Unpacking the Addis Ababan Exceptionalism—Living and Making Sense of Violent Protests in Ethiopia’s Capital

Violent protests and ethno-communal violence in Ethiopia have been on the rise since 2015. Whilst the country’s capital has mostly been spared, protests have occurred in areas surrounding the capital, its suburbs and, to a lesser extent, even in its city centre. This article aims to answer how Addis Ababa residents make sense of and (dis)engage from/in violent protests and ethno-communal violence where they occur. The article explores the perceptions and experiences of established middle-class residents in the capital, drawing on ethnographic fieldwork conducted between June 2020 and May 2021. The article finds that, due to the city’s multi-ethnic character and the residents’ strong metropolitan identity, ethnic identity has not constituted a powerful tool for political mobilisation of residents within the capital. Claims based on ethnicity made by non-residents towards the capital’s resources have largely been rejected by Addis Ababans. Instead, residents have collectively defended their right to the city. The federal state’s readiness to intervene in protests in the capital has further discouraged and countered public protests and communal conflicts. The article reveals that, whilst ‘exceptional’ at the first glance, the peaceful normality of Addis Ababa has relied on ignoring conflicts within Ethiopia’s ethnic-based political settlement and has hidden ordinary, everyday forms of structural violence and conflict.


Introduction
How do residents in Ethiopia's capital Addis Ababa understand the rare occurrence of violent protests and ethno-communal violence in their city? Since 2015, Ethiopia has experienced a rise in inter-group tensions and ethnic conflict, following large-scale public protests and ensuing political instability. Whilst public protests initially concerned the expansion of Addis Ababa into the surrounding Oromia regional state, they soon addressed larger political questions pertaining to the ruling coalition's undemocratic character and spread to other regions. Largely organised along ethnic lines, protests have been accompanied by rising ethnic tensions and conflicts countrywide (Ayalew, 2019;Jima, 2021). The most influential of these protest movements, the qeerroo movement, concerns the status of the Oromo, who constitute the biggest ethnic group in Ethiopia, making up around 35% of the population. Cities in Oromia regional state and cities in the border regions of Oromia have been the main theatre of the qeerroo protests, that have contested the marginalisation of the Oromo in the distribution of economic, social and political resources. Protests have been accompanied by ethno-communal conflict and large-scale displacement. Whilst the ownership of Addis Ababa, which simultaneously is the country's capital and the seat of the Oromia regional administration, has been a central question for the qeerroo, only on a few occasions have the protests been fought in and around capital. Moreover, ethno-communal violence in the capital city has remained low (Østebø & Tronvoll, 2020;Workneh, 2020).
According to the Ethiopian constitution, the residents of the capital city of the federal state have been granted the right to self-government, despite the fact that Addis Ababa is located in Oromia regional state. The city's administration is responsible to the federal government and Addis Ababa residents are represented in the House of People's Representatives. Whilst the special interest of Oromia regional state in Addis Ababa regarding the city's resources and its administration has also been constitutionally recognised (Proclamation No. 1, 1995, sec. 49), the capital has constituted an independent administrative unit. Whereas Ethiopia's governance system is based on the principle of ethnic federalism, Addis Ababa's multiethnic character has defied administration based on ethnic representation. Instead, residents have emphasised the metropolitan identity of Addis Ababans that unites residents from different ethnic backgrounds and have contrasted it to the ethnicised identities of rural residents. As Addis Ababa remained peaceful and calm during the 2015/18 political riots, residents have often emphasised the exceptional status of the capital in Ethiopian politics and society (Pellerin & Elfversson forthcoming).
Whereas state-perpetrated violence and repression have been a recurrent topic in previous research on state society relations in Addis Ababa during the EPRDF rule   (Di Nunzio, 2014Eyob, 2017), little research has focussed on violent protests and ethno-communal violence in Ethiopia's capital. This is the case, since the latter have largely constituted non-events, apart from some few exceptions like the 2005 post-election protests (Abbink, 2006). Previous research on violence in Addis Ababa has often emphasised the structural and 1 3 Unpacking the Addis Ababan Exceptionalism-Living and Making… physical violence exercised on individuals through the process of urbanisation, e.g. displacements caused by urban planning initiatives (Pedrazzini et al., 2014;Terrefe, 2020) and the life in informal settlements and in marginality (Di Nunzio, 2017;Nunzio, 2019). Research has primarily studied violence of urbanisation, rather than communal conflict and violence linked to the process of urbanisation and urban lifestyles in the capital. Moreover, little attention has been paid to the qeerroo protests in Addis Ababa, as incidents and levels of violence have been low compared to other major cities in Oromia and in its border regions. This article aims to explore the rare occurrence of violent protests and ethno-communal violence in Addis Ababa, through the prism of the perceptions and experiences of residents in the capital. Taking the qeerroo protests and violent communal conflicts in the capital as an entry point, the article explores how Addis Ababans themselves explain the sudden occurrence of violent protests and ethno-communal violence, make sense of and (dis)engage from/in it.
Analysing the rare occurrence of violent protests and ethno-communal violence through the lens of Addis Ababans, the article finds that ethnic identity has not constituted a powerful tool for political mobilisation of residents within the capital, due to the city's multi-ethnic character and the residents' strong metropolitan identity. Moreover, public protests and communal conflicts in the capital have further been discouraged by the federal state's readiness to intervene. Whilst exploring Addis Ababans' narrative of the 'exceptionally peaceful' city, the article reveals how the everyday peace in Addis Ababa does not account for everyday/ordinary forms of violence, such as ethnic-based discrimination.

Urbanisation, Violent Protests and Communal Conflict
An increasing canon of work has analysed the rise of urban protests in Sub-Saharan Africa (Branch & Mampilly, 2015;Chaturvedi, 2016;Eckert, 2017;Ellis & Kessel,2009;Fox & Bell, 2016). Some research has even noted a shift from rural rebellions to urban protests (Golooba-Mutebi & Sjögren, 2017;Raleigh, 2015), a phenomenon that seems to have particularly touched the urban peripheries (Gizelis et al., 2021). Studies have identified a variety of factors that have favoured the 'urbanisation of violence' (Raleigh, 2015, p. 90), as well multiple drivers of conflict in urban areas.
The question of why conflict and violent forms of contestation seem to have moved from rural to urban areas has often been answered referring to institutional changes and shifting political incentive structures. Studying the case of Uganda, Golooba-Mutebi and Sjögren (2017) find for example that rising costs of and decreasing returns from rural rebellions, coupled with improved state capacity to control them, the institutionalisation of dissent through multiparty politics and the rise and following repression of political opposition parties particularly in urban areas, jointly explain the decrease of rural rebellions and the rise of urban protests. According to Raleigh (2015), changes in political institutions and (incomplete) democratisation processes in Africa have led to a disenfranchisement of urban populations and a growth of spatial inequalities favouring rural over urban areas. Raleigh theorises that 'the rural bias' has caused grievances in urban areas and has subsequently favoured public protests.
Particularly in contexts with a history of civil wars, cities seem prone to the rise of violence (Beall & Goodfellow, 2013). The rise of 'fragile cities' (Raleigh, 2015, p. 90) that have become dominant sites of conflicts in the Global South has been attributed to demographic, socioeconomic and political factors. Urban growth, both through migration and natural growth, can lead to conflict between different groups over scarce resources. Goldstone (2002) for example finds that an unequal population growth amongst ethnic groups in urban areas can favour ethno-communal conflicts. Moreover, population growth also affects socioeconomic opportunities in cities and can lead to a mismatch between actually existing employment opportunities and the size of the labour force, potentially causing political instability because of ensuing economic hardship. According to Goldstone (2002), particularly growing segments of unemployed youth constitute risks for the emergence of violence. High and increasing levels of inter-group inequalities in urban areas have also been found to drive urban violence (Østby, 2016). Apart from population growth and economic hardship, the spatial growth of cities into surrounding rural farmland has been identified as a potential trigger for conflicts over resources such as land and water (Abate, 2019;Cash, 2014;Narain, 2016). Finally, poor governance and lack of sufficient service provision to urban residents have been found to trigger violent protests (Raleigh, 2015).
Mainly focusing on the occurrence of violence, research exploring the 'urbanisation of violence' hypothesis has paid little attention to the ways in which individuals in deeply divided societies avoid or even minimise conflict, creating and/or sustaining what Mac Ginty (2014, p. 549) has termed 'everyday peace'. Although urbanisation can be a driver of conflict, cities have the potential to promote community cohesion through the creation of shared urban identities. Whilst popular in recent research, the 'urbanisation of violence' hypothesis has further been criticised, as statistical evidence so far remains inconclusive (Buhaug & Urdal, 2013;Gizelis et al., 2021). Moreover, the dichotomic division between rural/urban has been questioned, as they fail to account for the complex geographies of violence and socio-spatial connections between rural and urban areas (Acker, 2018).
Addis Ababa constitutes an interesting case with regard to the topic, as it seems to defy the 'urbanisation of violence' hypothesis. Whilst Ethiopia has a long history of civil war (Zewde, 2001), violent protests in the capital city have remained the exception, not the rule, and rural armed insurgencies continue challenging central politics. Although the capital constitutes a stronghold for political opposition parties, violence around national and local elections has constituted the exception, not the rule (Abbink, 2006(Abbink, , 2017Melakou, 2008). Despite rapid levels of urban growth, growing inequalities between residents, the marginalisation of the poor and lack of social service provision, violent protests and inter-group violence have remained low (Di Nunzio, 2012;Pedrazzini et al., 2014). Moreover, research has shown relatively high levels of social capital even in slum areas in Addis Ababa, where residents from a variety of ethnic backgrounds have organised and engaged through traditional community organisations and self-help groups (Kassahun, 2015).
Analysing how residents themselves make sense of the rare occurrence of violent protests and ethno-communal violence and engage in/with such conflicts constitutes an important contribution in this field. Instead of exploring causal relations between urbanisation and conflict, it allows to reflect on people's experiences and understandings of conflict, but also peace. Who experiences the city as peaceful or respectively violent and what does 'everyday peace' (Mac Ginty, 2014) means in a larger context of conflict are important questions to be answered.

Studying Protests and Violence in Addis Ababa
To capture how residents understand and (dis)engage in urban protests where such events occur, the article takes one of the rare protest events in the capital as an entry point. It focuses on the protests that followed the assassination of the Oromo singer and political icon Hachalu Hundessa on the 30th of June 2020 in Addis Ababa. The protests following Hachalu's death constituted the first large-scale event of violent protests and ethno-communal violence in the capital since the outbreak of the qeerroo protests in 2014/15. Although outskirts outside of the city borders had been affected by protests and violence before, apart from isolated incidents, the qeerroo protests had not reached the city centre of the capital.
Whilst it is possible to observe protests, understanding the ways in which people make sense of, participate in or disengage from such events requires studying and understanding details which are not easily visible for the external observer. This article is based on ethnographic fieldwork in Addis Ababa, conducted between June 2020 and May 2021. Fieldwork was conducted both jointly as well as separately by both authors. Observations of protests and their aftermath, as well as formal interviews and informal conversations with citizens, civil society organisations, activists, government officials, bureaucrats, experts and international donor organisations, form the basis of the article.
To capture what happened after the assassination of Hachalu Hundessa in Addis Ababa and to reflect on how Addis Ababans understood the events and participated in them, as well as to capture the impact of the events on the lives of Addis Ababans, we conducted observations and interviews during and after the protests. Observations were primarily conducted in our respective areas of residence and the surroundings accessible on foot. As moving around the city was nearly impossible during the protests, information about events in other areas of the city was collected through phone calls to our respective social networks. Moreover, we also conducted formal interviews with residents from different areas in Addis Ababa between July and December 2020. Our interviewees were long-term Addis Ababa residents 1 and most had been born in the capital. Interviewees were from middle-class households, 2 lived in formal accommodation and worked in the private sector and for government institutions, as well as for religious and non-profit organisations. Interviews were semi-structured and covered the following topics: (a) the types of events and violence in the different areas of Addis Ababa; (b) the ways that residents reacted to the violence; (c) residents' perceptions about how the government handled the situation; (d) residents' understanding of and attitude towards the protests and (e) the ways in which the lives of residents were impacted by the events.
To capture the complexities of protests and violence in Addis Ababa, we made use of our existing knowledge of and networks in the city. Whilst the fact that we are both residents of Addis Ababa has helped us in our study, our respective positionalities have also posed limits regarding access to study participants. We both live in middle class neighbourhoods on the outskirts of the capital but within its city boundaries. One of us is Ethiopian and a native Amharic speaker, whilst the other one is an international resident and has good Amharic skills. We both have ethnically mixed social circles containing individuals supporting the Oromo protests and those contesting their legitimacy and their means. This has allowed us to conduct observations in different parts of the city and interview Addis Ababans from different backgrounds and featuring a wide range of different opinions on the topic. However, we capture the perceptions and opinions of the established Addis Ababa middle class and do not account for those of more marginalised groups or new arrivals in the city.
Contrary to the urban poor or new arrivals, who were not part of the sample, the established Addis Ababa middle-class households had a lot to lose in the protests, both in economic terms and regarding their social and political status. Whilst the middle class dominate political and economic capital in the city, their experiences and opinions do not reflect those of the entirety of residents in Addis Ababa. Consequently, future research should also reflect on the ways in which the urban poor and other marginalised groups understand and engage with violent protests.

The Right to the City: Addis Ababa and the Oromo Question
Given its status as national and regional capital, the question of ownership of Addis Ababa has been central to Oromo political movements that have struggled for the Oromo's political, social and economic emancipation since the region's forceful integration into the Ethiopian empire (Jalata, 1995). Whilst the capital city in theory serves both the citizens of Ethiopia and, regarding their regional affairs, the citizens in Oromia, Oromo movements have claimed that the Oromos have been excluded from accessing resources in the capital and have pointed at the displacements of Oromo caused by the expansion of Addis Ababa into Oromia regional state. Disagreements between the Oromia regional government and the Addis Ababa city administration about the legal grounds for the expansion of the capital and the administration of the growing city have also caused political tensions (Benti, 2002;Jima, 2021). The process of urbanisation in Addis Ababa has de facto led to the expansion of the capital into the Oromia regional state, provoking sometimes violent conflicts between the indigenous Oromo, the state and the Addis Ababa residents.
Rapid urban expansion has led to an increase of the urban area, encroaching into the surrounding rural areas a decrease of agricultural land and forest and the displacement of rural farming communities. The eviction of Oromo farmers from the capital's suburbs and surrounding towns has largely happened uncontrolled and favoured dispossession without compensation (Abate, 2019;Deribew, 2020;Zewdie et al., 2018).
To coordinate planning and service provision between the Addis Ababa city administration and the surrounding Oromia zone, the central government developed the so-called Addis Ababa Master Plan (World Bank, 2015). However, opponents of the plan held that it formalised the expansion of Addis Ababa into Oromia regional state, contesting the centralised planning process and top-down communication. The so-called Master Plan Protests staged by Oromo youth in 2014/15 eventually spread throughout Oromia region and later country-wide (Jima, 2021). The protests triggered the declaration of two subsequent states of emergency in 2016 and 2017 and eventually forced the ruling coalition to adopt reforms in 2018. Abiy Ahmed became the first Oromo to be elected Prime Minister of Ethiopia (Temin & Badwaza, 2019).
The Oromo youth movement that had been crucial in bringing about the political change is called the qeerroo movement. 3 The qeerroo are heterogenous in character and constituted of lots of different factions, some fighting for more regional independence, whilst others advance claims for more power at the federal government level (Østebø & Tronvoll, 2020). Forms of resistance and contestation by the qeerroo have taken a range of different forms, both non-violent and violent, such as stayat-home protests, peaceful demonstrations, riots and ethnic feuds. The qeerroo have emerged as a crucial political force since 2015, in some places effectively undermining state power and assuming quasi-state functions for certain periods of time (Davison & Leake, 2019). Cities in Oromia and bordering Oromia like Sheshemene, Dire Dawa, Harar, Hawassa and Robe have seen large-scale violence in connection to the qeerroo protests, as well as ethnically targeted attacks against non-Oromo (Addis Standard, 2020c). Whilst a crucial force in Abiy Ahmed's rise to power, parts of the qeerroo movement have progressively turned against the Prime Minister. According to some qeerroo, the new Prime Minister has not sufficiently advanced an Oromo political agenda. They have contested Abiy's discourse of 'Ethiopia first', holding that it fails to account for the historical marginalisation of the Oromo and the fact that Amhara national language and traditions make up the core of the so-called national culture of Ethiopia (Jima, 2021).
Compared to other cities in Oromia and cities in areas bordering Oromia, Addis Ababa has seen relatively low levels of disturbances and violence throughout the qeerroo protests. Apart from isolated incidents, the protests did not reach the city centre of the capital. However, the capital's outskirts that lie outside of the city borders and formally in Oromia special zone surrounding Addis Ababa like Sebeta, Alem Gena and Burayu, have seen large-scale destruction of public and private property as well as violent riots as part of the qeerroo protests (Addisstandard, 2019;Workneh, 2020, p. 319f). Residents in Addis Ababa have often described the city as 'a bubble' 4 , that remained peaceful and sheltered throughout the protests. However, in July 2020, protesters entered the capital, following the assassination of the Oromo singer and political icon Hachalu Hundessa. The protests turned violent, leading to mass destruction of public and private property as well as causing lives lost (BBC News, 2020).

The Death of Hachalu Hundessa and Violent Protests in Addis Ababa
On the 29th of June 2020, the Oromo singer and political icon Hachalu Hundessa was killed in the area known as Galan condominium in Addis Ababa. The news about his murder were shared via social media and amongst others by Jawar Mohamed, a famous Oromo activist. In the early hours of the 30th of June, Oromo youth entered the capital. 5 Initially proceeding to Galan condominium to grieve Hachalu's death, more qeerroos entered Addis later during the day and gathered at other meeting points in the city to protest against what was perceived as the government's involvement in the murder. Passing through the outskirts to gather in the city centre, the qeerroos caused significant disturbance, destruction and violence. Protesters destroyed windows of public and private buildings, looted shops and residences, blocked roads, burned tyres and clashed with local residents. Addis Ababa residents recounted that violence included 'throwing stones', 'crashing cars' and 'burning tyres and vehicles'. 6 In contrast to previous qeerroo protests, nearly all parts of the capital city were touched, including the city centre (see Fig. 1). Whilst physical conflict between the protesters and the Addis Ababans in the city centre remained the exception and violence largely involved the destruction of property and other belongings, in the outskirts of the city, violence included direct physical violence against individuals and had a clear ethnic undertone with non-Oromo being targeted.
Despite the fact that the Hachalu protests constituted the first event of large-scale violence in the capital linked to the qeerroo protests, residents had anticipated such a possibility. Soon after information about Hachalu's death and calls to the qeerroo to go to the capital to grieve his death had been circulated on social media, Addis Ababans warned each other via telephone and text messages about potential disturbances. 7 Many residents stayed at home for 2 days and did not leave the house. 8 Residents' anticipation of violence can be traced back to their awareness of violence linked to the qeerroo protests in other locations (Ayalew, 2019), political party 1 3 Unpacking the Addis Ababan Exceptionalism-Living and Making… discourses on the qeerroo portraying them as dangerous (Samuel, 2020;Titz, 2019) and the rare occasions of disturbances caused by the qeerroo in Addis Ababa and its suburbs prior to the Hachalu events, for example the protests that had followed the alleged removal of the special protection forces of the Oromo political opposition figure, qeerroo leader and Oromo nationalist Jawar Mohammed on the 23rd of October 2019 (Ayalew, 2019;Marks, 2019). The Addis Ababa-based political opposition party Balderas for Genuine Democracy 9 used the residents' fear of the qeerroo, to advance a political programme for the capital that questioned Oromia's constitutionally anchored, special interest in the city. The party leader claimed publicly that "Addis Ababa belongs to its residents and the rest of Ethiopia and no region should claim a special interest in the city" (Balderas For Genuine Democracy, 2019). Already before the Hachalu events, Balderas had organised youth in many parts of the capital, calling on the youth to defend Addis Ababa against the qeerroo (Addis Standard, 2020b). An Addis Ababa resident recounted how Balderas had prepared the Addis Ababa youth to defend the capital: When the Addis youth organised themselves. I think Eskinder Nega was involved.
[…] The youth got ready to fight. People were disseminating the wrong kind of information. You should mobilise. Even before a month ago there have been attempts. In Addis there have been informal groups. They have gathered for events such as football and sport. But the reason is to organise them against intrusions by the outskirts of the city. 10 When the qeerroo entered the capital after Hachalu's death, causing conflict and destruction, in many parts of Addis Ababa, the police did not intervene timely and instead residents had to defend their own neighbourhoods. Communities set up neighbourhood watches and supplied their guards with 'horns' and 'bells' 11 to warn the residents should the qeerroo attack. Addis residents armed themselves with tools available at home, which included sticks, cooking and cleaning utensils and gardening tools. 12 In many places, the qeerroos turned around when they encountered armed residents. However, in several places, the qeerroo clashed with the Addis Ababa youth who were out to defend what they perceived as their city (Addis Standard, 2020a).
Many of the qeerroo who came to Addis Ababa after Hachalu's death encountered difficulties to find their way around the city. Some residents reported to have intentionally pointed qeerroos towards the opposite directions of their agreed 9 Balderas for Genuine Democracy is an Addis Ababa-based opposition party founded by Eskinder Nega, renowned activist and politician. Balderas has been promoting national unity over ethnic federalism but has been accused of supporting Amhara nationalist tendencies draped behind claims for unity. The party has recently been associated with Amhara nationalist forces and for the 2021 general elections in Ethiopia Balderas has entered a collaboration with NaMA (National Movement of Amhara) (Sisay, 2021). 10 Informal conversation with resident from Kasanchis, 04.07.2020. 11 Fieldnotes, Pellerin, 12th of July. 12 Informal conversation with resident from Maganagna, 02.07.2020; informal conversation with resident in CMC, 01.07.2020. meeting spots, to prevent violent mass gatherings and destruction. 13 Many interviewees pejoratively described the qeerroo as 'peasants from the countryside'. 14 Contradicting claims of the qeerroo regarding the purpose and aim of the protests, Addis Ababa residents who held that they had talked to the qeerroo reported 'they [the qeerroo] didn't even know who Hachalu was' 15 and 'they were like twelve years old and had been sent here by those who want to instigate chaos'. 16 Such perceptions contradicted the claims made by the qeerroo, that their movement was based on legitimate grievances and represented broad-based demands of the Oromo. The perceptions also enforced the conviction of Addis Ababans, that the protests were part of a larger struggle for power between the federal government and selected politicians and prominent individuals in Oromia.
Whilst the majority of events reported by the interviewees involved the qeerroo coming from the outside causing destruction and violence, 17 the research Fig. 1 Locations affected by the violent protests June/July 2020. Source: Pellerin & Elfversson forthcoming. The Addis Ababa city boundary is marked in black. The arrows indicate the entry points of the qeerroo. The mapping is based on field interviews, as traditional and social media reports cannot be used as basis for systematic mapping of the events due to the general internet shutdown starting on the 30th of June and lasting for 3 weeks 13 Informal conversation with resident from Maganagna, 02.07.2020. 14 Informal conversation, fieldnotes, Author A, 1st of July 2020. 15 Interview with resident from Jemo, 19th of December 2020. 16 Informal conversation, fieldnotes, Author A, 2nd of July 2020. 17 Fieldnotes, Pellerin, 1st and 2nd of July 2020.

3
Unpacking the Addis Ababan Exceptionalism-Living and Making… revealed more complex dynamics, questioning the idea that ethno-communal violence amongst Addis Ababans did not exist. Patterns of ethnically targeted violence differed between the city's suburbs and its core. In some of the Addis Ababa suburbs and surrounding towns, young Oromo residents attacked residents from other ethnic backgrounds, whereas in parts of the capital's inner city, non-Oromo Addis Ababa youth attacked Oromo residents. Particularly Amhara residents were targeted in the city suburbs. Ethnically targeted violence against the Oromo in the capital city was exacerbated by the calls to arms by Eskinder Nega and his party Balderas, mobilising the Addis Ababa youth against the qeerroo and, more broadly, Oromo residents. 18 Finally, the chaos in Addis was used by different groups to loot shops, offices and private residences. Residents recounted for example: My friend is Oromo. He was chased and beaten by the Addis Ababa youth in Bulbula. He was on his way home when he was stopped by a group of youngsters. He spoke Afan Oromo with them. He thought they were qeerroo. But they were the Addis youth. So he had to run. 19 In Alem Gena, it was the qeerroo from Alem Gena who attacked Amhara residents. I have seen many of them before. They live in the neighbourhood. They know who is who. 20 There was looting everywhere. And it's hard to know who did it. In our neighbourhood it was also the kids from the neighbourhood who took the opportunity to participate in the chaos. One of the clothes stores was destroyed and robbed. And the next days you could see our shoe shiners in new outfits. 21 Unrests in Addis Ababa only persisted for two and a half days. After initial reluctance and delay to intervene, the government re-established order by force. The borders of the capital were closed and heavily controlled, to prevent the qeerroo from the outside from accessing the capital. A countrywide internet shutdown, lasting for

Exploring the Addis Ababa Exceptionalism
In Addis Ababa, levels of physical violence in inter-and intra-communal conflicts have remained low in the national comparison, as well as when compared with other urban areas in the region. Peace in the everyday lives of Addis Ababans has been enforced through a set of practices and norms, that has allowed residents to avoid and minimise conflict. Avoiding contentious topics, adopting an urban identity concealing ethnic origins, deferring blame for violence to outsiders of the capital and continuing polite neighbourly relations in the face of conflict, were but some of the strategies deployed to self-enforce 'peace' amongst Addis Ababa residents. Addis Ababans often describe the absence of physical violence and unrest as 'normal' and where violence and unrests have occurred, residents have collectively expressed a strong urge to 'go back to normal'. 23 However, sustaining as well as re-establishing 'the normal' in Addis Ababa, has often meant not acknowledging and not addressing existing grievances, structural forms of violence and discrimination. Despite residents' attempts to enforce peace, the government has been involved in fuelling ethno-communal tensions through negatively targeting some groups, as well as through promoting some groups over others.

Ethnicity and Ethnic Discrimination in Addis Ababa
According to the 2011 national census (Gregorian calendar), Addis Ababa is home to individuals representing over eighty ethnic groups in Ethiopia (Central Statistical Agency, 2007a). Although Addis Ababa is a multi-ethnic city, de-facto it has been dominated by the Amhara, leading to what Oromo nationalists have called the 'Amharisation' of the capital. According to the last national census conducted in 2011, 47% of Addis Ababans were Amhara, 19.5% Oromo and 33.5% from other ethnic backgrounds. 24 The dominance of the Amhara amongst Ethiopia's rulers until 1991, and the ensuing overrepresentation amongst national political and economic elites, has favoured their relative overrepresentation amongst the Addis Ababa residents, as well as in other important urban areas. 25 Whilst Addis Ababa is located in Oromia regional state, the official administrative and working language in the capital is Amharic and the majority of Addis Ababans, including many Oromo born and raised in the capital, do not speak Afan Oromo. 26 The relative marginalisation of the Oromo and the Oromo language and culture has caused grievances amongst Oromo nationalists, including the qeerroo movement. Whilst voiced publicly, amongst others by the qeerroo, many Addis Ababans do not believe that there exist legitimate claims of the Oromo to Addis Ababa. In 25 (Central Statistical Agency, 2007). 26 Interview with resident from Figa, 19.12.2020; informal conversation with resident from Sidist Kilo, 12.12.2020. 23 Informal conversation, resident from Haya Arat, 02.07.2020. 24 The census did not account for multi-ethnic backgrounds. In multi-ethnic families, until today, children are registered under the ethnicity of their father (Central Statistical Agency, 2007a, p. 25f). the interviews, the city has for example been described as a 'multinational territory', 27 'independent from Oromia'. 28 Moreover, many Addis Ababans, including residents of Oromo origin, hold that in the capital, 'ethnicity is not important'. 29 They describe their families as well as social and professional circles as being ethnically mixed and, consequently, many do not identify by their respective ethnicity. Residents explained for example: In Addis Ababa, it's a metropolitan area. Not one group dominates. It's mixed. 30 We are all alike. Ethnicity is not a strong sentiment. There is no difference. We are all the same. 31 According to many Addis Ababans interviewed, rather than by ethnicity, their identity is defined by metropolitan ideals and norms. When asked about their origin, many identify as Ethiopian, all the while acknowledging that 'ethnicity is important in the countryside'. 32 The idea of 'ethiopiawinet', ethiopianness, is pronounced amongst many Addis Ababans, often presented as modern and contrasted to the 'backwardness' of an ethnic-based identity. However, there is little recognition that outside of Addis Ababa, and especially amongst different Oromo nationalist groups, ethiopiawinet is often perceived as an idea and ideal dominated by Amhara culture, which does not account for the multinational character of the country. This is particularly true for Amhara Addis Ababans, many of whom de-emphasise their ethnicity 33 and do not recognise the Amhara dominance in Ethiopian history and its continued effects. An interviewee explained: Whilst many Addis Ababans de-emphasise ethnicity and refute claims to the capital based on ethnicity, hostility towards specific ethnic groups in Addis Ababa is not new, nor has it simply targeted the Oromo. One group that has been subjected to discrimination are the Tegaru. Around the 2005 national elections, public political debates in the capital revealed a clear hostility towards Tegaru. Common statements used were 'Ika wede Kebele, Tigray wede Mekele', meaning 'All things to the Kebele and Tigrayans to Mekele'. Addis Ababans and particularly its urban elites perceived the ruling coalition EPRDF as a surrogate of its Tigrayan member, the Tigrayan People's Liberation Front (TPLF), and associated the Tegaru in Addis Ababa with the TPLF. Fifteen years later, after the outbreak of an armed conflict between the TPLF and the ruling Prosperity Party in November 2020, Tegaru residents in Addis Ababa 27 Informal conversation with resident from Bulbula, 04.05.2021. 28 Informal conversation with resident from Hayat, 13.08.2020. 29 Interview with resident from Hayat, 15th of December 2020. 30 Interview with resident from Sarbet, 15.12.2020. 31 Interview with resident from Alem Gena, 16.12.2020. 32 Informal conversation with resident from Hayat, 02.08.2020. 33 Informal conversation with resident from Bole Bulbula, 01.11.2019; informal conversation with resident from Bole, 14.12.2020. 34 Informal conversation with resident from Hayat, 01.12.2020. were ethnically profiled, suspended from work, prevented from leaving the country and arrested and their bank accounts were frozen 35 (Marks & Abdi Latif, 2020). Addis Ababans of Tegaru origin also reported that they were harassed by neighbours of non-Tegaru origin (Freudenthal, 2020), whereas some non-Tegaru residents interviewed acknowledged that they supported the government actions. 36 Since the outbreak of the qeerroo protests, anti-Oromo sentiments have become more pronounced amongst Addis Ababans. Only few non-Oromo Addis Ababans interviewed have described the claims of the Oromo towards the capital as legitimate. Residents held for example that the Oromo special interest in the capital poses a threat to the existence of Addis Ababa as a multi-ethnic city. They further explained that the qeerroos' claim to resources threatens non-Oromo Addis Ababa residents, who might lose their property and investments. Finally, they noted that the qeerroos' demand to add Afan Oromo as a federal language would give the Oromo an unfair advantage over other ethnicities, with respect to their ability to claim jobs in federal government institutions. Rumours about the new government actively recruiting Oromo into the Addis Ababa city administration, 37 granting title deeds to Oromo based on false claims of ancestry, 38 illegally distributing government housing to Oromo 39 and handing out resident cards to Oromo who are not fulltime residents in the capital 40 , have further fuelled negative sentiments.
Since the outbreak of the qeerroo protests, many Addis Ababa residents have felt uncomfortable about arriving and newly settled Oromos. Not seldom they have called them 'local qeerroos' 41 and commented on their failure to 'speak proper Amharic', 42 'their laziness' 43 and their 'low level of education'. 44 In some neighbourhoods, conflicts have arisen between long-term residents and newly settled Oromos over land. In several cases, the kebeles (local government offices) are said to grant title deeds to arriving Oromo based on claims of ancestry, whereas in other cases, the new arrivals have simply fenced land without acquiring a title deed prior. Rumours have suggested that the Oromo kebele officials have been collaborating with the newly arriving Oromos, sharing profits from selling title deeds. 45 In many cases, the new arrivals from Oromia have built provisory housing structures, lacking financial resources for formal construction, and have tried to quickly sell the plots. However, a growing number of concrete constructions with a pointed roof structure are being erected, colloquially called 'qeerroo houses' (see Fig. 2). Given that the significant resources needed for 35 Interview with resident from Haya Arat, 18th of November 2020; informal conversation with resident from Hayat, 12th of December 2020, fieldnotes, 11th of November 2020. 36 Informal conversation with resident from Kotebe, 12.04.2021; informal conversation with resident from Haya Arat, 10.01.2021. 37 Informal conversation with resident from Kasanchis, 15.08.2020. 38 Informal conversation with resident from Mekanissa, 17.11.2020. 39 Informal conversation with resident from Kasanchis, 28.07.2020. 40 Informal conversation with resident from Tor Hailoch, 21.11.2020. 41 Fieldnotes, Pellerin, 29.11.2019. 42 Informal conversation with resident from Mekanisa, 01.12.2020. 43 Informal conversation with resident from Kasanchis, 31.03.2020. 44 Informal conversation with resident from Mekanissa, 19.07.2020. 45 Fieldnotes, Pellerin, 17.05.2021. such constructions, many Addis residents interviewed speculated about who was providing the finance for such constructions. Several concluded, that it was unlikely that children of displaced farmers claiming plots based on ancestry were able to gather the necessary finance by themselves. 46 Oftentimes, the land had been used by longterm residents prior to recent fencing of the plots, sometimes for purposes such as farming or recreational activities, whereas in others, long-term residents actually possessed title deeds for these plots but had not built yet on the land. 47 This has led to the emergence of conflicts over land, creating animosity between the new arrivals and established residents. 48 Whilst many non-Oromo residents in Addis Ababa openly hold negative stereotypes about the Oromo and perceive them as threat to the capital, long-term Oromo residents are not described or perceived as Oromo. Instead, non-Oromo interviewees have explained that 'they [the long-term Oromo residents] are one of us' 49 and 'they can't throw stones or destroy things'. 50 Rather than identifying them by their ethnicity, long-term Oromo residents have been identified by non-Oromo residents through the shared metropolitan identity. Living in the capital for an extended period of time has been taken as a sign that people drop their strong ethnic affiliation and ties and instead grow into the melting-pot. Moreover, it has been recognised that the long-term Oromo residents were also victims of the qeerroo protests in Addis Ababa and that the '[…] ethic Oromos in the communities they were scared like us'. 51 Long-term Oromo residents in Addis Ababa also emphasised how they and their families had become victims of the qeerroo protests. 52 Consequently, the qeerroo protests have been perceived as Oromo from outside of Addis Ababa threatening 48 Interview with resident from Lebu, 27.04.2021. 49 Interview with resident from Gerji, 19.12.2020. 50 Informal conversation with resident from Bulbula, 11.07.2020. 46 Informal conversation with resident from Bulbula, 05.06.2021. 47 Informal conversation with resident from Bole, 11.05.2021. 51 Interview with resident from Gerji, 19.12.2020. Fig. 2 Pictures of suspected land grabbing in Addis Ababa. Photo to the right, so called qeerroo house with characteristic roof structure. Photos in the middle and left: new fencings and suspected land grabbing by individuals from Oromo origin the capital's multi-ethnic and metropolitan population. The protests have challenged the everyday peace in Addis Ababa, which has been largely built on avoiding conflict, rather than addressing ethno-communal tensions.
Middle-class Addis Ababans described their metropolitan identity as defined by urban characteristics, attitudes and values. They held that their lifestyle differed significantly from rural Ethiopia and was similar to the lifestyle in metropolitan areas elsewhere in the world. Access to education and work in the formal sector, use of Western style clothing, possession of a smart phone and/or access to the internet and TV, long commutes to work, expensive and overcrowded accommodation and continuously rising living costs were but some characteristics mentioned. Urban attitudes, according to the Addis Ababans interviewed, were fostered through a strong spatial identification with the metropolitan area. Addis Ababans supported the idea of unified governance and national identity, instead of ethnic-based governance arrangements and regionalism. They were interested in national politics and the economy, rather than regional questions. Finally, metropolitan values, according to Addis Ababans, were often related to the importance attached to making it to middle class status. People judged that what was important in life was educating oneself, earning a good income and founding a family. Moreover, they deemed that stability was key for the economy to prosper and hence attached high importance to the state's ability to deliver a stable political environment.

Understanding the July 2020 Protests, the Addis Ababan Perspective
Most Addis Ababans interviewed described the July 2020 riots as an 'exceptional event', something that was not comparable to previous incidents in the city. Most commonly, it has been described as a conflict between qeerroo entering the capital from the surrounding Oromia region and Addis Ababa's multi-ethnic population. Many Addis Ababa residents interviewed have struggled to make sense out of the failure of the government to intervene timely; however, most have supported harsh measures taken after the initial lag in reacting. Whilst the re-establishment of the so-called normality has been welcomed by many, a persistent sense of insecurity amongst Addis Ababans has challenged the residents' ability 'to go back to normal'.
Whilst a few residents compared the post Hachalu riots to the 2005 protests that broke out after the general elections, they clearly emphasised that in 2005, the violence pitted 'people against the government' whilst the 2020 violence pitted 'people against people'. 53 In the riots following the Hachalu events, Addis Ababans described that they were targeted, whilst the 2005 protests targeted the incumbent government. The residents shared a sense of victimhood and betrayal by the government for failing to protect them from the qeerroo and even Oromo Addis Ababa residents felt uneasy and saw themselves as targets. An Oromo resident interviewed stated for example, "I am Oromo and I was scared for my family too". 54 Residents explained that it was unseen that the government allowed large-scale violent protests 53 Interview with resident from Hayat, 15th of December 2020. 54 Interview with resident from Figa, 19.12.2020.

3
Unpacking the Addis Ababan Exceptionalism-Living and Making… to break out in Addis Ababa, since it is the home of the federal government. In 2005, the government reacted swiftly and with force to the protests, causing many lives lost. In 2020, the police intervened with nearly 24 h of delay and some Oromo Addis Ababans explained that they had tried to serve as peacemakers and informed the qeerroo in Afan Oromo that their neighbourhood was filled with 'good people'. A resident explained: 2005 was the last time we saw this level of violence. Addis Ababans explained, when the qeerroo protests broke out in 2014/15, they mostly did not reach Addis Ababa. Although they occasionally touched the Oromia special zones surrounding Addis Ababa, similar to the 2005 events, the protests targeted the incumbent government and public property. 56 Mostly the protesters were described as youth living in the Addis Ababa suburbs who demonstrated against the authoritarian government. The government crushed the protests at the costs of lives lost. 57 The city centre of Addis Ababa remained largely a bubble of imposed peace and calm as the city's boundaries were closed by the government. Addis Ababa residents held that it was impossible to stage protests in the capital, as the government would not allow its power to be challenged right at the centre of its administration. A resident explained: Addis Ababa is a bubble. We are shielded from reality. The mengist [government] is here. No one contests the mengist in its home. He would not allow. 58 According to Addis Ababa residents, the 2020 riots following Hachalu's death clearly broke with previous patterns on several levels. First, the protesters were identified as the qeerroo entering the city, not residents from Addis Ababa and its suburbs. Second, the protesters targeted residential and commercial property, as well as government offices, civilians and the state's security forces. The perceived change in the protest and violence patterns meant that non-Oromo and Oromo Addis Ababans interviewed, even those few who initially supported the qeerroo protests, felt that targeting of civilians took away any potential legitimacy that the qeerroo protests might have had in the beginning. 59 According to Addis Ababans, instead of expressing legitimate grievances, the qeerroo were pawns in a larger struggle for power over federal politics. Many held that the majority of the qeerroo had been organised by Oromo nationalist opposition groups, to instigate chaos in Addis Ababa and to put pressure on federal politics to give more power to the Oromo. Addis Ababans 55 Interview with resident from Hayat, 15.12.2020. 56 Interview with resident from Alem Gena, 16.12.2020. 57 Informal conversation with resident from Kasanchis, 29.10.2016. 58 Interview with resident from Old Airport, 03.11.2016. 59 Interview with resident from Alem Gena, 16.12.2020; interview with resident from Meskel Flower, 15.12.2020. held that the qeerroo had been brought from the countryside by bus and given some money to participate in the protests: [The qeerroo]came from the outside. Those were brought by busses from Ambo. They asked where is Haya Hulet 60 ? They got dropped here with 300 birr. That pays a good lunch. That lunch could give them power for anything. 61 The qeerroo, most of them didn't even know what they were protesting for. 'it's not really about grievances. Some of them don't know what they are doing. They just blindly follow the calls for violence. 62 Addis Ababans nearly unanimously held that the government had failed to intervene, despite many signs indicating that violence would follow the murder of Hachalu. Whilst some residents suggested that the police had simply lacked the resources to intervene and keep the situation under control, 63 many suggested that parts of the police were actually supporting the qeerroo, 64 whilst even others held that the government waited to intervene to demonstrate to Addis Ababans the danger of the qeerroo and to justify state repression in order to control them. 65 Addis Ababans expressed that the riots could have been prevented, had the police intervened timely. Consequently, they ascribed responsibility to the government regarding the violence that unfolded in the capital: The police didn't intervene. They are the same. They are qeerroos in uniforms. They were contributing to the protests. They didn't intervene. They let it happen. They also participated. 66 I have a testimony from my mum. She met a police officer early in the morning. The police officer told her that they had been ordered not to do anything. In Sheshemene, Shimeles called the Mayor and told him to go home and sleep. He told him not to intervene. If Shimeles replies in this way this is all instigated by Abiy. The arrests of Jawar, Bekele and others. This was political imprisonment. The postponing of the elections caused problems. So they had to silence them. 67 The police here. They didn't have the resources. So they let the Addis residents defend themselves. 68 After the initial delay, the government intervened harshly. Addis Ababans interviewed widely supported government measures taken, including mass imprisonment of protesters, arrest of opposition politicians, forced closing of media stations allegedly supporting the violence and a countrywide internet shutdown. Many Addis 1 3 Unpacking the Addis Ababan Exceptionalism-Living and Making… Ababans described the qeerroo as 'barbaric ', 'uncivilised' and 'unpredictable', 69 reflecting the narrative the government portrayed about the qeerroo. Reports of riots in other part of Oromia that involved ethnically and religiously targeted attacks perpetrated by the qeerroo in other cities in Oromia only strengthened these perceptions. Addis Ababa residents shared news stories and stories they had heard through friends, family and other individuals or watched on government media, stating gruesome details on alleged incidents in the qeerroo riots following Hachalu's death. Addis Ababans explained: In Zeway they [the qeerroo] came to a house of Amharas. The husband begged them for mercy. But they came in and killed him. They put their weapons on his wife's belly. She was nine months pregnant. She died in hospital afterwards from the shock because she could not give birth. 70 They targeted non-Oromo and orthodox Christians. They slaughtered them like animals, dragged them through the streets and left them there. 71 Many Addis Ababa residents interviewed repeated the Rwanda metaphor used by government, 72 to emphasise the danger emanating from the qeerroo.
During the Hachalu events. In Sheshemene, people were attacked because of their ethnicity. I think this is genocide. People are attacked for their ethnicity, you can call it genocide. They called for the death of the Neftenjas. The Amhara were the victims in Oromia. Mostly the Amharas were attacked. I have a friend who have family in Sheshemene. They had two houses. They were both burned down. They have lived in Sheshemene for two generations. But they were Amhara. 73 Due to the internet blackout that lasted for 3 weeks, the collective processing of the event primarily relied on available government reporting as well as pieces of information gathered through social and community channels. Instead of objective and verified information, residents had access to rumours and one-sided reporting, further fanning fear of the qeerroo. This crucially shaped the understanding of Addis Ababans of the events. Rumours were influential in shaping the perceptions of Addis Ababans and were circulated through text message during the internet shutdown. One example was a rumour about the qeerroo having poisoned drinking water in Addis. The rumour was circulated over text message and spread so widely that the Addis Ababa Water and Sewage Authority had to go on TV to calm the citizens and diffuse the rumour. 74 Addis Ababans' perception of safety in the capital changed radically within a matter of a few hours. Residents drew hasty conclusions when they saw or heard suspicious events, individuals or noises, contributing to the diffusion of unverified 69 Fieldnotes, Pellerin, 11.07.2020. 70 Fieldnotes, Pellerin, 09.07.2020 Informal conversation with resident from Maganagna, 06.08.2020. 72 Summary from government new reporting, including a speech from Abiy Ahmed, 01.07.2020. 73 Interview with resident from Sarbet, 15.07.2020. 74 Fieldnotes, Pellerin, 03.07.2020. information about the riots and related events. 75 Even after the forceful re-establishment of law and order, Addis Ababans interviewed acknowledged their fear that similar events could repeat themselves. Several individuals explained that they had moved since the riots or that they had tried to protect their residences from further potential attacks: You should see the walls they built around their house. It looks like the US Mexican border. I don't want to know how it feels from the inside. Like staring at the Great wall of China. My friend told me 'it's a qeerroo protection wall'. They qeerroo proofed their house. No one can throw a stone over the wall even. 76 We moved into a gated community now. Closer to the city. It's a little bit safer. The compound is in front of the federal police camp. To kill me, they first need to kill all the federals. So it's safer now. But still, we cannot escape if it happens again. 77 Whilst life went 'back to normal', some Addis Ababans acknowledged that the calm was deceitful.
We don't know or understand anymore really what is going on. The city centre is so calm. Everything looks normal. People go about their lives as usual. But underneath, where we can't see, things are boiling. It looks calm. But it's really not. 78 The streets are full of people. Everything looks normal. Addis is a bubble. We don't feel anything here. Life continues business as usual while in other parts of the country people are dying. 79 The events also left impacts on residents' private lives. Many were unable to process the events with family and friends, due to disagreements about what had happened and the role of the government in the events. The internet shutdown and limited access to information exacerbated confusion amongst many residents, who were unable to assess what and whom to believe. Whilst Addis Ababans interviewed mostly agreed that the government had to re-establish order after the riots, some Addis Ababans questioned the government's narrative of the events as well as the government's way of managing the situation. Those residents who described themselves as deviating from the mainstream perception of the Hachalu events and their aftermath often held it was best not to openly voice their opinions to avoid conflicts. A sense of hopelessness and collective depression was tangible amongst those individuals, who did not buy in the government narrative that law, order and peace had been re-established.
Yesterday we got into a fight. I think a lot of families are fighting at the moment. People are worrying.

Unpacking the Addis Ababan Exceptionalism-Living and Making…
Even members of the same ethnicity disagree. There is so much argument about the political situation. 80 It's too emotional. And I don't want to have the conversations with some people. I have people in my circle from both extremes. And I don't want to discuss with them. I want to stay friends. But I know when we start talking about this, we might not. 81 A few months after the riots, life in the capital seemed to have gone back to normal. According to Addis Ababans, tensions within their communities had not increased as a result of the violence and destruction following Hachalu's murder. Instead, everyday peace seemed to be re-established with residents avoiding conflictual topics, emphasising their shared identity and upholding peaceful community relations. Living in the present, rather than looking back at the conflict, allowed Addis Ababans to return to what they described as 'normal' community relations. Some held: I think it was a momentary thing. Everything is forgotten already. We are like this. The Ethiopian middle class, we like to forget. It's a syndrome. … 82 After the incidents, in my neighbourhood, there was not more tension between the Amhara and the Oromo. You continue going out with your Amhara friend. You grab coffee. You socialise. Life goes on and people continue just as they were. … 83 Whilst normality was re-established on the surface, the events left a mark on the capital and its inhabitants. Despite claims that community cohesion in the capital had not been affected by the Hachalu events, election results for the 2021 national elections published by NEBE (National Election Board of Ethiopia) revealed that many Addis Ababans voted for Balderas and its 'Addis Ababa to the Addis Ababans' programme. Balderas came second after Abiy's Prosperity Party. Formally promoting national unity over ethnic federalism, Balderas has been accused of supporting Amhara nationalist tendencies and promoting an anti-Oromo agenda. The party has also formally been associated with Amhara nationalist forces, entering a collaboration with NaMA (National Movement of Amhara) for the 2021 national elections (Sisay, 2021). Such developments question the idea of the Addis exceptionalism, the absence of ethno-communal conflict and the idea of a truly multi-ethnic and multi-cultural city. Instead, they reveal structural violence and ethno-communal tensions and demonstrate how larger political developments in the country also affect life in the capital.

Conclusion
How do Addis Ababa residents make sense of and (dis)engage from/in violent conflict and ethno-communal violence? Drawing on the perceptions and experiences of long-term, middle-class residents, this article finds that ethno-communal 80 Informal conversation with resident from Arat Kilo, 13.07.2020. 81 Informal conversation with resident from Haya Arat, 17.07.2020. 82 Informal conversation with resident from Hayat, 01.12.2020. 83 Interview with resident from Alem Gena, 16.12.2020. violence and conflict between the city's residents constitute a rare occurrence. According to residents, ethnicity and ethnic-based grievances have not constituted powerful tools for mobilisation in the capital, given the city's multi-ethnic character and residents' strong metropolitan identity. Instead, Addis Ababans reported to have rejected ethnic-based claims made by non-residents towards the capital's resources and to have collectively defended their right to the city. Residents interviewed also emphasised the role of the government in preventing violent protests in the capital.
Whilst explaining the relative absence/rare occurrence of ethnic-based mobilisation in the capital, the narrative of Addis Ababans omits how the idea of a multiethnic and multicultural capital belonging to its residents has caused grievances amongst the Oromo and favoured ethnic-based mobilisation outside and in the peripheries of the capital. Interviews revealed that in the capital, long-term Addis Ababa's residents have been pitted against newly arriving Oromo in the city, as well as Oromo from outside of the capital. The occurrence of large-scale physical violence in the capital after the murder of Hachalu Hundessa has invited a more critical reflection on the idea of Addis Ababa being an 'exceptionally peaceful city'. Interviews revealed the existence of everyday structural violence and conflict, suggesting that Addis Ababa might in fact be more adequately described as 'ordinary violent'. Everyday peace in the capital has built on avoiding, rather than minimising conflict. Even after the Hachalu protests, Addis Ababans managed to re-establish peaceful community relations through avoiding addressing contentious topics. Whilst Addis Ababans and the government managed to prevent large-scale physical violence from re-emerging after the Hachalu protests, the events have rendered the fragility of the capital's 'normal life' visible and have exposed the risk of the bubble being pierced.
Analysing how residents themselves make sense of the rare occurrence of violent protests and ethno-communal violence and engage in/with such conflicts constitutes an important contribution to urban violence and conflict studies. Instead of exploring possible causal relations between urbanisation and conflict, the article addresses the complexities of conflict, but also peace, in an urban setting from below. It reflects on people's understanding of 'conflict'/'peace' and explores their everyday manifestations in citizens' lives. The article reveals that middle-class Addis Ababans interviewed largely understand the 'peaceful exceptionality of Addis Ababa' as signifying the absence of physical violence. However, this does not account for the existence of structural forms of violence and conflict. The article further demonstrates that peace in the everyday lives of Addis Ababans has relied on the adoption of a series of norms and practices, largely destined at avoiding, but not necessarily minimising, conflict: Avoiding addressing contentious topics, such as residents' ethnic backgrounds and the Oromos claims towards resources of the capital; blaming outsiders for causing conflicts and trying to live in the present rather than dealing with past trauma and conflict are but some of those. Interviews revealed the difficulties of navigating a multi-ethnic environment in a deeply divided society and reflect what 'everyday' peace (Mac Ginty, 2014) for people can mean in a larger context of conflict.

Unpacking the Addis Ababan Exceptionalism-Living and Making…
Focussing on the perspectives and experiences of long-term middle-class residents, the article captures how the capital's (relatively) privileged residents have understood the violent protests and ethno-communal violence around Hachalu Hundessa's death. Their residences and property constituted key targets in the protests, as they were perceived to be amongst those benefitting from current political arrangements. Whilst the article contributes important knowledge about lived experiences of violent protests and ethno-communal violence in Addis Ababa, it does by no means capture or reflect the perceptions and experiences of all of the capital's residents. Who experiences the city as peaceful or violent might very well depend on the respective positions of individuals in society. Future research should explore the perspectives of the urban poor and other marginalised groups, to analyse their understanding of and relationship to the violent protests and ethno-communal violence in the capital. Moreover, studying the perspectives of newly settled Oromo residents is crucial, given that some became targets of violence perpetrated by other Addis Ababa residents.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/.