‘Constructive investors’ versus ‘rent seekers’

After gaining power in 1991, the Ethiopian Peoples' Revolutionary Democratic Front (EPRDF)Footnote ¹ based its development policies on the 13 million smallholders involved in subsistence farming and, in particular, the ‘poor farmers’ who made up the vast majority.Footnote ² This had a threefold objective: increasing their productivity so as to lift them out of poverty; achieving national food security; and establishing a virtuous circle, Agricultural Development Led Industrialisation (ADLI). The increased wealth of farmers would lead to their demand for basic commodities that in turn would lead to industrialisation, which, among other things, would bring about an improvement in inputs, which again would end by increasing agricultural productivity. But, each year on average, emergency or permanent aid was (and still is) vital for about a sixth of the population (OCHA 2011). Industry remained (and remains) constant at about 13% of GDP (MoFED 2010). The culprit was this ‘pro-poor strategy’, which prevented ‘the farmer from escaping a hand-to-mouth and scattered way of life’, because the Front's goal was ‘to detach the farmers from commercialisation’ instead of ‘focusing on market economy as the centre of rural activities’ (EPRDF 2007). So the agricultural strategy was to undergo its biggest shift since 1991 and even 1974.

This became part of a global vision that emerged after the expulsion of the TPLF ‘left wing’ in 2001 and then consolidated after the unexpected outcome of the 2005 elections.Footnote ³ In ‘the time of globalisation’, Ethiopia had no other choice ‘to survive’ than to carry out a ‘structural change’ in order ‘to employ free market economy’ (EPRDF 2006a). This change would oppose the ‘constructive investors’ or ‘new entrepreneurs’ to the ‘rent seekers’ obstructing the emergence of the market and the competition that comes with it, in order to continue to benefit from the perks that went with their official position or their ‘corrupt relationship with officials’ (EPRDF 2007). At the same time, this antagonism was superimposed onto the opposition between democratic and anti-democratic forces: the rent seekers can only operate where there is an absence of transparency and accountability, and, on a wider scale, democracy. The creation of a market-based economy pushed by the developmental state is presented as inseparable from the move towards democracy.

‘Model farmer’ versus ‘acha farmer’

While agriculture may remain ‘the engine of growth’ (MoFED 2010), ‘the key actor in the sector's development will not be the semi-subsistence small farmers’. The priority is now ‘to capture the private initiative of farmers’ so as ‘to intensify marketable farm products’ (MoFED 2006) – hence the official division of the farming population into two unequally sized groups. On the one hand is the meagre elite of farmers, who have to be recruited as ‘model farmers’,Footnote ⁴ and as vanguard party members or ginbar kedem.Footnote ⁵ The support of the public authorities focuses on them. On the other hand are the ‘acha farmers’ – acha means something like ‘peer’ in Amharic. While, in theory, ‘improvement of pro-poor subsistence farming still needs to take place’ (ibid.), this mass of ‘poor farmers’ is in reality left to its fate, or more accurately to market forces. ‘Those who take advantage will prosper, and the rest will lose mercilessly’ (EPRDF 2006a).

Neo-liberalism versus ‘developmental state’ and ‘revolutionary democracy’

But this recognition of the primacy of the market and democracy does not go so far as to embrace the dominant neo-liberal ideology. Ethiopia is considered to be at the stage of ‘apprentice free market’, struck by numerous ‘failures’ (quotations from Meles 2006, EPRDF 2006a). Rendering the state ‘non-activist and non-interventionist’ and reducing it to a ‘night watchman’ role, thus ‘unleashing the market’ while expecting it to correct its own ‘failures’, would lead to a ‘dead-end’. The authorities have to assume an extended and essential function to bring this free market to ‘maturity’, i.e. ‘to replace the rent-collection political economy with a constructive developmental political economy’. The public authorities must therefore keep control of ‘the so-called commanding heights of the economy’. In fact the party-state still directly or indirectly controls the majority of the formal non-agricultural economy, through either public enterprises or ‘party-statal’ companies (see IMF 2011). This dominant position should also extend further with the process of ‘land leasing’, where the state, which owns the land,Footnote ⁶ plans to rent 7 million hectares of land to investors, mainly foreign, by 2015 (see Desalegn 2011). In this way the developmental state or ‘constructive government’ has the power needed ‘to engage to the extent possible, in areas where the private sector is unable to engage’, and ‘to reward or punish it’ by supporting ‘private constructive investors’, or, on the contrary, ‘to direct this sector away from rent seeking activities’, towards which it always tends to lean, in this initial phase of the market.

The supposed dead-end produced by the neo-liberal paradigm also led to the establishment of ‘revolutionary democracy’. In an emerging market, the rent seekers always have the upper hand, especially among a ‘backward’, ‘scattered’, ‘unorganised’ peasantry (quotations from EPRDF 2007). ‘This objective reality’ leads it ‘to fall into the nets of rent seekers’ and leaves it ‘with little capacity to fight anti-democracy’. As a result, the peasantry needs an avant-garde to exercise ‘strong revolutionary democratic leadership’, a function devolved to the EPRDF. This paradigm also has a major defect whereby ‘the political stage is occupied by a few politicians … who substitute for the public at large’. On the contrary, with ‘revolutionary democracy … the social forces of the peasants, labourers and majority city dwellers [are] consciously and un-interruptedly engaged’ in the decision-making process. Revolutionary democracy and developmental state, which fuse together in the ‘developmentalist democracy’, are thus the basis of ‘leadership by all strata of government’ over the peasantry by ‘mobilising’, ‘organising’ and ‘coordinating’ it, starting with its best ‘agents of development’. A key term, ‘to participate’ is supposed to join two antagonistic imperatives together in practice: the EPRDF leadership and democracy.

‘Model farmers’ and/or ginbar kedem

Until its major disillusionment in the 2005 elections, the regime was confident that the massive majority of poor farmers would continue to be its unshakeable constituency, carrying on from its fight against the Derg Footnote ⁷ and in recognition of everything it felt that it had done for them since 1991 (Berhe Reference Berhe2009; Young Reference Young1997). But, in the more liberal electoral climate of 2005, the rural elite, largely opposed to the regime (see below), could assume its role of opinion maker and therefore ‘election maker’. So, in order to re-establish its hegemony (Aalen & Tronvoll Reference Aalen and Tronvoll2009; Clapham Reference Clapham2009), the ruling party undertook a massive recruitment drive among this elite (Lefort Reference Lefort2007, Reference Lefort2010), using force if necessary, to turn them into ginbar kedem or ‘vanguards’. There was a dual objective: to deprive the opposition of its rural base, and gradually to win over this elite and turn it into its new constituency. As a result, the ‘model farmers’ automatically became members of the party, and vice versa, peasant party members often, if not systematically, became model farmers. In exchange for their support, or at least their political neutrality, the regime promised to open the way to their economic prosperity, particularly by reserving for them the almost exclusive support of the authorities, and by giving them elbow room to devote themselves entirely to their farms.

Main local features

The community surveyed belongs to smallholder agricultural Amhara society (see Lefort Reference Lefort2005, Reference Lefort2007, Reference Lefort2010),Footnote ⁸ with a standard of living clearly higher than the average and a greater contact with outside world, given the relative ease of communication in the area. It occupies one of the three sub-kebele of a kebele in the North Shoa Zone of Amhara State, an hour or two's walk from the centre of the kebele, a market town on the main road from Addis Ababa to the north-east. The kebele has about 6,000 inhabitants, mostly living in hamlets (gott). There are 1,133 land taxpayers,Footnote ⁹ over 600 landless and slightly fewer than 100 officially registered ‘young jobless’. The town's inhabitants include shopkeepers and civil servants; it has a primary school, a telephone office, a health centre, a police station and a Farming Training Centre.

The average area farmed by the land taxpayer is officially about 1.34 hectares.Footnote ¹⁰ The main traditional crops are teff,Footnote ¹¹ sorghum, barley, wheat and lentils. Farmers also own the usual small and large livestock. No NGO is active in this community, which is not eligible for any food-for-work or cash-for-work programme. A dual agricultural revolution is underway, the biggest since the egalitarian agrarian reforms of 1975,Footnote ¹² and the introduction of fertilisers a few years later. There has been a striking expansion of growing cash crops, starting with onions, followed by other vegetables, fruits and sugar cane, and recently khat Footnote ¹³ and mung bean. For the first time, some ‘average’ farmers are saying that the small scale of their farms is no longer an obstacle to a decent standard of living, at least as long as the prices of these cash crops remain as high as now.Footnote ¹⁴

Inequalities are also growing. Mainly due to annual population growth of 2.6% (Census 2007), shortage of land and an increasing vulnerability to the debt trap among the poorest, the number of landless people is rising, notably among young males. Meanwhile, rich farmers have grown richer, notably through rental or sharecropping agreements, to which the weakest are frequently forced to sign up. In the space of ten years, certain ‘strong farmers’ have multiplied the area they cultivate four or five times.Footnote ¹⁵ These agreements very often do not respect the regulations, but the authorities turn a blind eye. ‘Experiences of developed economies clearly show that as an economy grows there is a tendency for some small farmers to quit the sector and seek employment in other sectors, and there are others who accumulate enough capital to go big in the sector’ (Meles 2006).

One of these ‘other sectors’ is a growing migration to Arab countries, which is mostly legal for women – who often find work as maids – and illegal for men. The money sent home by unmarried girls or young wives, probably numbering between fifty and a hundred in the kebele concerned, is out of all proportion to the amounts brought in by domestic migrants, and can be equivalent to the income of a very rich farmer.Footnote ¹⁶ Recently, a few dozen males have also emigrated, overwhelmingly illegally, and a growing number plan to do the same, mainly from the young landless population.

The Agricultural Extension Programme

Ethiopia is probably the African country with the greatest state involvement in the agricultural sector, through the developmental state; it also has the highest level of state investment.Footnote ¹⁷ While a significant part of this commitment can be seen both upstream and downstream of agricultural production itself (such as roads, markets and research centres, but also health and education), the developmental state also affects the farmer's activities directly at the field level in four main ways. Cooperatives are responsible for supplying agricultural inputs (fertilisers, selected seeds, loans). A network of banks, mainly state-owned, provides rural loans. Participation in collective work, known as ‘communal work’, ‘development work’ or ‘social work,’Footnote ¹⁸ is compulsory, as are the hefty financial ‘voluntary contributions’ to local development programmes. These seek to build infrastructure (roads, schools, health centres, etc.) and improve the agricultural environment (terracing, water holes, reforestation, etc.) The final mainstay, largely interlocked with the former, is the Agricultural Extension Programme, probably the largest in sub-Saharan Africa (Davis et al. Reference Davis2010: 14). ‘The public sector is the single most important player in Ethiopia's agricultural innovation system, especially in terms of inputs, at the local level for smallholders’ (ibid.:10). The programme reasserts the constant aim of the regime: ‘to increase “modernisation” and revitalisation of agriculture through improved and new crops, livestock, and natural resource management technologies … and the increase in input use by farmers’ (ibid.:1).

One of its main components is the Development Agent (DA). The goal is to assign three DAs in each rural kebele to look after, respectively, crop production, livestock and natural resource management. The target is around 60,000 DAs, or one DA for near to 476 farmers, one of the highest ratios in the world (ibid.: 15). Each rural kebele is also expected to have a Farming Training Centre, with two hectares of demonstration fields. In fact, only about half the kebeles have one (ibid.). In the kebele concerned, the building exists but there are neither the demonstration fields nor the funding for it to operate. There are three development agents and one veterinary agent.

The local political context

This community consists almost exclusively of Orthodox Christian Amharas (see Lefort Reference Lefort2005, Reference Lefort2007, Reference Lefort2010). The great majority continue to perceive society as organised according to a strict hierarchy. God, at the top, is the sole master of the destiny of each and every one, from whom, ultimately, all power emanates. To submit to divine will leads to an acute sense of obedience to the authorities. ‘We are people to be ordered’, this majority often says. A small but growing proportion of the population – about 10% to 15% today – has a more secular outlook. This is found mainly among the rural elite, whose education, wealth and social influence put them at the top of the socio-economic hierarchy. The vast majority of this elite supports the opposition, for four main reasons.

First, until the 2005 elections, the authorities imposed a systematic and constant intrusion into the daily life of the peasantry, notably to ensure their ‘development’. The most obvious manifestations of this top-down approach were obligatory participation in collective work, payment of ‘voluntary contributions’, and compulsory meetings that were as frequent as they were interminable. It was almost impossible to avoid these without sanctions, including the risk of forfeiting access to state-monopolised services (e.g. modern agricultural inputs, credit, schools, health), and even, ultimately, land itself. While most peasants submitted to these impositions with a certain degree of fatalism, the rural elite abhorred them, considering them illegitimate and useless, and, above all, a major obstacle to devoting themselves entirely to their own farms, thus hampering their growth.

Second, with the general support of the population, this elite had supplied most of the local administrators for the form of local self-government which had been launched by the Derg after the agrarian reforms of 1975, earning them the nickname birokrasi (from ‘bureaucracy’). After 1991 the birokrasi was completely ostracised. The incoming regime imposed a new category of local administrators, generally poor young peasants who were judged to be naturally close to the new regime. The birokrasi was also constantly vilified and above all arbitrarily and heavily penalised during the land redistribution of 1997 (Ege Reference Ege1997). This elite increasingly considered the authorities in power as mere temporal rulers, while they aspired more and more to citizenship and respect for basic human rights.

Third, the Amhara had played a central and dominant role in the gradual building of the Ethiopian polity, but lost this role completely after the Tigrayan-led takeover in 1991. They criticise an ethnic bias of the regime in favour of the Tigrayans.Footnote ¹⁹ Having rapidly broadened its economic base after 1997, and having kept its position in society, this elite acted as opinion leaders during the 2005 elections, and were followed by about two thirds of the electorate from this community. A final factor of discontent is the return of rocketing inflation. According to official figures, it reached 40% in September 2011 – and 52% for food.Footnote ²⁰ Between half and two thirds of the peasants – and particularly the landless – are usually net purchasers of food. As they often say, ‘inflation is worse than prison’.

The surveys

This article is the last in a series of studies carried out since 2002 in the sub-kebele concerned, focusing on the relationship between the peasants and those in power (Lefort Reference Lefort2005, Reference Lefort2007, Reference Lefort2010). It is based on a survey carried out in December 2010. A sample of twenty-five respondents was randomly chosen among the registered model farmers. Interviews were also carried out with the Chairman of the kebele and the kebele manager,Footnote ²¹ the DAs, shopkeepers and peasants whom the author had met on previous stays and were not part of the sample. A second visit in May 2011 complemented and verified the previous work.

Formal recruitment criteria

There is no publicly available document that sets out the official, national line regarding model farmers.Footnote ²² Nevertheless, the author obtained a very rough document that is still in force at least for the woreda concerned, as well as a more complete document for Amhara State (CBO 2007). In addition, a document in force in Amhara state clarifies what is expected of model farmers, setting out criteria for dividing them into three hierarchical categories and rewarding the most worthy (Agricultural Bureau n.d.). The following quotations are taken from these documents.

Those eligible must respect two principal lines of conduct. The first is socio-economic: ‘gaining wealth’ by ‘developing well his land’ by ‘working hard’ and by ‘learning new technologies’ and ‘putting them into practice’. This involves being ‘in the forefront in the adaptation of farming innovations’ with the aim of increasing production of ‘market oriented crops’, to generate cash that can be saved in ‘rural banks’. Model farmers must set an example socially, in the widest sense, in terms of moral rectitude (starting with sobriety), sending all their children to school, keeping an orderly house, and following rules of hygiene. Just as important, they must display leadership in community life, whether by taking part in traditional self-help organisations, or, for older members, by helping to resolve everyday disputes.Footnote ²³ In other words, priority should be given to those who are both ‘strong farmers’ and ‘opinion leaders’. But the criterion of wealth creation is primordial. As the Chairman of the kebele summarised it, ‘a model farmer is simply a rich farmer’. Of the model farmers interviewed, several denied being such because they said they did not earn enough. This would primarily be an obstacle in itself, but would also prevent them being able to use the requested new technologies. Some mentioned they had not made the required sanitary improvements, or that model farmers are only those who have been rewarded.Footnote ²⁴

The second criterion is exclusively political. The model farmer must above all be a dedicated militant of the ruling party. He has to carry out his duties ‘as a vanguard’, keen ‘to improve his political efficiency’, active and at ease in the ‘political institutions’, committed ‘to change the farmers around him’, notably because he is willing ‘to administer [them] politically’, by organising their participation in ‘revolutionary activities of communal works’.

Effective recruitment

The recruitment drive for model farmers was essentially carried out in the autumn of 2005, at the launching of the regime's nationwide recovery strategy. The local authorities summoned slightly fewer than 300 peasants, including of course those who were already party members (Lefort Reference Lefort2010). The number of registered model farmers dropped to 272 in May 2011, officially due to natural departures. These were partly compensated for by new recruits, this time selected by the party cells (see below), who put their candidates forward to the DAs, who in turn submitted them to the Cabinet, the executive body of the kebele. As in 2005, this process was in the hands of the party-state alone.

As the title suggests, a model farmer should be a farmer. But four of the twenty-five interviewees did not themselves cultivate the land to which they had right of use.Footnote ²⁵ Two are guards, one is a butcher, the fourth has serious mental health problems. A fifth, mentally retarded, has neither land, contract nor a sharecropping agreement. All the remaining information derived from this survey is therefore based on a reduced sample of twenty model farmers. There are only two women in the sample (10%), whereas they represent 23% of the 1,133 land taxpayers. The average age is forty-three years. Only half have been educated beyond the second year of primary school, although this group is supposed to be particularly innovative, which usually goes with a relatively strong education. But comparisons are not possible because local age and education statistics are not available.

According to the official criteria, there are three levels of land taxpayer: rich, middle and poor. One might expect that the great majority of model farmers would be ‘rich’. Yet in fact the latter make up exactly half of the sample, which even includes three ‘poor’. The average level is just above the ‘middle’. Two other criteria can be used in determining wealth level, but with caution given the inaccuracy of the data: the area under cultivation and the livestock inventory, especially oxen.Footnote ²⁶ In this kebele, a land taxpayer, on average, has a user right over 1.34 hectares – the official figure – but many do not farm this land themselves. The average area farmed by each land taxpayer, who is effectively a farmer, is certainly higher. Sixty per cent of the sample farm as much or more, with an average of about 1·65 hectares. The ratio of 1:5 between those with the least land and those with the most is highly indicative of the heterogeneity of this group. The asset value of livestock kept by members of the sample was less unequal, with a maximum ratio of 1:3.5. The average value is hardly any higher than that for the land taxpayer in the kebele as a whole.

Are they really any wealthier?

One indicator used to measure whether the model farmers have effectively ‘gained wealth’ was the evolution of their basic assets, measured by their livestock and the area of the land being farmed, compared with the year in which they enrolled, which was almost always 2005. These assets seem to have diminished for 10% of the sample, stayed the same for half of them, increased slightly for 20% and considerably for 20%, mainly through an increase in the land under cultivation (contract and sharecropping). But, in the absence of available statistics, it is impossible to compare this increase in wealth with that of farmers as a whole.Footnote ²⁷ Another indicator looks at the use of fertiliser, which is considered by all farmers to be essential. All members of the sample use it. The cooperative, which holds the de facto monopoly over their sales, only supplied 750 farmers out of 1,133 land taxpayers. Despite the fact that not all land taxpayers are engaged in farming, their number is certainly above 750, which means that not all of them bought fertiliser. Three quarters of the sample grow cash crops. On average their last net annual income from cash crops represents over a quarter of the value of their whole production. But once again any comparison with the other farmers is impossible, in the absence of data.

The authorities have divided the model farmers into three categories.Footnote ²⁸ Category A includes those who, inter alia, ‘start to sell their products in the market and to become rich’; category B are those who, inter alia, ‘don't change fundamentally even if they improve their life to a certain extent’; category C are those who, inter alia, ‘don't improve their own way of life and are not ready to do so’. Only one third of model farmers in the sample are in category A (33%); the official guidelines put the expected figure at 30%. According to the authorities' own assessment, two thirds of model farmers are in category B (51%) (45% according to the official guidelines) or C (17%) (25% according to the official guidelines).

Why these discrepancies?

One high official in the woreda spontaneously confirmed that the number of what he called ‘true’ model farmers was probably not much more than sixty in the kebele. Why, then, enrol five times more, often by renouncing the official criteria? He replied, like other officials, that the authorities higher up had ordered 300 inscriptions, and that ‘some mistakes’ had been made. At the end of 2005, the ruling party was obsessed with restoring its hegemony, and particularly with winning the forthcoming 2008 local elections. The new electoral regulations had raised the number of members of the kebele council to 300. It was almost impossible for the opposition to muster that number, especially in such a repressive atmosphere. But the peasant elite – the backbone of the opposition – even silenced, continued ‘to want to have nothing to do with those in power’, as its members often repeated. Thus the regime used the enrolment of model farmers as a means of massive recruitment to the party.

This enrolment was presented to this elite, during meetings lasting weeks, as a purely economic advantage, offering attractive access to inputs, loans, etc. It was only during the last days of these meetings that the farmers learned that, without their knowledge, their enrolment as model farmers automatically made them members of the party. While the majority grudgingly accepted the situation, saying ‘it's impossible to live without carrying out the orders of the government, so I joined’, considerable pressure was put on those who refused, including imprisonment, until most of them finally gave in (Lefort Reference Lefort2010).

Later, members were recruited on a case-by-case basis. These new recruits were mainly farmers who did not meet the membership criteria, but whose vulnerability or submissiveness meant they could not disobey a recruitment order, or a few rare farmers who were successful and powerful enough to have resisted for that long. But their position became increasingly uncomfortable: they often managed to compromise their membership by being officially registered as model farmers and therefore ‘vanguards’, but tacitly dispensed from all but the financial obligations of party membership.Footnote ²⁹ Officially, the process aimed to promote the market economy and therefore the freedom to run a business, so that the model farmer could give his all as an entrepreneur; but from the very beginning it was subverted for political ends: restoring and then guaranteeing the permanence of the ruling power. Two thirds of model farmers (categories B and C), however, lacked any real market-oriented entrepreneurship, while part of the peasant elite managed to resist all the pressure put on them to become model farmers, even though there was more of this pressure than ever. The Chairman of the kebele said there were at least thirty of these recalcitrants in the kebele.

This leads to a situation where membership of the party prevails over being a model farmer, to such an extent that the latter fades into the background and even disappears behind the former. Ninety per cent of members of the sample claimed to be ‘vanguards’; 10% said they did not know whether they were or not. However, only 55% of model farmers interviewed said that they were, 15% said that they did not know, and 30% that they were not. Of the sample of twenty model farmers, only three more or less distinguished the economic function of membership from the political function of ‘vanguard’.

The reforestation zemacha – an example of massive waste

This rallying effort spread and intensified to an unprecedented extent, as seen in the last development campaign, known as a zemacha. Footnote ³⁵ At the end of 2010, officials from the woreda were mobilised for the imminent launch of a great zemacha, without being consulted about its objectives or modalities. In the first days of January 2011, category A and B model farmers were called to a meeting. On the second day they learned that it was going to last two weeks, full-time, despite the religious holidays. They risked sanctions if they did not attend. They protested and one participant went so far as calling the woreda official who chaired the meeting a ‘liar’ and then walked out of the meeting, followed by a dozen others.Footnote ³⁶ Participation petered out and militia were sent to fetch the absentees from their homes.

The goal was to combat climate change through reforestation. But for most of the participants, ‘God and God alone decides to provide sunshine or rain’. For the more enlightened, the problem of recurring droughts will certainly not be solved locally, perhaps not even regionally. Also, the reforestation project targeted common grazing land, which would reduce pastureland. This zemacha was supposed to extend over the five-year period of the Growth and Transformation Plan 2010/11–2014/15. In the final days of the meeting, the participants learned that they had to follow on with sixty uninterrupted days of reforestation work. They protested. A compromise was reached at forty-eight days, more or less continuously, with four hours’ work each day, to be started at daybreak. Presence was obligatory, but the question of sanctions and their nature was left vague. On top of this, they had to produce 300 indigenous tree seedlings in the year.

Once the reforestation project started, absenteeism gradually increased and the Chairman of the kebele put some fifty absentees in prison. The woreda authorities ordered them to be released, as the instructions from higher up were now to ‘persuade, not punish’. It turned out to be a botched job. The first heavy rains washed away the initial work, which had not been carried out properly, even though the participants had warned the authorities about this. The forty-eight days of work were supposed to end on 30 April. By that date only thirty days had been used. There was massive desertion. According to several interviewees, there were no more than a dozen participants by the thirtieth day. At the time of the latest survey, there had been no new call to work. There was no longer any mention of the seedlings.

The developmental state in practice

This example illustrates the manner in which those in power, at least at the local level, see their leadership and how it is executed. First, they and they alone feel entitled to define both the aim and the means of execution of their decisions, and to be answerable only to themselves, for at least two reasons. On the one hand, they consider that all organisations and forces other than those they control (like opposition parties and civil society movements) behave like ‘rent seekers’, solely in their own interests, whether mercantile or to gain power. Only those in power are intent on furthering the common interest.Footnote ³⁷ On the other hand, capacity is correlated with status, and so is at its greatest at the top of the power hierarchy and, whatever may be said in public, almost zero at the bottom. Socio-political and knowledge hierarchies overlap almost totally. Local officials almost universally regard the vast majority of the peasants as ‘backward’, ‘uneducated’ and even ‘like disabled persons’.

This attitude bolsters a strictly top-down approach, and, contrary to the official credo, precludes any form of participatory approach. Decisions taken at the top, and which middle-ranking and local officials must endorse, are deliberately presented in a maximalist manner, leaving a small margin for ‘discussion’ with those that are implicated, so that they can be scaled down and become less excessive. This marginal negotiation process, where only the means of implementation are partially up for discussion, allows the authorities to claim that the final decisions have been taken ‘by the people’.Footnote ³⁸ Yet these official attitudes also neutralise any form of criticism from those at the bottom of the hierarchy. They are irrelevant because they are ‘irrational’, as the local authorities stated.

Second, the mission of the developmental state is fulfilled by the ruling party and not the state, despite its title. But the ‘monolithic party-state’ alone does not explain this devolution. It is based upon a constantly repeated argument that ‘development is a political process first and an economic and social process later … It is the creation of a political set-up that is conducive to accelerated development that sets the ball of development rolling’ (Meles 2006). The main feature of the developmental state is that it mobilises the peasant in his capacity as ‘vanguard’ to an infinitely greater extent than as model farmer.

The first duty of a party member is ‘to develop his farm and his environment’.Footnote ³⁹ The ‘meetings’ organised by the party, which he has to attend one or two days every quarter, deal almost exclusively with questions of socio-economic development in the wider sense, while political issues are only occasionally raised. The same is true for the party cells, especially given that their remit now includes that of the former development group. The interviewees sum up the marginal political function of these meetings as just to report ‘problems’ or ‘misconduct’ to the authorities higher up. Finally, the only activities specific to the model farmers are those categorised as ‘training’, undertaken outside the ‘meetings’. They are very rare.Footnote ⁴⁰ The DAs hold two training session per year, and admit that they are poorly attended.Footnote ⁴¹ Two thirds of the interviewees say they have never been to one. They are not part of a formal system, but lie on an informal boundary between the demand for training coming from the peasants and the offer proposed by the DAs. All of these training activities are optional.

Third, the training that is given by the party-state to promote development among the peasants is ‘mostly theoretical’, as a woreda high official pointed out. It mainly consists of harangues, speeches, lectures, exhortations and slogans – it takes the form of words.Footnote ⁴² As one interviewee summarised the general view: ‘All the meetings are connected to each other and we are always told the same thing.’

Fourth, the goals are absurdly exaggerated – and thus unattainable –as well as being erratic and abandoned or applied in haste. They lose any credibility they may have had. Expecting a peasant to abandon his farm for sixty consecutive days to carry out communal work or to grow 300 indigenous tree seedlings in a year is unrealistic. The previous year, the target was to make compost (twenty-five cubic metres per farmer household). The year before that, it was digging water holes. This voluntarist rhetoric, rather like a mantra, recalls the former TPLF's Maoist ideology. As a result, the great majority of interviewees distance themselves from their status as model farmer, and even simply turn their back on it. Not only do they find it brings them no benefits – no ‘development’ as they said – but they also deplore the burdensome duties arising from the confusion between the status of model farmer and that of ‘vanguard’.

Eighty per cent of interviewees said that they had never received training that was of ‘any use’ whatsoever. Yet in answer to the question: ‘If you were free to stop being a model farmer, would you?’Footnote ⁴³ Seventy-five per cent of those who know they are model farmers replied ‘no’. This is because, as one interviewee put it, ‘something may come in the future’. The reasoning goes as follows: although membership does nothing for me at the moment, there are no specific obligations, and perhaps one day it will be beneficial, through some useful practical training, or even training in the district capital, with a per diem of 35 birr. This hope suggests that the supposed ‘passivity’ of the majority of the peasantry in the eyes of most of the local officials is false, at least for some of them, and proves their desire to learn in order to improve their lot. On the other hand, four fifths of the interviewees said that they would leave the party if they could. Of the reasons given, political opposition, in the strict sense, is not top of the list. Most cite the nature and burden of their ‘vanguard’ duties. In the words of one of the interviewees, ‘they do not offer any solution’.

The squeezed DAs

Since commercial agriculture became the priority, the Agricultural Extension Programme has been targeting model farmers, with two expected benefits. First, access to inputs, essentially fertilisers and seeds, depends solely, at least in the concerned kebele, on how much is available at the cooperative, and the farmer's purchasing power. But as a rule supplies are inadequate and so the model farmers are given de facto priority. They are also disparaging about the second supposed benefit –the training dispensed mainly by the DAs. When asked about the main difficulties they encounter, the DAs put their poor working conditions first (low salary, lack of resources, etc.), then the lack of experimental fields at the Farming Training Centre, and the refusal of the great majority of peasant farmers to call on them. ‘The peasants do not come to our training sessions’; ‘they appoint me [in their field] but they don't come’. The peasants themselves say they do not have enough money in any case to implement the training they would receive, and the DAs are in no position to provide any material assistance. The peasants already feel overwhelmed by their party obligations, and have no time for optional training. The DAs, though, think that ‘the peasants only believe in punishment’. But the main explanation they give for this resistance is that the farmers ‘are not ready to learn’. They are convinced that ‘our fate is in the hands of God, not in ours’, said the DAs. DAs never question the relevance of the training they are offering, in terms either of content or the way it is delivered. They see no mismatch between the training they offer and the needs and means of the peasants.Footnote ⁴⁴

Only five of the twenty model farmers interviewed said they had benefited from genuine training, two of them with a DA coming to their field. Three felt that the training they received was useful, the DAs concerned being judged as competent and able to put what they taught into practice themselves.Footnote ⁴⁵ Meanwhile, the explosion in growing cash crops, starting with onions, represents the great production breakthrough in the last two decades. Three quarters of the interviewees took part in this, to varying degrees. All were pleased with the results and want to increase their production. But they all had the same story about how this new vegetable came to be grown: the first ones had learned about it from farmers in a neighbouring woreda, with a strong market garden tradition, and then passed on the know-how by word of mouth alone. None said they had received any kind of training in this area through the agricultural extension programme.

The first argument used by interviewees to explain their disdain for the DAs is their incompetence: ‘the farmers are better than them’; ‘they don't even know how to handle a plough!’ They also reproach them for their lack of availability. Whatever the reason, the vast majority want to keep their distance. But while the DAs may be short of practical skills, they do possess some knowledge that many peasants could benefit from, as proven by three members of the sample. Conversely, the claim that peasants ‘are not willing to learn’ can be disproved by at least three indicators: the spread of cash crops; the significant proportion of the interviewees who wish to remain model farmers; and an obvious thirst for exchanging knowledge with other farmers. This last point is the only use that some interviewees find for the innumerable meetings, and generally occurs outside the formal discussions.

Why this divide between DAs and peasants?

The divide between rulers and ruled probably originates in the perception of the overwhelming majority of peasants, at least in the kebele concerned, of the political order in which they are living, which is inseparable from a deep respect for hierarchy. This perception and meaning are even more significant, given that they are completely validated by the way this order is manifest in concrete terms. An almost impassable divide traverses the rural society concerned. 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 and, at least in past centuries, between exploiters and exploited. This relationship has, for ages, led peasants to do all they can to keep the mengist at a distance.

The DAs, and beyond them the entire programme, are perceived by almost all peasants as one of the main cogs in this mengist machinery, both statutorily and operationally. As interviewees stated, in all meetings, the DAs sit with the mengist on a podium beside the officials, while the peasants sit facing them. They are the main leaders when it comes to the compulsory execution of development work, and are therefore seen as also being responsible for all the constraints it brings with it. They draw up official production figures, which all the peasants – and local officials – know are seriously overestimated. But the authorities rely on these figures to declare as unjustified any requests for aid that peasants may make, when there are obvious food shortages.Footnote ⁴⁶ Finally, some interviewees mentioned that the DAs take part in the party recruitment campaigns and propaganda and in collecting the land tax. So the DA/peasant relationship, instead of being one of technical cooperation, is part of the hierarchical relationship with the authorities in the broadest sense. Hence the final judgement of almost all interviewees about the DAs: ‘we don't trust them’.