Researching knowledge-based aid

Setting the scene

Since 1996 there has been a remarkable growth within development co-operation agencies of interest in knowledge-based aid. Most agencies have launched projects that seek to make their work better grounded in the knowledge that they already possess within their organisations and to explore more effective ways of acquiring external knowledge related to development. At the same time, there has also been a growth in emphasis on more effectively disseminating this knowledge – to other agencies, to their own civil societies, to their partners in the South, and to the billions of poor people who are the stated beneficiaries of the whole intertwined aid and development project. Equally, there has been a revisiting of old notions that the poor are poor in large part because of their lack of appropriate knowledge. To the old account, expressed in many colonial and missionary texts, are added the new dimensions of globalisation (as the force shaping the knowledge needed) and information and communications technologies (ICTs – as an important new set of tools in the dissemination of this knowledge).

This book is the first that seeks to examine this phenomenon as a result of academic research in a series of agencies. It does so through a detailed analysis of what the new knowledge-based aid means at the level of discourse and practice in four leading development co-operation agencies.

In so doing, we are mindful that this knowledge-based aid contains a language that suggests that the lessons of past aid and development mistakes have been learned and that a new ethics of aid is an important aspect of the language of the new approach. However, we are also aware of both the continuing critiques of aid practices that suggest that there is much more business-as-usual than transformation, and the continued questioning of the theoretical underpinnings and practical impacts of aid and development.

In writing this book, we inevitably had to engage with the literatures on aid and development, and have sought to add to these. Accounts of agency policy and practice tend to polarise between 'official versions' (e.g. Kapur et al. 1997) and polemical attacks (e.g. Hancock 1989), and have been heavily focused on the multilateral agencies. Some, however, have more successfully attempted critical engagement, through an analysis that uses national case studies of the relationship between stated policy and its operationalisation (e.g. Mosley, Harrigan and Toye 1991), through a sensitive and reflective negotiation of access to a single agency in a single sector (e.g. Jones 1992), or through an analysis of the nature of development discourses and their playing out in a national context (e.g. Ferguson 1994). Whilst Crewe and Harrison (1998) talk of an ethnography of aid, even their work does little to get inside bilateral or multilateral agencies. Valuable though their approach is, their case studies are essentially of a large NGO (with real insider insights, as one of them worked for this NGO) and of a project of a multilateral agency. This present book is an unusual attempt (cf. King 1991) to look in depth across a group of bilateral and multilateral agencies in a way that allows for critical dialogue with the agencies. It is also an attempt to produce a sociological reading that moves between the realms of text and practice.

Agency accounts of aid are typically ahistorical. In their rush to develop new ideas and to gloss over past failures, agencies have tended to construct a collective amnesia about the past. It is inevitable, therefore, that current aid discourse recreates many elements of this past without recognising them. One of our concerns in this book will be to provide some historical depth. It is particularly important that we explore where knowledge-based aid has come from. We shall preview two strands of this context here before returning to them in some depth in the next two chapters.

Knowledge-based aid is only a small part of the broader changes that have taken place in development co-operation since the fall of the Berlin Wall. The World Conferences of the 1990s, the growing importance of the Development Assistance Committee (DAC) of the Organisation for Economic Co-operation and Development (OECD) and its promulgation of six International Development Targets, and a new architecture of donor co-ordination mechanisms have furthered a broader ideological convergence of agencies.

The new aid agenda brings with it new knowledge needs. However, it also brings a new importance for knowledge as a major theme of development and co-operation. The new focus draws heavily on wider arguments about the centrality of knowledge to economic success and about the connective power of new ICTs. Through the interweaving of these accounts in a literature and practice of knowledge management, agencies have begun to look at internal patterns of knowledge use as a key response to the critique of their effectiveness. At the external level, the term 'knowledge sharing' has become attractive as a way of distancing agencies from the widespread critique of conditionalities, while at the same time seeking to ensure that agency positions have influence over Southern countries' policies. Concurrently, arguments about the importance of knowledge economies have been directly translated into the development context to argue that knowledge is the key determinant of development (World Bank 2002a).

Research questions

However, this agency fascination with knowledge is in need of careful questioning. What does the emergence of knowledge-based aid amount to in practice? What explains its emergence? Why do agency approaches differ and how significant is this? Whose knowledge and whose visions of development are prioritised and whose marginalised? Does knowledge-based aid make for more efficient and effective agencies? Does knowledge-based aid make for more efficient and effective aid? This book seeks to address these questions.

As knowledge-based aid develops further it will also be important to research its impacts on the supposed beneficiaries in the South who are to be helped out of poverty through its operations. However, this is beyond the scope of our study.

A new way of researching; a new way of working

Almost at the very moment that the 1998–99 World Development Report on Knowledge for Development (World Bank 1998a) (see chapters 3 and 4 of this book) was published, the British Economic and Social Research Council (ESRC) announced a new research programme, 'Future Governance', that would seek to develop new knowledge about policy processes internationally. One of its particular interests was in the way that policy ideas spread across national boundaries. In the Centre of African Studies at the University of Edinburgh we were coming to the end of a research grant and were thinking of new proposals. One of these was for a comparative study of a set of development co-operation agencies. The almost simultaneous appearance of the World Development Report and the ESRC programme led us to rework our idea somewhat. Not really yet aware of how the knowledge interests of the World Bank were beginning to spread across the bilateral community, we decided to include the knowledge-for-development focus within our overall proposal to examine what appeared to be the emergence of new ways of working amongst development co-operation agencies. It was thus, perhaps through serendipity, that we embarked on a research project that quickly became primarily focused on the discourse and practice of knowledge-based aid (King 2000).

Thus, from a very broad concern with changes in agency discourse and practice, it was through subsequent research on agencies that we came to a narrower focus on the ways in which knowledge had become an important element of this discourse and practice. In this sense, the focus of our theoretical and empirical explorations emerges primarily from its grounding in the data, although the importance of the focus could have been derived from an extrapolation of trends external to development co-operation agencies, given the emergence of accounts of knowledge economies and knowledge management.

For the purpose of the funding application, we had to select a set of agencies in order to explore in detail how the new ways of working were emerging, if that was indeed the case. Whilst both of us had knowledge of, and working relations with, a number of agencies, we did not have the current and detailed knowledge to select those agencies that would be the most interesting case studies of the new approach to aid. Nonetheless, based on the knowledge that we did have, and the practicalities of running a research project, we selected four case-study agencies. From the multilaterals, we decided to select only the World Bank, although we were very aware of the growing importance of the European Union as a major provider of official development assistance (ODA). The Bank was not only the biggest player in development co-operation but also often the trendsetter, including in knowledge-for-development. We decided on the British Department for International Development (DFID), both as an agency with which we had good contacts and as one that was undergoing a process of rapid transformation (including a name change) under the new British government. We selected the Swedish International Development Co-operation Agency (Sida) as Sweden appeared to be a leader in thinking and practice about key trends in development cooperation such as sector-wide approaches and nationally led development partnerships. Our fourth case study was the Japan International Cooperation Agency (JICA). Japan had been the largest bilateral donor throughout the 1990s in absolute terms. More importantly, however, JICA was the one major agency that came from outside the West. Thus, it offered the possibility of examining practices and discourses that emerged from a radically different cultural context.

A first phase of interviews showed that issues about knowledge and development did have resonance for all four agencies. It also importantly indicated some major divergences as well as convergences of language and activities. Crucially, this phase of research began a process of dialogue with the agencies rather than traditional data-gathering. Thus we found ourselves invited to a number of agency-oriented fora to participate in discussions about knowledge and development, and to contribute to agency development of their policies and programmes in these areas. As a result, our research began to move early on from mere observation into active participation in the construction of the field we were studying.

Knowledge theory and knowledge research

It is important to consider the effect that immersion in the literature on knowledge had for our researching of knowledge practices and discourses. Coming from a comparative education perspective, much of the current knowledge debate serves to reinforce the central importance of context within that discipline. The British pioneer of comparative education, Michael Sadler, writing more than a century ago, emphasised the danger of decontextualising knowledge through careless borrowing from other experiences:

We cannot wander at pleasure among the educational systems of the world, like a child strolling through a garden, and pick off a flower from one bush and some leaves from another, and then expect that if we stick what we have gathered into the soil at home, we shall have a living plant. (Sadler 1979 [1900]: 49)

At the end of the twentieth century, similar sentiments were to be expressed by Joe Stiglitz, then Chief Economist of the World Bank, two years prior to receiving a Nobel Prize for his work on knowledge and information: 'The overwhelming variety and complexity of human societies requires the localisation of knowledge' (Stiglitz 2000 [1999]: 7).

Thus, knowledge theory reinforced our professional training in stressing the importance of context. Comparative work faces the challenge of paying more than lip service to this notion. Of crucial importance here is the extent to which it is possible to understand multiple contexts, particularly within the constraints of a time-limited research project. In our other disciplinary home of African Studies, much of the debate in this regard centres on the issue of language. As Mbembe puts it: 'It should be noted, as far as fieldwork is concerned, that there is less and less. Knowledge of local languages, vital to any theoretical and philosophical understanding, is deemed unnecessary' (Mbembe 2001: 7). Our limitations in this regard must be noted. Whilst it can be argued that English is the language of development, particularly knowledge-for-development, it is clear that our lack of competence in Japanese or Swedish had a double impact on our research. First, it made inaccessible those parts of the 'archive' of Sida and JICA written in their national languages. Second, it limited the extent to which we could claim to understand the culturally embedded meanings behind texts and practices. None the less, the experiences of this project convince us of the worth of attempting such research. This points to the importance of acknowledging such limitations and seeking to address them honestly and openly. It became particularly important to check meanings with agency staff, academics and students from Japan and Sweden, to seek to understand perceptions of how discourses and practices were culturally embedded, and to explore with the assistance of colleagues some of what remained untranslated amongst agency documents.

There are a number of other elements of knowledge (and learning) theory that also have implications for research such as this. First, this theory emphasises the networked, social and distributed nature of knowledge. Second, work on the difference between tacit and codified knowledge points to both the importance of the former, and the difficulty of accessing it. Third, Argyris and Schon's (1978) work on organisational learning has made the important distinction between 'espoused theory', that to which an organisation is officially committed, and 'theory-in-use', that which appears to be manifested in its practices.

These point to the need to get beyond the conventional case-study blend of analysing official documents and backing this up with a series of interviews with key informants within the organisations. Rather, it raises the challenge of focusing clearly on where knowledge is inscribed in policies and practices, and exploring this further through the virtual, team-based and networked nature of the contemporary organisation. Repeat interviews, many of them with more than one informant and with different combinations of informants, often reflected the ways in which staff chose to organise and present themselves for interactions, but also shed valuable light on the fragmented and sometimes contradictory nature of discourses and practices. Meeting some of the same agency staff in interviews, at workshops within their agency and at the interagency level, and more informally in conferences, corridors and canteens enriched this process.

Argyris and Schon's work also pointed to the importance of seeing large organisations such as these as having an internal architecture, which makes the bridge between a study of the organisation-as-monolith and the organisation-as-individuals. Grounded in organisational theory as well as learning theory, their account highlights the nature of organisations as a collection of departmental or divisional fiefdoms, increasingly overlaid by a series of cross-departmental structures that often operate with close reference back to the dominant departmental model. Thus it was imperative to examine agencies also at the level of some of these structures. This took us into a variety of departments and units and into mini case studies of particular knowledge projects. Foucault (1972) has already popularised the notion of an 'archaeology of knowledge' in his study of disciplinary discourses. However, we wish to re-use the notion in a somewhat different way, which relates our study of knowledge to some of the practices of the field archaeologist. Part of the challenge for the archaeologist is to understand how the site relates to its broader context, both spatially and theoretically. The site will often be excavated through identification of potentially significant locations within it, where a series of test pits and trenches will be dug; some of these will prove unrewarding and will be quickly abandoned. This is similar to our approach. The overall picture, significance and context were essential, but we also chose to explore each individual agency through a focus on what appeared to be significant projects, activities and departments. These were each specific to the agency in question. This was a comparative project at the level of overall methodology and research questions, not at the narrow level of common and rigid research tools and interview schedules.