This article analyses the Rwandan elite's visions and ambitions for a wide-ranging re-engineering of rural society. The post-1994 political elite has few links to rural society and the peasant way of life, and sees little room for small-scale peasant agriculture in Rwanda's economic future. The article shows how current Rwandan policy makers aim to realize three social engineering ambitions: first, to transform the agricultural sector into a professionalized motor for economic growth, centred on competitive and commercial farm units; second, to artificially upgrade rural life by inserting ‘modern’ techniques and strategies into local realities, while hiding true poverty and inequality; and, finally, to transform Rwanda into a target-driven society from the highest to the lowest level. The article points to the (potential) dangers, flaws, and shortcomings of this rural re-engineering mission, and illustrates how the state as the engineer ‘hovers’ above the local without consulting those affected. It concludes that contemporary polices are unlikely to be conducive to poverty reduction.

After a devastating four-year civil war and an apocalyptic genocide in 1994, Rwanda's post-conflict reconstruction process certainly has its merits. The state was rebuilt at surprising speed, taking up responsibilities in terms of service delivery in education, health, and infrastructure. Rwanda is today cited as a country with relatively low levels of corruption, and, according to the World Bank's Governance Indicators, both political stability and government effectiveness have improved from 2002 to 2007.1 The gacaca courts, used to try those accused of acts of genocide, were initially seen as an inventive, locally embedded form of restorative justice that would allow for the peaceful resolution of discord. Economically, donors have invested massively in the reconstruction and development of the Rwandan economy.2 Indeed, recovery of the Rwandan economy has been exceptional and, after a spectacular post-genocide economic boom, national income has continued to rise steadily.3

At the same time, criticism has mounted, pointing to the weak points of the reconstruction process. Political liberties have been constrained,4 the gacaca process has been far from a uniformly positive experience,5 and economic growth has not been accompanied by commensurate poverty reduction. While in percentage terms poverty decreased from 60.3 percent of the population in 2001 to 56.9 percent in 2006, in absolute terms it increased from 4.82 to 5.38 million people (based on the national poverty line). Moreover, inequality increased, from a Gini coefficient of 0.47 in 2001 to 0.51 in 2006.6 In rural areas, certainly, progress has been limited and has remained concentrated in the hands of a small class of agricultural entrepreneurs, while the majority of Rwandan peasants are confronted with increasingly difficult living conditions.

Rwandan policy makers are aware that poverty is a mounting problem. A preliminary government report reviewing progress in living conditions from 2001 to 2006 acknowledges that the reduction of poverty levels has been uphill work ‘because growth over this period has been accompanied by increasing inequality’.7 The government's vision of how to achieve economic progress and poverty reduction is set out in its ‘Vision 2020’ document, and has been further elaborated and operationalized in the first Poverty Reduction Strategy Paper, implemented during 2002–5, and in the new Economic Development and Poverty Reduction Strategy, finalized in 2007 and to be implemented between 2008 and 2012.8

The overall aim of the current government is to transform Rwanda from ‘low’ to ‘medium’ human development, as defined by the UNDP's Human Development Index. This is to be done through a radical modernization of the overall social structure, particularly by moving the agricultural sector away from subsistence and towards a more commercial and diversified economy. In this radical vision of transformation, Rwandan policy makers can be seen to conform to what Pound has defined as the engineering mission of law and policies.9 According to this view, social arrangements can be manipulated and managed through conscious human control using law as an instrument. As Ellerman describes it, ‘If we use the metaphor of the doers as trying to work their way through a maze, then the helpers [development agents] as social engineers see themselves as helicoptering over the maze, seeing the path to the goal, and supplying directions (knowledge) along with carrots and sticks (incentives) to override the doers’ own motivation and push the doers in the right direction.’10 In other words, the social engineers ‘hover’ above the local, providing strategies and knowledge, laying claim to the ‘whole picture’ without consulting those affected.

In the Rwandan context, the engineering mission appears in many different domains. Reyntjens, for example, cites the rather picturesque, but nonetheless intrusive prohibition of the use of plastic bags, as well as the enforcement of massive consecration of marriages according to state law, and the implementation of an ambitious modernization policy.11 This modernization exercise has been most apparent in Kigali, where skyscrapers and massive buildings have arisen with great rapidity, but, increasingly, the ambition to re-engineer Rwanda also touches the rural setting. Ingelaere, in particular, points out the importance and dimensions of the social engineering aspect of the post-genocide Rwandan regime in multiple domains of rural life.12

Overall, the elite's re-engineering mission within the rural setting is characterized by three main objectives. First, policy makers aim to transform the agricultural sector into a professionalized motor for economic growth, with little scope left for traditional smallholder agriculture. Second, policy makers seek to upgrade rural life by inserting ‘modern’ techniques and strategies into local realities, while hiding the extent of poverty and inequality. Finally, they hope to transform Rwanda into a target-driven society from the highest to the lowest level. Taken together, these three social engineering ambitions amount to a top-down developmentalist agenda with a central role for the state as the engineer that shapes and reshapes the rural environment.

In this article, I first explore how political developments – with special reference to 1994 – have brought to power a political elite whose identity (both ethnic and spatial) differs profoundly from that of the overall majority of the population. I also show how this elite differs from the pre-1994 elite, both in terms of their profile and in terms of their attitudes and ambitions for rural development. The main part of the article analyses how Rwandan policy makers aim to realize their engineering ambitions for rural society. I then draw some conclusions about the (potential) dangers, flaws, and shortcomings of the current re-engineering mission. To capture the discourse of Rwandan policy makers, I draw on 26 interviews conducted between May and July 2007 with persons closely involved in poverty reduction, agricultural, and land policies. The interviewees included high- and lower-ranking officials of the three ministries centrally engaged in rural development: the EDPRS (Economic Development and Poverty Reduction Strategy) Department within the Ministry of Finance and Economic Planning; the Ministry of Agriculture and Animal Resources; and the Ministry of Land, Environment, Forests, Water, and Mines. Alongside secondary data, these interviews provide a comprehensive picture of the present rural development discourse within government circles.13

What is old, what is new? Characteristics of pre- and post-1994 elites

Rural development is not a purely technical issue. As Bebbington argues, ‘Neither patterns of asset distribution nor institutional conditions in rural areas are accidental. Indeed, they each derive from the broader relationships between politics, economy, and society that drive and undergird the overall patterns of rural development.’14 The World Development Report of 2008 similarly suggests that agricultural policy making results from a political bargaining process driven by the power dynamics between citizens and politicians.15 In other words, the elite–peasant relationship is crucial to an understanding of the political elite's rural development discourse and policy.

The elite–peasant relationship in Rwanda is rooted within various layers of identity, and although historical research has focused principally on ethnicity, other factors such as regional background, kin, social class, occupation, and gender have profoundly influenced these identities. The relative importance of each factor has also been subject to change over time within Rwandan historiography.16 Many studies have elaborated extensively on ethnic cleavages, focusing on the strong animosity between Hutu and Tutsi, and often expose very divergent and contradictory theories on the origins of ethnic differentiation.17 It is clear, however, that the Hutu–Tutsi rift evolved over time from a flexible social class indicator towards an increasingly institutionalized division. Belgian colonial policy also used ethnicity to accomplish its own political agenda. Just prior to independence in 1962, ethnic Tutsi dominance in the political centre was reversed, and for the following 30 years a Hutu president ruled Rwanda. During this period, several episodes of inter-ethnic tension and violence led to significant Tutsi emigration.18

Geographic origin(s), along with ethnicity, equally underlie disparities. The ruling Hutu elite, though often urban-based, retained important links with their rural areas of origin, and tried, unduly if not surprisingly, to advance their interests. The first Hutu president, Grégoire Kayibanda, favoured the central region. After a military coup in 1973, the new president, Major-General Juvénal Habyarimana (a Hutu, native of the north-west province of Gisenyi), began to grossly favour the country's north.19 Similarly, Newbury and Newbury documented regional disparities in the distribution of government development funds during this period, while de Lame pointed to the importance of the Kigali political elite's ties with their specific hills of origin.20

The new Rwandan political elite differs from that of the preceding period, both in terms of ethnicity and in terms of its relationship to peasants and the countryside. After the genocide in 1994, the Tutsi-based Rwandan Patriotic Front (RPF) seized power, and the previously ruling (Hutu) elite disappeared from the political power centre. Although the leaders of RPF typically portray themselves as the main force behind the ‘liberation’ of Rwanda, post-1994 Rwanda has in fact been ruled by an elite whose identity, in terms of ethnicity and origin, differs markedly from the majority of the population. The rebel army was transformed into the RPF political party, including both Tutsi and Hutu. For the top positions, however, ethnic background has come to play an important role. Table1 illustrates the ethnic bias in the composition of the presidency and the government. Although 85 percent of the Rwandan population are Hutu, they occupy about 50 percent of presidential and government positions, an equal share to the Tutsi. The majority of the new Tutsi elite are ex-refugees, and most returned from Uganda (including President Kagame) and the Democratic Republic of the Congo after having lived outside Rwanda for decades, or even generations. Following the RPF's military victory, the new elite installed themselves in the capital. While this was partly for security reasons (the countryside was still unstable in the immediate post-1994 period), it also reflected the fact that a considerable number of the returnees had lost their ties with the ‘hill of origin’ and had little incentive to go to the rural areas.21

Table 1. 

Composition of the Presidency and the Government

 Ethnicity RPF / non-RPF 
 Tutsi Tutsi (non)      
 ex-refugee ex-refugee Hutu N.a. RPF Non-RPF N.a. 
10/01/1998* 14 18  20 15 
08/02/1999* 18 19  26 17 
20/03/2000* 12 11  19 11 
31/05/2001* 15 12  21 11 
10/08/2002* 18 14  21 10 
20/06/2003* 16 15  23 10 
24/06/2004* 15 17 23 
22/06/2005* 15 15 23 
23/05/2006** 12  15  
01/06/2007** 11  15  
01/04/2008* 10 11 5*** 17 
 Ethnicity RPF / non-RPF 
 Tutsi Tutsi (non)      
 ex-refugee ex-refugee Hutu N.a. RPF Non-RPF N.a. 
10/01/1998* 14 18  20 15 
08/02/1999* 18 19  26 17 
20/03/2000* 12 11  19 11 
31/05/2001* 15 12  21 11 
10/08/2002* 18 14  21 10 
20/06/2003* 16 15  23 10 
24/06/2004* 15 17 23 
22/06/2005* 15 15 23 
23/05/2006** 12  15  
01/06/2007** 11  15  
01/04/2008* 10 11 5*** 17 

* Ethnicity of president, ministers, secretary of state and secretary general (of all names available); ** Ethnicity of president and ministers only. *** Four are tutsi.

Source: F. Reyntjens and S. Marysse, Annuaires L’Afrique des Grands Lacs, 1998-1999, 1999-2000, 2000-2001, 2001-2002, 2002-2003, 2003-2004, 2004-2005, 2005-2006, 2006-2007, 2007-2008, (L’Harmattan, Paris).

The over-representation of Tutsi ex-refugees in official posts and their detachment from the rural setting was frequently raised during interviews, especially by non-government interviewees. A representative of a European donor country highlighted that the characteristics of the current elite explain, to a considerable extent, their relation with the countryside:

Many of the government officials have never known the Rwandan countryside. They came from refugee camps, and when they took over power, they often left their parents behind in Umutara [a province in the north-east of Rwanda with many large cattle farms]. Moreover, in the past, there was still a lot of insecurity in the countryside, so people preferred to live in the city of Kigali. That is where they are concentrated now, it explains why they have limited knowledge and understandings of how peasants live.22

In addition to the ethnic and refugee dimensions that shape the identity of the current political elite, there is a clear language dimension. Before 1994, the second language after Kinyarwanda was French. In the post-1994 period, under the influence of the Ugandan diaspora, English is gradually becoming the dominant language within governmental circles. This favours the promotional chances of former exiles. As a representative of civil society put it,

The government really adopts a policy to exit the ‘Francophony’ and to enter the ‘Anglophony’… . Within the government, we see the same trend. Those who speak French and have the right competences are not taken into consideration [when assigning someone to a post]; and when they are, they keep quiet because they do not want to create any problems. It is nonetheless mostly the French-speaking who master the rural setting.23

This interviewee referred to the fact that those with more experience in rural development are often Francophones who were present in Rwanda before 1994. They have a longer and more in-depth experience with the problems of the rural environment, and often have direct ties with the countryside where their families still reside. Accordingly, they are more familiar with the logic of subsistence farming, but in the current context their lack of English is a barrier to a government post. According to several interviewees, even if Francophones do learn English (as many now do), their chances of promotion are still lower due to a lack of political connections. As one representative of a European donor country mentioned, this is not merely an issue of English- versus French-speaking; ‘It is not that much about bilingualism; it is clear that people with Ugandan roots are favoured. The language issue is simply the consequence of the fact that more people from Uganda are pushed forward.’24

Overall, we can conclude that the characteristics of Rwanda's current political elite contrast strongly with those of the majority of Rwandans. The former are mainly Tutsi, nearly always urban-based, and often born and raised in a neighbouring country. The latter are mostly Hutu, rural peasants, and born and raised in Rwanda. Likewise, the identity of the new elite also differs from that of the pre-1994 era, when many were Hutu and still had personal connections with the countryside, with their hill of origin.

Visions and attitudes towards the peasantry

A key question accordingly becomes whether this difference in identity translates into a different attitude towards the peasantry. At the level of political discourse, the contrast between past and present opinions about the potential of the peasantry in the development of the economy is dramatic. The Habyarimana administration (1973–94) relentlessly championed the culture of an agrarian society. Verwimp shows how Habyarimana in his speeches often ‘glorified the peasantry and pictured himself as a peasant’.25 An intriguing example can be found in his speech on the 25th Anniversary of the Republic of Rwanda in 1987:

If in the 25 years of our independence Rwanda has known a lot of success in its struggle for progress, if it has been able to take a number of important steps, it is in the first place our farmers who made this happen… it is their total devotion to their work, every day… their fabulous capacity to adapt, their pragmatism, their genius, their profound knowledge of our eco-systems that allowed them to extract an amazing degree of resources from their plots of land… .26

While the rhetorical language of the Habyarimana administration referred to the peasantry as the ‘Rwandan ideal’, this has profoundly changed under the Kagame administration. The influential ‘Vision 2020’ document outlines the main challenges and policy priorities of the Rwandan government up to 2020, and states in a strikingly different tone:

Rwanda's economic policies since independence are said to have targeted agriculture as the main engine of economic growth. However, the agricultural sector has continued to perform poorly, with consistently declining productivity. It will be necessary to formulate and implement realistic developmental policies that move beyond past delusions of viable subsistence-based agriculture.27

The Strategic Plan for Agricultural Transformation identifies as a main challenge the ‘transformation of subsistence agriculture into commercial agriculture with all its involvements in terms of institutional, social changes of behaviour and distribution of roles and responsibilities between different stakeholders’.28 The land policy takes this even further, arguing that ‘the Rwandan family farm unit is no longer viable… . The reorganization of the available space and technological innovations are necessary in order to ensure food security for a steadily and rapidly increasing population.’29 Alison Des Forges refers to this process as the government's ambition to ‘winnow out the chaff’ in the agricultural sector.30

While there is clearly some awareness among current policy makers of the many institutional constraints that small-scale peasants face and that serve to keep them in subsistence agriculture, the solution to the poverty problem is often reduced to adopting a ‘good mentality’. In his inaugural speech in 2000, for example, President Kagame referred to the many development challenges facing Rwanda, and continued:

I do not believe that we should lose hope and surrender ourselves to lives of poverty. If we can utilize the resources that God has given us to good effect, we can eradicate poverty… . We would like to urgently appeal to the Rwandese people to work. As the Bible says, ‘he who does not work should not eat’.31

In subsequent years, President Kagame has frequently spoken of the burdens of the past, while emphasizing the responsibility of each citizen to overcome his/her own poverty. Similar sentiments were expressed by some interviewees, referring to an ‘awareness problem’ among Rwandans. One official, for example, maintained that

[We have to convince the people] to change radically and to become part of a society that can take care of itself, that can survive on its own, and that does not have to beg… . One should not wait until one comes to help you as if you are a little baby. The head of state is angry with this spirit. Instead of depending upon others, one has to do things on one's own… . We really have to convince everyone to be with this national slogan that everyone has to go forward in life.32

The tone of this quote suggests an underlying assumption that poverty is, in fact, a state of mental dependence, somehow a deliberate choice. Getting out of poverty accordingly becomes a question of adopting the right strategy and ‘putting one's mind to it’. Thus, a district official in Southern Province highlighted the mentality of the people in the region as one of the major explanations of the province's limited performance in relation to key indicators of well-being: ‘You talk to them and you think they listen, but the people do nothing with the good advice you give them. They say “yes” because they are tired of you and your speeches, but they are never convinced… . They are resistant, they are really difficult.’33 By no means the general sentiment among the interviewees, this statement is nonetheless symptomatic of the current elite's view that poverty is partly due to the ‘wrong peasant mentality’. A further illustration is found in the Strategic Plan for Agricultural Transformation, which refers to the problem of peasants’ ignorance and resistance to adopting recommended productivity-enhancing measures that go beyond traditional subsistence farming.34 As such, peasants’ perceived lack of capacity to embrace ‘modernized’ farming is at least partly attributed to a ‘lack of vision’, a view that totally disregards the institutional barriers they face.

These examples illustrate that there is indeed a difference in discourse between the pre- and post-1994 elites with regards to the potential of the peasantry in overall national development. However, the picture deserves more nuance. Despite its pro-peasant rhetoric, Habyarimana's policies also displayed at times a strong anti-rural bias. Verwimp has illustrated how Habyarimana's regime failed to respond to early warnings after crop failures in 1989.35 Similarly, Pottier has shown how agricultural officers in the pre-1994 period, who functioned as brokers between the central state and local peasants, often served as ‘imposers’ of state policies. He points to the excessive formality of public meetings that gathered together Rwandan cultivators and agronomists, and argues that this stood in the way of true dialogue and that interaction boiled down ‘to a one-way, dogmatic delivery of textbook instructions’.36 According to Newbury and Newbury, agricultural officers were chosen on the basis of their educational profile rather than actual experience. They conclude that, ‘not only does “state agriculture” become a coercive field, but much local knowledge (e.g. local variations of crops, soils, pests, labour practices, etc.) is lost, in the name of standardizing and “rationalizing” agriculture’.37 Indeed, the disparity between rural and urban settings deteriorated during the Habyarimana period, and by 1990, about one out of six people in urban areas lived below the national poverty line, while in rural areas this applied to over 50 percent of the population.38

In short, a top-down, state-centred governance structure is not new to Rwanda; nor are the rural–urban gap, the anti-rural bias in policy making, and the state–society cleavage specific to the post-1994 period. Instead they appear to be structural features of elite–peasant relationships in Rwanda, and beyond.39 That said, the argument of this article is that the current vision and ambition of the Rwandan elite to socially engineer rural society goes further than these previous attempts at reform, and that the potential implications are more problematic given that policy makers see no role for small-scale peasants. The social engineering ambitions of the Rwandan elite are analysed in more detail below.

What place for small-scale peasants and the unskilled labour force?

Overall, Rwandan policy makers see very little role for small-scale peasants in economic development, despite the fact that small family farms make up over 90 percent of all production units, with an average of less than one hectare in size.40 Instead, policy makers favour drastic transformation of the economic structure of Rwanda, where peasant agriculture plays a minor role. According to a donor representative, during the EDPRS discussions, there was even considerable debate over whether agricultural development should still be a core issue; ‘There has been quite a battle on this issue… . However, a degree of realism returned, certainly for the plans in the short and medium term.’41 The question remains what type of agricultural transformation should be aimed for. A considerable number of the interviewed policy makers supported investing in rapid modernization and professionalization of the agricultural sector, with a strong focus on maximum productivity and output growth. As I have analysed elsewhere, the government's focus on professionalization and modernization favours competitive and commercial farmers, and this disregard for equitable wealth distribution threatens further to enfeeble small-scale farmers.42 In fact, the structure of subsistence peasants’ farms often prohibits the necessary risk taking that would allow them to invest in new high-potential production systems.43 In the words of one human rights activist,

How will a peasant exploit his land in a ‘professional’ manner with the little he has and without the support of the government? In fact, they need training, specifically oriented towards the exploitation of small surfaces, because this is the reality of agriculture in Rwanda.44

Most government officials interviewed, however, linked the professionalization and commercialization of the primary sector to the necessity for larger farm units. Some see this happening by consolidating current farms through collective ownership, as expressed by this high-ranked government official of the Ministry of Land:

We will not take someone's land. The consolidation objective has the aim to intensify productivity; this is not equal to taking away land from people. When Minagri [Ministry of Agriculture] is talking about large farms, they do not mean that these farms would belong to one person… . Households will consolidate in terms of land use, not in terms of land ownership.45

However, a high-ranked government official in the Ministry of Agriculture stated:

They say that agriculture is the productive sector, but it isn't in Rwanda… . In fact, we should stop calling it the productive sector; it is at this point the survival sector. At this point, most people are not earning because the pieces of land they have access to are too small… . We have to get more people off the land, as we cannot continue a system with small pieces of land… . When people get off the land, there will be more land in the hands of fewer people, which will allow a better planning of the system.46

Government officials regularly referred to the aim and need to set in motion a ‘green revolution’ within the agricultural sector, but most saw no role for smallholders in this process. By comparison, the recent World Development Report stresses the importance of smallholders in a green revolution for sub-Saharan Africa. The report recognizes that modern agricultural evolution can undermine smallholder farming, but points to the potential of policy instruments to enhance smallholders’ competitiveness.47 Rather than increasing the competitiveness of peasants, one of the principal goals of ‘Vision 2020’ is to decrease the population dependent upon agricultural activities from 85 to 50 percent, that is, a decrease of 35 percent by the year 2020. This goal was frequently mentioned by the interviewees, and while it is clear that a decrease of the population dependent on agriculture might be necessary due to resource constraints, none of the interviewed policy makers had a clear vision of the employment alternatives available to the expected ‘surplus’ population about to enter the economy as off-farm wage labourers. As a representative of an international donor organization remarked, ‘employment in other sectors will appear, but the question remains at what pace, what scale – and what support this will get?’48

Some government officials had a rather fanciful view of the potential for off-farm employment: ‘We will build factories that work twenty-four hours. And this is not only in Kigali, but also in other centres of economic interest.’49 The first Rwandan PRSP even explores the opportunity ‘to leap-frog the stage of industrialization and transform her [Rwanda’s] subsistence economy into a service-sector driven, high value-added information- and knowledge-based economy that can compete on the global market’.50 Most interviewees were more realistic with regard to the immense challenge of absorbing the growing labour force within the off-farm sector. A high-ranking official in the Ministry of Agriculture pleaded for fewer people in agriculture, while highlighting the importance of developing smaller-scale industries and services.51 Several interviewees mentioned the importance of adequate training, arguing that ‘People cultivate because they did not have the chance to be educated. We have to give the people training. They should not leave the agricultural sector without alternatives.’52

Next to the need for a skilled labour force there is also a need for local demand for the goods and services produced by the non-farm sector. At the present time, the limited purchasing power of subsistence farmers means that there are already unemployed carpenters and masons in the countryside. In the words of a human rights activist, ‘if you walk in the hills, and even in the city, you see people outside of the agricultural sector who have nothing to do’.53 An independent consultant similarly reflected:

We should not dream. Where will we put all these people? If we would find something that could employ 40 to 60 percent of the population, at that moment we could count on a trickle-down effect. But with a range of activities that can give revenue to 2 to 5 percent of the population, we will never be able to create a trickle-down effect of which the benefits will reach the other 85 percent.54

The danger is that by focusing on output maximization and concentrating on highly productive farm units, the rural policies of the Rwandan government may do little to alleviate poverty, and may in fact aggravate destitution among some groups. If the population dependent on agriculture is to be reduced from 85 to 50 percent, the question of alternative livelihoods for this group becomes crucial. At present, there are few signs that off-farm employment and economic opportunities will be readily available on such a scale, nor does past experience suggest that any economic growth effects would trickle down quickly enough to assist the remaining population.55

An artificial upgrade of rural life

A second ambition of the Rwandan elite is to upgrade rural life by inserting ‘modern’ techniques and strategies into local realities, while hiding the extent of poverty and inequality. Ingelaere, for example, refers to a system of fines, imposing certain ‘measures improving general wellbeing’.56 In my own interviews officials often stated that ‘we have to teach farmers, we have to show farmers, we have to bring farmers in contact with modern techniques’ while referring to poverty as ‘at least partially a problem of mentality’. The most striking expression of this social engineering ambition is the government's villagization policy. Though this policy was originally designed to resettle Tutsi refugee households in compound villages, it has also been more widely adopted as a measure to increase the efficiency of land use.

The overall results of the policy have been disappointing. According to Human Rights Watch,57 people have been obliged to destroy their homesteads – traditionally scattered on the hills – to live in compact villages (imidugudu). Many households are now worse off, both in terms of housing quality and land possession. The policy has also failed to decrease pressure on available land holdings, and in many cases has even created or deepened land conflicts.58 Despite these negative experiences, the policy has not been annulled. On the contrary, it plays an important role in a new wave of ‘anti-poverty’ measures. A high-ranked government official advanced villagization as a way to stimulate activities in the off-farm sector, adding that

making people live communally in a village, could increase trade and exchange which is more difficult if people live scattered on the hills. It will also facilitate access to school infrastructure, health centres, etc… . The interaction between people is very important, and this is not optimal when people live scattered on the hills.59

Another government official raised this as a crucial issue in the environmental debate,

A big exercise will be to design the land use master plan. At this point, people cultivate everywhere and live everywhere. But the goal is to determine in great detail the purpose of each plot of land. We will plan for imidugudu, grouping people around centres with infrastructure. This will allow us to recover the land that is best for agriculture. It will allow us to cultivate large land surfaces with specific crops, and most importantly to implement anti-erosive measures on those surfaces.60

The continuing relevance of the villagization policy (since the administrative reforms of 2005) is manifest in the terminology that refers to the lowest administrative units. The term umudugudu (plural: imidugudu) was originally reserved for ‘villages’ created by the villagization policy, but the term is now separate from this policy and has replaced the more neutral term cellule, previously used for this administrative level. In drafts of District Development Programmes (set up by the district officials in mid-2007) there are references to the ambition to go forward with the villagization policy as the ‘liberation of space will enable the recuperation of arable land and an intensive industrial exploitation of the areas, the regrouping of the population in cooperatives active in production, transformation, distribution, savings and credit’.61

Another way in which Rwandan policy makers seek to upgrade the rural setting is through environmental policies that ban polluting activities. While defendable in principle, the way in which such policies are imposed is astonishing. For example, the Rwandan government has banned the traditional brick-baking activity, and the self-made ovens that required quite a lot of wood are no longer allowed to operate. In numerous local settings, this has had a huge impact on living conditions, particularly for the non-agricultural labour force. In addition, the price of bricks, as well as tiles, has increased dramatically as a result of this policy. The government has now re-allowed brick-baking, but only with modern ovens, significantly increasing the cost of production. As a result, hardly any of the local entrepreneurs have the necessary means; they are surpassed by external actors who are much less embedded within the local context.62

Social engineering ambitions can also be seen in the attempt to change rural life by introducing certain social obligations or prohibitions. In 2006, Twizeyimana mentioned the obligation to wear shoes, to be clean, use mosquito nets, adhere to the health insurance guidelines, wear school uniforms, construct toilets, make compost pits, and dry dishes on a table instead of on the grass.63 In the six field research settings visited by the author in June–August 2007, local inhabitants mentioned an obligation to ‘walk with shoes’. Initially this policy was only enforced in urban settings, but increasingly it has become common in rural areas. In several locations, people were not allowed to attend public meetings or gatherings (go to the market, for example) without wearing shoes or flip-flops. Several people reported that when arriving at the market without shoes, their food money was taken from them forcibly by the local authorities to buy them shoes. A similar but less enforced policy is for people to wear ‘decent clothes’. In several of the imidugudu, respondents also complained about the obligation to build stables for their livestock, and the prohibition on grazing cattle outside the homestead. In order to comply with this policy, farmers need sufficient space for a stable and for grazing, something that only better-off farmers can afford. An unintended effect of this modernization policy may be the disappearance of a solidarity mechanism through which better-off households lend a cow to somewhat poorer households, who then occupy themselves with raising and guarding the cow.

Taken together, these policies seem to be primarily concerned with the appearance of poverty, rather than alleviating the actual experience of poverty. It is, in other words, an ‘imposed modernity’ that seems to result in nothing more than the cosmetic upgrading of rural life while hiding the true extent of poverty. In contemporary Rwanda, it seems to have become ‘prohibited by official policy’ to be poor.

Social engineering through performance targets

Finally, the engineering ambitions take a very concrete form in the policy makers’ eager race towards performance targets at all levels. In terms of management, the definition of targets may not be a bad idea. However, the risk is that in striving to reach predefined targets, policies are enforced without an eye to local implications. In other words, blind policy enforcement in the name of targets can produce perverse effects in local settings. As an international donor representative cautioned, ‘indeed there is a danger in focusing too much on targets and not on the process. The results are important, but as important are the processes through which the results are achieved.’64 Ironically, however, the donor community strongly encourages target-driven development. Clear criteria provide donors with the necessary monitoring and evaluation tools to assess aid effectiveness. What is often omitted in Rwanda is a careful assessment of the impact of these targets on actual poverty reduction, which is more difficult to measure and to evaluate.

Target-drivenness takes very concrete forms within agricultural production, often with a detrimental impact on poverty reduction. For example, I came across an umudugudu in the Southern Province where the district's agronomist had set firm targets with respect to coffee plantation. If caught uprooting coffee trees, local peasants had to pay 100 Rwandan francs per tree destroyed; and if they could not pay, they stated that they would be imprisoned until the fine was paid. There are other records of peasants being obliged by local authorities to adopt specific production techniques. In 2006, officials urged peasants in the Eastern Province to plant their crops ‘in rows’ and adopt monocropping. In the autumn of 2006, local administrators in certain districts pulled out crops when peasants had not followed the guidelines (they had planted beans in between banana trees).65 In September 2006, the mayor in charge of Muhanga (in the Southern Province) urged the population to replace their banana trees with other cash crops, flowers or pineapples. After a critical broadcast by the BBC, the recommendations were suspended.66 In early January 2007, the Governor of the Eastern Province, Mr Mutsindashyaka, imposed a ban on sweet potatoes. His decision was later revoked by the Minister of State for Agriculture.67 In one of our own interviews with a district official (Southern Province), he maintained that specialization will be a major goal in the rural development strategy of his district. He added that the people fear the risk but, ‘they are obliged to obey: either they invest in this technique, or they will have to leave their land and work for someone who is willing to invest’.68

The issue of monocropping is controversial for several reasons. Authors have long pointed to mixed cropping patterns as a strategy of both risk minimization and profit maximization in particular contexts.69 Indeed, the extreme variety in soil types – even within the same locality – and in climatic conditions makes it difficult for local administrators in Rwanda to assign the ‘right’ crops to administratively defined regions. Rwanda's history of policy-imposed cultivation underlines this; both the Belgian colonial administrators and the Habyarimana government practised ‘forced cultivation’ with little success.70 Given this history, the current scale of crop planning, the blind belief in technical solutions, and the degree of force used during the implementation are again major reasons for concern. The danger is that pre-defined targets are allowed to override such concerns, to the detriment of small-scale farmers.

The risk of a stampede to meet targets is equally high with the new Economic Development and Poverty Reduction Strategy (EDPRS) framework. The sector ‘logframes’ mention very detailed targets, some of which are extremely ambitious. By July 2007, these targets were already (partly) communicated to the district offices, who had to consider them when designing their District Development Plans.71 In addition, the district mayors have signed a performance contract (imihigo) with President Kagame, specifying the key targets that individual districts are to attain within one year, in line with the government's national priorities.72 This makes the district the central unit in the decentralization policy, and the core level for national policies and targets to be re-stated in local plans.73

In principle, making local authorities responsible for the implementation of national policies could improve the process of translating national targets into the local context, making them more democratic and adapting them to the needs of the population. However, there are some major constraints. First, at the most important local administrative levels (district, sector and cellule), the main decision-making power lies with an administrative person (the executive secretary) who is appointed by the central administration, and thus not elected by the population. In other words, this person is not directly accountable to the local population, which he/she is supposed to represent, but to the national government. The person's position depends, to a great degree, upon compliance with and implementation of national priorities, regardless of the burden on the local population.74 As a human rights spokesperson stated,

Free expression is currently a utopia. Laws are implemented in an authoritarian way and the population cannot say ‘no’ to the authorities. In fact, the current leaders receive their instructions and have to implement. I do not believe in the district responsibles; they are commissioners of the RPF. They are the link between the RPF and the population, not chosen but imposed. And they receive good salaries for it. They are not at all close to the population; often they are not even from the region they rule. The local population had to accept, but it created tensions. This is not a stable base for peace. And the same goes for the sector and cellule level. The executive secretary is appointed by the RPF. The person responsible for security is often an ex-military person. And the [elected] coordinator, say the conseiller in the previous system, is not paid. Only the people appointed are paid. At the lowest level of the imidugudu, local authorities do not really have a decision capacity; their role is more symbolic.75

As an illustration, a district official in Southern Province mentioned that he had been ‘sent’ to the district when the Ministry of Agriculture was reorganized. He was clearly not satisfied with his new position outside of Kigali, having to go out to work and return every day. He explained, ‘I do not like to look at poor people and deal with them. In fact, when I worked in the ministry, I did not have to look at the poor. That was the level of policies and decisions. It is now in this new function that I am directly confronted with the poor.’76

It is clear then, that target setting has not brought the administration closer to the local population. The central administration, as well as the local peasants, regard local authorities as purely an implementing body for national strategies, without sufficient influence to translate or reinterpret these strategies so as to make them suitable for the local setting. When asking peasants about their opinion on specific policies, we got reactions like ‘one cannot discuss with the state’; ‘one cannot refuse the law that is given by the state’; ‘a peasant cannot neglect the ideas of the state’ and ‘generally, the peasant is always in favour of the authorities’ [being without any choice in the matter]. Going into more detail, someone mentioned, ‘our own umudugudu coordinator has no power; and as for the executive secretary (at the sector level), he might say that we are being disobedient towards the government if we protest’.

These remarks strongly suggest that the peasants interviewed by no means saw the local authorities as their representatives. The central role of the local administration in implementing national policy is also highlighted by central government officials. As an interviewee of the PRSP coordination committee (within the Ministry of Finance and Economic Planning) explained, ‘The ministries will play a monitoring role, while the biggest part will happen at the district level… . The districts will be evaluated based on the key indicators. If they do not meet them, they need to explain the reasons. It is of course not a crime. But the national priorities have to be implemented at the local level.’77 This view was reinforced by a Ministry of Agriculture consultant, who commented that ‘while targets at the national level were already considered to be ambitious, the targets at the local district level are set even higher than the national targets’ in the framework of the agricultural strategy.78

Indeed, in the preliminary drafts of some district development plans, we found extremely detailed references to crop production targets, to the percentage of soil that should be terraced, to the percentage of households that should be living in imidugudu. Such rigid targets seem to ignore what popular support there is (if any) for planning targets at the local level. The lack of grassroots participation and bottom-up reflection upon the usefulness and adequacy of targets will become all the more problematic given the government's current ambition to expand the ‘performance contracts’ (imihigo) down to the household level.79

Conclusion: implications for rural policy

The Rwandan case illustrates how rural development policies are not a purely technical issue, but one closely related to the position of elites and their relationship(s) with the peasantry. The current Rwandan elite is mostly Tutsi, urban-based and often born outside Rwanda, while the Rwandan peasantry is mostly Hutu, rural-based and born in the country. The physical, ethnic, and mental gap between their worlds profoundly shapes the chances for successful rural development.

The social engineering ambitions of the Rwandan government officials reveal a very top-down developmentalist agenda without much room for grassroots participation or for bottom-up feedback. Instead, the elite approaches law and policy as tools for ‘shaping’ society, and often neglects to consider the institutional and environmental conditions in which the new law(s) will operate. In addition, decentralization has not increased the voice of the rural poor in policy making. On the contrary, it has allowed the centre to extend its influence to the local level in a very authoritarian way. In such a context, target-based assessments to evaluate national policies are potentially dangerous monitoring and evaluation tools.

The elite believes in a rapid modernization and professionalization of the agricultural sector, and strongly rejects subsistence-based agriculture, although it remains the way of life for the majority of the rural population. The elite pleads for larger farm units and consolidation of land holdings, objectives that are to be achieved either by grouping farms in a type of collective ownership, or by transferring land into the hands of fewer people. Their goal is to reduce the proportion of the population that depends upon agricultural activities. There is, on the other hand, no clear public vision of what alternatives will be available to those who leave agriculture.

The image of a countryside consisting of amalgamated large-scale farms is not likely to be conducive to poverty reduction – at least in the immediate future. Moreover, there still is considerable unexplored productive potential in the population of small-scale farmers. Not only could they play an important role in growth strategies, but growth created through their hands would be more broad-based, and would more easily trickle down to the remainder of rural society. The main constraint for rural development is therefore not the lack of potential of small-scale farmers. It is the lack of political will among the Rwandan elite to orient rural policies directly towards the rural poor.

World Bank, ‘Worldwide Governance Indicators: 1996–2007’, <http://www.govindicators.org > (6 October 2008).
S. Marysse et al., ‘The aid “darlings” and “orphans” of the Great Lakes region in Africa’, European Journal of Development Research19, 3 (2007), pp. 433–58.
A. Ansoms, ‘Resurrection after civil war and genocide: growth, poverty and inequality in post-conflict Rwanda’, European Journal of Development Research17, 3 (2005), pp. 495–508.
F. Reyntjens, ‘Post-1994 politics in Rwanda: problematising “liberation” and “democratisation”‘, Third World Quarterly27, 6 (2006), pp. 1103–17; F. Reyntjens, ‘Rwanda, ten years on: from genocide to dictatorship’, African Affairs103, 411 (2004), pp. 177–210.
Ingelaere argues that the gacaca courts cannot be considered a straightforward success due to the fact that, among other things, they suffer from ‘a too extensive social and legal engineering campaign’. B. Ingelaere, ‘The gacaca courts in Rwanda’ in L. Huyse and M. Salter (eds), Traditional Justice and Reconciliation after Violent Conflict: Learning from African experiences (International IDEA, Stockholm, 2008), p. 57; L. Waldorf, ‘Mass justice for mass atrocity: rethinking local justice as transitional justice’, Temple Law Review79, 1 (2006), pp. 1–87.
Government of Rwanda (GoR), ‘Economic Development and Poverty Reduction Strategy 2008–2012’ (Ministry of Finance and Economic Planning, Kigali, preliminary draft, June 2007); GoR, ‘Poverty Reduction Strategy Paper’ (National Poverty Reduction Programme, Ministry of Finance and Economic Planning, Kigali, 2002).
GoR, ‘Preliminary Poverty Update Report’ (National Institute of Statistics, Rwanda, Kigali, 2006), p. 7. While Rwandan policy makers recognize the poverty problem and the fact that poverty in absolute terms did not decline, they are extremely sensitive to criticism. A recent UNDP report expressed strong concerns about Rwanda's limited achievements in terms of poverty reduction. This report was signed by Finance Minister Musoni, who was also the chairman of the steering committee overseeing the report's formulation. However, two weeks after the launch of the report, Musoni expressed public regret that he had not read the final edited version of the report, accusing the Swedish editor of adding unfounded and misleading interpretations of the figures. He stated that the government had blacklisted the editor and a Rwandan researcher involved in the report's formulation. This incident indicates the sensitiveness of the Rwandan government to criticism on poverty and inequality-related policy performance. See UNDP, ‘Turning Vision 2020 into reality: from recovery to sustainable human development’ (United Nations Development Programme, Rwanda, Kigali, 2007) and ‘I didn't read the UN report before launch – Musoni’, New Times, Kigali, 24 August 2007.
See GoR, ‘Poverty Reduction Strategy Paper’; GoR, ‘Economic Development and Poverty Reduction Strategy’; GoR, ‘Vision 2020’, (Ministry of Finance and Economic Planning, Kigali, 2000).
R. Pound, Social Control through Law (Archon, Hamden, CT, 1968).
D. Ellerman, Helping People Help Themselves: From the World Bank to an alternative philosophy of development assistance (University of Michigan Press, Michigan, MI, 2005), p. 9. Ellerman pleads for an alternative, decentralized social learning model in which local experiments and self-learning can flourish and peer-to-peer exchange through local networks is encouraged.
F. Reyntjens, ‘Chronique politique du Rwanda 2005–2007’ in S. Marysse et al. (eds), L’Afrique des Grands Lacs Annuaire 2006–2007 (L’Harmattan, Paris, 2007), pp. 1–19.
B. Ingelaere, ‘Political transition(s) and transitional justice – case study on Rwanda: a view from below’ (unpublished manuscript, 2006). This study, based on in-depth research on a particular hill, provides a detailed illustration of how the engineering ambition is instrumentalized in actual policy.
The majority of interviewees (17) worked for the Rwandan government. Interviews with donors (4) and civil society representatives (5) reiterate some points made in the article, but cannot be considered representative of the entire donor community and civil society. In addition to the interviews, the author conducted research in six different rural localities in the Southern Province on local people's perceptions of their own poverty. Although the data from these qualitative focus group interviews (at least 12 per setting) is not core information in this article, it is sometimes referred to when illustrating particular points.
A. Bebbington, with L. Dharmawan, E. Farmi, and S. Guggenheim (2006) ‘Local capacity, village governance, and the political economy of rural development in Indonesia’, World Development34, 11 (2006), p. 1963.
World Bank, Agriculture for Development (World Bank, Washington, DC, 2007).
D. Newbury and C. Newbury, ‘Bringing the peasants back in: agrarian themes in the construction and corrosion of statist historiography in Rwanda’, American Historical Review105, 3 (2001), pp. 832–77.
For an overview, see D. de Lame, A Hill among a Thousand: Transformations and ruptures in rural Rwanda (University of Wisconsin Press, Wisconsin, 2003); Newbury and Newbury, ‘Bringing the peasants back in’; J. Vansina, Le Rwanda Ancien: Le royaume nyiginya (Karthala, Paris, 2001).
D. Newbury, ‘Returning refugees: four historical patterns of “coming home” to Rwanda’, Society for Comparative Study of Society and History47, 2 (2005), pp. 252–85.
de Lame, A Hill among a Thousand.
Ibid.; Newbury and Newbury, ‘Bringing the peasants back in’.
On the conceptualization of the meaning of ‘coming home’ for Tutsi refugees, see D. Newbury, ‘Returning refugees’.
Interview, representative of a European donor country, May 2007, Kigali.
Interview, civil society person working in the domain of human rights, May 2007, Kigali.
Interview, representative of a European donor country, May 2007, Kigali.
P. Verwimp, ‘Peasant ideology and genocide in Rwanda under Habyarimana’ in P. Verwimp, Development and Genocide in Rwanda: Political economy analysis of peasants and power under the Habyarimana regime (KU Leuven, unpublished PhD thesis, 2003).
Ibid., p. 16.
GoR, ‘Vision 2020’, p. 17.
GoR, ‘Strategic Plan for Agricultural Transformation in Rwanda’ (Ministry of Agriculture and Animal Resources, Kigali, 2004), p. 33.
GoR, ‘National Land Policy’ (Ministry of Lands Environment, Forests, Water, and Mines, Kigali, 2004), p. 16.
A. Des Forges, ‘Land in Rwanda: winnowing out the chaff ’ in F. Reyntjens and S. Marysse (eds), L’Afrique des Grands Lacs: Annuaire 2005–2006: Dix ans de transitions conflictuelles (L’Harmattan, Paris, 2006).
‘Address to the Nation by H. E. Paul Kagame on his inauguration as President of the Republic of Rwanda’, 22 April 2000, <www.gov.rw/government/president/speeches> (6 October 2008).
Interview, high-ranked Rwandan government official from the Ministry of Finance and Economic Planning, June 2007, Kigali.
Interview, Southern Province district official, June 2007, Gitarama.
GoR, ‘Strategic Plan for Agricultural Transformation’, pp. 6–17.
P. Verwimp, ‘Agricultural policy, crop failure and the “Ruriganiza” famine (1989) in southern Rwanda: a prelude to genocide?’ (discussion paper, June 2002, Center for Economic Studies, Leuven, 2002).
J. Pottier, ‘Intolerable environments: towards a cultural reading of agrarian practice and policy in Rwanda’ in J. Parkin and L. Croll (eds), Bush Base: Forest farm (Routledge, London, 1992), p. 151.
Newbury and Newbury, ‘Bringing the peasants back in’, p. 856.
UNDP, ‘Turning Vision 2020 into reality’.
See M. Mamdani, Citizen and Subject: Contemporary Africa and the legacy of late colonialism (Princeton University Press, Princeton, NJ, 1996).
GoR, ‘Strategic Plan for Agricultural Transformation’, p. 10.
Interview, representative of a European donor country, May 2007, Kigali.
A. Ansoms, ‘Striving for growth, bypassing the poor? A critical review of Rwanda's rural sector policies’, Journal of Modern African Studies46, 1 (2008), pp. 1–32.
For more general literature on the topic, see F. Ellis, Peasant Economics: Farm households in agrarian development (Cambridge, Cambridge University Press, 1988).
Interview, civil society person working in the domain of human rights, May 2007, Kigali.
Interview, high-ranked Rwandan government official from the Ministry of Land, Environment, Forests, Water, and Mines, July 2007, Kigali.
Interview, high-ranked Rwandan government official from the Ministry of Agriculture and Animal Resources, June 2007, Kigali.
World Bank, Agriculture for Development.
Interview, employee of an international donor organization, July 2007, Kigali.
Interview, Rwandan government official from the Ministry of Finance and Economic Planning, June 2007, Kigali.
GoR, ‘Poverty Reduction Strategy Paper’.
Interview, high-ranked Rwandan government official from the Ministry of Agriculture and Animal Resources, June 2007, Kigali.
Interview, civil society person working in the domain of human rights, May 2007, Kigali.
Interview, Rwandan consultant working for the Ministry of Agriculture and Animal Resources, July 2007, Kigali.
Ansoms, ‘Striving for growth, bypassing the poor?’
B. Ingelaere, ‘Living the transition: a bottom-up perspective on Rwanda's political transition’ (Discussion Paper 2007.06, Institute of Development Policy and Management, Antwerp, 2007). Ingelaere analyses the upgrade of rural life and the mechanisms to achieve this (performance contracts and fines) through a detailed description of one hill in central Rwanda. Ingelaere, Political Transition(s) and Transitional Justice, pp. 29–91. These findings informed the fieldwork undertaken by the author in rural Rwanda in 2007.
Human Rights Watch, ‘Uprooting the rural poor in Rwanda’ (Human Rights Watch, New York, NY, 2001).
See also J. Pottier, Re-Imagining Rwanda: Conflict, survival and disinformation in the late twentieth century (Cambridge University Press, Cambridge, 2002); S. van Hoyweghen, ‘The urgency of land and agrarian reform in Rwanda’, African Affairs98, 392 (1999), pp. 353–72.
Interview, high-ranked Rwandan government official from the Ministry of Land, Environment, Forests, Water, and Mines, July 2007, Kigali.
Interview, Rwandan government official from the Ministry of Finance and Economic Planning, June 2007, Kigali.
District of Kamonyi, ‘Plan de développement de district 2008–2012, draft, July 2007’ (unpublished document).
A. Ansoms, ‘Rural poverty in Rwanda: views from below’, reworked version of paper presented at the African Studies Association 50th Annual Meeting ‘21st Century Africa: Evolving Conceptions of Human Rights’ (New York, 2007).
A. B. Twizeyimana, ‘Rwanda: La propreté à marché forcée’, Infosud Belgique 29 June 2006, <www.infosud-belgique.info/article.php3?id_article=258> (6 October 2008).
Interview, representative of an international donor organization, July 2007, Kigali.
Reyntjens, ‘Chronique politique du Rwanda 2005–2007’.
M. F. Cros, ‘Rwanda. Développement: de gré ou de force?’, La Libre Belgique, 5 October 2006.
‘No ban on sweet potatoes’, New Times, Kigali, 26 January 2007.
Interview, Southern Province district official, May 2007, Gitarama.
See, for example, H. Ruthenberg, Farming Systems in the Tropics (Clarendon Press, Oxford, 1971); C. C. Webster and P. N. Wilson, Agriculture in the Tropics (Longman, London, 1966).
J. Pottier, ‘Intolerable environments’; Newbury and Newbury, ‘Bringing the peasants back in’.
Interview, high-ranked Rwandan government official from the Ministry of Finance and Economic Planning, June 2007, Kigali.
For the ‘traditional’ and ‘new’ interpretation of the imihigo concept, see B. Ingelaere, ‘Living the transition’.
Interview, Rwandan government official from the Ministry of Local Government, May 2007, Kigali.
On the topic of local governance and (the chain of) accountability, including a schematic representation of the local governance structure, see Ingelaere, ‘Living the transition’.
Interview, civil society person working in the domain of human rights, May 2007, Kigali.
Interview, Southern Province district official, May 2007, Gitarama.
Interview, high-ranked Rwandan government official from the Ministry of Finance and Economic Planning, May 2007, Kigali.
Interview, foreign consultant working for the Ministry of Land, Environment, Forests, Water, and Mines, May 2007, Kigali.
‘Performance contracts to be signed at household level’, New Times, Kigali, 19 November 2007.