Confinement and development in Ethiopia The uses of prison in public policies
Translated and edited by Cadenza Academic Translations
Confinement and the politicisation of development
This chapter, which is based on research that has been underway in Ethiopia since
2010,1 examines how confinement is used in the development sector in order to politically control the population, through both policy and its implementation, what we call ‘development action’. By exploring the implementation of disciplinary and punitive state-led policies, we seek to understand what motivates its use and its social effects. We analyse the relationship, lying at the very heart of authoritarianism, between the use of force and the level of domination by observing various confinement practices - some legal, others merely punitive, part of the routine exercise of state domination - and the way that they are interrelated. To do so, we examine the problem of what is described in the Ethiopian countryside as ‘fertiliser debt’, where development policies compel farmers to purchase agricultural inputs, fertiliser mostly, on highly restrictive repayment terms (Planel 2014). We discuss how this debt becomes criminalised, its use in the daily exercise of state action, and its impact (in particular its political impact) on the peasantry' as a whole. Shifting our focus away from the prison itself, we respond to the recent call for scholars to make distinct the origins, practices, and formalisation of imprisonment in Africa (Morelle and Le Marcis 2015), and then emphasise these in authoritarian situations.
Rather than focusing on prisons themselves, which are difficult to gain access to in Ethiopia, this chapter examines their imprint on local societies. Studying rural societies enables us to pay ‘renewed attention to the social and spatial 
inscription of prison within its environment (particularly the urban environment), and the influences that they have on each other’ (Darley et al. 2013, p. 12). While we explore forms of state penal practices, we look beyond appeals to law alone, focusing on punitive practices as a whole, whether carried out by civil servants or by their peasant representatives. State punishment has many different forms and meanings. These help us to understand how popular illegalism - and its potential legal classification - results from modifications and constraints that are related to the very application of public policies. We can finally recognise, then, the truly political dimension of the prison system, and of the exercise of public action more broadly, particularly through the various ways in which bureaucracy is encountered (Lipsky 1980). We emphasise the punitive dimension of public action within an authoritarian situation and point out the degree to which this results from wider socio-political decisions taken in Ethiopia under the influence of partisan concerns and within the context of a developmental system.
We then examine the consequences in terms of social polarisation of incorporating a punitive system into development action (Fischer and Spire 2009) by observing the situation of those populations who are viewed (and who view themselves) as potential prisoners. This chapter does not focus on the minority segments of the population that are actually imprisoned but on how prison affects the government of peasant populations (see also Wacquant 2009). Rather than examining prison itself, this chapter looks at the carceral continuum, meaning ‘the influence of prison on social imaginaries’ (Morelle and Le Marcis 2015). We observe how state practices of punishment leading (or not) to confinement shape political societies and reproduce relations of domination both between the state and the peasantry' and within local societies. We thereby offer an analysis of the punitive and coercive forms of confinement in state-led action, examining their development, the diversity of ways that they are used, and their main effects outside prison walls.
An evolving system: Development, political control, and punishment
In Ethiopia, agricultural modernisation is a pillar of the development system which became widespread in the Marxist-Leninist period (1974-91) and has continued to evolve since then. Inspired by the model of the Green Revolution,
agricultural development policies link the problem of food security to that of agricultural yields and promotes new farming techniques - particularly the use of chemical fertilisers and improved seeds, distributed on credit - to increase yields. This system educates, monitors, and mobilises the peasantry', borrowing from a political practice that has been continuously reproduced in Ethiopia, one that shapes the contact between the state and the peasantry' into a highly asymmetrical power relationship based on the issue of development.
In the ‘developmental state’ model proposed by the country’s former strongman, Meles Zenawi, the state oversees economic development and the ‘democratic’ mobilisation of the masses, which it organises for ideological and political purposes (De Waal 2018). State-led development policy is used to supervise, control, and monitor the population (Villannuci and Fantini 2016), particularly the 85 per cent living in rural areas. Such surveillance ensures that Ethiopians remain faithful to the governing coalition (Lefort 2012) and limits political dissent by restricting it to fragmented ethno-federal projects, as demonstrated by recent mobilisations. Most prosaically, it draws the rural population into the party' apparatus, increasing the ruling party’s membership (Vaughan 2015).
Such population-control mechanisms are techniques of government (see Foucault 2004). They have largely been inherited from the socialist period, relying primarily on the spread of ‘development groups’ that, under party' control, operate in a central-democratic manner to mobilise both rural and urban populations by providing them with work. These techniques of control skilfully combine social regulation and political mobilisation. Similarly, the small-scale, bureaucratised state apparatus, which is constantly expanding in rural societies, strengthens the control imposed on populations. Finally, such control constructs and reproduces a domination whose forms continually adapt to the new development requirements that encourage the economy to open to the market. The strengthening of power asymmetries among the peasantry' is thus accompanied by an increase in inequality', due in particular to a coercive practice of financial inclusion (Lc Saout and Segas 2011, Planel 2016).
The party-state’s development policy defines new ways of behaving, disseminates them, and punishes deviant behaviours deemed to present a challenge. Prison is thus used in two ways: to punish and to produce consent. While day- to-day public action is becoming increasingly judicialised, its coercive practices (threats, blackmail, the use of force) are being supplemented by non-carceral and therefore non-penal forms of confinement. All these practices are legitimised by an ethic of subsistence, which guarantees food security for small peasantries and strengthens the authority of the state. This long-standing social contract between the state and the peasantry has recently undergone a major shift, due to the partial liberalisation of development policy, which not so long ago was entirely under public control. The use of imprisonment has, consequently, been transformed considerably.
Ethiopia’s agricultural development policies can be divided into two stages: before and after the mid-2000s, with the turning point being the 2005 general election. The first stage - the distribution of fertiliser and improved seeds on credit - was characterised by the coercive implementation of development measures. Farmers were forced to take on loans that both the creditors and debtors knew could not be repaid. There were no discussions over the highly rigid repayment process. And, when they were unable to pay, farmers were jailed for insolvency, potentially serving sentences of several years’ imprisonment. This stage was heavily politicised and its extreme bureaucratisation, involving a quota approach, continually reinforced the trend of peasant engagement with the state becoming increasingly formalised, contractualised, and judicialised.
The second stage involved a pseudo-withdrawal by the party-state: regional microfinance organisations took over the management of fertiliser debt. The repayment terms were relaxed, with the amount repaid varying depending on the liquidity of the debtor. These organisations arranged gradual seizures of assets with the cooperation of the local authorities who had previously been in charge of overseeing repayment. Prison sentences became shorter and rarer, and conflated different types of debt. Fertiliser debt became part of a broader, more composite peasant debt (heavily characterised by the promotion of microfinance) that had, to a larger degree, been taken on willingly, at least in the eyes of the public.
While prison sentences and punishments more generally have always served to educate rural populations, the meaning of this ‘education’ has evolved. Initially, the purpose was primarily punitive and sentences were relatively long. Today, sentences have become shorter and imprisonment is now more pragmatic, with the aim to improve and accelerate the repayment process. The sentences usually only last a few months (usually six months, more for repeat offences) and they are stopped once the debt has been paid in full, something that happens frequently. Imprisonment for debt actually comes at the end of a lengthier but failed repayment process that has been organised, at different times, by rural cooperatives, officials from local branches of the Ministry of Agriculture, and employees of microfinance organisations.
Prison sentences for misuse of fertiliser then seem to have declined. Estimating this is a difficult task, however, as there now exist new fraudulent practices (contraband and corruption), which are punishable by a sentence of similar length and a fine. While it is increasingly rare to see debtors in prison, there is still a large number offertiliser prisoners. In reality, the two categories often refer to the same peasants, as some farmers sell small quantities of fertiliser in order to quickly obtain the cash needed to repay their ‘fertiliser debt’.
The confinement process up close
Our reflection on the political uses of confinement - penal, punitive, and coercive - leads us to a qualitative interpretation of the practices, spaces, and uses of it. This juxtaposes prison sentences for debt, contraband, and corruption;   administrative practices of confinement (which occur long before the offence is committed); and seizures of assets.
The schedule for distributing fertiliser is spread across the whole year, but there are three moments in particular that bring out tensions, identifying those individuals who cannot or do not wish to do what the government wants. On closer inspection, we find that each of these moments relates differently to confinement.
The most politicised of the three moments is when those receiving fertiliser are registered: the Ministry of Agriculture’s extension workers set out to fill their quotas, compiling a list of those peasants who are happy to buy fertiliser and those who are reluctant, and estimating the required volume. This stage is characterised by the bureaucratisation of government action and the substantial mediation carried out by the party’s regional branches. During this stage, confinement is not used for penal or legal purposes - or it is at least uncommon (I encountered it only in Tigray).11 In Ethiopia, every municipality has a strong administrative centre, characterised by the presence of numerous public buildings that house the offices of the kebele,u the police, and the Ministry of Agriculture. In July 2012, peasants who continued to refuse to sign on for the next fertiliser purchase were taken to and detained in these offices at night. If their arrest was not enough to intimidate them and make them sign on to the ‘fertiliser lists’, they spent the night locked up and generally signed on the next day, having no idea how long they might otherwise be held. They were confined for a number of reasons. First, the illegalism here being punished is perceived as an attack on the authoritarian regime: it represents a public challenge to the state’s ability to govern its subjects and to correctly identify the beneficiaries of a public policy. Second, such confinement is enforced by a state administration, making it legitimate enough to dispense with using the law. One Sunday afternoon in a small cafe in Adigrat, two development agents from the Ministry of Agriculture talked about the good old days and joked about how docile the farmers used to be: they recalled how they drew up false reports and forced peasants to do as they said by threatening and imprisoning them. As soon as I introduced myself and told them about the focus of my work, they reproached me for not having introduced myself sooner, insisted that these practices were now in the past, and immediately emphasised their status as state representatives.
The second moment, which begins with the harvest, is that of the criminali- sation of debt. This stage has been highly influenced by the judicial sphere and the ongoing legal reforms: it is both the most formalised and the one that has changed the most. Today, it is marked by a fall in the number of imprisonments. In recent years, debt has been penalised less harshly and judges are instead encouraged to remind offenders of the law. As a judge from Gununo told me: ‘The farmer is asked to pay and, if he can’t, he is given an extension to pay later and a meeting date is set. If the repayment still hasn’t been made by that date, we make a second judgement, authorising Omo Microfinance to seize the assets: money, crops, livestock’. In answer to my question, he added: ‘We don’t send them to prison. If they hide money, though, they go to prison until they pay ... [they are sent] to the little prison at the police station, and only for twenty-four hours’. These remarks - spoken by a young, newly appointed judge who is largely sympathetic towards the citizens - reveal the asymmetry' of the state-peasantry' relationship. They highlight both the distrust on the part of local authorities, who often interpret debt as a refusal to pay, and the reality of confinement practices, where prisoners are held in cells attached to police stations for periods that, despite what the judge said, can last a number of weeks, going beyond the legal limits of police custody. Finally, the Gununo judge gave his own opinion on jailed farmers: ‘Before going to prison, the people from the kebele say that you have to pay, that y'ou have to pay quickly. Those who get locked up [in the Sodo zonal prison] are those who resist’.
In 2015, there was nobody imprisoned for debt in the Sodo zonal prison (Wolaita), but around 20 prisoners were in there for corruption or contraband connected with the fertiliser trade. According to interviews with a small number of these prisoners, in which they assessed the distribution of sentences and the scope of imprisonment among the peasantry, many were indeed ‘fertiliser prisoners’. The fertiliser prisoner is a prominent social figure in the peasant imagi?nary, central to their experience of state domination and marginalisation. It also perpetuates a fear of the state and provides one reason among others for rural workers to obey the state.
The third moment can take place at any time in the agricultural calendar, but it typically occurs in the period before the crops are sown. This moment is the black-market resale of fertiliser, on which the state has a monopoly. This illegal sale of fertiliser has expanded in recent years, and it results from the fact that its distribution is compulsory (Planel 2014). Although it is common knowledge that peasants resell some of their fertiliser, such transactions are carried out with the utmost discretion. As an old peasant in Tigray explained: ‘It is done secretly, using spoken agreement, not on the market and not with traders - otherwise you get caught and sent to prison’. In Wolaita, the fertiliser is transported at night and left with a friend or relative living close to one of the many rural or urban markets, but the sale itself may take place during the day, in the back rooms of local traders. Meanwhile, peasants hoping to appeal to the judges’ sympathy permit some, mostly elderly women, to sell the fertiliser in broad daylight. However, in Tigray, peasants are very aware of the six-month sentence they would receive: T know the price is lower [on the black market], but I’m not going to use it, because I know that the government’s plan is for evert' farmer to use the fertiliser. So I have to use it, I have to obey the government’s orders. ... They warn us: if you sell it on the market, you go to prison’.
All of the past and present uses of confinement and legal classification contribute to a fluctuating, rather closed interpretive horizon. The local community judges of the ‘social court’ (mehabemwi ferde bet) are clear on this point:
We have official instructions for legal procedures regarding fines. If it was an accident, the fine is 30 ETB [ 1 euro] if it was the first time, the fine is between 30 and 100 ETB, and 100 ETB if it was done deliberately. But we, the people from our own community, we have our own rules: 50 ETB if the act takes place during the day and 100 ETB if it takes place at night. ... A farmer who knows the law might complain.
From a peasant’s point of view, these practices are understood according to a fairly widely shared interpretation that the state will always view its citizens as uncooperative and punish them. Punishments give material form to the subjective - and subjectivising - experience that peasants have of public authorities. The aim of these punishments is to make an example. Imprisonment and confinement thus publicly signal a breach of the obedience that all Ethiopians, and especially peasants, owe the state - part of a social contract that generally relates to an ethic of subsistence, and that, while tacit, is regularly reaffirmed by the party-state.
Prison beyond the walls
The effects of prison are not limited to what takes place within its walls: the greatest effects extend far beyond them, for the peasantry' at least. The threat of imprisonment is a powerful tool for Ethiopian development policies. Ethiopian peasants well understand the highly political dimension of fertiliser distribution. Most of them but mainly the poorest are afraid of the consequences of going into debt and try' to avoid the demand to use agricultural inputs. Nonetheless, official figures indicate that almost every farmer does indeed receive fertiliser. Our own research shows clearly that the threat of imprisonment or the confiscating of the means of production is enough to quell any resistance by farmers.
This carceral continuum is thus explained by the extension of the state apparatus and its representatives into focal society. It is peasants themselves who identify which of their fellow peasants are unwilling to cooperate or are at fault, who arrest them, and, sometimes, who pass judgement on them (in the case of the social courts). The disciplinary' nature of confinement and its effects on the control of the population are all the more operative in this case because they are established by individuals who might be judges, guards, or militia members one day, but who might possibly disobey the next. The peasant militia members are temporarily integrated into the state apparatus and its punitive practices, and are responsible for making arrests, seizing assets, and potentially guarding livestock for weeks at a time. Wolaita residents complain in particular about numerous abuses by militia members: charging high fees for guarding livestock, allowing private use of livestock, mistreating and even stealing livestock, making insults, and engaging in corruption. Probably for similar reasons, such practices no longer exist in Tigray. The social court secretary explained:
Most of the time, we send the peasants to prison rather than [confiscating and] selling their livestock at auctions. We do this for a number of reasons. Firstly, as long as the money hasn’t been given back [i.e., the livestock is still being guarded bv a militia member], the cattle cannot be used by other farmers. And the auctions take time, and that can impoverish the farmer [who pay's daily fees for the guards]. That might lead him to sell his goods at a bad price. When lie’s in prison, the farmer is happy to pay, and most of the time he pays off his debt within the first six months of his sentence.
While the proximity of guards and potential prisoners facilitates the ‘authoritarianism of the dominated’ (Geisser 2008) and the associated capture of resources, this everyday micro-discipline also allows for forms of resistance. For some, prison can appear preferable to having their assets seized, and imprisonment may seem the best conditions under which to make the necessary' repayments.
In Ethiopia, the effects of confinement on social differentiation (Wacquant 2009), especially in the poorest segments of the population, are very' mixed. The coercive implementation of development policies causes the impoverishment of the most vulnerable and forces them to sell off their means of production at low cost. Today, however, imprisonment for debt is apparently reserved for the richest debtors. A representative of the Tigray regional parliament living in the country'side explained the situation of the poorest:
People who have a credit with Dedebit [a microfinance organisation in Tigray] of about 1500 ETB over three y'ears and who end up using it all up, they have to pay. In those cases, they have to go to prison, because they used up all the money. Farmers with livestock go to prison instead of paying. They reject the law, so they go to prison. But if they’re very' poor they never go to prison. Because they don’t have any assets.
All farmers feel themselves to be affected by the potential for imprisonment. Recently, however, it is the poorest who have been illegally reselling their fertiliser and receiving prison sentences. In addition to this increase in domination over the poorest, we must also consider gender-related imbalances. While women are generally absent from this highly patriarchal development system, their vulnerability leads to greater fear of confinement. This becomes very clear when they become responsible for a farm after they are widowed or divorced, and they attempt to resell fertiliser on the black market. Sale prices are directly proportional to the seller’s social status. In summer 2017, a quintal of fertiliser sold officially by the cooperatives in the woreda of Ganta Afeshum was worth around 1,300 ETB. According to our research, resale prices on the black market were between 600 and 900 ETB per quintal. When the sellers were single women - often highly dependent on their neighbours for growing crops - prices could fall as low as 400 ETB.
As we have examined, in the implementation of Ethiopian development policies, prison serves as a powerful tool for governing the behaviour of the population, not because of the discipline imposed on bodies during confinement, but as an instrument of mass mobilisation that pushes the peasantry to accept the demands of the government, whether these demands are to use fertiliser or to repay debts.
The effectiveness of governing behaviour in this way owes much to the actors who introduce it into rural societies: state representatives, minor local elites, and peasants who have momentarily been picked out by the party framework. In Ethiopia, the criminalisation of what is considered as a wrong participation in public policy is spreading to an increasing number of actors and domains, even as the legal field seems to gradually be disengaging itself from public action, at least in the development sector.
This interweaving of the disciplinary and developmental registers of public action produces significant effects of social stratification. When this is combined with forced integration into the market economy, it leaves the poorest segments of the population more vulnerable than before. Ideological condemnations (neoliberal and partisan) of poverty are thus reaffirmed, even as poverty becomes more strictly criminalised, reinforcing relations of domination.
Bernard, D., 2018. Introduction: Pensee penale, pensee politique. Revue Interdisciplinaire D’Etudcs Jundiques, 80( 1), 43-48.
Darley, M., Lancelevee, C. and Michalon, B., 2013. Ou sont les murs? Penser l’enfermement en sciences sociales. Cultures et Conflits, 90(2), 7-20.
De Waal, A., 2018. The Future of Ethiopia: Developmental State or Political Marketplace? Somerville, MA: World Peace Foundation, https://sites.tufts.edu/r einventingpeace/files/2018/08/The-future-of-ethiopia-20180817.pdf.
Fischer, N. and Spire, A., 2009. L’etat face aux illegalismes. Politix, 87(3), 7-20.
Foucault, M., 2004. Securite, Territoire, Population. Corns Au College de France. 1977-1778. Paris: Gallimard Seuil.
Geisser, V., 2008. L’autoritarisme des “domines”: Un mode paradoxal de l’autoritarisme politique? In: O. Dabene, ed. Autoritarismes democratiques: Democratiesautoritairesau XXIe siecle. Paris: La Decouverte, 181-212.
Lefort, R., 2012. Free market economy. Developmental state’ and party-state hegemony in Ethiopia: The case of the Model Farmers’. Journal of Modern African Studies, 50(4), 681-706.
Le Saout, R. and Segas, S., 2011. La domination politique par les dispositifs financiers: L’exemple de la dotation de solidarite communautaire (DSC). Politix, 93(1), 141-165.
Lipsky, M., 1980. Street-Level Bureaucracy: Dilemmas of the Individual in Public Services. New York: Russell Sage Foundation.
Morelle, M. and Le Marcis, F., 2015. Pour une pensee pluridisciplinaire de la prison en Afrique. Afrique Contemporaine, 253(1), 117-129.
Planel, S., 2014. A view of a bureaucratic developmental state: Local governance and agricultural extension in rural Ethiopia. Journal of Eastern African Studies, 8(3), 420-137.
Planel, S., 2016. Le developmental state ethiopien et les paysans pauvres: Economic politique du developpement rural par le bas. Politique africaine, 142(2), 57-76.
Vaughan, S., 2015. Federalism, revolutionary democracy and the developmental state 1991-2012. In: G. Prunier and E. Ficquet, eds. Understanding Contemporary Ethiopia: Monarchy, Revolution and the Legacy of Meles Zenawi. London: Hurst, 283-312.
Villannuci, A., Fantini, E., Perier, M., 2016. Sante publique, participation communautaire et mobilisation politique en Ethiopie: La women’s development army. Politique africaine, 142(2), 77-99.
Wacquant, L., 2009. Punishing the Poor: The Neoliberal Government of Social Insecurity. Durham, NC: Duke University Press.
-  This chapter is based on a series of studies carried out in the Ethiopian countryside, beginning in 2010. More specifically, it presents the results of two research projects carried out aspart of the Economic de la Peine et de la Prison en Afrique (ECOPPAF) program: one inthe southern region ofWolaita, in February 2015, and one in the northern region of EasternTigray, in November 2017.
-  Translator’s note: Unless otherwise stated, all translations of cited foreign-language materialin this article are our own.
-  Political both in terms of policy making and raising political awareness and civil education.See Bernard (2018).
-  This model was largely developed in India and Southeast Asia in the 1970s, and then exportedto other developing nations, primarily by the Food and Agriculture Organization of theUnited Nations. It offered a way out of food insecurity by increasing agricultural productivity (through irrigation and the use of inputs). Much like the Industrial Revolution, this wasthought to encourage development.
-  Hierarchically structured development groups (in Amharic, limat budin) organise cells forinspections and mass mobilisation by the party. Each leader in the cell reports the actions oftheir group to the higher-level group and communicates to their cell the party’s orders. Since2010-11, the smallest of these cells has contained five people.
-  I.e., in places other than in prison buildings.
-  Officially known in Amharic as ‘afetsat’sem'.
-  According to Omo Microfinance, Head Office (25 February 2015, Damot Sore, Gununo), in2015 it lasted about two years.
-  By contrast, accusations of‘corruption’ lead to far longer sentences of around five years.
-  These political processes make it possible to settle individual or political accounts. It is common for this to occur within the fertiliser distribution chain, leading to prison terms ofseveral years for scapegoated or wrongly accused farmers. Firings may be ordered by ananti-corruption commission that monitors rent-seeking behaviours that are outlawed by thegovernment.
-  The dominant ethnic group in the ruling coalition comes from this northern region.
-  The equivalent of a municipality, this level includes a devolved administration - particularlyin agriculture - and decentralised government bodies.
-  Conversation on 28 October 2017 with two former development agents from Wukro, partially translated into English by Aleksander Asmelesh.
-  The small capital city' of Eastern Tigray.
-  An interview with a judge, court of Damot Sore, Gununo, 26 February 2015, originallycarried out in English.
-  The zone is an administrative level between regional states (kilil) and districts (woreda).
-  Interviews with the head of the prison in Sodo, 25 and 29 February 2015, carried out inEnglish and translated into Wolaita by Guebre-Michael Kuke.
-  This does not mean that resistance is completely absent, but there are no uprisings and mobilisations are partial, localised, and often abortive. This lack of mass protest is produced to alarge extent by the authorities currently in power and reproduced by the balance of powerwithin the social body.
-  The sowing period is crucial in agricultural economies in crisis because it is at such momentsthat peasants take risks.
-  Menguistu, woredn of Ganta Afeshum, Eastern Tigray, 23 October 2017.
-  Desta, Kushet likamember, Ganta Afeshum, Eastern Tigray, 31 October 2017.
-  A small family court, with four or five judges from the local population.
-  Eastern Tigray, woreda of Ganta Afeshum, 28 October 2017, secretary of the social court.
-  Many peasants stated that the seizure of assets is worse than a prison sentence.
-  Representative of the Tigray House of Parliament, Eastern Tigray', 26 October 2017.
-  One quintal is approximately equivalent to 100 kilograms.
-  The resold quantities are, however, far lower, from a few kilograms to a half-quintal.