Amhara region into Tigray region in the north. When the federal government attempted to arrest the chairman of this committee, the protests erupted. Protests in Oromia continued also in 2016, escalating in the aftermath of a traditional religious festival in Bishoftu town in October 2016, when security forces killed around 50 festival-goers. All in all, conservative official numbers are that 669 were killed and 26,130 detained across the country since protests began in November 2015.Footnote ⁴ The government introduced a nationwide state of emergency on 9 October 2016, finally lifted on 4 August 2017, prohibiting all kinds of public gatherings and giving wide-ranging powers to a national command post dominated by defence and security.

Although the protests in Ethiopia’s two largest regions were not coordinated, they represented a number of shared grievances. In essence, they asked for genuine regional self-government, both in political and economic terms. In this way, the flaws of the federal system, exacerbated by the centralizing developmental policies, were at the heart of the protests (Lefort Reference Lefort2017). Apart from the land issue, they complained about the narrowing of the political space, for opposition politics in general, and against the TPLF/Tigrayan domination and the power asymmetry within the EPRDF in particular. They were also about shared economic grievances. In the context of rapid economic growth, rising inequality, and unemployment among a large population of educated youth, protesters asked for a larger share of economic benefits and jobs.Footnote ⁵

The protests led to internal self-evaluation within the ruling party, and an attempt to reinforce the principle of revolutionary democracy and democratic centralism to ensure a shared analysis and solution to the problems. As during previous crises – such as in 2001 when Meles Zenawi was challenged by his own party comrades – the EPRDF launched a massive renewal or reform process. This has traditionally entailed reshuffles in the party and substantial re-education programmes among the party’s rank and file. This time, the EPRDF launched what it termed the ‘Deep Renewal’. In line with revolutionary democratic ideas, the EPRDF leaders identified corruption and rent-seeking, ‘narrow ethno-nationalist’, ‘anti-democratic’ outlooks and bad governance as the reasons for the protests.Footnote ⁶ Officials in the whole party apparatus, at all levels, were involved in self-evaluation processes in the following year. Arrests were made on charges of corruption, involving around 100 low- and middle-rank government officials. Thus far, this followed the patterns seen in similar situations in the past.

A significant departure from previous reform processes within the EPRDF is that while the party agreed on the reasons behind the protests, there was no shared view on the right remedy. For the first time in the EPRDF’s history, there was no consensus on how to address chauvinism/narrow nationalism, anti-democratic attitudes or bad governance.Footnote ⁷ It was thus no longer possible for a unified vanguard to lead the party and the masses as one coherent body. Instead, a new and assertive Oromo wing leadership put forward claims of genuine regional autonomy and ‘economic revolution’ instead of revolutionary democracy (OPride Reference OPride2017). This was an expression of a party wanting to diminish its deficit of political legitimacy, which the Oromo party had suffered from since it was created by the TPLF during the last days of the civil war. Emboldened by the anti-TPLF sentiments among the protesters, the Oromo leadership wanted to show that Oromo politics was no longer automatically aligned with that of the TPLF. In autumn 2017, a new alliance between the Oromo and the Amhara wings emerged. In a meeting in the regional capital of Amhara, the Amhara and Oromia regional presidents made statements expressing strong pan-Ethiopian solidarity. Although their speeches did not attack the TPLF directly, they said bluntly that the current political system fostered ‘a hate environment’, but from now on, ‘no group’ should come in between their newly forged alliance (Addis Standard 2017).

The failure to impose cohesion after the protests is not only a product of the contradiction between regional and central policy dynamics. It is also a delayed outcome of the power changes that took place after prime minister and TPLF leader Meles Zenawi died in 2012. Meles Zenawi was the party’s main ideologue and proponent of revolutionary democracy, instrumental in the TPLF’s move towards Marxism in the mid-1980s (Young Reference Young1997), the liberalization and launch of ethnic federalism during the transition in the early 1990s, and also behind the turn towards the developmental state in the mid-2000s (De Waal Reference De Waal2013; Zenawi Reference Zenawi2012). Meles Zenawi’s successor, chairman of the Southern Ethiopian People’s Democratic Movement, Hailemariam Dessalegn, was seen as a weak, compromise candidate. The new prime minister had no fighter background and represented southern Ethiopia, an area that had never before had a prominent seat at the national political table. With the death of Meles Zenawi, TPLF lost its unquestionable dominant position within the party coalition. But its continued control of the leadership of the security and defence forces and its manipulation of the new prime minister allowed the TPLF to hold its position for several more years. The protests and the EPRDF rupture, however, ultimately undermined TPLF’s monopolizing power. In this way, it is possible to argue that recent events are expressions of what normally happens when the dominant leaders of a civil war die or are eradicated. When the political and military leader of a victorious rebel group dies, the legacies of war are altered and become less of an asset for the remaining movement (Gandhi and Przeworski 2007). In Ethiopia, although delayed, this may bring opportunities for new actors to take the stage and redefine post-war politics into post-TPLF politics.

The protests and the rupture in the EPRDF created unforeseen and spectacular changes to the Ethiopian governance in 2018, the outcomes of which we do not know yet. As a consequence of the EPRDF’s Deep Renewal reform process, Hailemariam Dessalegn resigned in January. This was followed by a surprise appointment of one of the reformers within the Oromo wing of the ruling party, Abiy Ahmed, as the party chairman and prime minister. His first months in tenure included a range of reform moves, which appeased protesters as well as opponents. He pardoned and released political prisoners, introduced an amnesty law, lifted the state of emergency and exposed systematic human rights abuses committed by the government (Tronvoll Reference Tronvoll2018). Finally, he opened peace talks with Eritrea. His public speeches focused on national reconciliation and unity, leading to new hopes of reform in the country. It is still unknown, however, where the new leadership stands on major ideological issues linked to revolutionary democracy. Abiy Ahmed is himself a product of the party and its revolutionary democratic practices and organizational structures, having held positions in the national security and intelligence system in the past. Whether he will implement reforms that actually open up for genuine political pluralism or if he will struggle to maintain the EPRDF’s monopoly on power is still not known. What is clear, however, is that revolutionary democracy no longer provides the ideational resource for the party to conceal policy contradictions and conflicts.