This essay ‘from the archive’ was written for my 1997 MA (Econ) Development Administration and Management at the Institute of Development Policy and Management (IDPM) of the University of Manchester.
This essay will explore a number of approaches to rural development and its planning, with reference to the widely contrasted notions of ‘blueprint’ and ‘process’. It will argue that, while process approaches share an appreciation of the uncertainty surrounding the development process that distinguishes them from blueprint approaches, process approaches themselves differ fundamentally in the extent to which they also embrace uncertainty with regard to the goals of development. Those that at least implicitly share the blueprint approach’s economistic assumption of development goals as fixed and certain will be termed ‘instrumental process’ approaches. Those that adopt an empowerment perspective, and so accept the goals of development to be intrinsically variable and uncertain, will be termed ‘teleological process’ approaches.
It will be argued that, while process approaches may and indeed have been synthesized with blueprint approaches, the extent to which a synthesis, or indeed any approach, may be considered effective can only be assessed in the context of its underlying assumptions regarding the goals of development. Where the goal is empowerment for the increased well-being of the rural poor as defined by themselves, and so intrinsically variable and uncertain, no economic measure of effectiveness can substitute for a wholly participative teleological process approach in which ends as well as means are defined and redefined through experimentation and learning with the poor themselves. Furthermore, given the political will, the accountability constraints of governments and donors that have been used to explain their widespread reliance on economistic approaches to date need not preclude such an empowerment-oriented approach.
The blueprint and process approaches will be introduced with reference to some of the writers who have contrasted them, and their assumptions relating to uncertainty surrounding the development process will be identified as a key factor by which they may be distinguished. The process approaches of Korten (1980) and Rondinelli (1993) will be contrasted in relation to their assumptions regarding the goals of development, and identified as teleological and instrumental respectively. An explicit attempt to synthesize blueprint and process, Brinkerhoff & Ingle’s (1989) ‘structured flexibility’ approach, will be introduced. It will be argued that, in common with blueprint and instrumental process approaches, its effectiveness in empowerment terms cannot compare with that of a teleological process approach. A case study of integrated rural development in Ethiopia will be presented to illustrate the feasibility of a truly participatory teleological process approach, even in the context of official development assistance.
Blueprint and process contrasted
The dominant approach to rural development planning of the early growth-oriented national development strategies became known as the ‘blueprint’ approach to reflect its emphasis on the project preparation process as they key to successful intervention. According to Gittinger’s 1982 text on agricultural project planning:
“Perhaps the most difficult single problem confronting agricultural administrators in developing countries is implementing development programmes. Much of this can be traced to poor project preparation” (Gittinger 1982, 3).
Moris has likened this preparation to an architect preparing his blueprint, to “generate specifications for components and to map their points of linkage into a common structure” (Moris 1990, 28). Once the blueprint is accepted, producers of the various components are expected make their respective contributions accordingly, the plan serving effectively as a substitute for management. As the focus of national development strategies shifted in the 1970s to redistribution and rural poverty alleviation, this blueprint approach was identified as an impediment to effective rural development, and contrasted with an alternative ‘process’ approach that was found to be characteristic of more effective interventions. Sweet and Weisel first drew this distinction in their 1979 study of a 1973 review of 36 field programmes (Moris 1990, 27).
Korten has characterized the blueprint approach by inter alia its emphasis on careful and detailed pre-planning, and its conceptual and actual separation of planning from implementation. He underlines the inappropriateness of such an approach to the task of rural development:
“Where knowledge is nearly non-existent, the blueprint approach calls for behaving as if knowledge were nearly perfect… Where the need is for a close integration of knowledge building, decision-making and action-taking roles, it sharply differentiates the functions and even the institutional locations of the researcher, the planner and the administrator” (Korten 1980, 497).
Rondinelli has also underlined the inappropriateness of the assumptions of the blueprint approach that exhaustive analysis will aid the understanding of complex problems, and that there will be a direct relationship between government policy, action and outcomes (Rondinelli 1993, 3). In fact, according to Long and Van der Ploeg, a major source of the uncertainty inherent in the development process is the human agency of the various social actors involved. They have described the mechanistic assumptions of causality inherent in the blueprint model as:
“a gross over-simplification of a much more complicated set of processes which involves the reinterpretation or transformation of policy during the implementation process” (Long & Van der Ploeg 1989, 227).
Thus, the assumption of the blueprint approach that uncertainty can be reduced by gathering more data and expanding the project design phase fails to recognize, as Rondinelli observes, that many constraints remain hidden until implementation (Rondinelli 1993, 17).
In order to deduce elements of approaches that work in rural development, Korten studied the cases of five Asian ‘success stories’ of rural development intervention. From the wide variation in programme and organizational variables evident among the cases, he concluded that the determinants of the programmes’ success were to be found not in their programme or organizational content, but in the process by which these had developed:
“Each project was successful because it had worked out a programme model responsive to the beneficiary needs at a particular time and place and each had built a strong organization capable of making the programme work” (Korten 1980, 496).
He describes this as a process of achieving ‘fit’ between task, context and organizational variables, such that the organization’s distinctive competence fits the programme’s task requirements, the programme outputs fit the beneficiary needs, and the organizational decision-making process fits the beneficiaries’ means of expressing their needs. He refers to this bottom-up process of programme and organizational development as a ‘learning process approach’ observing that, rather than blueprints, leadership and teamwork were the key elements of these successes – they were not planned and implemented but rather evolved through a process of learning in which programme personnel and villagers worked together to create both the programmes and organizations that effectively responded to beneficiary needs.
Korten describes an idealized representation of this learning process as comprising three successive stages: ‘learning to be effective’, achieving a good fit between beneficiaries, programme and organization; ‘learning to be efficient’, reducing the ratio of input requirements to output; and ‘learning to expand’. Rondinelli advocates an incremental process of four stages of project development: experimental projects of a moderate scale, and flexibly implemented, for when problems and conditions are unknown; pilot projects, for when objectives have been well-defined, to test the results of the experiments under various conditions and to determine their relevance, transferability and acceptability; demonstration projects to exhibit the effectiveness and increase the acceptability of the project; and replication projects to test full scale production technology (Rondinelli 1993, 24-26). Both emphasize the integration of planning and management in an iterative process of learning and capacity building to achieve developmental goals in conditions of change and uncertainty.
Empowerment and the uncertainty of goals
The broad consensus among advocates of process approaches that their flexibility and learning orientation are keys to greater effectiveness in achieving development goals in conditions of change and uncertainty conceals fundamental differences in their conceptions of the nature of those goals against which effectiveness is to be measured. Rondinelli is explicit in his expression of an orthodox economistic conception of development goals:
“the essence of development is expansion of participation in economic activities through the creation of social and economic systems that draw large numbers of people into processes of production, exchange, and consumption” (Rondinelli 1978, cited Rondinelli 1993, viii).
As indicated by the title of his book ‘Development Projects as Policy Experiments’, the principal intended agents of his process approach, and concomitant learning and capacity building, are the official development administrations of governments and international assistance agencies. The goals of development are understood as essentially economic and unproblematic, and the means of their accomplishment essentially a problem of technique. In this his process approach concurs with the blueprint approach, or what Hulme has called the orthodox approach, where rural development interventions are viewed as “activities… in pursuit of the achievement of a known and stated objective or set of objectives” (Hulme 1994, 213). Although he criticizes the quantitative methods of the orthodox blueprint approach as “ineffective precisely because they tried to clarify and make technical those issues that were inherently complex and political” (ibid., 19), his process approach, regarding projects as social experiments to reduce uncertainties and unknowns, is essentially a technical solution to the problem of achieving a known goal. Thus, Rondinelli’s may be identified as an ‘instrumental process’ approach.
Korten’s approach, in contrast, centres on community organizations as the principal agent of a learning process approach with empowerment as its goal. His 1980 article, aptly entitled ‘Community Organization and Rural Development: A Learning Process Approach’, addresses the question of how to foster
“effective community controlled social organizations as important if not essential instruments… [for] the rural poor… to give meaningful expression to their views, mobilize their own resources in self-help action, and enforce their demands on the broader national political and economic systems’ (Korten 1980, 480).
Chambers more explicitly addresses the problematic nature of such a development goal in his 1993 call for a ‘new professionalism’ with Korten’s learning process approach at its core. He writes that “development is not movement towards a fixed goal but continuous adaptation to maximize well-being in changing conditions”, and its goal “is not growth as defined by normal professionals, but well-being as defined by the poor for themselves” (Chambers 1993, 10). Similarly, Uphoff writes of Korten’s learning process approach that “it presumes that neither the ends nor the means of… interventions can be fully known in advance, and that understanding and consensus on them must be built up through practical experience” (Uphoff 1993, 12). Thus, Korten’s may be identified as a teleological process approach.
From such an empowerment perspective, the critical determinant of the effectiveness of a rural development initiative will not be the extent to which the route of the journey is pre-planned, or even the extent to which its destination is decided in advance. Indeed, some goal setting and pre-planning may be a pre-requisite of any purposeful action. More important will be who participates in the planning and deciding, and how responsive they are to the changing aspirations of beneficiaries as well as unforeseen obstacles and opportunities along the way. While Korten’s process approach puts the self-expressed needs of the rural poor firmly at the centre, shaping both programme and organizational variables, participation of beneficiaries in Rondinelli’s approach is essentially instrumental and extractive, and any empowerment incidental.
Rondinelli calls for the reorientation of official development administrations “to cope more effectively with the inevitable uncertainty and complexity of the development process” (Rondinelli 1993, 5), and regards the only certainty as “that the course of development is uncertain” (ibid., 156). Korten in contrast regards effectiveness as a product of “building a supporting organization around the requirements of the programme, or… adapting the capabilities of an existing organization to fit those requirements” (Korten 1980, 497), while the programme evolves to fit beneficiaries’ needs. While Rondinelli advocates the capacity building, and integration of the planning and management functions, of official development administrations, Korten advocates the capacity building of local community-based organizations, and their full participation in an integrated learning approach to planning and management.
Synthesized approaches and structured flexibility
A number of authors have written of the possibility of synthesizing blueprint and process approaches. Scoones, for example, writes that “these two options are obviously not mutually exclusive” (Scoones 1994, 6), and Chambers notes that, “although they are presented… as dichotomies, the blueprint and learning process approaches can be and have been combined in many ways” (Chambers 1993, 84). He cites the ODA’s ‘planning by successive approximation’, as one example, and the ‘structured flexibility approach’ as another (ibid.).
As articulated by Brinkerhoff and Ingle (1989), the structured flexibility approach is an explicit effort to integrate the blueprint and process approaches in order to overcome “the lack of fit between the precepts of the process model and the current modalities by which the vast bulk of international development assistance is provided”, which have “restricted its use and potential for expansion” (Brinkerhoff & Ingle 1989, 489). They argue that, despite its weaknesses in terms of performance, a blueprint approach is widely favoured over the process approach for its control and specificity, because donor accountability requirements serve as a disincentive for experimentation, and because political concerns over the balance of power lead decision-makers to fear the empowerment potential of the process model.
The structured flexibility model retains the blueprint orientation to analysis, planning and specificity, but applies its analytical tools, in a participatory, process mode, to “facilitate people’s accurate assessment of opportunities and choices, and the potential actions based upon them, within a structured framework that encourages feedback and learning” (Brinkerhoff & Ingle 1989, 490). The capacity-building goal of the process model is complemented by short term product or service delivery targets, satisfying accountability requirements for measurable delivery while building long-term capacity by developing problem-solving skills in addressing concrete immediate needs:
“Management by structured flexibility… means repeating the action-research cycle of reconnaissance, design, implementation, and learning with the aim of generating initial performance gains while building the indigenous capacity required for sustainability and replicability” (Brinkerhoff & Ingle 1989, 493).
Brinkerhoff and Ingle offer three case studies of the structured flexibility approach in action to illustrate its effectiveness . A brief review of these reveals their assumptions of certainty regarding their goals. In all three cases, programmes and organizations had been created by governments and aid agencies, and their effectiveness has been assessed in relation to goals that had been defined top-down, in advance and assumed unproblematic. In the case of the Farming Systems Research and Development project of the Caribbean Agricultural Research and Development Institute (CARDI), for example, CARDI had been established by member states of the Commonwealth Caribbean Community, and the project was designed by ‘key actors’ from USAID and CARDI. “Participation was expanded to include farmer representatives” only after the goals of “building CARDI’s institutional capacity plus production of specific technical outputs over the 5-year life of the project” had been defined (Brinkerhoff and Ingle 1989, 498). An evaluation report is cited as concluding that “the project is making good progress toward its objectives”, including “realization of the End of Project Status” (ibid.).
In terms of Korten’s concept of fit, CARDI’s effectiveness relates to the fit achieved between the organization’s competence and the programme’s requirements. However, the fit between its decision-making process and the beneficiaries’ means of expressing their needs, and between those needs and the programme’s outputs, have been assumed a priori as unproblematic. While beneficiaries may participate in the learning process of programme management, they have been excluded from the fundamental stage of project identification. Due to its responsiveness to uncertainty and change, the structured flexibility approach may indeed have been more effective in achieving the programme’s goals than a blueprint approach would have been. To the extent that USAID’s accountability requirements were better satisfied, it may have been a more effective means of attracting finance than Rondinelli’s process approach. It cannot, however, be expected to have been as effective from an empowerment perspective as had a programme and organization “emerged out of a learning process in which villagers and programme personnel shared their knowledge and resources to create a programme which achieved a fit between needs and capacities of the beneficiaries and those of the outsiders who were providing assistance” (Korten 1980, 497).
It has been argued that full empowerment requires not only the ‘power to’ prevail in decisions affecting one’s interests, but also ‘power over’ defining the agenda of decisionable issues, and ‘power within’ to transcend conventional ways by which one’s interests are perceived and known (Kabeer 1994, 224-7). This latter element, the ‘power over’, is irreducibly subjective and therefore cannot be achieved except through experiential recognition of strategic as well as practical needs and interests. From such a perspective, while empowerment may be facilitated through intervention, it must ultimately be self-generated. Any approach to planning rural development interventions, be it blueprint, process or a synthesis, in which participation is not central even to project identification, will be relatively ineffective from such a perspective.
The South Wollo Community Empowerment Programme
Chambers has written that: “the essence of a good learning project is good staff put in the field and sustained for periods of months or, more likely, years, exploring and learning from and with the local people and trying to see how better they can gain what they want and need” (Chambers 1993, 86).
Brinkerhoff and Ingle’s rationale for seeking a middle-ground between blueprint and learning process approaches is that the former is ill-suited to the complexity and uncertainty of development, while the latter is ill-suited to the accountability requirements of international assistance bureaucracies. Their response, adapting the process approach to bureaucracy, fails to address the root of the problem – that ‘development’ agencies remain primarily accountable to their financiers and not their beneficiaries. This accountability structure represents a greater impediment to effective empowerment than any rural development planning or management procedure.
One example of bureaucracies transcending this accountability constraint, the case of a teleological process approach being adopted by a national government bureaucracy in partnership with an official donor agency, is offered by Bergdall (1995). He writes that in 1993 the Swedish and Ethiopian governments agreed to collaborate on a long term integrated rural development programme in South Wollo Zone of Ethiopia, and selected four districts as pilot areas. Convinced that the programme should be built from the bottom-up, based on beneficiaries’ own perceptions of their development needs, rather than designed as a blueprint by professional planners, the project Steering Committee of Zonal officials opted for a process approach. The intention was that a first phase, involving Community Participation Workshops (CPWs) in 24 sample areas of the four districts, would generate the community needs data to be input to a more conventional programme in Phase II, involving various ministries in typical sectoral activities.
Expatriate specialists in community participation were contracted to help design the CPWs and train the facilitators, in recognition of the steering committee’s lack of practical experience in such an approach. Traditional community self-help associations called kires were chosen as the organizational unit for the CPWs, in preference to the governmental Peasant Associations that tended to be larger, more political, and still strongly associated with earlier centrally planned development efforts involving coerced mobilization of labour. Twenty-two staff from the Departments of Health, Education, Natural Resources, Agriculture and Planning were assigned as facilitators for a six-month period of intensive field work. They were hosted by villagers in their homes, and traveled by foot, mule and public transportation.
Representative groups of men, women and youth were included in each CPW. The workshops had three main objectives:
- “to provide information about local perceptions of development needs and priorities as a basis for defining sectoral projects that could be implemented by the line ministries…
- to provide experience for the development of future community participation strategies… [and]
- to provide an opportunity for people in a kire to think about ways of solving their own problems and to regain the initiative for their own development, thereby building a foundation for sustainable development in the long term future” (Bergdall 1995, 16).
During the workshops, each kire identified and prioritized what they considered to be their most important development needs, and drew up an action plan for addressing the top priority need through their own efforts, and using local resources.
The facilitators returned to each kire after two to three months to review progress and assist in planning new initiatives. The results of the local initiatives found during these visits led to a reconsideration of the earlier plan for Phase II. Bergdall cites examples of kires whose accomplishments exceeded their action plans, had involved broad participation of the community residents and support from previously ‘negligent’ Peasant Associations, and had instilled renewed pride and self-confidence in the communities and their capacities to accomplish projects for their own benefit. Instead of going ahead with the centralized planning of conventional sectoral programmes for line ministry implementation, the Steering Committee decided to launch a process-oriented Community Empowerment Programme with the objective “to build the capacity of rural communities to initiate and sustain their own development activities” (Bergdall 1995, 15).
In order to expand the impact of the programme, the Committee will seek to build on the experience of bottom-up development activity based on local initiative by integrating additional support components. These may include: provision of technical assistance to the local communities; facilitating community access to additional external resources; ensuring effective collaboration of government, NGO and private sector; adaptation of existing extension services to support the bottom-up approach; and fostering wider institutional linkages (Bergdall 1995, 24). Bergdall writes that diverse solutions to these issues will be allowed to evolve over time, with experimentation, reflection and learning as key operative guidelines.
Any discussion of effectiveness is meaningless except in relation to a goal or set of goals. What Hulme (1994) has called orthodox models of planning and managing rural development interventions are based on economistic assumptions of utility and welfare, from which are derived the certain and known economic goals of development action. This assumed certainty of goals is a feature that Rondinelli’s process approach and Brinkerhoff and Ingle’s Structured Flexibility Approach share with the conventional blueprint model. What Hulme has called political models recognize the role of power imbalances and conflicts of interest in the reality of development planning and management. Neither the choice of planning approach nor its outcome can be fully explained except in its political context.
As Hulme has observed, in the search for improved project methodologies,
“it is not simply a question of blueprint versus process… but a question of which form of blueprint or process, in which circumstances, and even of what means may be used to integrate blueprint and process approaches” (Hulme 1994, 230)
Perhaps most importantly, the question must be raised, ‘for what purpose?’ Certainly there may be projects and circumstances for which a blueprint or synthesized approach may be well suited – for example, “the construction of a large scale, physical infrastructure where the task is well defined, the outcomes terminal, the environment stable and the costs predictable” (Korten 1980, 497), or solving developing countries’ problems of “economic stagnation and poor productivity, resource gaps and debt burdens… etc.” (Brinkerhoff & Ingle 1989, 487). As Hulme & Limaoco have observed, however,
“to empower beneficiaries to be their own agents of development is more significant than any number of roads built, and subsequently not maintained” (Hulme and Limcaoco 1991, 232).
Given the remarkable record of economic development in generating and exacerbating rural poverty on a global scale, such empowerment is likely to also be more significant to the poor than any improvement in economic indices. Empowerment is an inescapably political process, however, and those who enjoy power and embrace a narrow economistic conception of their own welfare, including governments, official aid agencies and the institutions and people to whom they are accountable, may be expected to resist it. That the Community Empowerment Programme of South Wollo has emerged from within official aid bureaucracy may be regarded as a breakthrough. Only time will tell to what extent the programme may overcome potential political constraints to empowering the rural poor of South Wollo, but those who truly seek to serve the interests of the rural poor would do well to start similarly by adapting their organizations and programmes to their beneficiaries’ self-expressed needs.
Bergdall, T (1995) “…and miles to go before we sleep: closing the rhetorical gap in South Wollo, Ethiopia” in Forests Trees and People no. 26/27, 17-24
Brinkerhoff, DW & Ingle, MD (1989) “Integrating blueprint and process: a structured flexibility approach to development management” in Public Administration and Development 9, 487-503
Chambers, R (1993) Challenging the Professions IT Publications, London
Gittinger, J (1982) Economic Appraisal of Agricultural Projects
Hulme, D (1994) “Projects, Politics and Professionals: Alternative approaches for project identification and project planning” in Agricultural Systems 47, 211-233
Hulme, D & Limcaoco, J (1991) “Planning integrated rural development projects in the Philippines: from blueprint to process?” in Project Appraisal 6/4, 223-233
Long, N & Van der Ploeg, JD (1989) “Demythologizing planned intervention: an actor perspective” in Sociologia Ruralis XXIX – 3/4, 226-249
Kabeer, N (1994) Revered Realities: Gender Hierarchies in Development Thought Verso, London
Korten, DC (1980) “Community Organization and Rural Development: A Learning Process Approach” in Public Administration Review September/October 1980, 480-511
Moris, J (1990) What organization theory has to offer third world agricultural managers, mimeo, IDPM library
Rondinelli, D (1993) Development Projects As Policy Experiments Routledge, London
Scoones, I (1995) “New Directions in Pastoral Development in Africa” in Scoones, I (Ed.) Living With Uncertainty IT Publications, London
Uphoff, N (1993) Learning From Gal Oya Cornell University Press, Ithaca NY