Rwandan developmental ‘social engineering’: What does it imply and how is it displayed?

Abstract

Rwandan development policy is frequently referred to as ‘social engineering’. As this concept is not analytically clear and carries negative connotations, this article elaborates on its meaning. An analytical framework is developed and applied to selected Rwandan policies and programmes, as depicted in government documents, academic writing and field work. The analysis points to far-reaching social engineering in terms of visionary design, depoliticized managerialism, firm and partly coercive implementation, state–population separation and paternalism. In the conclusion, impressive results as well as problematic aspects are discussed, as are the negative connotations and the legitimate use of social engineering.

Keywords

Development development policy developmental engineering rural development Rwanda social engineering

I Introduction

Rwanda is a noteworthy contemporary case of a country that, despite poverty and aid dependence, is determined to set its own development agenda (Hayman, 2008). After the 1994 genocide and some years of reconstruction, the government has since around 2000 persistently driven an ambitious, radical and firmly implemented development policy, throughout demonstrating a strong preference for national rather than international solutions (GoR, 2010: 30; Minocefin, 2000: 9). This autonomous and determined approach has attracted due attention and evoked divergent reactions among international scholars (Booth and Golooba-Mutebi, 2012: 384; Jones, 2012: 228).

With high growth rates, Rwanda is acclaimed for doing better than countries with more favourable conditions. Life expectancy increases, school enrolment is high, and not least strong progress on child and maternal mortality rates and low levels of corruption have impressed (Booth and Golooba-Mutebi, 2012: 385, 392; Crisafulli and Redmond, 2012; Minocefin, 2013: 7; UNDP, 2013; World Bank, 2013: 2). International donors, too, are largely favourable. Rwanda has been commended for the use and accounting of aid resources, and for the results achieved (Dfid, 2013: 3), and in 2011 was one out of two countries receiving the top score on the operationality of its development strategy (OECD, 2011: 30f, 2012: 4).

Rwanda is also recognized as a potentially emerging ‘developmental state’ (Routley, 2012: 42, 45; cf. Straus and Waldorf, 2011: 10), associated with state-led developmentalism in the East Asian ‘tiger economies’. According to Johnson (1982: 302ff) who coined the concept, the developmental state’s main characteristic is its ability to mobilize the nation around economic development. With power concentrated to a small political leadership, an efficient state bureaucracy, and close relations to business and financial interests, the state can effectively manage non-state economic interests (Evans, 1989; Leftwich, 1995). The Rwandan government is clearly inspired by and aspires to follow this example. Its ambitious ‘Vision 2020’ concludes that ‘The development experience of the East Asian “Tigers” proves that this dream could be reality’ (Minocefin, 2000: 25). The economic transformation currently aimed at is exemplified by ‘the fast growing East Asian economies’, and the explicit intent is to continue deepening the economic ties with this region (Minocefin, 2013: 16ff). State officials also visit the region and are trained on East Asian development, while ministries have been re-modelled and re-named following the example (Kimberley Brandt, 2009: 36ff).

Alongside positive accounts, however, the Rwandan approach is criticized for being authoritarian, repressive and exclusionary (Desrosiers and Thomson, 2011: 436; Straus and Waldorf, 2011: 10ff), with Reyntjens (2004, 2011: 2) claiming that there is consensus among international scholars that ‘Rwanda is run by a dictatorship’. There is also concern that Rwandan policy contributes to worsening rural living conditions and exacerbating a rural–urban divide (Ansoms, 2009, 2013; Desrosiers and Thomson, 2011; Hintjens, 2013; Pritchard, 2013; Purdeková, 2012). While urbanization is consciously promoted and facilitated, 85 per cent of the population still live in rural areas and income inequality is very high, with a Gini coefficient of 53.1 and a rural poverty rate of 48.7 per cent, as compared to 22.1 per cent in urban areas (Minocefin, 2013: xi, 8, 31f, 37; UNDP, 2007: 17f, 2013). Rural–urban inequality is only slowly decreasing (World Bank, 2013: 36f) and Ansoms (2009: 299) finds policy making anti-rural, while Desrosiers and Thomson (2011: 438) assert that structural violence continues to widen the socio-economic chasm, and Hintjens (2013: 82) warns for polarization between the haves and have-nots.

Among critical scholars, Rwandan policy and practices are frequently referred to as ‘social engineering’. According to Ingelaere (2009: 454, 2011: 68) this has become ‘the modus operandi in state-society relations’, with the local state being restructured as part of its ‘larger social engineering campaign’. Ansoms (2009: 299, 2011: 240) finds ‘social engineering ambitions’ particularly in the rural sector, while Thomson (2011: 333, 2012: 96) interprets the policy of national unity and reconciliation as an ‘ambitious social engineering project’, and Reyntjens (2004: 196) talks about villagization as an ‘ambitious security-driven form of social engineering’. Although, as Verwimp (2011) illustrates, ‘social engineering’ is not limited to the post-genocide period, it is considered to have been intensified and taken further (Ansoms, 2009: 299; Newbury, 2011: 226) which the rich collection of critical analyses provided by Straus and Waldorf (2011) illustrates. In the volume, the two editors as well as Newbury refer to Scott’s elaboration of high modernism, and find that all its aspects are found in Rwanda, including the belief in the need for social engineering (Newbury, 2011: 229; Straus and Waldorf, 2011: 10ff).

While social engineering is not an analytically clear concept, it has strongly negative connotations. Scott, while illustrating a range of disastrous effects, does point out that social engineering practices have been widely used in the West as well as in international development cooperation, and had much positive impact on large populations, for example within public health, education, urban planning and transportation/ communication systems (Scott, 1998: 60ff, 96f). Associated with authoritarian modernism, central planning, manipulation and contrived collectivities, however, ‘social engineering’ has always evoked strong criticism (e.g., Hayek, 1944; Popper, 1957)¹ and tends to induce discomforting connotations of ‘Orwellianism’ and totalitarianism (Alexander and Schmidt, 1996: 5f). The referring to Rwandan policies as social engineering is also commonly made in connection to strong criticism, such as, when Straus and Waldorf (2011: 15) maintain that donors, through passivity, enable ‘RPF’s repression, exclusion and social engineering’. Meanwhile, Booth and Golooba-Mutebi (2012: 380), who are positive about the Rwandan government’s approach, argue that the concept of ‘high modernism’ is adapted to supporting a critical view in the analysis of country policies.

Considering the frequent use, the lack of analytical clarity, and the negative connotations of ‘social engineering’, this article aims to investigate what it implies and how it is displayed in Rwandan development policy and practice. It will do so by delineating an analytical framework and applying it on selected policies and programmes. After this introduction, the main approaches and assumptions of social engineering are outlined. The framework thus formulated is an elaboration of one previously used when analyzing international development cooperation in a different part of the world (Hasselskog, 2009). This is followed, first, by a presentation of selected Rwandan development policies and programmes and, then, by the analysis. In concluding, along with impressive results, some problematic aspects of the developmental social engineering found are discussed, as are the negative connotations and the legitimate use of social engineering.

II Social engineering

Though there is no clear definition of the concept, the approaches and underlying assumptions of social engineering, and of the closely related high modernism, have been scholarly explored, which is elaborated in the following.

1 Transformation based on vision and design

As Fein (2001) notes, social engineering commonly refers to large-scale efforts to reshape social structures with the aim of reordering society in desirable directions, based on an assumed ability to correctly understand a problem, and to implement appropriate solutions. Scott (1998) eloquently illustrates how social engineering/high modernist practice aspires at complete transformation, emanating from a vision of the future. This ambition is based on firm beliefs in the power of planning and in the possibility, desirability and rightfulness of intervening in order to change or replace prevailing features and thereby create a certain societal change (Hasselskog, 2009). An assumption is thus that ‘man’ could and should, according to his desires and through the design and implementation of grand schemes, shape the future society (Ikeotuonye, 2002: 71f).

2 Experts rather than politics

The aspiration of man-made transformation links social engineering to the idea of development as ‘intentional’ rather than ‘immanent’ (Cowen and Shenton, 1996: 117f). The notion, as elaborated by Ikeotuonye (2002: 72), is that of a controllable linear universe, in which intentional development resolves problems caused by arbitrary events and laissez-faire. This reflects a non-political view of societal development, grounded in utilitarian logics and optimistic beliefs in science and technology. There is an assumption of rational solutions that will create progress and order. Such logics result in a technical approach, with a key role assigned to ‘experts’ – rather than politicians – such as, technicians, planners and bureaucratic intelligentsia. After their diagnosis and prescription follows intervention, which is expected to set in motion a predictable step-by-step procedure with controllable outcomes (Scott, 1998: 89f).

Social engineering is thus a matter of depoliticized managerialism. As discussed by Kantola and Seeck (2011), there is a trend of managing countries in technocratic, functional ways, which includes the notion of a pool of resources that can be rationally engineered, focusing on end results and striving for national competiveness. Stipulated goals, expert solutions and management schemes are presented as objective, in need of no democratic process. Rather, governing elites display their management style to legitimize themselves, demonstrating that their exercise of political power is functional and competitive. Similar ideals are illustrated by Crisafulli and Redmond (2012: 4), stating that their book (on the Rwandan model) is not a political one, but one on economic development, governance and leadership. Accordingly, the sphere of politics is narrowed, political differences downplayed and scientific or economic rationality come to replace political arguments (Kantola and Seeck, 2011; Scott, 1998: 13f, 51; cf. Dar and Cooke, 2008).

3 Coercion rather than choice

The rationalistic conception comprises the population, and the assumed existence of scientifically correct solutions makes human choice unwarranted. Alexander and Schmidt (1996: 1, 15) argue that social engineering is associated with a perception of human beings as raw material, dehumanized and possible to manipulate, and Scott (1998: 346) illustrates how people are treated as bricks to be used for end means and expected to behave in predictable ways. With such a rationalistic view also of human resources, everything is assumed possible to design, not only a city or a scientific forest, but also a revolution, a shift of livelihood systems and human thinking and behaviour. Dehumanization, denunciation of democratic decision-making, and the orientation towards end results may also lead to the use of coercive means. Firm implementation of correct solutions, including or not physical violence, may be regarded as necessary in order to reach desired outcomes and, according to Alexander and Schmidt (1996: 2), social engineers have often resorted to violent measures in order to break the will of those whom they wanted to change.

4 External legibility assumed and created

Scott argues that large-scale visioning and rational planning build on a separation between planners and what is to be transformed. Diagnoses and prescriptions are made by distant experts with only scarce contact with the area, which must therefore be legible from outside. Schematic knowledge, such as mapping and statistics, creates an overview but necessarily implies simplification, since only those factors are included that are observable and appear relevant to external experts (Scott, 1998: 11ff, 76, 219). As Li (2007: 270) elaborates, by articulating resources and factors in a seemingly clear way, a technical domain is constructed of what is possible to know at a distance, and accordingly to manage and improve. Scott (1998: 187), meanwhile, analyzes the creation of ‘state spaces’ where society and economy are reconfigured.

Separation and simplification result in standardized solutions being implemented across local settings and ensuing reduction of local variation. Accordingly, in addition to social engineering being based on assumed legibility, its practice tends to create increasingly uniform terrains. Through rational design, something is gradually moulded that can be centrally monitored, facilitating further transformation. The extreme of this is ‘miniaturisation’, that is, the creation of micro-orders such as model cities, villages and farms (Scott, 1998: 24, 81f, 257ff).

5 Central control, moral authority and paternalism

External visioning and planning indicate inherent power relations. Everett (2006: 5, 19) finds that social engineering refers to extremes of hierarchical leadership and one-way influence, denoting that a small number of individuals seek to influence large numbers of people.

Such top-down command is commonly exerted with good intention. As Alexander and Schmidt (1996: 3) point out, the presumed ability to define problems and prescribe solutions includes the ability to tell good from bad. Social engineering is based on assumed moral authority, implying that the engineers should be trusted to know what is best for society, linking it to the notion of trusteeship (Ikeotuonye, 2002: 73). The assumption of own abilities is combined with scepticism about local skills and practices, indicating a patronizing attitude. Benevolence, assumed moral authority, and low regard of the population give rise to a paternalistic approach ingrained in social engineering. The assumption is that leaders and experts have, not only knowledge of what is needed and how to attain it, but also an obligation to do so, which may require guiding as well as pressuring of the population (Kantola and Seeck, 2011: 32; Scott, 1998: 96, 234).

The aspects of social engineering delineated above provide an analytical frame that will be applied to selected Rwandan development policies and programmes, which will accordingly be explored with regards to approaches and assumptions related to (a) transformation based on vision and design, (b) experts rather than politics, (c) coercion rather than choice, (d) external legibility assumed and created and (e) central control, moral authority and paternalism.

III Rwandan development policy

The analysis of Rwandan development policy and practice builds on government material, scholarly writing and field work. Policies and programme information are largely available on ministries’ home pages, and activities are recounted by government-friendly newspapers accessible on the web. Scholarly texts provide different perspectives on proceedings and practices and, though not a main source, reference is also made to field work in Southern Rwanda. During July 2011, January 2012 and July 2013 interviews, in most cases repeated over the rounds of research, were conducted with 14 officials in one district, two sectors, two cells and two local communities and with 70 citizens in the two communities.

Before the analysis, the policies and programmes that have been studied will be presented. They were selected due to their centrality, development relevance, and concern for those involved. The document ‘Vision 2020’ (Minocefin, 2000) provides the frame for Rwandan development planning, confirmed and partly updated in the ‘Government Programme 2010–2017’ (GoR, 2010), and complemented by three poverty reduction strategies (Minocefin, 2002, 2007, 2013). With 85 per cent of the population living in rural areas, and most poverty found there, development policy is largely concerned with agriculture and rural welfare and the article, therefore, focuses on these policy areas. A strategic plan for agricultural transformation has been formulated (Minagri, 2004, 2009) as well as a policy on human settlement (Minifra, 2009). Also included in the analysis are a number of more specific programmes intended to improve rural well-being, which during field work were found to be of much concern to local officials and residents.

1 Overall ambitions and approach

‘Vision 2020’ outlines the government’s main development ambitions. The overarching goal is to ‘transform Rwanda to a middle-income country’, also turning it into a ‘sophisticated knowledge-based society’ and away from aid reliance (Minocefin, 2000: 9). ‘Vision 2020’ is a well-utilized document, frequently referred to in other government documents and forming the basis for an elaborate planning framework, including sector policies, strategies and time-bound plans, along with concrete programmes.

In addition to intricate planning, Rwanda is known for effective implementation (Beswick, 2011: 1929), largely ensured through a hierarchy among local officials, with the appointed and most influential ones responsible for executing national policies (Ingelaere, 2010: 288; Silva-Leander, 2008). Execution is, since 2006, further safe-guarded through a rigorous system of performance contracts, signed between the president and local authorities. These contracts, stipulating detailed objectives and measurable indicators, are closely monitored and, depending on their performance, appointed officials may be promoted, relegated or dismissed (GoR, 2008: 59; Minocefin, 2007: 107f; RGB, 2013a; Versailles, 2012).

2 Rural programmes

The government vision entails that subsistence farming is replaced by a productive, commercial agriculture sector, envisaged to provide an initial growth engine, later to be replaced by the industry and service sectors (Minocefin, 2000: 23, 2002: 30, 36f, 2013: 43f). This transformation is to be achieved through modernization, professionalization and private sector involvement. Land consolidation, that is, rearrangement of smaller land parcels into larger holdings, is intended to promote market oriented production. Meanwhile, stipulating that neighbouring farmers grow the same crop, and focusing on priority crops, are to help replace customary multi-cropping, and enhanced use of modern inputs, along with mechanization and irrigation, are to increase yields without extending the cultivated area (Minagri, 2004, 2009, 2011: 13ff; 2012; cf. Ansoms and Rostagno, 2012: 435f, 441; Musahara and Huggins, 2005; Pritchard, 2013).

Land consolidation also requires resettlement in congregated villages. The idea, which goes back to the 1993 Arusha Accords, was initially to provide emergency housing for returning refugees. With the Human settlement policy in 1996, however, the aim was to have all rural inhabitants settled in designated settlements, as compared to the traditional pattern of scattered dwellings. This is expected to bring about more rational land use, stimulate market development and off-farm income generation, facilitate service delivery and promote security, as well as foster reconciliation (Minagri, 2011: 15; Minocefin, 2002: 53, 2013: 41; Minifra, 2009: 26f; cf. Newbury, 2011).

Related to the settlement policy is the bye bye nyakatsi programme, which started in 2008 and implies the eradication of grass thatched houses. By late 2009 almost 125.000 grass thatched houses had been identified, families categorized according to the kind of support needed, and the programme accelerated, with the Rwanda Defence Forces and National Police involved in removing remaining houses (Minaloc, 2010; The Rwanda Focus, 2011). Construction of new houses is partly organized through community work, umuganda, which also commonly implies cleaning and maintaining community infrastructure, and which is followed by meetings led by local leaders (Minaloc, 2008: 13; RGB, 2013b).

3 Endowments and contracts

Living standards are also to be improved through a range of endowments to selected households, with selection partly based on a general socio-economic classification. Since 2011 every household is placed in one out of six socio-economic categories, and the data entered into a data base to be annually updated (Minocefin, 2001: 4, 23f; 2002: 14f; RLDSF, 2011, 2012: 52f, 75). In addition, there are special selection criteria for particular provisions.

Most well known among the endowment programmes is the ‘one cow per poor family’ (girinka), introduced in 2006. Poor people may receive a cow if they fulfil certain criteria such as practicing good farming activities and having planted animal fodder and constructed a shed (Minagri, 2013a). A more comprehensive social support programme, ‘Vision 2020 Umurenge’,² started in 2008 and has gradually expanded. Depending on people’s socio-economic status, they may qualify for one of the programme’s components of public work, micro-credits and direct cash support (VUP, 2011). Another main subsidy to selected households is free inclusion in the national health insurance. Inclusion is compulsory³ and the fee, which varies with socio-economic category, is for the poorest two categories to be settled by the government (MoH, 2010).

Moreover, people who are defined as genocide survivors may receive government support, mainly for education, but also housing and cows for the most vulnerable (Farg, 2011). There is special support for demobilized soldiers too. After being medically screened, severely disabled ones may be selected to receive housing – what type depends on the extent of disabilities – while others may qualify for vocational training (RDRC, 2011).

Alongside such provisions, household performance contracts are expected to increase rural welfare. Having been tried in some areas, the practice was comprehensively launched in 2012. By signing a contract, a household commits to, within a specific timeframe, achieving particular targets, such as having a vegetable garden, rearing domestic animals, eliminating grass thatched houses, owning a radio, subscribing to the health insurance and practicing family planning. Performance is to be monitored by local leaders on a monthly basis (The New Times, 2012a).

A system of fines is also intended to make people adopt stipulated practices, with fines for example for not participating in community work or delivering at a health clinic, or for allowing cattle to graze (IRIN, 2009; ODI, 2012; The Independent, 2011; cf. Ansoms and Rostagno, 2012: 74; Ingelaere, 2011: 74).

Rwandan development policy thus embraces clearly articulated ambitions, comprehensive programmes, and various endowments and incentives, intended to improve living standards. While the approach has been commanded as elaborate and functional, it has also been condemned as authoritarian and referred to as social engineering. Next, therefore, the approaches and assumptions of selected policies and programmes will be analyzed in relation to those of social engineering.

IV Analysis: Rwandan developmental engineering

In this section, the analytical scheme outlined earlier will be applied. Thus, in order to trace aspects of social engineering, policies and programmes will be analysed with regards to (a) transformation based on vision and design, (b) experts rather than politics, (c) coercion rather than choice, (d) external legibility assumed and created and (e) central control, moral authority and paternalism.

1 Transformation based on vision and design

The key document, ‘Vision 2020’, explicitly emanates from the perceived necessity to ‘clearly define the future of the country’ (Minocefin, 2000: 3) and outlines a forthcoming society very different from the current one. Envisioning rapid and radical change from a subsistence economy to a knowledge based society, the document sets a development agenda that reaches far beyond reconstruction. Rwandan development policy and practice are thus guided by a clearly articulated vision, implying thorough transformation. The envisioned shift is also represented by a new flag, national anthem, motto, seal and language of instruction (Kimberley Brandt, 2009: 32; Straus and Waldorf, 2011: 9).

The agricultural transformation, to be achieved through large-scale schemes of land consolidation, regional specialization and crop intensification, entails thorough rearrangement of land holdings, substitution of crops and methods, and replacement of subsistence livelihood by commercial production and market reliance. Meanwhile, the human settlement scheme radically alters settlement patterns, and with them social relations and identities. These programmes thus imply thorough restructuring of the agricultural sector, rural areas and social life, confirming the aspiration to design and reshape.

Stipulated targets are ambitious, such as not relying on foreign aid, 70 per cent of land being ‘consolidated’, all of the rural population living in grouped settlements, and 95 per cent being included in the health insurance (GoR, 2010: 30f, 40, 52). As noted, determined efforts are also made to reach the targets. Repeatedly announcing the grand goal of ‘shifting Rwanda from a poor country to a middle-income country’ (GoR, 2010: 10), and tirelessly employing resolute approaches – anticipating to ‘improve, strengthen, develop and ensure quick implementation of programmes to multiply achievements made so far’ (GoR, 2010: 9) – the Rwandan government demonstrates commitment and confidence, displaying a clear break from previous regimes’ policies as well as independence from donors. The articulation of a far-reaching development vision and a striking development agenda also clearly reflects the assumption that ‘man’ can and should shape the future according to his desires. The formulation and implementation of radically remodelling schemes are ostensibly based on a firm belief in ‘the power of planning’ and the possibility, desirability and rightfulness of intervening in order to achieve a certain kind of societal change. This can be asserted, since without those assumptions, it would be difficult to motivate such practices. It can thus be concluded that ‘Vision 2020’ and a number of large-scale rural programmes comprise approaches and assumptions that reflect social engineering ambitions to transform society, based on vision and design.

2 Experts rather than politics

Overall vision and design are followed by an intricate array of meticulously formulated sector policies, strategy papers and plans, guided by several sets of quantified targets. In the government programme (GoR, 2010: 8), for example, reference is made not only to ‘Vision 2020’ and national EDPRS targets, but also to the Millennium Development Goals. Local performance contracts further specify what goals to be achieved and what indicators to assess, which along with close monitoring endorses execution. Such detailed planning and such firm and target-driven implementation indicate an assumption that societal development can be closely steered by human intent. Various experts are assigned to delineate and resolve problems, and their ostensibly correct solutions of, for example, intensification, commercialization and land reallocation are firmly employed, with certain predicted production and livelihood systems expected to emerge. This reveals the assumption that, based on expertise, a situation can be assessed and an intervention crafted, which will work as anticipated and achieve predicted outcomes in a controllable manner.

The diverse means of rural programmes, such as consolidating plots, concentrating on selected crops, specializing in what a region is most suited for and intensifying production by increased use of modern inputs, also all indicate the ambition to make more efficient use of limited natural resources. According to ‘Vision 2020’, ‘Rwanda’s land resources are utilized in an inefficient /…/ manner’ (Minocefin, 2000: 16), and the human settlement policy explicitly aims at ‘complete control’, ‘rational restructuring’ and ‘more rational use’ of scarce land resources (Mininfra, 2009: 17, 27). The intent of determining the most productive use indicates, apart from an emphasis on rationality and efficiency, the perception of a pool of resources to be elaborated with.

Letting experts conclude how to allocate land and organize agriculture, what crops to grow, by what methods, and where to live, also indicates depoliticization. With no recognition of possible political differences, major development concerns are reduced to technocratic management tasks, and devised expert solutions treated as objectively rational, in no need of a democratic process. This notion is illustrated by the comparison of the Rwandan president to a corporate ‘Chief Executive Officer’ (Crisafulli and Redmond, 2012: 92). With goals and targets stipulated in policies, plans and performance contracts, focus is on end-results, while the practice of performance contracts, aimed at ‘making the public agencies more effective’, also serves to encourage competitiveness among local officials (RGB, 2013a; The Independent, 2012). Their strict focus on set goals and the widely announced evaluations with ensuing rewards and retributions, clearly illustrate a results-based, managerialist approach.

It can thus be concluded that detailed planning and the comprehensive system of performance contracts, based on experts’ rational resolution and aimed at efficient use of resources, demonstrate a depoliticized, results-based management approach, reflecting social engineering assumptions of societal development as intentional and technical.

3 Coercion rather than choice

The intention to resettle people in designated villages reflects expert conclusions of how to most efficiently use, not only land but also labour resources, in addition to facilitating rational service provision. ‘Plentiful supply of cheap labour’ is stated to be one of Rwanda’s comparative advantages, and an aim is to turn the population into a ‘productive and efficient workforce’ (Minocefin, 2000: 10, 13). Compulsory community work also demonstrates the authorities’ command over the population, and underscores the ambition to make the most out of it (Purdeková, 2012: 201). Meanwhile, the meetings held in connection to such work, and other forms of civic education, aim at sensitising the population. The need for a ‘change of mindset’ is frequently mentioned in government documents (Minagri, 2013b; Minaloc, 2008: 19) and, as noted, an explicit aim of household performance contracts is to instil a culture of competitiveness and discipline (The New Times, 2012a). Everybody from the age of seven is to be educated on values of ‘hard work’ and ‘good behaviour’, youth are to be ‘sensitized’ and helped to ‘change their behaviours’, media is to disseminate information promoting a certain mindset, a productive middle class is to be created, and Rwandans to ‘develop into learned and competent entrepreneurs’ (GoR, 2010: 13, 19, 21, 50; Minocefin, 2000: 11). The rationalistic conception thus overtly includes the population, which is regarded as part of available resources to be transformed and efficiently used, and the practices fit with Purdeková’s (2012) analysis of how suitable development subjects are to be created.

Competition and discipline are also related to programmes being vigorously executed, stipulated practices made mandatory and non-compliance corrected through penalties. Once correct solutions have been determined, ensuring efficient employment is apparently deemed more important than allowing citizens a democratic choice. As noted, the police and the military have been involved in tearing down grass-thatched houses, and fines are used to reinforce policies of for example zero-grazing, community work, and giving birth at a health clinic. There are also reports of other forms of coercion, with non-prescribed crops being uprooted and farmers punished for growing traditional crops along with the stipulated ones, compelled to purchase fertilizers, arrested for selling or applying it to non-commercial crops, and threatened with land confiscation and eviction unless they invest in certain techniques (e.g., Huggins, 2009: 299f). Such coercive practices may not always be sanctioned from the top, but may be caused by competition among local officials and fear for the consequences if targets are not met, which are in turn consequences of disciplinary managerialism and strict result orientation.

It can thus be concluded that the population is regarded and treated as a resource to be rationally used and transformed, which includes coercively making people adopt stipulated practices.

4 External legibility assumed and created

As noted, the local officials in Rwanda responsible for executing policies may, depending on performance, be promoted or relegated. This implies frequent transfers between geographical areas and, accordingly, the most influential local officials do commonly not come from the area where they rule.⁴ This notion of a state–population separation is reinforced by Jones’ (2012) analysis of the Rwandan government as composed almost entirely of people who returned after the genocide after having spent all their lives outside Rwanda, and who have few established links with the local population and a fundamentally different vision of the state (cf. Ansoms, 2009; Ingelaere, 2010: 288).

Decisions included in large-scale rural programmes, such as what crops to grow, which inputs to use, and where to establish villages, are necessarily based on quite rough classifications and fairly general knowledge, providing the sort of overview that makes planning from a distance possible. Designed solutions of consolidating land plots, increasing the use of modern inputs, concentrating on certain crops, and gathering the population in congregated settlements, are based on what experts define as relevant and reflect external rationalities of for example maximising annual output. Meanwhile, local variations in, for example, soil fertility, erosion and micro-climate, and local rationalities, such as securing access to food throughout the year, evening out labour peaks and protecting against crop failure, are too complex to be considered (e.g., Hintjens, 2013: 96f; van Damme et al., 2013). Standardized systems of agriculture and settlement are thus implemented across diverse local settings of fragmented plots, small-scale multi-cropping and scattered settlement, and since the new systems are more easily monitored and mapped than what is replaced, the terrain and the population become gradually more legible.

As goals and targets stipulated in local authorities’ performance contracts are also necessarily standardized, efforts across regions are concentrated on reaching similar targets. The combination of local authorities striving for standardized goals and being closely monitored results in a clear and continuously updated overview of their performance. Household performance contracts, too, include some few selected and observable indicators, while a range of more complex factors relevant to people’s well-being are excluded. People are thus induced to adopt the same practices in terms of agriculture, housing standards, domestic animals etcetera, which – along with local officials’ regular monitoring – makes the population increasingly legible along those indicators. This notion was confirmed by a District Mayor as the system of household performance contracts was launched: ‘we will follow each and every family on a daily basis’ (The New Times, 2012b). The socio-economic categorization also necessarily builds on limited factors, and defining every household as belonging to one out of six categories, and entering the information into a database, provides the central level with a clear overview of the population.

It can thus be concluded that selected practices reflect aspects of social engineering of separation and distance between planners and what is being planned for. Policies and programmes are based on limited and schematic knowledge, resulting in standardized agriculture and settlement patterns, which, along with performance contracts and categorization, make the terrain and the population increasingly legible.

5 Central control, moral authority and paternalism

Scholars widely agree that contemporary Rwanda is run by a small and resolute group of leaders. The presidency is strong and Paul Kagame’s personal leadership style has been deemed determined ‘to the point of ruthlessness’ (Booth and Golooba-Mutebi, 2011: 15), while Jones (2012) analyzes how policy is driven by a closely linked leadership circle around the president (cf. Beswick, 2011: 1925). The central appointment of local officials responsible for executing development plans ensures that implementation, too, is centrally controlled, and the use of performance contracts reinforces upward accountability by binding local authorities to specific targets in line with national priorities. Although the formal institutions are recent, hierarchical relations of dependence reflect a long history of patronage with patron–client ties having become less personal and affective (Newbury, 1978). The sense of discipline and competition that recurrent evaluation inculcates among local officials⁵ further facilitates central management, and the recent extension of the system of performance contracts to the household level, completes the chain of control and upward accountability.

Policy makers’ development commitment and good intentions are clearly demonstrated in visionary and thoroughly elaborated policy documents. Along with benevolence, however, explicit phrasings as well as the underlying tone reveal a general disdain for the rural population’s practices and abilities. Firm statements are made along the line that ‘the most important issue retarding Rwanda’s agricultural development is /…/ low productivity associated with traditional peasant-based subsistence farming’ and associated practices of farm fragmentation, multi-cropping and diversification (Minagri, 2004: 21, 35, 2009: 46, 2011: 11; Minocefin, 2000: 18; cf. Ansoms, 2011: 242f; Desrosiers and Thomson, 2011; Huggins, 2009: 298). As noted, various policies include the ambition to transform people’s behaviour and mindset. The president in his inaugural speech after re-election in 2010 urged Rwandans to develop a ‘promptness culture’, and the population is to be sensitized on a range of issues, including patriotism, self-reliance, ‘the culture of being quick and do quality work’, the importance of decent housing, and the use of clean water and hygiene, with ‘cleanliness to become a habit’ (GoR, 2010: 8, 19, 51f, 54; The Rwanda Focus, 2011). Implicit notions are for example that people currently do not have appropriate understanding and attitude, do not work hard enough, and do not practice proper hygiene, and the patronizing attitude thus revealed indicates a sense of moral authority.

Meanwhile, detailed stipulations of how to conduct oneself, such as having a latrine, rearing domestic animals and delivering at a health clinic, indicate an assumption that people do not know what is best for them, but need to be told. Interviews with local officials also show that poor people who receive supposedly untied cash support get close ‘guidance’ on how to use it. Making promoted practices mandatory and using penalties for non-compliance, as well as basing the distribution of endowments on who is deemed able and deserving, for example, by ‘practicing good farming activities’ (Minagri, 2013a), also point to a notion that the population needs to be persuaded, through force as well as awards. Centrally determined and firmly implemented efforts to transform citizens thus indicate not only a rationalistic conception of the population and a patronizing attitude, but also a paternalistic sense of mission, implying a perceived responsibility – and rightness – to make people adopt certain behaviour and mindset, which requires and justifies close guidance and firm pressuring.

It can thus be concluded that, in Rwanda, a small leadership group is trying – out of conviction and benevolence – to influence a large population. Practices reflect social engineering tendencies of top-down, centrally controlled planning and implementation, based on assumed moral authority and a paternalistic sense of mission.

V Conclusions and discussion

The analysis of selected Rwandan policies and programmes points to numerous instances of far-reaching social engineering. Ambitious goals, strong commitment, hard work, firm implementation and focus on selected targets have contributed to impressive results – at least when read from what policy makers have defined as relevant – which have attracted due attention and recognition.

The analysis, however, also illuminates problematic aspects of Rwandan developmental engineering, many of which concern the notion of central expertise in combination with denunciation of local practices. Negative views of the majority population, as discerned in the tone and phrasing of government documents, may reflect a firm conviction of knowing what is needed, while the use of coercive means may reflect strong commitment to achieving it, along with an urge among local officials to reach centrally stipulated targets. Both a patronizing attitude and an imposing approach, however, risk reinforcing the state–population separation. The conviction of knowing what is needed, including that local practices need to be abolished, is also related to the assumption that external experts are in a position to devise solutions, and the ensuing approach of large-scale, standardized systems replacing locally varying ones. Meanwhile, outlawing poly-cropping, uprooting non-stipulated crops, tearing down poor quality houses, fining those who let their cattle graze and threatening to expropriate plots if modern techniques are not applied, may be efficient ways to make people behave as stipulated and reach certain targets. However, history – not least aid history – is full of examples of development efforts failing due to ignorance and overruling of local conditions, expertise and rationalities (Richards, 1985). In Rwanda, agricultural conditions significantly vary over short distances and there is rich evidence of fine-tuned strategies and ‘micro-niche management’ adapted over centuries (van Damme, Ansoms and Baret, 2013; Huggins, 2009). Also, the encounter between newly introduced practices and locally prevailing ones always gives rise to a dynamic interplay with uncountable unanticipated consequences and ensuing sidetracking (Hasselskog, 2009; Olivier de Sardan, 2005: 145). Scott (1998: 309f) eloquently shows how big schemes have led to local improvisation and adaptation, which cannot be planned, but without which the schemes would not be sustained. Rwandan peasants’ strategies of growing their preferred crops alongside the stipulated ones, finding some fodder for their cattle, and using affordable construction material, may be what make them survive when prescribed practices do not provide enough, centrally determined cash crops do not ripen soon enough, or high-quality housing is too expensive. Therefore, aiming to abolish and hindering by coercion people to use such local improvisation is not rational. Rather, when aiming at increased food security, facing significant variations in agricultural potential, and having so many living on the margin, it would make sense to actively safeguard and make conscious use of locally perfected practices, and deliberately allow room for using all available opportunities. That would, however, imply abandonment of key aspects of social engineering approaches and assumptions delineated above.

Still, and without disregarding highly problematic features, undeniable development achievements call into question the negative connotations of social engineering. As noted in the introduction, Rwandan policy and practice show many similarities with the ‘developmental state’. Such similarities are in no way surprising. Developmental states are socially engineering, and social engineering is – at least intentionally and potentially – developmental. The ‘developmental state’ generally has more positive connotations than social engineering, but is also known as authoritarian, which has sometimes been deemed necessary and justifiable (Leftwich, 1995: 416). Gaining legitimacy from developmental performance rather than democratic nature, it has however also been emphasized that, not least in Africa, a developmental state must be democratic (Meyns and Musamba, 2010: 10, 27; Robinson and White, 1998). Meanwhile, contemporary Rwanda is increasingly judged as authoritarian, and the central control and restriction of political space are by some observers considered to have contributed to development achievements (Beswick, 2011: 1925; Booth and Golooba-Mutebi, 2012; Samset, 2011). Whereas Rwandan developmental engineering may thus be regarded as motivated and functional, repression remains a severe problem. An apparent problem, too, with the partial success of an authoritarian but development-oriented government, is that it may increase the acceptance of authoritarianism, which will be problematic in case a government is, or becomes, less developmentally oriented.

This, in turn, points to the issue of the legitimate use of social engineering. As noted in the introduction, the Rwandan government makes a firm point of not allowing donors to set the agenda. As also mentioned, aspects of social engineering have been apparent throughout the history of international development cooperation (Scott, 1998: 223ff, 247, 342; Hasselskog, 2009). As much of earlier aid approaches have come to be avoided, inherent social engineering has become less obvious. Large-scale schemes of villagization and agricultural revolution, blatantly paternalistic approaches and crude conditionality are for example no longer endorsed. However, with ‘development partners’ deeply involved in radical ‘reconstruction’ of recipient states’ economic, political and administrative systems, and with ‘policy advice’ being potentially as firm as conditionality, aspects of social engineering are still there, though apparently avoided through ideals and ambitions of partnership, ownership, alignment and harmonization. In Rwanda, on the other hand, and as has been illustrated, instances of social engineering are strikingly overt. Not least when it comes to human resettlement, agricultural transformation, and ‘changing people’s mindset’, judgements and intentions are clearly stated and radical policies consequently executed. Such an open approach could be considered more honest than subscribing to inherently contradictory ideals. As indicated, it may also be highly ‘efficient’. Moreover, for a national government to pursue radical restructuring policies and practices is reasonably more legitimate than for foreign donors who are not even formally accountable to recipient populations.

Footnotes

Acknowledgements

This article is based on work supported by the Swedish International Development Cooperation Agency, Department for Research Cooperation. I am also grateful to two anonymous reviewers.

Notes

References

Alexander

Schmidt

J. K. H. W.

1996: Social engineering: Genealogy of a concept. In Podgórecki

Alexander

Shields

, editors, Social engineering. Carleton University Press, 1–19.

Ansoms

2009: Re-engineering rural society: The visions and ambitions of the Rwandan Elite. African Affairs, 108, 289–309.

Ansoms

2011: Rwanda’s post-genocide economic reconstruction: The mismatch between elite ambitions and rural realities. In Straus

Waldorf

, editors, Remaking Rwanda: State building and human rights after mass violence. University of Wisconsin Press, 240–51.

Ansoms

2013: Large-scale land deals and local livelihoods in Rwanda: The bitter fruit of a new agrarian model. African Studies Review, 56, 1–23.

Ansoms

Rostagno

2012: Rwanda’s vision 2020 halfway through: What the eye does not see. Review of African Political Economy, 39, 427–50.

Beswick

2011: Aiding state building and sacrificing peace building? The Rwanda-UK relationship 1994–2011. Third World Quarterly, 32, 1911–30.

Booth

Golooba-Mutebi

2011: Developmental patrimonialism? The case of Rwanda, Africa power and politics programme, Working Paper 16. ODI.

Booth

Golooba-Mutebi

2012: Developmental patrimonialism? The case of Rwanda. African Affairs, 111, 379–403.

Cowen

Shenton

1996: Doctrines of development. Routledge.

10.

Crisafulli

Redmond

2012: Rwanda, Inc.: How a devastated nation became an economic model for the developing world. Palgrave Macmillan.

11.

Dar

Cooke

, editors, 2008: The new development management. Zed Books.

12.

Desrosiers

M.-E.

Thomson

2011: Rhetorical legacies of leadership: Projections of ‘benevolent leadership’ in pre- and post-genocide Rwanda. Journal of Modern African Studies, 49, 429–53.

13.

DFID 2013: Operational Plan 2011–2015. DFID Rwanda, https://www.gov.uk/government/uploads/system/uploads/attachment_data/file/237402/Rwanda.pdf, last accessed on 2 January 2014.

14.

Evans

1989: Predatory, developmental, and other apparatuses: A comparative political economy perspective on the third world state. Sociological Forum, 4, 561–87.

15.

Everett

M. A.

2006: From social engineering to social movement: Power sharing in community change in New York’s Hudson Valley and Catskill Mountains. PhD dissertation, Erasmus University of Rotterdam.

16.

Farg 2011: Farg citizen’s charter. Republic of Rwanda.

17.

Fein

M. L.

2001: Social engineering in context: Some observations on turner. Sociological Practice: A Journal of Clinical and Applied Sociology, 3, 121–25.

18.

GoR 2008: Rwanda: Joint governance assessment report. Republic of Rwanda.

19.

GoR 2010: Government programme 2010–2017. Republic of Rwanda.

20.

Hasselskog

2009: Development intervention on the ground. Inherent rationales of aid and their encounter with local dynamics in three Cambodian villages. Ph D dissertation, University of Gothenburg.

21.

Hayek

F. von

1944: The road to serfdom. Chicago Press.

22.

Hayman

2008: Rwanda: Milking the cow. Creating policy space in spite of aid dependence. In Whitfield

, editor, The politics of aid: African strategies for dealing with donors. Oxford University Press, 156–84.

23.

Hintjens

2013: Land reform, social justice, and reconstruction: Challenges for post-genocide Rwanda. In Veney

C. R.

Simpson

, editors, African democracy and development: Challenges for post-conflict African nations. Lexington Books, 81–108.

24.

Huggins

2009: Agricultural policies and local grievances in rural Rwanda: Peace Review. A Journal of Social Justice, 21, 296–303.

25.

Ikeotuonye

F. C. R. A.

2002: Lateral shades of social engineering: A critical exploration of ‘interest representation’, ‘state’ and ‘development’. Oxford university Press and Community Development Journal, 37, 69–79.

26.

Ingelaere

2009: Living the transition: Inside Rwanda’s conflict cycle at the grassroots. Journal of Eastern African Studies, 3, 438–63.

27.

Ingelaere

2010: Peasants. Power and ethnicity: A bottom-up perspective on Rwanda’s political transition. African Affairs, 109, 273–92.

28.

Ingelaere

2011: The ruler’s drum and the people’s shout. Accountability and representation on Rwanda’s hills. In Straus

Waldorf

, editors, Remaking Rwanda: State building and human rights after mass violence. University of Wisconsin Press, 67–78.

29.

IRIN 2009: Rwanda: Cows take up residence in bid to stop soil erosion. 3 December, http://www.irinnews.org/Report/87308/RWANDA-Cows-take-up-residence-in-bid-to-stop-soil-erosion, last accessed on 29 June 2013.

30.

Johnston

1982: MITI and the Japanese miracle: The growth of industrial policy, 1925–1975. Stanford University Press.

31.

Jones

2012: Between Pyongyang and Singapore: The Rwandan state, its rulers, and the military. In Campioni

Noack

, editors, Rwanda fast forward: Social, economic, military and reconciliation prospects. Palgrave Macmillan, 228–46.

32.

Kantola

Seeck

2011: Dissemination of management into politics: Michael Porter and the political uses of management consulting. Management Learning, 42, 25–47.

33.

Kimberley Brandt

2009: An Asian miracle in Africa. On the institutional context of developmentalism as a determinant factor for a successful development strategy. A case study of Rwanda. MA Thesis, Institute of Social Studies, The Hague.

34.

Leftwich

1995: Bringing politics back in: Towards a model of the developmental state. The Journal of Development Studies, 31, 400–27.

35.

T. M.

2007: The will to improve: Governmentality, development, and the practice of politics. Duke University Press.

36.

Meyns

Musamba

, editors, 2010: The developmental state in Africa: Problems and prospects. University of Duisburg-Essen.

37.

Minagri 2004: Strategic plan for the transformation of agriculture in Rwanda: Main document. Republic of Rwanda.

38.

Minagri 2009: Strategic plan for the transformation of agriculture in Rwanda – Phase II (PSTA II). Final report. Republic of Rwanda.

39.

Minagri 2011: Strategies for sustainable crop intensification in Rwanda: Shifting focus from producing enough to producing surplus. Republic of Rwanda.

40.

Minagri 2012: Farm land use consolidation in Rwanda: Assessment from the perspectives of agriculture sector. Republic of Rwanda.

41.

Minagri 2013a: One Cow per Poor Family Program ‘GIRINKA’. http://www.minagri.gov.rw/index.php?id=28, last accessed on 6 June 2013.

42.

Minagri 2013b: Crop Intensification Program (CIP). http://www.minagri.gov.rw/index.php?id=618, last accessed on 6 June 2013.

43.

Minaloc 2008: Community development policy. Revised version. Republic of Rwanda.

44.

Minaloc 2010: Terms of references for joint task force shelter scheme on eradication of ‘nyakatsi’. Republic of Rwanda. http://www.minaloc.gov.rw/fileadmin/documents/Minaloc_Documents/ToRs%20Nyakatsi%20eradication%20signed%20dec%202010.pdf, last accessed on 29 June 2013.

45.

Minocefin 2000: Rwanda vision 2020. Republic of Rwanda.

46.

Minocefin 2001: National poverty reduction programme: Participatory Poverty Assessment (PPA). Republic of Rwanda.

47.

Minocefin 2002: The government of Rwanda poverty reduction strategy paper. Republic of Rwanda.

48.

Minocefin 2007: Economic development & poverty reduction strategy paper 2008–2012. Republic of Rwanda.

49.

Minocefin 2013: Economic development & poverty reduction strategy 2013–2018: Shaping our development. Republic of Rwanda.

50.

Mininfra 2009: Updated version of the national human settlement policy in Rwanda. Republic of Rwanda.

51.

MoH 2010: Rwanda community based health insurance policy. Republic of Rwanda.

52.

Musahara

Huggins

2005: Land reform, land scarcity and post-conflict reconstruction: A case study of Rwanda. In Huggins

Clover

, editors, From the ground up: Land rights, conflict and peace in sub-Saharan Africa. Institute for Security Studies.

53.

Newbury

M. C.

1978: Ethnicity in Rwanda: The case of Kinyaga. Africa, 48, 17–29.

54.

Newbury

M. C.

2011: High modernism at the ground level: The Imidugudu policy in Rwanda. In Straus

Waldorf

, editors, Remaking Rwanda: State building and human rights after mass violence. University of Wisconsin Press, 223–39.

55.

ODI 2012: Delivering maternal health. Why is Rwanda doing better than Malawi, Niger and Uganda?, ODI Briefing Paper 74. http://www.institutions-africa.org/filestream/20120611-odi-briefing-paper-74-delivering-maternal-health-why-is-rwanda-doing-better-than-malawi-niger-and-uganda-v-chambers-and-d-booth-may-2012, last accessed on 29 June 2013.

56.

OECD 2011: Aid effectiveness 2005–10: Progress in implementing the Paris Declaration. OECD Publishing: http://effectivecooperation.org/files/resources/2011%20Report%20on%20Monitoring%20the%20Paris%20Declaration%20ENGLISH.pdf, last accessed on 21 January 2015.

57.

OECD 2012: Aid effectiveness 2011: Progress in implementing the Paris Declaration – Volume II Country Chapters. OECD Publishing: http://www.oecd.org/dac/effectiveness/Rwanda%205.pdf, last accessed on 2 January 2014.

58.

Olivier de Sardan

J.-P.

2005: Anthropology and development: Understanding contemporary social change. Zed Books.

59.

Popper

1957: The poverty of historicism. Routledge.

60.

Pritchard

M. F.

2013: Land, power and peace: Tenure formalization, agricultural reform, and livelihood insecurity in rural Rwanda. Land Use Policy, 30, 186–96.

61.

Purdeková

2012: Civic education and social transformation in post-genocide Rwanda: Forging the perfect development subjects. In Campioni

Noack

, editors, Rwanda fast forward: Social, economic, military and reconciliation prospects. Palgrave Macmillan, 192–209.

62.

RDRC 2011: Report on medical screening and categorization of disabled former war combatants 2008–2011. Republic of Rwanda. http://demobrwanda.org.rw/reintergration.html, last accessed on 25 June 2013.

63.

Reyntjens

2004: Rwanda, ten years on: From genocide to dictatorship. African Affairs, 103, 177–210.

64.

Reyntjens

2011: Constructing the truth, dealing with dissent, domesticating the world: Governance in post-genocide Rwanda. African Affairs, 110, 1–34.

65.

RGB 2013a: Imihigo. http://www.rgb.rw/governance-innovations/imihigo/article/imihigo, last accessed on 21 January 2015.

66.

RGB 2013b: Umuganda. http://www.rgb.rw/governance-innovations/umuganda/article/umuganda, last accessed on 21 January 2015.

67.

RLDSF 2011: Citizen’s charter. Republic of Rwanda.

68.

RLDSF 2012: RLDSF Annual activity report, 2011/2012. Republic of Rwanda.

69.

Richards

1985: Indigenous agricultural revolution: Ecology and food production in West Africa. Unwin Hyman.

70.

Robinson

White

, editors, 1998: The democratic developmental state: Politics and institutional design. Oxford University Press.

71.

Routley

2012: Developmental states: A review of the literature. ESID Working Paper No. 03, http://r4d.dfid.gov.uk/PDF/Outputs/ESID/esid_wp_03_routley.pdf, last accessed on 2 January 2014.

72.

Samset

2011: Building a repressive peace: The case of post-genocide Rwanda. Journal of Intervention and Statebuilding, 5, 265–83.

73.

Scott

J. C.

1998: Seeing like a state: How certain schemes to improve the human condition have failed. Yale University Press.

74.

Silva-Leander

2008: On the danger and necessity of democratisation: Trade-offs between short-term stability and long-term peace in post-genocide Rwanda. Third World Quarterly, 29, 1601–20.

75.

Straus

Waldorf

2011: Introduction: Seeing like a post-conflict state. In Straus

Waldorf

, editors, Remaking Rwanda: State building and human rights after mass violence. University of Wisconsin Press, 3–21.

76.

The Independent 2011: Umuganda under the spotlight. 15 August, http://www.independent.co.ug/News/regional-news/4505, last accessed on 29 June 2013.

77.

The Independent 2012: IMIHIGO, Six years of transforming Rwanda. 10 October, http://www.independent.co.ug/supplement/117-supplement/6597-imihigo-six-years-of-transforming-rwanda, last accessed on 25 June 2013.

78.

The New Times 2012a: Family performance contracts fuel hope. 13 March, http://www.newtimes.co.rw/news/index.php?i=14930&a=51269, last accessed on 29 June 2013.

79.

The New Times 2012b: Family contracts to improve welfare. 21 June, http://www.newtimes.co.rw/news/index.php?i=15030&a=55052, last accessed on 11 October 2013.

80.

The Rwanda Focus 2011: Nyakatsi eradication moves 125,000 people into decent housing. 11 May, http://focus.rw/wp/2011/05/nyakatsi-eradication-moves-125000-people-into-decent-housing/, last accessed on 29 June 2013.

81.

Thomson

2011: Reeducation for reconciliation: Participant observations on? Ingando. In Straus

Waldorf

, editors, Remaking Rwanda: State building and human rights after mass violence. University of Wisconsin Press, 331–39.

82.

Thomson

2012: Peasant perspectives on national unity and reconciliation: Building peace or promoting division? In Campioni

Noack

, editors, Rwanda fast forward: Social, economic, military and reconciliation prospects. Palgrave Macmillan, 96–110.

83.

Turner

J. H.

2001: Social engineering: Is this really as bad as it sounds? Sociological Practice: A Journal of Clinical and Applied Sociology, 3, 99–120.

84.

UNDP 2007: Turning vision 2020 into reality: From recovery to sustainable human development. National human development report, Rwanda 2007. UNDP.

85.

UNDP 2013: Human development report 2013. The rise of the South: Human progress in a diverse world. UNDP.

86.

Van Damme

Ansoms

Baret

P. V.

2013: Agricultural innovation from above and from below: Confrontation and integration on Rwanda’s hills. African Affairs, 113, 108–27.

87.

Versailles

2012: Rwanda: Performance contracts (imihigo), Country learning notes. ODI.

88.

Verwimp

2011: The 1990–92 massacres in Rwanda: A case of spatial and social engineering? Journal of Agrarian Change, 11, 396–419.

89.

VUP 2011: Vision 2020 Umurenge programme. Annual report 2009/10. Republic of Rwanda.

90.

World Bank 2013: Rwanda economic update. Maintaining momentum. With a special focus on Rwanda’s pathway out of poverty. World Bank.