Indigenous knowledge: A false dawn for development theory and practice?

I Introduction

Over the last 20 years or so, interest in indigenous knowledge as a field of research and practice has moved from the periphery of development debates to a rather more central position where it has been taken seriously by some as a development intervention of promise. It has been adopted by development agencies as a ‘good thing’, not least by the World Bank (1998: 2007), as it is seen to value and hence empower the local. However, there have been frustrations and disappointments on the way, perhaps most effectively encapsulated by Sillitoe (2010: 12) when he writes: ‘[A]fter two decades or so, the indigenous knowledge (IK) in development initiative has not, frankly, had the success that some of us expected.’ There is a sense of frustration and disappointment that the promise of indigenous knowledge has not been realized. Indigenous knowledge has not succeeded in making the leap from a set of conceptual and empirical propositions to a position where it is deployed as a matter of course in development interventions, nor has it succeeded in becoming mainstream in development practice despite all the optimism.

What makes the above quotation particularly compelling is that Paul Sillitoe is one of the most experienced and well-respected researchers and writers in the field of indigenous knowledge. If Sillitoe is feeling disappointment, then something serious must surely have gone wrong. Yet there was considerable optimism among researchers in the 1990s and for much of the 2000s, and there appeared to be much about which to be optimistic (see, for example, Leach and Mearns, 1996; Reij et al., 1996; Ellen et al., 2000; Pottier et al., 2003). The transfer of Western science and technology, often uncritically, had failed to transform the lives of the majority of people in the global south, and especially so in sub-Saharan Africa. Indigenous knowledge systems, by valuing and working with local people and their own knowledge, seemed to offer a radical alternative for the successful economic transformation of people’s lives. Escobar (1995: 98) had written that ‘[t]he remaking of development must start by examining local constructions, to the extent that they are the life and history of the people, that is, the conditions for and of change’. This was in a sense a call to arms for those who wished to rethink ‘development’, a view which encouraged the promotion of indigenous knowledge at the local level as one way of addressing (rural) poverty in many parts of the global south.

It is not as though much of the writing on indigenous knowledge has been unrealistic or that it has taken a particularly romanticized view of indigenous knowledge. In truth, much, but not all, of this work developed a critically informed view of indigenous knowledge with theoretically driven interpretations of such systems which recognized the very real challenges of using indigenous knowledge in development practice (Briggs, 2005). Nonetheless, we seem to find ourselves at an important moment in time for the future of indigenous knowledge approaches in contributing to poverty reduction. But how have we found ourselves at this point, and, indeed, is Paul Sillitoe justified in feeling a sense of disappointment? It is the intention of this article to attempt to address some of these issues, firstly, by reviewing what have been the main areas of research and interest in indigenous knowledge over the last few years, before going on to suggest tentative ways of moving forward to promote the use of indigenous knowledge in development theory and practice. This is not an easy task, as there is also a sense of disillusionment in the practitioner community about the value of indigenous knowledge in poverty reduction strategies. Indeed, a senior development worker from one of the major bilateral aid agencies recently said to me: ‘You guys [proponents of indigenous knowledge] have led us to the top of the hill, and now you are leading us down again’. McFarlane (2006), in a rather different context, perhaps captures the essence of the problem when he raises the challenge of how to locate and especially draw upon local knowledge in local, national and international development policy. Scaling up becomes a problem which challenges how indigenous knowledge can be effectively deployed. But, first, how have we got into the current position?

II Recent debates in indigenous knowledge

It is not the intention of this section of the article to cover old ground in indigenous knowledge debates, but rather to attempt to take the debate forward by considering what appear to be three broad, but related, themes which have developed over the last few years, albeit in some cases from debates which have evolved from earlier discussions. The first of these, both a key challenge and, paradoxically, a strength at the same time, of indigenous knowledge is that it is invariably locally and geographically specific, something which has been recognized in the literature for some considerable time (see, for example, Agrawal, 1995). There has been, however, a trend to continue to build up our repertoire of examples of good practice in indigenous knowledge from as wide a range as possible of different geographical, environmental, economic and cultural settings as possible. As this repertoire of good practice builds up, so potentially, it seems, do the opportunities to transfer examples of good practice across geographic space into quite different economic, environmental and cultural settings. After all, Western scientific method has done precisely that, albeit with marked uneven development consequences.

There continues to be a vast amount of work undertaken on such locally specific situations, much of it very scholarly and rigorously researched. For example, Mapinduzi et al. (2003) in their study of ecological knowledge in Tanzania have shown the acute awareness of Masai cattle-herders of the effects of grazing and cropping on local ecologies. On the evidence base of their work, they were able to make a case for such indigenous methods of landscape evaluation being the starting-point for a more subtle assessment of rangeland biodiversity. In the different setting of the Ethiopian highlands, Mekoya et al. (2008) have drawn attention to the differences in use by farmers between local and exotic species of fodder trees. Even though the introduced exotics are more productive, local varieties are preferred because they are multifunctional, have a longer lifespan and are rather more compatible with the local cropping system. What is important here is to recognize that these farmers make choices not just on feed value but on a whole mix of multiple objectives. Also in Ethiopia, Watson (2009) has traced the ways in which indigenous terrace landscapes have evolved, highlighting how these processes are deeply embedded in the broader Konso political and cultural landscapes. In a south-east Asian context, Rerkasem et al. (2009) have shown the ways in which local indigenous knowledge has contributed to improving forest management and in promoting biodiversity conservation, by constantly adapting to and managing the challenges associated with the tensions between forestry and cropping practices.

These are examples, among many, of high-quality studies very much at the local level which make compelling cases for the value of indigenous knowledge within these particular geographic settings. However, the development practitioner, unless he/she is involved in a development intervention in these specific areas, has difficulty in recognizing their utility. The very reasonable arguments that these studies make important contributions to the depth of local understandings, or that they expand our repertoire of knowledge for future exchanges of such knowledge over geographic space, cut little ice with development practitioners who seek effective, and preferably immediate, interventions and solutions for poverty reduction. This, indeed, speaks to some of the disappointments and frustrations being felt by proponents of indigenous knowledge, as it is unclear to many development practitioners how such indigenous knowledge can inform development practice and interventions beyond the immediate area in which the research was undertaken. However scholarly and careful the individual piece of research, there is still left the challenge of how such knowledge can be used more widely, beyond that immediate study area. For development practitioners, it still leaves unanswered the question of how to ‘scale up’ effectively. The enduring, perhaps intractable, challenge is how to use ideas, concepts, thinking and results derived from one specific local area in another, perhaps quite different, area. There is a clear problem of universality here, or, to put it in development practice terms, and to use one of the practitioners’ current buzz-phrases, the challenge of scaling up from the specific ‘local’ to the more general ‘regional/national’. Without being able to do that, indigenous knowledge may well remain at the periphery of practice, and continue to disappoint.

A second broad theme in the research agenda over the last decade or so has been work on the ways in which indigenous knowledge might be usefully and purposively integrated with formal science to produce sets of hybrid knowledge, based on the premise that farmers use knowledge which is useful to them, regardless of its provenance (Briggs et al., 2007). There were many pointers to this in the 1990s, but it is really only in more recent years that there has been a much more deliberate and systematic attempt to make the link. A significant amount of this work has been in the field of ethnopedology, a research field situated at the interface of the natural and social sciences (Barrera-Bassols et al., 2006), and which places emphasis on the holistic nature of such hybrid knowledge, a view which chimes closely with that expressed by Berkes and Berkes (2009). There is a serious attempt in such work to value indigenous knowledge as an equal to formal scientific knowledge, and it is probably no accident that it is in the area of soils where this has been most fruitful, in that soils and their use are an essential part of everyday practice for farmers in the global south. However, there have also been links made between indigenous knowledge and formal Western science in the areas of range management (Homann et al., 2008; Thomas and Twyman, 2004; Walker et al., 1999), branching off into ethnobotany (Blanckaert et al., 2007; Lado, 2004); of more effective coastal systems management (Foale, 2006); and of water and climate more generally (Liwenga, 2008; Marin, 2010; Weatherhead et al., 2010).

However, the challenge of integrating the two ways of making sense of the world remains a concern for many (see, for example, Homann et al., 2008; Liwenga, 2008; Mercer et al., 2009), but it is nonetheless seen to be a worthwhile challenge. To support this, Lado (2004: 281) calls for an interweaving of modern and indigenous knowledge ‘to produce a more realistic and sensitive understanding and management of natural resources for sustainable development’, and active attempts have been made to find a way forward. Mercer et al. (2010), for example, in wishing to develop a framework for reducing community vulnerability to environmental hazards in Papua New Guinea, suggested that a four-step approach might be developed along the lines of (a) community engagement; (b) identification of vulnerability factors; (c) identification of indigenous and scientific strategies; and (d) the development of an integrated strategy. Whilst this offers a realistic and interesting potential, it is still rather silent on a number of matters. How will members of the community engage? How will the power relations play out, both within the community and between the community members and the science ‘experts’? And how will that integration, so crucial to Step d, really be achieved? In the different environmental context of the Kalahari Desert in Botswana, Reed et al. (2007) attempt to show how local and scientific knowledge can be combined to develop strategies to reduce land degradation, or, at least, to adapt to it in meaningful ways. Like Mercer et al. (2010), they also suggest a four-stage approach involving a desk literature search (the science input); semi-structured interviews with stakeholders (the indigenous knowledge input); focus group discussions with stakeholders on the various scientific and indigenous options; and, finally, the production of assessment guides from these discussions to provide locally relevant management options. Reed et al. (2007) come to the conclusion that this combination of knowledge and the involvement of local stakeholders produces more relevant results than either knowledge system could have done separately.

The above examples, and many others, of this integrative kind of work potentially seem to offer a way of dealing with the location-specific nature of much, if not all, indigenous knowledge, in that this approach seemingly reduces, or dilutes, the location-specific nature of indigenous knowledge by drawing on the greater universality of formal scientific knowledge which are much less constrained, it seems, by being location-specific. There is an attraction of adapting and integrating best practice and understanding from both traditions, conveniently uniting the assumed rigour of formal science with the everyday realities of indigenous knowledge at the local level. Nonetheless, however encouraging the results of this bank of work may appear, there is still the troubling issue of power relations, and, in this case, not just within and between various stakeholder groups, but also between the two knowledge systems themselves (see Briggs and Sharp, 2004). There is a sense in two of the examples above that formal science and indigenous knowledge are on the agenda as equals, but this is rarely the case in reality. Indeed, there is a body of literature which tends to reinforce this unequal power relationship by focusing on the validation of indigenous knowledge by formal science, as a means of legitimizing the former. By doing so, of course, this privileges the latter. Indeed, there is a sense that the validation of indigenous knowledge by science is often seen by planners as crucial (Fairhead and Scoones, 2005), or, as Jacobson and Stephens (2009, 159) put it, ‘[e]ven when it is recognized that local knowledge is insightful, an expectation exists that it must be contrasted against science for it to be considered true’. There still exists that pervasive sense that indigenous knowledge cannot be trusted, or even valued, unless it is given the stamp of approval by formal science.

In supporting this line of thinking, there has been no shortage of recent research which seeks to validate indigenous knowledge by an authorizing Western science, and this again has especially been prevalent in the context of pedology (see, for example, Hillyer et al., 2006; Mairura et al., 2008; Niemeijer and Muzzacato, 2003; Saito et al., 2006; Schuler et al., 2006; Trung et al., 2008; Vigiak et al., 2005). It would seem, although often quite implicitly, that indigenous knowledge can only be taken seriously if it has the approval of science. Verlinden and Dayot (2005), for example, point to the scepticism of scientists about indigenous knowledge which consequently makes it ‘difficult to institutionalize the approach [indigenous knowledge]’ (Verlinden and Dayot, 2005: 166). But where such indigenous knowledge of soils can be shown to have scientific ‘rigour’ to support it, such as by standard scientific measures of soil quality, including pH, available nitrogen, total organic carbon (Mairura et al., 2008; Saito et al., 2006), or by the evaluation of land units (Hillyer et al., 2006; Schuler et al., 2006), then suddenly indigenous knowledge has currency. Interestingly, this does not appear to work the other way round.

Such legitimization by formal science may be a useful tactical device to give indigenous knowledge greater currency among both development academics/researchers and development practitioners, but it does therefore inevitably devalue it as a knowledge system in its own right. Sillitoe (2006) has expressed unease about this, as he takes the view that the prioritization of Western science can distort understanding, by producing what he terms an ill-informed and decontextualized knowledge, something which is ultimately unhelpful, as it ignores local social, cultural and even economic priorities. This apparent lack of confidence of proponents in indigenous knowledge itself, as evidenced by the desire by many to legitimize it through formal Western science, has not ultimately helped the cause, and may, surreptitiously, have contributed to the lack of willingness to use indigenous knowledge in development practice. Formal scientific understandings seemingly still hold the cards.

A third recent theme over the last decade is that of what might be seen as the co-option or appropriation of indigenous knowledge into the prevailing discourse of neoliberalism. At one level, this may seem to be a logical next step if indigenous knowledge is indeed to contribute in a meaningful way to poverty reduction; indeed, the World Bank has made several attempts to incorporate indigenous knowledge into its approaches (see, for example, World Bank, 1998; 2007), perhaps what might even be seen as a ‘professionalisation’ of indigenous knowledge. If indigenous knowledge can be co-opted in this way, then this surely is to its advantage as it becomes part of the development agencies’ armouries in development interventions. However, as we have already seen, indigenous knowledge is still widely perceived as a poor relation of formal Western scientific knowledge, and is rarely seen as an equal partner. But this is only part of the problem. The value of indigenous knowledge is that it is indeed culturally and economically situated within the communities in which it is found, as it is produced through practices which are spatially and materially relational (McFarlane, 2006). The danger of a co-option of indigenous knowledge is that this importantly grounded nature of indigenous knowledge will be lost in the desire to generate a universalizing knowledge which is aspatial, in the sense of having lost the ‘local’, and therefore having a more general applicability, with a capability of being ‘scaled up’ from the local to the regional/national and even beyond. The irony is that one of the very strengths of indigenous knowledge, its groundedness in the local, can become significantly reduced.

However, as well as a means of more effectively reducing poverty, there have been suggestions of other reasons for the co-option of indigenous knowledge. One such argument is that the commoditization of such knowledge can be seen to serve the needs of capitalism and neoliberalism (Busingye and Keim, 2009; Laurie et al., 2005; Nightingale, 2005). Following this argument, indigenous knowledge runs the risk of following a path similar to that of modern scientific knowledge in the sense that its acceptance has less to do with its indigenous scientific power, and more to do with how it can serve the needs of capitalism. This has been argued particularly in the context of intellectual property rights and patents, and especially so in regard to commercially valuable knowledge (Rao, 2006), or what Oguamanam (2008) has termed a bioprospecting framework. For example, this can be seen in the context of medicinal plants, where it has been argued that asymmetrical power relations between ‘sophisticated external intermediaries and custodians of local knowledge’ (Oguamanam, 2008: 29) can represent a real threat to the economic security and wider aspirations of an indigenous people. Vermeylen (2008) has shown how San communities in South Africa have tried to come to terms with such a commodification of their knowledge of medicinal plants, with some arguing that it can be a liberating act and others that it is little more than a pragmatic choice driven by economic hardship. Either way, this has changed San cultural meanings of those plants now that they have been ‘captured’ as commodities (Vermeylen, 2008).

Central to all this, of course, is the question of power relations and the extent to which the co-option of indigenous knowledge is little more than a regularization of both knowledge and social relations in the interests of the market (Laurie et al., 2005). Indigenous knowledge, therefore, can be useful, as long as it fulfils this function, and, following this argument, the legitimization of indigenous knowledge by formal science therefore becomes immaterial. And that it is where strength of indigenous knowledge apparently lies in the power game. But what indigenous knowledge must not do is threaten the development process which seeks to produce ‘people who live like “ethnic” majorities or dominant groups who define national culture’ (Marschke et al., 2008: 486). It would seem, therefore, that any co-option of indigenous knowledge can only go so far, in that it must remain compliant with the existing dominant discourse of neoliberalism, and it cannot be seen ultimately to threaten the dominance of modern science which both underpins that discourse and the nature of the current development ‘industry’.

So, in the context of these three recent research themes, where does this now leave us? Despite the wealth of well-researched and convincing material developed over the last two decades or so at the local scale, the current status of indigenous knowledge may be characterized as something that is seen to be interesting, curious and perhaps even quirky. But it can also be seen to be desirable in development interventions at the community level, at least in some cases, but without necessarily having a clear understanding as to how indigenous knowledge might be reasonably used as a poverty reduction strategy more broadly. Even the legitimization of indigenous knowledge by science does not appear to have done the trick, and the co-option of indigenous knowledge, even if only seemingly half-hearted, still does not see it as an equal to Western science. Indigenous knowledge remains the Cinderella of development.

III Where now indigenous knowledge?

Despite all the continuing excellent and scholarly work being undertaken on indigenous knowledge systems, there remains still a pervading sense of disappointment and frustration, best expressed by Sillitoe (2010), and which worries many indigenous knowledge researchers, that ‘the indigenous knowledge in development initiative’ has not succeeded, that indigenous knowledge research has indeed been nothing more than a false dawn for development theory and practice. It is therefore the intention of the remainder of this article to offer some tentative suggestions as to how indigenous knowledge research might move forward beyond these frustrations, but suggestions which require a change in how we view knowledge and development interventions more generally.

In an interesting contribution, Berkes (2009a) may be moving us helpfully in this direction by arguing that traditional knowledge, as he calls it, should focus less on its continuing interest on content, and instead focus rather on process. In this view, whereas content emphasizes information that can be passed from one person to another, process puts the emphasis on ways of observing, discussing, questioning, analyzing and making sense of information, whether it is new or received. That is, the focus now becomes indigenous ways of knowing, with a sharper focus on the epistemology of indigenous knowledge systems. In a sense, this is not new; Briggs and Sharp (2004) called for just such a focus in an attempt to break indigenous knowledge out of its content constraints. This is not to deny the importance of information and knowledge, but if deploying indigenous knowledge systems is to take its place at the centre of development practice then it is essential that it extends beyond the culturally and spatially bound restrictions of current indigenous knowledge thinking. There is a danger here, of course, if we are not careful, because, as Sillitoe (2010) reminds us, indigenous knowledge is not globally situated, it is not culturally disembedded, nor is it overly systematic or abstract. It is very much rooted in the here and now, and within particular cultural, economic, social and environmental settings. However, in their work on indigenous technical knowledge among Jamaican farmers, Barker and Beckford (2006) identify ways in which these farmers adopt a particularly dynamic way of evaluating and using both indigenous and external knowledge and innovations. Although the precise knowledge itself is specific to yam farmers in Jamaica, the authors identify the nature of the process which is informed by economic advantage to the farmers themselves. But in making this shift from a focus on content to one on epistemology, there emerge significant tensions and challenges.

The problem has been that we may have spent rather too much time focusing in one direction, content, to the relative neglect of another, process or epistemology. Anxious to have indigenous knowledge accepted as an equal to formal science, much work has tried to do just this – and clearly for the right reasons. If indigenous knowledge is to be taken seriously by the development practitioner community, then it apparently needs to be the equal of formal science with the same universalizing logic. However, the challenge remains that indigenous knowledge cannot be easily extracted from its local context, as we have been forcefully reminded on many occasions (see, for example, Agrawal, 1995; Davis, 2005; Sillitoe, 2004, 2006; Sillitoe and Marzano, 2009; Briggs et al., 2007). Indeed, as Berkes (2009a: 154) starkly puts it: ‘Not taking knowledge out of its cultural context is one of the biggest challenges of indigenous knowledge research’. Following this line, it can be argued that the wealth of research on spatially specific indigenous knowledge systems has fulfilled a vitally important function in determining both the validity and value of indigenous knowledge as an alternative way of knowing. This, however, has necessarily been within particular cultural contexts which have made the wider, universalizing applicability of such knowledge so problematic, and this is where Berkes’ (2009a) conceptual distinction between content and process becomes helpful. However, the use of ‘process’ as a term may still imply a degree of abstraction which may be unhelpful to the practitioner, perhaps suggesting in the minds of some a degree of separation of the thought processes from the actual outcomes of indigenous knowledge. In this context, the term ‘practice’ might be more appropriate in that this implies an indigenous knowledge that is grounded and rooted in a particular context and is a clearly integral part of the everyday practice of production. It also makes it clear that indigenous knowledge is indeed embedded within particular contexts. Ferguson et al. (2010) suggest that there exists what they term a latent knowledge in communities which sees knowledge not as a tool, but rather more as a perspective on development, as a way of knowing, and importantly recognizing that this happens within particular contexts. This line of reasoning shifts the focus of indigenous knowledge research from a focus on content to a focus more on practice.

Nonetheless, this still leaves the unresolved issue as to how development practitioners are going to make use of indigenous knowledge in their own everyday activities, and especially so in the light of the above if we accept indigenous knowledge as being spatially-bounded, with a wider applicability beyond the local context being difficult at best. However, this misses the point in that a culturally-embedded indigenous knowledge provides possibilities for new ways of working, new ways of thinking and new ways of making development interventions, and hence new ways in which the relationships between development practitioners and members of communities have to be re-thought. Berkes and Berkes (2009) do not argue for indigenous knowledge to replace scientific measurements, but for it to make additional enriching contributions to environmental understandings, by providing locally-derived insights at the local or micro-scale. Hence, it can help in the formulation of relevant research questions which have a particular resonance and relevance within and for the communities. They argue further that it is the diversity of the two rather different knowledge systems which bring strength to the debates, and that there is in fact little to be gained from working towards developing formal hybrid knowledge systems. However, to be successful, such debates have to be conducted in an atmosphere of mutual respect at the local level, where indigenous knowledge is no longer seen as ‘non-knowledge’ (Nygren, 1999: 271), and where indigenous knowledge can be seen to enrich and broaden current scientific understandings of scientific technology (Sillitoe, 2004). This type of approach foregrounds the local cultural, economic and environmental contexts, and creates the sense of indigenous knowledge being an integrated part of everyday practice at the local level (Lauer and Aswani, 2009; Parrotta and Agnoletti, 2007), the opposite of a disembodied, aspatial and universalizing Western scientific knowledge.

There remains the further challenge, however, of how to effect a change in the mindset of development practitioners, many of whom remain committed, quite understandably, to the hegemony of Western science and technology as a means of reducing poverty. As a way of achieving this, Berkes (2009b) argues perfectly reasonably for the establishment of co-management comprising a partnership of equals between local communities and government, recognizing that there are multiple actors – importantly, there is neither a unitary state nor a homogenous community, which creates yet another challenge in the sense of understanding the power relations not only between these two groups, but also within them. Indeed, there is an argument that real progress in poverty reduction can be achieved only when people from communities in the South fully participate as both contributors as well as users (Sen, 2005). With regard to such engagements, Wilson (2007) reminds us of the importance of ‘learning with’, rather than ‘learning from’, and if this is undertaken in an active spirit of mutual engagement, this can promote the expansion of the boundaries of what is already known. However, for such outcomes to become a reality, there need to be developed novel and imaginative ways of engagement between all stakeholders, with such engagements being central and rooted in right from the start of any development intervention. Indigenous knowledge, if it is to be effective, cannot simply be a convenient ‘add-on’, to which only lip-service is paid.

This is of course easy to say, but the problem is that this change of mindset involves the development ‘expert’ conceding power, because, as Laurie et al. (2005: 477) put it: ‘in contemporary development paradigms local specialist knowledge detracts from development expertise’, and conceding power is never easy. Consequently, a re-ordering of power relations becomes central to the debate and the challenge becomes one of convincing experts to think in new ways, and in so doing, to concede some, or perhaps even a large element, of their power (MacKinnon, 2006). However, this may raise the danger, as discussed earlier in this article, of the co-option of indigenous knowledge where the development practitioner community subsumes indigenous knowledge, which, in turn, becomes part of the received wisdom of development knowledge, but with little regard for local sociocultural and economic realities. The power and seduction of a universalizing knowledge remains.

To add to the mix, there is the additional danger of the ‘quick-fix’ and target mentality of many development agencies which is not necessarily consistent with the longer-term process of understanding the local sociocultural context within which a particular development intervention is to take place. Indigenous knowledge requires grounded, time-consuming work to understand the logic and reality of every day practices of communities at the local level, the level at which development actually means something (Escobar, 1995). But this deep, ‘slow-burn’ approach is something which sits uneasily with much contemporary development practice. The development ‘industry’ is driven by short-term, measurable outcomes to demonstrate effectiveness and value for money, or, as Sillitoe and Marzano (2009: 15) capture it, the use of indigenous knowledge ‘presents a problem in development contexts with politically driven, short-term demands for quick results’. This leads to inevitable tensions between a results-driven approach to development, as promoted by politically driven development agencies, and the rather more ‘slow-burn’ but deeper research implicit in indigenous knowledge research. This is compounded further by the emphasis on the demands for scaling-up by many development agencies, seeking generic solutions to development problems which have a universalizing capability. As we have seen, indigenous knowledge, as currently understood and practiced, does not naturally lend itself to this approach.

IV Some final thoughts

There is little doubt that the effective deployment of indigenous knowledge in development practice is problematic, and that it consequently may indeed represent a false dawn for development theory and practice. This may have come about for several reasons, perhaps because indigenous knowledge is indeed ‘heterogeneous and complicated which is an inconvenience for development’ (Sillitoe and Marzano, 2009: 14), as well as it being place-specific creating challenges for scaling-up. Its use also requires long-term engagement with communities, and the inclusion of members of such communities as equal participants in the research and development process from inception through to implementation. Hence, if indigenous knowledge is to contribute to successful development interventions, then it cannot be a last-minute add-on, however well-meaning. It must be central, grounded and integrated into the intervention programme from the start. However, as we have seen, this is not always a realistic proposition for development agencies with political pressure from above to deliver positive outcomes, preferably quantifiable ones, in the shortest time possible.

However, such finger-pointing only takes us so far. To address Sillitoe’s (2010) disappointment, the indigenous knowledge research community for its part needs to re-assess where it is going. For good reasons, there has been a strong focus on content in indigenous knowledge research. Without doubt, this has fulfilled a vital function by demonstrating the importance and relevance of indigenous knowledge repertoires, and the ways in which local people in communities have been able to develop and use such knowledge in their everyday practice. In the literature, including the ‘grey’ literature, an impressive and convincing library of case studies of indigenous knowledge in practice has been built up, such that few people can now realistically question the legitimacy, value and authenticity of local people’s indigenous knowledge systems. This work has been, and will continue to be, important to ensure that the legitimacy of indigenous knowledge continues to be recognized as an alternative way of knowing.

However, it is also time to move on. Earlier in the article, there was a call for a major change in how we, both researchers and practitioners, look at the relationship between knowledge and development interventions. Following this, a future focus for indigenous knowledge research might be one that moves away from research on continuing to build up local knowledge repertoires, on legitimizing indigenous knowledge by formal Western science, even on revisiting binary tensions between the two knowledge systems, and so on, to one where the focus is indigenous knowledge as process, not content, where communities become increasingly empowered through self-confidence, derived from their own knowledge, to inform development interventions which directly affect them in their everyday lives. This is not to discourage continuing work on content, indeed far from it. Rather, it is to recognize that indigenous knowledge may have reached a position where, if it is to have a wider impact on development practice and poverty reduction, different ways of thinking have to evolve. The focus on process, or practice, as outlined above may offer us a way forward, to develop our own knowledge about indigenous ways of knowing, to understand better the complex power relations associated with knowledge at the local level, to think about ways in which the power which currently exists, both implicitly and explicitly, in formal science and technology knowledge and how these can be negotiated, and how individuals at the community level can be involved in the process, where the emphasis is on empowerment without the naiveties of the participation era (there is not the space here to discuss these at length, but see Cooke and Kothari (2001) for a full discussion, and Hickey and Mohan (2005) for more nuanced interpretations of the complexities of participation in development). This is a challenging agenda, but one which, if successfully addressed, may result in us not looking at a false dawn after all, and for indigenous knowledge to make a meaningful contribution to poverty reduction strategies more generally.