Untitled Document

Coffee and the State in Rural Ethiopia

Sarah Howard (Goldsmiths)


Coffee drinking is fundamental to social life in Ethiopia. Based on research in eastern Amhara Region between 2011 and 2015, this paper explores the omnipresent buna ceremony during which coffee is prepared and served, and its role in the lives of rural government workers as an occasion for building group solidarity as protection against the hardships they face. While Ethiopian society is commonly portrayed as highly authoritarian and hierarchical, this ethnographic account of the social lives of low-level officials complicates the picture of a strict divide between state and society, and is a contribution to calls for attention to the ways in which material practices continually constitute the state as a reality.

Winner of the 2017 Christine Wilson Award
from the Society for the Anthropology of Food and Nutrition (SAFN)


If I ever complained of a headache during my research, I would be advised to drink coffee. Frequent coffee drinkers like many Ethiopians, the people in my research area were well aware of the physical attributes of coffee, including caffeine withdrawal headaches. However, the non-material properties of coffee far outweigh its stimulant effects: recent ethnographic work from a coffee-producing area of northern Ethiopia has shown that commensality, or the sharing of substance, is a primary social and moral idiom among the Amhara, both in the religious sphere of Orthodox Christianity and in the secular social world (Boylston 2013). The buna ceremony, the quotidian and omnipresent way in which coffee is prepared and served, works to ‘establish, deepen and anchor social connections’ (Knowles 2014: 163), as an index of hospitality and as a venue for both constructive discussion and hurtful gossip.

In this paper I will look at the role of buna as an instance of commensality among rural government workers. During around fourteen months of ethnographic research between 2011 and 2015, mainly in one kebele1(administrative unit) in eastern Amhara Region, I was interested in the disconnect between the reach and control that the Ethiopian government is popularly supposed to hold over its citizens, and the lived reality of low-level state workers, ‘the marginal officials who constitute the bureaucratic frontier’ (Singh 2012: 120-1), who are charged with exerting this control. Coffee drinking was not my focus, but as an unavoidable feature of social life, I was struck by its potential as a lens through which to look at the government workers’ everyday lives. I hope that a close look at coffee consumption practices in the villages where these workers are posted will illuminate social dynamics in which they both wield power and are dependent on others. In doing so, I will add to the debate between those who posit that Ethiopian society is structurally hierarchical (Poluha 2004; Vaughan 2011) with an authoritarian and autocratic political culture (Lefort 2012; Abbink 2014), and those who see relations between the Ethiopian state and its citizens as ‘outcomes of particular negotiations, conflicts, and arrangements of relationships of force’ (Di Nunzio 2014: 436; see also Hagmann 2006), or who question the prevailing association of hierarchy with dominance (Malara and Boylston 2016).

Popularly associated with togetherness and social unity across religious and ethnic boundaries, ‘the communal sharing of coffee bolster[s] collective empathy and solidarity even in conditions of utmost poverty’ (Frehiwot 1998: 181). Dropped into the mouths of infants from a few months old, Don Seeman defines ‘buna practice as a privileged site of culture’s transaction’ (2015: 737) for Ethiopians at home and abroad. Ethnographic literature documents buna drinking as a site of specifically female bonding (e.g. Yedes et al. 2004), as well as being notorious for engendering gossip about those not present (Seeman 2015). Coffee is both symbolic of the nation, the probable original source of the wild coffee plant (McCann 1995: 149), and vital for the national economy: the domestic market consumes more than half of total production (ICO 2017), and the remainder makes up around a third of all exports (Farley 2013). Much more than just a cash crop though, coffee plays a vital role in marking occasions of all kinds, and is a material medium through which to make contact with the spiritual (Boylston 2012: 181), as well as an ordinary part of daily life for the majority of people. However, some evangelical Christians and reformist Orthodox Christians refuse to drink it. Their reasons include its addictive properties, which are associated with harmful zar and, less commonly, buda spirits,2 and its association with malicious gossip, loss of self-control, and the ‘laziness’ imputed to the long buna ceremony (Boylston 2012). In urban areas, others reject it not as part of a group identity, but as a type of individual self-making practice. Refusal is a way of setting themselves against what is construed as old-fashioned and rural-inflected, partly influenced by development discourse on ‘backward’ attitudes to work. As I have witnessed, some refuse not the substance of coffee itself – they may drink coffee in cafes or prepare it at home for themselves – but avoid buna ceremonies, either through fear of hurtful talk or a wish to use their time differently (Seeman 2015).

In this paper, however, I will be focusing on its routine usage in the everyday lives of rural public sector workers, who are both representatives of the state and the ‘ordinary people’ through whose ‘lives and practices the idea of a state takes shape’ (Navaro-Yashin 2002: 122). Following Philip Abrams’s (1977) seminal article questioning the existence of the state as an unambiguous empirical fact, anthropological writing about the state has emphasised its constructed and contingent nature, showing how the state is reified as separate and as encompassing the local (Ferguson and Gupta 2002), how ‘state effects’ can be associated with globalised entities such as NGOs (Trouillot 2003), and the futility of trying to distinguish mundane material practices from an abstracted notion of an ideal state (Mitchell 1991). Such analyses posit state formation as a process of continual and improvised construction by a heterogeneity of actors. However, Thomas Bierschenk and Jean-Pierre Olivier de Sardan argue, ‘little empirical attention has been devoted to actual “state apparatuses” in Africa’, as opposed to ideas about the state and activities at the margins of the state (2014: 4). As they perceive it, there is a focus on seeing and imagining the state shorn of social and historical context rather than ‘doing the state’: the often improvised and contested nature of bureaucrats’ banal daily practices. Answering Akhil Gupta’s call for anthropologists to pay attention to ‘the broad base of the bureaucratic pyramid’ (1995: 376), there have recently been a number of ethnographic accounts that illustrate the nature of the quotidian mundane practices carried out by state officials in Africa (e.g. Blundo and Le Meur 2009; Körling 2012). This paper aims to contribute to that emerging body of work, concentrating on the government workers not as bureaucrats trying to implement government policy, but as social actors in the places where, often unwillingly, they are sent.

Government Worker Solidarity

I will start with an ending. When I left my rented room in the rural area of my fieldwork site, I organised a final coffee ceremony in the compound, as I knew was only right and proper to mark all events and occasions. It caused me considerable social anxiety, as I still wasn’t sure about the correct etiquette for invitations despite hosting ceremonies before. On this occasion, as previously, I sensed it would have been a faux pas to make too much of it. After all, coffee drinking was an absolutely ordinary and unremarkable part of normal life, and advance notice was not usually needed; whoever was around or passing by at the time would come, or so I thought. Not wanting to make a big deal of my leaving (which in any case was only to the nearest large town), I expected anyone nearby to attend, and also thought word would spread organically among the government employees who had kindly included me in their own coffee ceremonies, after I casually mentioned it to a few of them the day before. That turned out to be a mistake, as word did not spread to everyone. Over the following days before I left I encountered some reproachful remarks from those I should have asked personally, such as the head teacher, and worse: others too shy to say anything showed a more distant demeanour towards me, even when I returned several months later. I realised that I had been found guilty of favouritism to those I had asked directly: unlike farmers I knew in the village, whose non-invitation was not noticed or marked as a rejection, the government workers constituted a social group and should have been treated as such.

What my error reveals about social relationships is that, as much as it is a site of conviviality and inclusive sociality – ‘coffee is the substance of togetherness’ (Pankhurst 1997: 536) – buna consumption can also be a productive way to think about exclusion. While certain categories of person – for example, craft workers, some ethno-religious minorities, or those believed to be carriers of harmful buda spirits – are entirely excluded from sharing not just coffee but all substance with others, including a ban on intermarrying (Freeman and Pankhurst 2003), the exclusion I am considering here is one-sided, and neither acknowledged nor disputed. While the government workers do drink coffee on occasion with their farmer landlords in the main living spaces of the famers’ houses, they do not invite the farmers to participate in the coffee drinking they host, which circulates from cramped room to courtyard mattress. I will argue that this is part of what creates ‘government worker’ as a social category. Coffee in the village is never drunk alone – there is always an element of ceremony, which I will describe in more detail below – and the coffee-drinking friendship group, while elastic, is a way of building a group identity. In fact, it is crucial to social acceptance: ‘one who does not maintain their visiting networks, particularly as expressed by the coffee ceremony, cannot truly be capable of achieving full personhood in any Amhara community’ (Boylston 2012: 198). However, I do not see the government worker-exclusive coffee drinking groups as an indication of their superior status, but instead as a function of the necessity to build short-lived zemed- or kin-type relationships between the government workers who are living away from their own families, and whose material lives, from food to housing, depend upon the farmers whom they are sent to live amongst. The ‘blurred boundaries’ between ‘state’ and ‘community’ described by Gupta (1995) are here being partially resisted, not due to hierarchical ideas about government worker superiority or a ‘societally pervasive respect for the state’ (De Waal 2013: 473), but as a means of solidarity to offset their marginal social status in the village.

The Coffee Ceremony

The Ethiopian coffee ceremony is a complex event and requires a certain amount of equipment and time, as well as considerable skill to prepare, almost always by women. Even casual daily coffee contains most elements of the following description of an ideal-typical buna ceremony. Preparation of the coffee starts with the raw green beans, which are rubbed roughly between the hands in water to remove husks and dust. They are then roasted in a flat iron pan, preferably over charcoal. It takes experience to roast the beans to an even dark colour, with neither burned black patches nor raw green ones. When ready, the roasted beans are circulated around the room by the person preparing them, who swishes the smoke towards everybody present so they can take a deep draft of the delicious scent. The beans and optional spices of choice are then pounded to powder in a pestle and mortar before being boiled in a clay coffee pot with a round bottom, called a jebena. Bringing the coffee to a boil, leaving it to settle, tasting, perhaps boiling again or adding more hot water – all these actions show the skill of a coffee maker, and young girls will check with older women how to proceed.

The aesthetics of the ceremony are also crucial. All the accoutrements for service are kept in a miniature lidded table called a rekabot: the small sini cups and saucers, teaspoons, sugar bowl, and the crocheted cover, or dantel. Incense is essential for the ceremony, either a small piece of fragrant wood or a commercial brand of joss stick manufactured in India for the Ethiopian market, with the unfortunate brand name in English of Shitto (meaning ‘scent’). There must also be some form of snack, however nominal: popcorn or roasted grain are traditional, manufactured biscuits a cheap alternative. Grass, leaves, and on special occasions flowers are spread on the floor around the rekabot, or are substituted by a plastic mat with green fringed edges. The female coffee maker should cover her head; preferably, she would be wearing traditional clothes made of embroidered white cotton.

Showing the ‘widespread idealization of coffee as a naturalized constituent of Ethiopian national culture’ (Seeman 2015: 737), one of the most popular decorations in rural houses is a poster depicting such an ideal ceremony, with a beautiful smiling young woman in a traditional white dress with an elaborate hairstyle inviting, ‘ኑ! አቦል እንጠጣ!’ ‘Nu! Abol entata! [Come! Let’s drink the first cup together!]’. The first cup is important, as each pot produces three rounds, with the first round being the strongest and most prestigious. Gender hierarchy in the serving of coffee is pronounced: women are supposed to make the coffee and offer the first round to men in order of age, followed by women and children. Although these gender divisions are generally adhered to among both farmers and government workers, from necessity there is some fluidity among the latter concerning both cooking and coffee making in private spaces. In a coffee ceremony performed to mark an occasion in a public space, however, this would be unthinkable. For example, the buna ceremony held at the local school to mark International Women’s Day in 2014 followed speeches about how men and women can, and should, perform the same work, but was served by two of the female teachers with the customary displays of bodily deference: a covered and bowed head, delivering the coffee cup in a semi-curtsey with one hand while the other hand cups the elbow. Age, wealth, and parenthood are also status markers, and as government workers were all young and poor, and mostly unmarried, they would not be offered the first cups of buna, and even the men may be expected to hand cups around to senior farmers.

Government Work in Context

The existence of salaried state workers in Ethiopia began only in the mid-twentieth century, through the modernisation of education and bureaucracy, and largely consisted of urbanites from the north and central highlands. Until the mid-1960s, the civil service provided unrestricted access to well-paid state employment, ensuring ‘the smooth absorption of educated Ethiopians into the expanding state apparatus’ (Markakis and Nega 1986: 49), until saturation point was reached and recruitment curtailed in the 1960s, especially after economic retrenchment following the Suez crisis in 1967. Disenchanted government workers became radicalised and played a role in the revolution that overthrew the Imperial regime in 1974: teachers were ‘protesting pay and conditions, and their opposition to the education reforms proposed that would limit education to primary and practical level’ (Markakis and Nega 1986: 82). The turbulence of the military Marxist Derg (‘committee’) regime era saw the state enact massive land and structural reform, including large-scale resettlement, implemented by new and often zealous cadres; for example, in 1975 ‘1,000 men – plus a single woman – were appointed as administrators, development agents and land reform bureau staff, called “apostles of change”, to dislodge the imperial provincial bureaucracy’ (Desalegn 2009: 142). Despite many other ideological and policy differences, the doctrine of state-led development and many of the same structures persist under the Ethiopian People’s Revolutionary Democratic Front (EPRDF), who toppled the Derg in 1991 and remain in power in 2017. However a significant increase in the influence of global financial and development institutions led to an ‘IMF-mandated reform [that] required a significant downsizing of the public sector’ –public sector employees fell from 65% of the male labour force in 1990 to about 30% in 1997 (Mains 2012b: 10). The Derg-era administrators, known as birokrasi, were thus replaced by ‘a new category of local administrators, generally poor young peasants who were judged to be naturally close to the new regime’ with its roots in mass mobilisation of rural populations (Lefort 2012: 687). While there has been a massive increase in access to education and training in the last twenty-seven years, the EPRDF retains a rural focus: Ethiopia has the largest agricultural extension programme in Africa (Lefort 2012: 686), which has seen a shift in ideology in recent years from the prevention of rent-seeking to the promotion of market expansion (Chinigò and Fantini 2015).

Far from crumbling, retreating, or becoming a ‘ghost’, as diagnosed by many writing about the state in Africa (e.g. Graeber 2007: 157-180), the Ethiopian state is commonly accepted to be expanding its reach down to ever smaller units of the population, through policies such as the ‘1 for 5’ scheme, which gives each group of five people one direct supervisor to ensure adherence to development policies, whether in agriculture or latrine building, or within schools, universities, and government departments. After the opening and swift closing of political space leading up to and following the contested and violent 2005 elections, there was an ‘expansion of the local administration councils from about 600,000 members to 3.5 million… and the enlargement of the EPRDF party membership base from about 760,000 in 2005 to more than 5 million’ (Tronvoll 2010: 132), with membership of the ruling party linked to access to goods and services (Lefort 2012). The ways in which this substantial growth of the state apparatus is implemented, its effect on people’s lives and livelihoods, and the potential for resistance against it, are beyond the scope of this paper. However, the co-optation of rural people into the work of ‘doing the state’ echoes the mobilisation of youth in inner city Addis Ababa, and the plan to ‘create a tighter chain of command between the elaboration of policy and political strategies at the top and the activities of political mobilisation and the implementation of development programmes at the lowest levels of the state bureaucracy’ (Di Nunzio 2014: 441). The extension of responsibility for government objectives to large numbers of citizens seems to complicate prevailing ideas about the deep divide between state and society in Ethiopia, as found in the following quote from René Lefort, based on long-term ethnographic research in a different rural area of North Shewa:

On one side, there is “it”: the mengist, a term that signifies simultaneously the authorities as such and the various bodies through which they exert their power – the government, the ruling party, the kebele Cabinet, the state apparatus – and also “them”, all of its members and civil servants. On the other side is “us”: the gebäre, the peasants. But relations between mengist and gebäre are, in essence and concretely, deeply unequal relations of power – relations of command and subordination between dominators and dominated (2012: 699).

This characterisation does not ring true when applied to the daily life of the low-level government workers in my research area. In the next section I will return to the specific situation in which they live, in an undesirable location at the end of a bad road, where there is ongoing serious conflict between farmers and pastoralists, the weather is uncomfortably hot for at least a third of the year, and the infrastructure is poor: water is collected from the river, and electricity only arrived in mid-2014. Obviously these conditions obtain for all those who live there, both farmers and officials, but they are more difficult to navigate for the temporary residents.

Everyday Life

The government workers were not there by choice, and there was huge turnover: between my first visit in 2011 and 2015, only four teachers from a total of up to thirty state employees remained. State workers are assigned to positions, as they had been assigned to their studies. A medical degree is considered the pinnacle, and subjects descend down the hierarchy of desirability to diploma-level primary teaching. This is not an option anyone I met consciously chose, but the result of disappointing grades at preparatory level, and a fate to be rued and cursed: it could mean being sent to a truly remote single-room school as the only teacher. Freshly qualified workers are sent to rural areas like my field site: how long they stay and where they move onto will be determined by ambition, personal characteristics, and social connections. Being assigned to an urban area could take five or ten years of service, but there are many different paths to reaching the end goal, as expressed by most, of a job in a large town. Many change places every year or two within rural areas, trading off various factors: a long walk to the nearest road (eight hours in one case), the availability of desirable food, the degree of isolation, and the weather: hot and dusty as in my research site being deemed the worst. Some employ unconfirmed but suspected strategies, such as family connections or sexual affairs, or deploy unfortunate events, consciously or not, such as an HIV diagnosis or a suspected attack of zar spirits. Many also leave state work for the private sector: running a small business, teaching in non-government schools, or drivinga bajaja three-person vehicle used for short trips.

While in the village, government workers live in rooms that they rent from the richer farmers; there were five or six households that rented out rooms, many shared by two men or a married couple. The rooms have un-plastered mud walls on top of their wooden structure, and tin roofs. They are small and narrow, with enough room for a foam mattress or two, a table or stool for storage, and a bucket for water. Rats and bedbugs are a menace: the hotter the weather, the worse the bedbugs, and in the hot season of April to June, some would rather sleep outside among the livestock. During 2014 a handful of government workers moved into rooms in a newly completed concrete house belonging to a successful delala, or broker, who had made a lot of money sending people illegally to Saudi Arabia via Yemen, including some ex-government workers. However, even the lucky few in the new house live in cramped conditions and all – unlike their landlords – sleep, cook, and socialise in the same small and basically unfurnished room. With no families to provide food in the form of staple grains and pulses, no transport to bring any back from rare visits home, and no place to store it even if so, their access to food, water, accommodation, and other goods depends on the farmers. Coffee ceremonies to which only government workers are welcome are one of the ways in which they share information and come to group decisions about what treatment they will or won’t accept, and how to negotiate their place in the village. They also behind one of the conflicts over access to goods, namely sugar.

The central state periodically places controls on staple commodities, most commonly oil and sugar, to prevent price fixing by merchants. Sugar – an integral part of coffee drinking – is often in short supply and sold on the black market for high prices. In the village, the government workers have to reluctantly pay to buy sugar from farmer-merchants if they run out of their allotted amount. Newly arrived workers, who have not had the chance to build social networks, often struggle to access even basic supplies. One fresh young teacher asked our mutual landlady for an empty bottle so he could buy state-subsidised oil during its limited distribution session, but she said no and he walked off, looking embarrassed and frustrated. A longer-term government worker who had witnessed the event whispered to me, ‘I remember when I was new and no one would help me. It was awful’. The solidarity built during coffee drinking allows the government workers to push back against illegal price fixing by the famers or to put on a display of resistance against rules set by higher authorities, such as when they all refused to buy the subsidised oil because they were restricted to two litres, which was difficult to measure in a standard three litre bottle. One day I came across two teachers leaning close in the narrow pathway between the yellow rock walls of neighbouring compounds, using their notebooks to shade the glare from their eyes while dust swirled around the hems of their long skirts. They were conferring urgently, upset after a recent purchase of grain. ‘She [the farmer-merchant] increased the price by fifty cents a kilo’, said one, the newlywed with a sweet round face and frizzy hair, looking stricken. Later during coffee, she and the others were horrified to discover that on top of these food price increases, I was being charged double the normal rent for my room – not upset for my sake, but because of the risk that the other landlords might find out and increase all their rents. Discussion of living costs dominated coffee that day, with the conclusion that it was not fair but just about acceptable to charge more from a foreigner, but they would not stand for it for themselves and would stick together to resist it. Being private and exclusive to government workers only, buna ceremonies were the only occasions that I heard such complaints discussed openly.

Dissent, in the form of cynicism or critique of their work or of government programmes, was also occasionally aired during these coffee ceremonies, ‘the shared irreverence and defiance’ that Herzfeld (2005: 1) maintains enables the state to continue as a concept and a reality (see also Navaro-Yashin 2002: 155-187). While they may be ‘constrained by the sense of collective identity, however evanescent, that they help to create’ (Herzfeld, 2005: 9), the collective identity and social dynamic they create through their coffee drinking helps them to cope with the vicissitudes, small humiliations, and even dangers of their lives. For example, in 2010 a male teacher was reportedly attacked after being wrongly suspected of having an affair with a local woman; and in 2014, a female student alleged she had been hit by a newly qualified female teacher, leading to threats from the girl’s family. This is a quote from my notes at the time:

The daughter of the man killed by Afar [a pastoralist community]3 accused this new teacher of hitting her – the teacher says she didn’t [do it]…. The girl has a red eye, aggravated by her crying for her father, and her uncle – the brother of the dead man – came to the school and threatened her [the teacher] in front of everyone. He called her a prostitute and said he would kill her in the next five days. So she [the teacher] wants to leave and resign her job, and says it’s better to die of hunger from no salary than to stay there and be killed.

Zemed-, or non-residential kin-, type relations are created among government workers through the reciprocal sharing relationships involved in hosting coffee, for ‘the continuation of a relationship with one’s zemed is premised on continued reciprocity’ (Hendrie 1999: 48). The difference between coffee preparation by farmers and by government workers was striking. Farming families would have their own supplies of coffee and sugar and all the necessary equipment to make it. They would make coffee at least once every day and serve it casually to neighbours, tenants, or whomever was passing by, with an only loosely reciprocal expectation of being hosted in return. For them an individual coffee ceremony was not a notable occasion. Whereas, as I showed in the case of my own socially maladroit leaving buna ceremony, an invitation to drink coffeewas much more important for government workers, as a means through which to build social ties amongst themselves, and also as a relatively infrequent treat. They would use it to mark a special occasion or to pass time on a long slow weekend afternoon – an activity other than washing clothes and listening to the radio. Replacing social mechanisms in place for more settled residents, on the death of one teacher’s grandfather in her home village, all the others contributed three Birr4 so that she could host the necessary ceremony. Overall, money was pooled to buy the coffee, sugar, and snacks in small quantities just for that occasion. This was often the cause for bantering and prevarication over who should pay for what, whose turn it was to ask to borrow equipment, or who – which woman – should be responsible for the preparation. The government worker group was not homogenous, however, and coffee could also be an occasion to show status. For example, one teacher with a sister working as a domestic servant in Kuwait was able to host and pay for more frequent and well-provisioned coffee ceremonies, with several types of snacks and the draw of new music videos on her fancy smartphone.

Despite the tensions between farmers and government workers mentioned above, the ad hoc invitations to drink coffee extended from farmers to whomever is passing show the other side of these relations. Daily interactions between farmers and state workers in the main ranged between cordial and warmly friendly, and there were occasional more lasting examples of trust and reciprocity, including long-term friendships and committed relationships. The situation was far from contentions such as that, ‘where relations between individual peasants and state officials are concerned, these are always vertical, entrenching a hierarchy that prevents trust or interdependence from developing’ (Poluha in Vaughan and Tronvoll 2003: 45). The desirability of government work that Mains (2012a) found in Jimma, and Di Nunzio (2012) in Addis Ababa, was not shared by the farmers from this relatively rich agricultural area – nor by the government workers themselves, for the most part. The farmers openly pitied the government workers for their low salaries despite the hardships of the work, their remove from stable networks of support, and the fact that state employees are prohibited from access to land. Government workers were not distant, high-status figures to be deferred to and respected as hierarchical superiors. As David Graeber writes, hierarchical social relations are characterised by the treating of those higher up as ‘somehow abstract, sacred, transcendent, set apart from the endless entanglements and sheer physical messiness of ordinary physical existence’ (2007: 13), manifested in physical sensibilities that avoid bodily contact between ‘higher’ and ‘lower’ or display overt deference. This was certainly not the case for the government workers I am discussing here. It was common to find a male policeman, teacher, or administrator in close physical contact with a male farmer, in the same way as with other friends, perhaps with their legs entangled or arms wrapped round each other stroking an earlobe. Female workers meanwhile could be victims to physical over-familiarity, such as when I witnessed a clearly uncomfortable first grade teacher being grabbed and her dress pulled down her chest by a relative of her landlord.


A recent development affecting coffee consumption in the village – and across the country – is the return of large numbers of migrants from the Middle East.5 Some female returnee migrants have opened businesses where hospitality in the form of coffee and food is starting to be monetised. Shortly before I left, the landlord’s daughter in one of the larger compounds where government workers rent rooms, herself a returnee from Saudi Arabia, had opened a cafe with signs on the walls warning people that they were not to sit without consuming anything and that they cannot consume without paying. Whether or not this business venture will succeed, it is an example of attempts to break with previous forms of sociality, in which she and other family members would freely offer food and drink to their tenants. Similarly, in the nearest small town, where travellers have to change buses and often wait hours to get to and from the village, another enterprising returnee with a tea shack had purchased unusually large tea glasses and increased the price correspondingly: a small sum, but noted by the financially struggling bureaucrats. Encouraging small-scale entrepreneurship is a government priority (Chinigò and Fantini 2015), so while supposedly the ones taking on the role of ‘shaping capitalist subjectivities’ (Singh 2012: 119) among rural people, low-level government workers may find themselves at further disadvantage, compounding their status as struggling rather than upwardly mobile.

My overall contention in this paper has been that the everyday lives of low-level government workers in rural Ethiopia belie the notion that they are part of strict vertical power relations as envisaged in the social science literature, or that they show a deep division between ‘state’ and ‘society’. While the Ethiopian state is commonly acknowledged to be expanding its reach and control over the population, in contrast to other parts of Africa, state workers in my research area were not treated with deference as the representatives of a hierarchical power structure. Through the medium of coffee practices, I hope to have shown a range of forms of sociality between government workers and farmers, encompassing closeness and trust as well as the material and social disadvantages faced by the bureaucrats, including recent attempts to monetise coffee-based sociality as mentioned above. Although coffee is a material substance that is enjoyed for its taste, smell and physical attributes as a stimulant, it is the social and symbolic aspect of coffee drinking that is important throughout Ethiopia. As I have shown, the social relations that are built and strengthened for state employees through buna drinking in my research area help to mitigate their vulnerability, while, by constituting a government worker group, being part of what makes the idea of the state a reality. Government worker-exclusive coffee ceremonies help separate them as a social entity from farmers, although this distinction does not serve to place them in a position of superiority. At the same time, buna ceremonies are important in building kin-like relationships of reciprocity among the workers, and this familiarity also allows for them to critically reflect on their position in ‘the mundane and uncertain process of producing the nation’ (Mitchell, 2002: 181).

About the Author

Sarah Howard (sarah.howard@gold.ac.uk) is a doctoral candidate in the Anthropology Department at Goldsmiths College.


As per convention, Ethiopian authors are listed by their own (first) name, rather than by their father’s (second) name.

Abbink, Jan. 2014. ‘Democracy Deferred: Understanding Elections and the Role of Donors in Ethiopia’, in Jan Abbink and Tobias Hagmann (eds.). Reconfiguring Ethiopia: The Politics of Authoritarian Reform, pp. 213-239. London: Routledge.

Abrams, Philip. 1988 [1977]. ‘Notes on the Difficulty of Studying the State’, Journal of Historical Sociology 1(1), 58-89. https://doi.org/10.1111/j.1467-6443.1988.tb00004.x.

Bierschenk, Thomas and Jean-Pierre Olivier de Sardan (eds.). 2014. States at Work: Dynamics of African Bureaucracies. Leiden: Brill.

Blundo, Giorgio and Pierre-Yves Le Meur (eds.). 2008. The Governance of Daily Life in Africa: Public and Collective Services and Their Users. Leiden: Brill. https://doi.org/10.1163/ej.9789004171282.i-347.  

Boylston, Tom. 2012. The Shade of the Divine: Approaching the Sacred in an Ethiopian Orthodox Christian Community. PhD Thesis, London School of Economics.

———. 2013. ‘Food, Life, and Material Religion in Ethiopian Orthodox Christianity’, in Janice Boddy and Michael Lambek (eds.). A Companion to the Anthropology of Religion, pp. 257-273. Oxford: Wiley-Blackwell. https://doi.org/10.1002/9781118605936.ch14.  

Chinigò, Davide and Emanuele Fantini. 2015. ‘Thermidor in Ethiopia? Agrarian Transformations between Economic Liberalization and the Developmental State’, EchoGéo 31. https://doi.org/10.4000/echogeo.14141.

Dessalegn, Rahmato. 2009. The Peasant and the State: Studies in Agrarian Change in Ethiopia, 1950s - 2000s. Addis Ababa: Addis Ababa University Press.

de Waal, Alex. 2013. ‘The Theory and Practice of Meles Zenawi: A Reply to René Lefort’, African Affairs 112(448), 471–475. https://doi.org/10.1093/afraf/adt022.

Di Nunzio, Marco. 2014. ‘“Do Not Cross the Red Line”: The 2010 General Elections, Dissent, and Political Mobilisation in Urban Ethiopia’, African Affairs 113(452), 409-430. https://doi.org/10.1093/afraf/adu029.

———. 2012. ‘“We Are Good at Surviving”: Street Hustling in Addis Ababa's Inner City’, Urban Forum 23, 433-447. https://doi.org/10.1007/s12132-012-9156-y.
Farley, David. 2013. ‘Coffeeland’, Afar Magazine, 4 August.
Available at: http://www.afar.com/magazine/coffeeland (accessed 12 September 2016).

Ferguson, James and Akhil Gupta. 2002. ‘Spatializing States: Towards an Ethnography of Neoliberal Governmentality’, American Ethnologist 29(4), 981-1002. https://doi.org/10.1525/ae.2002.29.4.981.

Freeman, Dena and Alula Pankhurst. 2003. Peripheral People: The Excluded Minorities of Ethiopia. London: Hurst.

Frehiwot Tesfaye. 1998. Food Security and Peasants’ Survival Strategy: A Study of a Village in Northern Shewa, Ethiopia. PhD Thesis, University of Toronto.

Graeber, David. 2007. Possibilities: Essays on Hierarchy, Rebellion and Desire. Oakland: AK Press.

Gupta, Akhil. 1995. ‘Blurred Boundaries: The Discourse of Corruption, the Culture of Politics, and the Imagined State’, American Ethnologist 22(2), 375-402. https://doi.org/10.1525/ae.1995.22.2.02a00090.

Hagmann, Tobias. 2006. ‘Ethiopian Political Culture Strikes Back: A Rejoinder to J. Abbink’, African Affairs 105(421), 605–612. https://doi.org/10.1093/afraf/adl020.

Hendrie, Barbara. 1999. “Now the People Are Like a Lord” – Local Effects of Revolutionary Reform in a Tigray Village, Northern Ethiopia. PhD Thesis, University College London.

Herzfeld, Michael. 2005. Cultural Intimacy: Social Poetics in the Nation State. London: Routledge. https://doi.org/10.4324/9780203826195.

International Coffee Organisation (ICO). 2017. Historical Data on the Global Coffee Trade. Total production and domestic consumption (2016/7). Available at: http://www.ico.org/new_historical.asp (accessed 21 June 2017).

Knowles, Caroline. 2014. Flip-Flop: A Journey Through Globalisation’s Backroads. London: Pluto Press.

Körling, Gabriella. 2011. In Search of the State: An Ethnography of Public Service Provision in Urban Niger. Uppsala: Acta Universitatis Upsaliensis.

Lefort, René. 2012. ‘Free Market Economy, “Developmental State” and Party-State Hegemony in Ethiopia: The Case of the “Model Farmers”’, The Journal of Modern African Studies 50(4), 681-706. https://doi.org/10.1017/s0022278x12000389.

McCann, James. 1995. People of the Plow: An Agricultural History of Ethiopia, 1800-1990. Madison: University of Wisconsin Press.

Mains, Daniel. 2012a. Hope is Cut: Youth, Unemployment and the Future in Urban Ethiopia. Philadelphia: Temple University Press.

———. 2012b. ‘Blackouts and Progress: Privatization, Infrastructure, and a Developmentalist State in Jimma, Ethiopia’, Cultural Anthropology 27(1), 3-27. https://doi.org/10.1111/j.1548-1360.2012.01124.x.

Malara, Diego Maria and Tom Boylston. 2016. ‘Vertical Love: Forms of Submission and Top-Down Power in Orthodox Ethiopia’, Social Analysis 60(4), 40-57. https://doi.org/10.3167/sa.2016.600403.

Markakis, John and Nega Ayele. 1986. Class and Revolution in Ethiopia. Trenton: Red Sea Press.

Mitchell, Timothy. 1991. ‘The Limits of the State: Beyond Statist Approaches and Their Critics’, The American Political Science Review 85(1), 77-96. https://doi.org/10.2307/1962879.

———. 2002. Rule of Experts: Egypt, Techno-Politics, Modernity. Berkeley and Los Angeles: University of California Press.

Navaro-Yashin, Yael. 2002. Faces of the State: Secularism and Public Life in Turkey. Princeton and Oxford: Princeton University Press.

Pankhurst, Rita. 1997. ‘The Coffee Ceremony and the History of Coffee Consumption in Ethiopia’, in Katsuyoshi Fukui, Eisei Kurimoto, and Masayoshi Shigeta (eds.). Ethiopia in Broader Perspective. Papers of the 13th International Conference of Ethiopian Studies, Kyoto, 12-17 December 1997, pp. 516-539. Kyoto: Shokado Booksellers.

Poluha, Eva. 2004. The Power of Continuity: Ethiopia Through the Eyes of Its Children. Uppsala: Nordiska Afrikainstitutet.

Seeman, Don. 2015. ‘Coffee and the Moral Order: Ethiopian Jews and Pentecostals Against Culture’, American Ethnologist 42(4), 734-748. https://doi.org/10.1111/amet.12167.

Singh, Sarinda. 2012. ‘Bureaucratic Migrants and the Potential of Prosperity in Upland Laos’, in Liana Chua, Joanna Cook, Nicholas Long, and Lee Wilson (eds.). Southeast Asian Perspectives on Power, pp. 119-34. London: Routledge.

Tronvoll, Kjetil. 2010. ‘The Ethiopian 2010 Federal and Regional Elections: Re-establishing the One-Party State’, African Affairs 110(438), 121–136. https://doi.org/10.1093/afraf/adq076.

Trouillot, Michel-Rolph. 2003. Global Transformations: Anthropology and the Modern World. New York: Palgrave.

Vaughan, Sarah. 2011. ‘Revolutionary Democratic State-Building: Party, State and People in the EPRDF’s Ethiopia’, Journal of Eastern African Studies 5(4), 619-640.

Vaughan, Sarah and Kjetil Tronvoll. 2003. Ethiopia: Structures and Relations of Power. Stockholm: Sida.

Yedes, Janet, Robbin Clamons, and Amal Osman. 2004. ‘Buna: Oromo Women Gathering for Coffee’, Journal of Contemporary Ethnography 33(6), 675-703. https://doi.org/10.1177/0891241604269321.

Young, Allan. 1975. ‘Why Amhara Get Kureynya: Sickness and Possession in an Ethiopian Zar Cult’, American Ethnologist 2(3), 567-584. https://doi.org/10.1525/ae.1975.2.3.02a00130.


1 A kebele is the lowest formal level of administration, in this case an area with around 4,500 inhabitants.

2 Accounts of the nature, results of, and reasons behind possession by spirits are ambiguous and disputed (Young 1975; Boylston 2012); however, it is well known that zar spirits can be appeased by the smell and consumption of coffee (Seeman 2015).

3 The Afar are a pastoralist people from the neighbouring region, whose border is very close to the village, and with whom the local farmers are in a persistent state of violent conflict.

4 About ten pence, as of May 2017.

5 Female migrants are mainly employed as domestic servants in the Middle East, and some manage to return with savings, especially if they were working legally. Male migrants have no access to legal work, and the jobs they are able to get – shepherd, gardener, cafe worker – are badly paid, so they often come back without savings.

Anthropology Matters Journal ISSN: 17586453 Publisher: Anthropology Matters url: www.anthropologymatters.com