INTRODUCTION
In 1974, the long reign of Emperor Haile Selassie I was brought to an end by a succession of protests, mutinies, and demonstrations that evolved into ‘sub-Saharan Africa's one undoubted social revolution’.Footnote 1 A quick succession of changes, including the establishment of a military government, the abolition of the monarchy, and wide-ranging land reform, radically reconfigured Ethiopia's political structure and social relations. In light of the extensive violence of the following years, many Ethiopians soon forgot that the radical changes that swept across the country in 1974 had initially been celebrated as Abyot yale menim dam – a revolution without [the shedding of] any blood. Amid the turmoil generated by the Ethiopian revolution, unprecedented political opportunities emerged: initiatives were launched to empower hitherto marginalised groups, fierce public debates took place over the path to a more equitable society, and new political movements were formed. Yet, accompanying and punctuating these changes were increasing instances of armed confrontation and state violence against dissenting groups. In urban Ethiopia, such conflict culminated in the Red Terror of 1977 and 1978, in which the country's new military government, known as the Derg, sanctioned and facilitated the systematic extrajudicial repression of opposition organisations. A decade after its end, a prominent Derg official described the Red Terror as ‘violence which had no parallel in Ethiopian history’.Footnote 2
It is certainly true that even in a century of Ethiopian history that has seen violent colonial occupation, border wars, decades of civil war, and repeated famine, the Red Terror stands out as one of the most egregious instances of systematic human rights abuses.Footnote 3 Tens of thousands were murdered, the precise number elusive because killings were often secretive and decentralised.Footnote 4 Tens of thousands more were extrajudicially detained and heavily tortured. Hundreds of thousands found their lives transformed by the period's violence: losing children, siblings, relatives, or friends, or living with physical and psychological impairments caused by torture. Many were forced into an exile so common that it contributed to the formation of a permanent Ethiopian diaspora, while those who remained had to acquiesce to an authoritarian regime under which political freedoms collapsed alongside real wages. The political opportunities and widespread political activism of the Ethiopian revolution were undone and eclipsed by state violence during the Red Terror. The result was a politics far removed from the ideals of 1974 and more closely resembling that of many other African states in the wake of the 1973 oil crisis, marked by authoritarian rule, economic mismanagement, and ethnonational insurgencies.
The Terror gained infamy for the forms as much as for the scale of its violence: bodies of executed regime opponents were displayed on the streets for days, bearing placards denouncing them as ‘counter-revolutionaries’. Parents were prohibited from mourning their slain children, prevented from performing burial rites, and in some recorded instances – where officials had required more than one bullet for an execution – required to pay for the surplus ammunition used to kill their children.Footnote 5 The Red Terror was overseen by the Derg, whose chairman Mengistu Haile Mariam publicly called for large-scale extrajudicial killings of those who opposed his regime's leadership of the revolution.Footnote 6
The scale and brutality of the Red Terror reflected the threat to military rule posed by civilian Marxist opposition parties, which disputed the Derg’s right to lead the revolution and demanded an immediate transition to civilian rule. In 1976, the most significant of these groups, the Ethiopian People's Revolutionary Party (EPRP), augmented its small-scale rural insurgency with an urban guerrilla tactic. This strategy revolved around the targeted assassination of regime members and alleged collaborators, especially supporters of the rival All Ethiopia Socialist Movement (Meison). Widely remembered as ‘the white terror’ – an infelicitous epithet, as the Marxist EPRP also regarded its violence as revolutionary and hence as ‘red’ – this anti-regime violence raised the stakes for the Derg, making the dismantling of the EPRP's cell structure, social bases, and operational capabilities a priority rivalling the fight against Eritrean rebels in the north.Footnote 7 In a revolutionary situation perceived as a dangerous atrocity environment by many of its leading figures, Red Terror was conceived, organized, and justified by Derg radicals as a matter of public security, revolutionary progress, and national interest.
How the Ethiopian Red Terror should be understood and explained remains heavily contested. The historiography of the Ethiopian revolution has at times served as a platform for the continuation of political rivalries long settled on the ground, and continues to be replete with arguments intended to either justify or denounce particular actions or groups.Footnote 8 Yet despite the diversity of viewpoints within this literature, the analytical areas of focus and the narrative frameworks employed to explain revolutionary violence have been surprisingly similar. Three main concerns or approaches dominate the scholarship: first, a growing literature examines the dynamics of the student movement, which grew in size and militancy through the 1960s, creating divisions and factions that energised the Red Terror.Footnote 9 Second, several analyses have concentrated on the party politics and fragmentation that grew out of this student movement, dominated by the EPRP and Meison.Footnote 10 And third, some studies have traced the influence of power struggles within the Derg on the evolution of Ethiopia's revolutionary violence.Footnote 11 The few studies that have treated the Red Terror as a subject in its own right, rather than merely as a violent aftereffect of the revolution, have combined all three of these approaches in an attempt to make sense of its violence.Footnote 12
These three foci are important and indeed necessary for explaining the Ethiopian Red Terror. However, they are also insufficient. Their common limitation concerns the level of analysis at which they operate: all three approaches presuppose the precedence of affairs of state or of political movements that sought to revolutionise or capture the state apparatus, over alternative arenas of agency, meaning, and tension.Footnote 13 The result has been the assumption of a top-down model of state terror, whereby the meanings of violence originate only with actors and conflicts at the ‘political centre’ and all local violence is treated as reducible to these actors’ decisions and interests. Where the sheer diversity and lack of discernible logic defining the terror on the ground preclude such reasoning, analysts have fallen back on a common analytical binary identified by Stathis Kalyvas.Footnote 14 If violence cannot be explained by macro-political tensions and interests, it must be senseless and anarchic, the outworking of an unintelligible Hobbesian chaos.Footnote 15
The persistence of such binary approaches to violence in the historiography of the Ethiopian revolution is both unfortunate and puzzling in light of insights developed by scholars working on conflicts elsewhere on the African continent. In contexts as diverse as the Kenyan Mau Mau Emergency, the 1972 reprisals in Burundi, the Angolan civil war, and the Rwandan genocide, historians have highlighted the extent to which political violence was shaped by the synergy of actors that operated at different ‘levels of analysis’ – from national government to local neighbourhoods – but within shared violence-facilitating institutional settings. Footnote 16 In unfortunately characteristic isolation from wider developments in African Studies, historical work on the Red Terror in Ethiopia lags behind these approaches, limiting itself instead to the reproduction of interpretative frames within which local perpetrators and institutional contexts of violence can be treated as subsidiary and interchangeable.Footnote 17 Nor are such frames limited to academic scholarship: in 2010, Ethiopia's most significant memorial to the period's violence, the ‘“Red Terror” Martyrs’ Memorial Museum’, opened in Addis Ababa's central Meskel Square. Its permanent exhibition represents the Red Terror as a straightforward if particularly brutal instance of state violence, designed and executed by the Derg and its security apparatus against a young generation that agitated for greater freedom and justice in Ethiopian society. There is no sense in this narrative that political violence in revolutionary Ethiopia might have sprung from other types and arenas of antagonism or have been made possible by concomitant social and institutional changes, nor indeed that large numbers of civilians became complicit in the performance of this violence.
In order to counter such narratives, this paper demonstrates how revolutionary Ethiopia's political violence was shaped by agents, by institutional contexts, and by social tensions in milieux that have hitherto been overlooked or mistakenly regarded as reducible to established interpretative frames. It was the negotiated interaction between actors operating in different spheres and at different levels of influence, rather than merely an imposition of state violence ‘from above’, that shaped and defined the Red Terror. In urban Ethiopia, these interactions were particularly concentrated and consequential within two settings: on the one hand, in the labour movement with its associated struggle over control of trade unions, and on the other in the newly formed local administrative structures of the Ethiopian state in urban areas, the Kebele. As this article argues, the histories of trade unions and Kebeles, and of the diverse actors who worked within these institutions, need to occupy a central role in our understanding of political violence in revolutionary Ethiopia.
Source materials for examining these histories are as rich as they are heterogenous: the extensive archives of the Confederation of Ethiopian Labour Unions (CELU) are yet to be brought to bear on the historiography of the Red Terror, as are the surprisingly revealing files in the archive of the East German Ministry of State Security, which collaborated closely with the Derg from 1977.Footnote 18 This article triangulates these archival materials with oral interviews as well as with materials familiar to scholars of the Ethiopian revolution – from government newspapers and radio broadcasts over clandestine pamphlets to government proclamations and diplomatic correspondence – to rethink the history of the Ethiopian Red Terror.
Read alongside each other, these sources suggest the need to rethink the history of revolutionary Ethiopia's political violence in at least the following three ways, explicated throughout this article: first, the scale, forms, and targets of the Red Terror were closely tied to developments within trade unions and Kebele administrations, institutions which shaped this violence whilst also being radically and consequentially reshaped by it. Second, much of the period's political violence developed in marked continuity with pre-existing patterns of tension and conflict ‘from below’, among actors and groups that have not been accorded sufficient attention or recognition in the historiography. And finally, the history of the Red Terror needs to be recognised as the history of conflicts, synergies, and negotiated alliances between these actors and the more familiar cast of protagonists in the Red Terror, including the leaders of the Derg. It was through these shifting alliances, marked by the mutual appropriation of the other's interests and resources, that the large-scale violence of Ethiopia's early revolutionary years was produced, with lasting consequences for social and state-society relations in Ethiopia.
REVISITING THE BERHANENA SELAM KILLINGS
The perhaps best-known killings of the Red Terror offer a revealing vantage point onto the need to rethink the period's violence. The events surrounding the extrajudicial killing of nine workers of the Berhanena Selam Printing Press are widely recounted: On 23 March 1977, the workers were arrested at their workplace in Addis Ababa's Arat Kilo neighbourhood before being detained, tortured, and, later that night, shot dead. Among the executed was Daro Negash, a mother of eight who was due to deliver her ninth child within weeks. Their remains were found in a riverbed the following morning. Responsible for the assassinations was an Abyot Tebeka [‘revolution defense’] squad commanded by Kebele chairman Girma Kebede, an enthusiastic participant in the Red Terror. Even in revolutionary times, the killing and inhumane disposal of a heavily pregnant mother triggered such public outrage that Derg leaders felt compelled to distance themselves from this violence; ten days later, Girma Kebede and five of his aides were themselves publicly executed.Footnote 19
In the historiography, the murder of Daro Negash and her colleagues is held up as the paradigm case of ‘gratuitous killings by psychotics’ set loose by the Terror, an act betraying a mindset that ‘only a psychiatrist could fathom’.Footnote 20 Maaza Mengiste's novel Beneath the Lion's Gaze sees a protagonist describe the murder as the epitome of the Terror's senseless violence: ‘All she did was work at a printing press [and they] killed her. A pregnant woman!’Footnote 21 Since the killings of Daro Negash and her colleagues do not seem to fit with the known rationales for political violence in revolutionary Ethiopia, they are instead ascribed to the realm of individual psychopathology.Footnote 22
Yet when the background to the Berhanena Selam killings is more closely examined, palpable rationales for this violence begin to appear, pointing to the importance of understanding arenas of agency and contestation that have until now featured insufficiently in the historiography of the Red Terror. For whereas it is true that Girma Kebede may have had personal reasons to embrace his role with such abandon – rumours abound in Addis Ababa to this day about his traumatic childhood, social anxiety to fit in, and experience of having been bullied at school – he was also following general directives and acting on sound information when he ordered his squad to carry out the executions. Upon closer inspection, the Berhanena Selam killings illustrate not so much the empowerment of irrational psychopaths as the aggregate logic of the Red Terror, shaped by the interplay between local information and initiative on the one hand and regime policy on the other. For whereas in public memory Daro Negash's identity as a mother and employee has rendered her other identities invisible, her executioners were well aware of her varied roles and activities. The Berhanena Selam killings were carried out in line with a clear, if horrific, political logic.
In the years preceding her murder, Daro Negash had emerged as an important trade union leader at the Berhanena Selam printing press. She was also a militant activist and Central Committee member in the clandestine Ethiopian Workers’ Revolutionary Union (ELAMA), an affiliate of the EPRP. Eighteen months earlier she had signed a Confederation of Labour Unions (CELU) resolution condemning the policies of the Derg and threatening a general strike.Footnote 23 Subsequently, she had ensured that workers of her union would be among the few who honoured the strike call.Footnote 24 Daro Negash's residence had hosted secret ELAMA Central Committee gatherings, and she had previously been arrested for disrupting a meeting that sought to organise a pro-regime industrial federation.Footnote 25 All victims of the 23 March 1977 murders had been denounced in a workers’ discussion forum – a common event across Ethiopian workplaces during the Red Terror, designed to expose regime opponents among workers and workers’ representatives – by two colleagues, who had summoned Girma Kebede and his Abyot Tebeka squad from the local Kebele for the occasion.Footnote 26 The Berhanena Selam killings, then, do not illustrate the gratuitous and meaningless violence of the Red Terror. Instead, they highlight the need to understand the joint production of the Terror by actors operating at different levels of spatial and political influence, and especially the need to do so in relation to the violent struggles over the Ethiopian labour movement and in the context of the Kebele structures, which mediated much of this interaction.
THE LABOUR MOVEMENT AND THE RED TERROR
In view of the important role played by the labour movement in the politics of early revolutionary Ethiopia, it is surprising that the sizeable CELU archives relating to the movement have not been more closely examined by researchers; this is probably one reason why connections between labour activism and Red Terror have hitherto not been rigorously scrutinised. These connections took three main forms: first, the labour movement represented an important arena of struggle. Its ‘capture’ was important for instrumental reasons: not only was it seen to bestow legitimacy by verifying the Marxist and proletarian credentials claimed by all conflicting parties, but the Confederation of Ethiopian Labour Unions and its constituent unions were regarded as organisational assets, whose membership needed to be mobilised, or, barring that possibility, at the very least pacified. An augmenting factor was the disruptive power that workers showcased in the initial stages of the revolution. As an arena in which revolutionary struggle and Red Terror were played out, the labour movement was caught up in a process that would fundamentally transform it.
A second connection was that the workers and institutions of the labour movement were prime targets of revolutionary violence. Where such institutions did not prove sufficiently malleable, or even appeared openly hostile to the Derg, they were actively repressed. The violence directed at the labour movement was at once particularly intense and exceptionally focused. Whereas other urban residents risked becoming victims of Kebele squads, of search teams, or of the various security arms of the state, workers were subject to these risks as well as to the dangers posed by designated factory militias, factory Abyot Tebeka squads, and workplace Red Terror committees. Both the Derg and oppositional political movements invested significant political and organisational resources into the struggle over the labour movement, resulting in levels of violence that were exceptional even in the years of Red Terror.
A third connection between the labour movement and the Red Terror concerns the role of workers as agents of revolutionary violence. Numerous workers participated in armed squads formed at their workplaces, which were designed specifically to recruit workers. In addition to workplace militias and Abyot Tebeka squads, these also included armed groups organised by ELAMA.Footnote 27 The labour movement and its membership thus represented an arena, an object, and a subject of revolutionary violence. To demonstrate the extent to which political violence was rooted in and shaped by the labour movement, it is necessary to briefly trace the history of the movement's tensions and struggles up to their peak in the early revolutionary period.
By the time the Ethiopian revolution broke out in 1974, organised labour in the country – apart from the syndicate of railway workers – had only existed for a decade and a half. From the outset, the labour movement had been beset by conflicting dynamics. On the one hand, state actors had brought their power to bear on the movement so as to ensure the prevalence of a moderate ‘business unionist’ leadership, which in turn spawned and supported the emergence of moderate and loyal leaders in basic unions. At the same time, however, the deferential policies of such leaders was opposed by a militant undercurrent which grew progressively stronger through the 1960s and early 1970s. This labour opposition took a number of expressions: it attempted to have the moderate leadership outvoted in congresses of the General Assembly, and it had a number of representatives serving in influential posts. It was also a dominant force in many basic unions, where it was the driving force behind wildcat strikes that took place across the country.Footnote 28 Although moderates retained the top positions within CELU, the militants gained increasing influence within central bodies and basic unions over time. This eventually pushed the Confederation into adopting an increasingly militant stance.
Propelled by the groundswell of militant energies among workers and constituent unions, CELU was involved in two highly successful strikes in 1973 and 1974. First, it organised solidarity measures with striking bank workers, thereby forcing the government to concede to the latter's demands. Then, after the executive leadership had subverted and stalled the General Assembly decisions to call a general strike for several years, a successful strike was finally called in March 1974, again forcing the government to concede. This strike, which fed into the growing momentum of the 1974 revolution, led to a surge in prestige for CELU as well as in the number of organised workers. Their officially recognised number grew from around 80,000 before the strike to around 125,000 a few months after, although the number of workers who participated in informal networks and practices, including the strike, was probably far greater.Footnote 29 The strike also discredited the moderate leadership, spelling the end of its incumbency. With the militant labour opposition ascendant, CELU was becoming an increasingly vocal force, on a collision course with state authorities. When, in September 1974, the Derg seized power and suspended democratic rights, CELU's General Assembly protested and threatened to call another strike. Over the coming year, CELU and the Derg would be engaged in a mutually hostile tug-of-war, with labour leaders arrested and released, offices closed and reopened, and expressions of workers’ discontent actively repressed. Regime efforts to re-establish a loyal and deferential Confederation by administrative means alone proved futile, forcing the Derg to increasingly resort to the use of violence instead. When ELAMA, the dominant organisation within the radical labour opposition, reciprocated this move, the stage was set for a militarised showdown.
Violence was not, however, a novel feature of labour struggles that emerged only after the revolution. Mere months after CELU's formation in 1963, state actors had first intervened to repress and replace its elected leadership. In the aftermath, veteran labour activist Aberra Gemu was found shot dead.Footnote 30 Over the following decade, violent repression was a constant threat to labour activists. This included casual harassment, arrests, and physical force against union leaders and organisers, police brutality against strikers and unruly workers, and even incidents involving security forces opening fire on groups of protesting workers.Footnote 31 But workers also initiated and engaged in violent contestation. This could take the form of physical assault on representatives of management.Footnote 32 More commonly, it consisted of violence within unions as different factions jockeyed for dominance.Footnote 33 Yet, although violence had been a feature of the labour movement from its inception, it intensified substantially following the revolution and the establishment of military rule.
In 1975, the Derg declared the forced dissolution of CELU. This was followed by protracted efforts to set up new trade union structures loyal to the regime. The basic unions were retained, although they were subjected to pressure to elect new, regime-supporting leaders. This strategy encountered great difficulties, as oppositional workers displayed a tendency to re-elect the same ‘undesirable’ representatives whom the Derg was seeking to remove. Moreover, clandestine structures, most prominent among which was ELAMA, were set up to fight back within both the basic unions and the new structures. The resistance of workers to the reorganisation schemes imposed from above caused such frustration to regime officials that, when violence began to be unleashed on the wider urban opposition in late 1976, the labour opposition constituted a primary target.Footnote 34
The centrality of the labour movement in the process leading up to the Red Terror can be illustrated with reference to the gradual escalation of violent events involving labour activists throughout the years following the revolution, concentrated around the annual celebrations of the revolution's anniversary and of May Day. Thus, in September of 1974, the newly established Derg administration came into immediate conflict with CELU. After the latter threatened a general strike, the Derg clamped down on the Confederation, arresting workers participating in the strike as well as union leaders. Over the coming two years, similar events would occur each September, marking intensifications of the conflict. In 1975, CELU's General Assembly passed a resolution deeply critical of the Derg, outlining a series of radical demands and threatening a strike should these not be met.Footnote 35 By this time, the Derg was prepared to respond with lethal force to such expressions of dissent, killing an estimated one hundred workers and workers’ representatives while arresting an estimated sixteen hundred others.Footnote 36 The aftermath saw a state of emergency being declared, which banned strikes, slowdowns, public meetings, and leaflet distribution, while according security forces the right to shoot suspected offenders on the spot.Footnote 37 This sequence of events was closely paralleled by yet more dramatic dynamics in September of 1976: ELAMA called a general strike, which coincided with the EPRP's first assassination attempt on Mengistu Haile Mariam. This strike attempt was to be the last offensive move by oppositional workers, as the repression that followed included the execution of hundreds of strikers and protestors.Footnote 38
May Day celebrations in 1976 and 1977 similarly escalated into violent confrontations between demonstrators and the Derg. In 1976, around twenty workers who had been carrying oppositional banners were fired on and killed by security forces.Footnote 39 The following year, demonstrations called by ELAMA in collaboration with the EPRP's Youth League for the eve of May Day became subject to the most intense urban violence of the entire revolutionary period, as machine guns were trained on demonstrators, thousands were arrested, and hundreds of detainees were executed overnight.Footnote 40 This progression of events demonstrates the gradual escalation of confrontations over the labour movement, which culminated in and substantially shaped the Red Terror of 1977 and 1978. It began with state actors’ intolerance of autonomous organisation, progressing to the banning of institutions and the arrest of leaders who refused to be coopted, before it graduated to the deployment of lethal force against dissenters and finally to the mass execution of civilians.
Files from the CELU archives demonstrate how, once the Red Terror was proclaimed, unions and union officials were integrated into networks of repression through new violence-facilitating institutions such as the workers’ discussion forums, the Abyot Tebeka squads, and workplace Red Terror committee. Union officials cooperated in the organisation, arming, and drafting of workers into squads and militias, and they frequently cooperated in identifying targets. The formal structures of the labour movement were increasingly conjoined to those of the new repressive institutions and initiatives, from Abyot Tebeka squads to workplace meetings designed to elicit denunciations. The statutes of Abyot Tebeka squads established at factories across the country included stipulations that they work closely with the local trade union, that only members of the trade union would be eligible to join, and that the squads’ commanding committee include a trade union representative.Footnote 41
This entanglement led to cases in which trade unions took part in the purges of their own members and representative structures. In 1978, for example, the national confederation of trade unions, now renamed All Ethiopia Trade Union (AETU), reported that an unspecified number of ‘right roaders’ – a common derogatory term for Meison sympathisers – in the leadership of its local union at the Metahara sugar factory had been identified, subjected to the workplace Red Terror committee, and executed.Footnote 42 In other cases, unions intervened and pleaded the innocence of detained workers to authorities and ad-hoc committees. In 1977, for example, AETU President Temesgen Madebo, who himself was to meet a violent death later that year, wrote to the Derg asking that a factory worker who had killed an alleged EPRP squad member in the Akaki industrial district be released from detention.Footnote 43 In some cases the authorities intervened, while in others the revolutionary guards and Red Terror committees themselves acceded to such requests. Invariably, however, the release of detainees resulted in a formal announcement being sent to concerned institutions, including the detained workers’ Kebele. Such announcements raised suspicions with local authorities, which would increase the risk of further persecution. If unions were not always involved in the meting out of violence, they could not escape their centrality as arenas of the Red Terror. Neither could they escape becoming targets of violence, as the movement bifurcated and the divisions deepened, requiring each basic union to pick a side.
The division between the moderate and militant factions of the labour movement, which had been one of its features since the early 1960, thus escalated and intersected with the dominant conflicts of the early revolutionary era to such an extent that it became a major fault line in the wider Red Terror. By 1977, trade union activism was an exceptionally dangerous activity, even by the altered standards of the time. Among the hundreds of union officials, representatives, and workers who paid with their lives were the last President of CELU, Markos Hagos, as well as the first two Presidents of AETU, Tewodros Bekele and Temesgen Madebo. Yet as the preceding discussion demonstrates, the violence that occurred within the labour movement cannot – as is routinely the case within the historiography – be reduced to the familiar division between the Derg and its allies on the one hand and the EPRP on the other.Footnote 44 Instead, the intensity of these divisions itself needs to be viewed in light of the preceding decade of building tensions and pressures within and over the labour movement. Behind the Berhanena Selam killings and similar executions, in other words, lay a history of profound struggle over the control and use of trade unions, which fueled a determination among regime loyalists at the press to resolve these tensions by eliminating the militant strand of the labour movement, which Daro Negash and her colleagues were seen as embodying. It was to this end that Girma Kebede and his notorious Abyot Tebeka squad were summoned from the local Kebele.
THE KEBELE AND THE RED TERROR
The importance of rethinking the role of the labour movement in energising and shaping the Red Terror is paralleled by the need to reexamine the pivotal role played by the Kebele administrations in revolutionary Ethiopia's urban violence. Within the context of the Kebele, as within the labour movement, the Red Terror was shaped by histories and agents that have been accorded insufficient attention in narratives about the period, yet whose interactions with the much-studied politics of state and opposition parties were decisive in generating and defining the political violence of the early revolutionary years. Established in the aftermath of the revolution as the most local administrative structures of the Ethiopian state, the Kebele gradually became central to the Derg’s repression of urban opponents. The influence of the Kebele on the scale, forms, and meaning of the Red Terror, though variously acknowledged in the literature, has hitherto not been closely examined.Footnote 45 Nor has the extent to which the Kebele themselves were transformed through their role in the Red Terror, with lasting consequences for Ethiopian politics, been sufficiently recognised. The following section discusses several important connections between Kebele and Red Terror, arguing that the history of each needs to be rethought in light of these connections to the other. By drawing on and triangulating an exceptional breadth of sources, from oral interviews to foreign archives, it highlights the agency and interaction of a diverse range of actors in leveraging the Kebele for their purposes, giving rise to a synergy that fueled and shaped the Red Terror at the local level.
The structure of the Kebele, groups of which were organised into Keftegna [‘higher’] administrations, was established by the Derg in August 1975.Footnote 46 Over the preceding weeks, numerous urban properties had been nationalised. The Kebele were initially founded with the twin goals of encouraging grassroots political mobilisation and participation as well as of administering the distribution, rent, and upkeep of these nationalised properties.Footnote 47 During their first year, many Kebeles began to organise thriving Youth Associations, some established Women's Associations, and all held mandatory political education sessions and Weyeyet Kebeboch [‘discussion circles’] for residents. The early Kebeles functioned with an exceptional degree of democratic practice, holding open elections that saw leadership positions distributed to an eclectic mix of respected community elders, covert opposition sympathisers, and outspoken supporters of the Derg. At the same time, the administrative evolution of the Kebele began to make the potential of these structures as coercive governance tools apparent to Derg officials: households were numbered, residents were closely monitored by local Kebele officials, and many were successfully mobilised for rallies and political education events. Gradually, the structure of Kebele and Keftegna began to afford the Derg a degree of local information and control over urban society with few parallels in post-colonial Africa.
The coercive potential of the Kebele was furthered a year later, in October 1976, when the Derg radically expanded the powers invested in the officials staffing these structures. Administrators whose jobs had initially revolved around menial administrative tasks and efforts at political mobilisation were now charged with the control of ‘counter-revolutionary activities’ within their neighbourhoods, widely displacing the regular police forces.Footnote 48 The reforms came in response to the EPRP's launch of an urban guerrilla strategy a month earlier, which had claimed its first high profile victim in prominent Meison member Fikre Merid on 1 October.Footnote 49Meison leaders and their allies within the Derg demanded that actors within the Kebele be armed in response to this growing oppositional threat, and the October 1976 reforms began to meet these demands.Footnote 50 Renewed Kebele elections were held a week after the reforms were introduced, and were probably manipulated to ensure that all Kebeles were staffed by personnel loyal to the military government.Footnote 51 Henceforth, significant resources were put at the disposal of Kebele officials. These included makeshift jails, built for the detention and interrogation of suspected government opponents across all Keftegnas and most Kebeles, as well as the formation of Abyot Tebeka militias, systematically armed by the Derg from March 1977. Kebele officials themselves were effectively turned into violence workers, encouraged to forcefully eliminate opposition activists from their districts. Since these Kebele officials were also afforded substantial autonomy and impunity in exercising their new powers, the degree of repression and the resultant forms of Red Terror that emerged across urban Ethiopia's neighbourhoods over the following months varied widely and were rarely reducible to Derg interests or initiatives.
The spatial pattern and varying forms of the Red Terror were particularly determined by two factors: first, neighbourhoods with a sizeable EPRP membership tended to be those in which Kebele officials and their Abyot Tebeka squads established the most intense and public reigns of terror. Thus in Addis Ababa, the areas around Merkato, Gulele, Tekle Haymanot, and from Piassa north-eastwards to Arat Kilo and Sidist Kilo became particular hotspots of revolutionary violence. This was not merely a result of regime directives, which tended to generically call for the ‘weeding out of counter-revolutionary elements’ and left much room for the agency of local actors in shaping the Terror.Footnote 52 Rather, those who worked for the Derg within the Kebele structures of these neighbourhoods were perceived as particular threats to the EPRP and therefore regularly became targets of the party's assassination campaign.Footnote 53 With their own lives constantly threatened and their personal security depending on the rapid eradication of EPRP activists in their neighbourhoods, Kebele officials and their militias in these areas were prone to establish brutal and public reigns of terror, which usually consisted of widespread detentions marked by routine torture, extrajudicial executions, and the public display of dead bodies.Footnote 54
A second factor influenced the diverse forms and intensity of the Red Terror in different areas: since Kebele officials tended to be elected from within neighbourhoods where they were socially embedded, their new positions and authority often allowed them to redress local histories of societal tension and structural violence. It was not uncommon for Kebele officials, as well as for other agents of Red Terror at the local level, to personally know their victims: not only did they live and often grow up in the same neighbourhoods, there were also cases in which they belonged to the same iddir [‘mutual help associations’] or even, in rare instances, to the same households.Footnote 55 Such local ties and their histories widely affected how the Red Terror was expressed, and whom it targeted, on the ground. In some instances, Kebele officials used their positions to protect suspected activists.Footnote 56 In others, they appropriated the rhetoric and resources put at their disposal by the Derg to act upon local animosities or personal grievances. This latter dynamic is apparent in the well-documented case of Kelbessa Negewo, the notorious chairman of Keftegna 09 during the peak of the Red Terror.
Like other Keftegna and Kebele officials at the time, Kelbessa was charged with suppressing and dismantling oppositional networks within the district under his control; like them, he relied on a strategy of detaining numerous local suspects in order to interrogate them, regularly resorting to torture. Elicited confessions and denunciations allowed for the mapping of clandestine cells and triggered further searches, arrests, and in many cases executions. Such efforts at ‘weeding out counter-revolutionaries’ served the interests of the Derg, to which he also regularly applied for firearms and ammunition to intensify his activities.Footnote 57 Kelbessa's detainees, however, were chosen according to criteria that were shaped by the local history of social relations in the district as much as by government directives. Kelbessa's efforts targeted particularly those whom he knew to have benefitted from a privileged upbringing under the imperial regime. Even decades later, when three of his victims sued him for torture in a U.S. court, he disparagingly described them as ‘the feudal children’. Class did not necessarily map onto opposition to the Derg in this Gulele neighbourhood.Footnote 58 But there were personal reasons why this became a criterion that defined Kelbessa's actions: his own upbringing and prospects had been shaped by the structural inequalities of pre-revolutionary Ethiopia. When Kelbessa earned a scholarship to study Public Health in the United States, for instance, his scholarship had, in his view, been lost to the nepotistic machinations of a senior official in the imperial regime, whose nephew went in his place.Footnote 59 His Keftegna position in the context of the Red Terror allowed Kelbessa, like many other perpetrators in the Red Terror, to redress and reverse such power relations at the local level.
In some cases, the Red Terror was yet more transparently shaped by the personal motives of Kebele officials. Girma Kebede, who ordered the Berhanena Selam killings, is said to have killed his former chemistry teacher over a dispute.Footnote 60 Other officials had particular residents detained in order to confiscate desirable property: ‘If you had a nice car,’ one informant recalled about the Kebele in his Tekle Haymanot neighbourhood, ‘there was the danger of being accused of it being used for EPRP robberies and confiscated, with you being taken to prison.’Footnote 61 What these examples demonstrate is that the Red Terror cannot be reduced to the familiar macro-political divisions and policies of the Derg. Instead, its violence was also substantially shaped by the motives and agency of actors at Kebele level, and by their interaction with the interests and guidelines of the Derg. Local histories, animosities, and social relations need to be understood as central factors, rather than merely circumstantial ones, in the making of a terror that was jointly produced by a diverse range of actors. This perspective, which has been neglected in the historiography, is invariably confirmed in oral interviews by those who either experienced or perpetrated Red Terror-related violence within the context of the Kebele, and it features prominently in several of the memoirs recently published by survivors of the Red Terror.Footnote 62
Continuities between local histories of everyday violence and the Red Terror as perpetrated at the Kebele level also came to the fore in tensions between Abyot Tebeka squads and armed EPRP units in the Merkato neighbourhood. Both groups recruited members from criminal gangs that had operated in the area since the late imperial era. One gang, known as the ‘China Group’, had built its influence by extorting protection money from local merchants, resembling ‘a mafia in the making’.Footnote 63 Its criminal activities and violent feuding with local rivals, notably the aptly named ‘Genghis Group’, is reflected in the records of the police chronicle Polis Enna Ermeja and gained notoriety across the capital.Footnote 64 In the context of revolutionary Ethiopia's escalating political conflicts, the ‘China Group’ appears to have been politicised, throwing its weight behind the clandestine EPRP. The gang henceforth carried out its activities in the name of the party and of its political goals. The ‘Genghis Group’ underwent a parallel transformation, but instead allied itself with Meison and was drafted into local Abyot Tebeka militias, which received arms and training to repress the EPRP.Footnote 65 One former EPRP activist noted that the ‘China Group’, like other local criminal gangs including the ‘Kimbibit Group’ and the ‘Urael Group’, was ‘taken over, disciplined and organised for the revolution’ by the EPRP; putting their skills honed as criminals into the service of larger political projects, gang members ‘became wonderful revolutionaries’.Footnote 66 Yet behind the political conversion and successful mobilisation, celebrated and instrumentalised by the EPRP and the Derg, there were significant continuities with pre-existing local histories: gang members’ activities continued to revolve around the extortion of money from local traders and the conflict with rival gangs.Footnote 67 As in other cases in which the means of state violence devolved to civilians for counter-insurgency purposes, such violence was fueled and reshaped by the social ties and interests of its local agents, even as their community relations were renegotiated in light of the dominant political conflicts of the day and the monopolisation of discourse that attended them.Footnote 68
The Kebele became central to the Red Terror because it was on these recently established institutions that the attention of actors operating at very different levels of influence was focused. The extent to which this was true not only for the local actors discussed thus far and for government leaders but also for intelligence agents and even for some of the Derg’s international allies in the context of the global Cold War is revealed by an unlikely and hitherto ignored set of sources: the archives of the East German Ministry of State Security (MfS, commonly known as ‘the Stasi’). Due to the close diplomatic relations developed at the time of the Red Terror between the Derg and socialist bloc states, including the German Democratic Republic, these archives contain rich materials on Ethiopia's secret services and local violence.Footnote 69 The new diplomatic ties between Ethiopia and these socialist states famously resulted in the decisive military intervention of the Soviet Union and Cuba in the Ogaden war, as well as in a broad range of economic partnerships.Footnote 70 What is less well-known is that they also created close collaborations between Ethiopian intelligence services and their Soviet, Czech, Hungarian, Bulgarian, and especially East German counterparts.Footnote 71
The partnership between the intelligence apparatuses of Stasi and the Derg ran along two tracks: the first concerned the provision of modern equipment and training to the Ethiopian services. In August 1977, East Berlin dispatched to Ethiopia an initial group of Stasi advisors alongside materiel, chiefly state of the art radio communication equipment and communication interception technology.Footnote 72 By the end of the year, equipment worth more than 12 million marks had been gifted to the Ethiopian authorities as ‘material solidarity’, a figure that would rise further over the following two years.Footnote 73 The second area of collaboration concerned the radical reconfiguration of Ethiopia's intelligence services ‘after the example of the socialist states.’Footnote 74 Under the reign of Emperor Haile Selassie and under the early Derg, Ethiopia's intelligence apparatus had been built on an American model, trained and equipped by Israelis as well as by Americans.Footnote 75 Rival intelligence services of the Army, of Internal Security, and of Central Investigation all worked independently and were partly responsible for controlling and balancing each other. In collaboration with the Stasi, this model was to be jettisoned in favour of a centralised intelligence service under a new Ministry of State and Public Security. While the reorganisation would not be complete until 1979, first steps towards centralisation had already occurred by August of 1977.Footnote 76
The immediate objectives, in both cases, revolved around local intelligence gathered at Kebele level. There remained much room for gaining better information about local opposition cells and activities, but above all there was an urgent need to coordinate and respond to such intelligence as was being gathered through interrogations, denunciations, and confessions. The rapid and extensive investment in radio communication technology, for instance, was intended to equip mobile and stationary task forces that were formed in October across all twenty-three of Addis Ababa's Keftegnas in order ‘to combat hostile activities [within the Keftegnas]’.Footnote 77 Initial efforts to centralise the work of the intelligence services likewise focused on the Kebele and Keftegna: the transmission, collection, and processing of locally gathered information was successfully improved, which allowed opposition networks to be mapped and, through greater coordination of local interventions, to be systematically dismantled.Footnote 78 To this end, Stasi agents stationed in Ethiopia worked directly with several Keftegna chairmen as well as with agents of the Ethiopian Central Investigation Department.Footnote 79 Whereas Kebele officials such as Kelbessa Negewo had acted with significant autonomy in seeking to repress oppositional activities in their districts during the early stages of the Red Terror, they increasingly found themselves working within centralised command structures by late 1977.
The effects of these changes were soon palpable on the ground: opposition activists found their phones tapped, their comrades disappearing, their infiltrators in government institutions exposed, and remaining cell structures dissolved.Footnote 80 The violence of the Red Terror became more selective and, as a result, more effective. As an internal Stasi memo noted, ‘the “white terror” fanned by the counter-revolution is being crushed by the “Red Terror” organised by the Kebele.’Footnote 81 By late October, it could be reported that ‘while counter-revolutionaries continue their hostile activities to discomfit the population, they have become significantly less active.’Footnote 82 The EPRP, which over the preceding years ‘had attracted many of the most educated and talented young Ethiopians’ and had rapidly grown a substantial following, was left with a decimated leadership and unable to coordinate, let alone protect its members.Footnote 83 The party's armed squads ceased to effectively deter collaboration with the Derg, which had hitherto been one of their chief objectives.Footnote 84 Moreover, confessions and denunciations received by Kebele officials increasingly came from EPRP members themselves, who lost their confidence and hope in the party or simply ran out of options.Footnote 85 As one disillusioned EPRP activist wrote, by the close of 1977 ‘the EPRP was fully repressed’.Footnote 86
By this point, Meison had abandoned its support for the Derg and gone underground, turning members of the organisation into additional targets of the Red Terror. The swift and effective repression of Meison occurred despite the organisation's popularity among Kebele officials, and against objections from the regime's East German allies, who had regarded Meison as a ‘progressive’ and moderating influence on the Ethiopian government.Footnote 87 The Kebele had by this point become effectively integrated into the command structure of the revolutionary Ethiopian state, bringing an end to the earlier decentralised stages of the Red Terror.Footnote 88 By early 1978, Meison too was de facto eliminated, and by the end of the year an internal Stasi memo would note that ‘there is currently no oppositional group that would be capable of carrying out an organised subversive activity against Mengistu and his followers.’Footnote 89
The structure of the Kebele was central to the tragic success of the Red Terror, and it was itself transformed beyond recognition by its developing role as a hub of state violence. As a broad range of actors – from local residents over government figures to foreign security experts – brought their knowledge and resources to bear on these institutions, acting synergistically despite a great diversity of motives, they jointly produced both the violence of the Red Terror and the Kebele as a powerfully coercive institution. Established to solve problems of urban administration and political mobilisation thrown up by the revolution, the Kebele swiftly developed into arguably the single most effective structure of state control over society in sub-Saharan Africa, a tool of governance so compelling that – with some changes – it outlasted not only the Red Terror but even the Derg regime itself.
CONCLUSION
Analyses of the Ethiopian Red Terror have followed a well-worn path, entrenching a now hegemonic set of approaches and narrative patterns. This article has emphasized the need to rethink the history of this political violence, paying closer attention to actors, institutions, and interactions that have been insufficiently featured in the historiography. Trade unions and the newly established Kebele administrations in particular, are shown to have acted as arenas within which the histories, struggles, and interests of very different sets of actors – from local residents and militant activists to government leaders and their international allies – collided or converged, synergistically producing the period's political violence. In the form, targets, and evolution of this violence, the Red Terror reflected its diverse sources and the agency of its multiple architects. When approached from the vantage point of this joint production within contested institutions, defining episodes of Red Terror such as the Berhanena Selam killings appear in the new light of their social histories as products of struggles over the labour movement that were made possible by the militarisation of the Kebele.
As this paper has demonstrated, the history of collective violence in the Red Terror is also the history of consequential socio-political and institutional changes. Much as the Red Terror was shaped by the struggle of diverse actors within trade unions and the Kebele administrations, so these institutional settings themselves were radically transformed by these tensions and by the violence that brought an end to them. The Red Terror not only exterminated the most vibrant organisations in Ethiopian civil society, among which the labour movement featured prominently; it also played a role in bequeathing to Ethiopia a powerful and intrusive state presence at the heart of urban communities, frequently staffed by officials who had played prominent roles in the death, torture, or detention of sizeable population groups within their districts. These effects of the Red Terror point to a wider dynamic: the exceptional degree of knowledge and influence over urban society at the local level achieved by Ethiopian state actors, frequently ascribed to the country's distinctive history over the longue durée, also needs to be understood in light of the institutional transformations and political violence of the 1970s.Footnote 90 Rethinking the Red Terror, then, is necessary for the sake of understanding Ethiopia's consequent history, and in order to make sense of these violent years as a deeply significant period in its own right.
