Introduction

On a recent trip to the Sida (Swedish International Development Agency) office in Stockholm I was prompted to reflect on the relationship between development and history. My visit, which was to present a piece of research on the local power structure in Bangladesh, took place a few days after Sida had moved to its new offices. There were boxes of documents piled everywhere, and the place was in an understandable state of disarray as this massive and complex logistical task was being completed. People that I met were full of stories about the move and the various difficulties it had entailed, but one comment in particular struck me as interesting: faced with the enormous task of clearing through the years of papers and documents in their offices, people had been advised that they could discard anything that was more than 2 years old, and only needed to keep the rest. Although I am sure that this was not a case of Sida jettisoning decades of institutional memory and learning about development (I assume that a centralized library keeps track of all such material somewhere), it did strike me as symptomatic of a wider tendency within many development agencies to live in what I term the 'perpetual present'. This is a state characterized by an abundance of frequently changing language and 'buzz-words' (cf. Cornwall and Brock, 2005), by frequent discussion of new approaches that promise better chances of success than those currently in use, and by a strong – and in many ways understandable – sense of wanting to look forward rather than back.

Anthropological analyses of policy processes have been relatively new and far between, but arguably have much to offer development (Shore and Wright, 1997; Lewis, 2008a, 2008b).¹ My main argument in this paper is that anthropologists of development are particularly well positioned to restore a stronger sense of development history than is currently apparent within the constantly changing ideas, knowledge and practices of the aid industry. An anthropological approach can usefully anchor international development policy and practice more firmly within wider histories both at the levels of institutions and individuals. One of the key elements of neo-liberal policy orthodoxy that has increasingly come to dominate thinking about development – as it has many other aspects of social and economic life – is a tendency to insist on what is at best a limited sense of historical perspective, and at its worst an active suppression of historical depth and distance. This idea is central, for instance, to Scott's (1998, p. 95) analysis of 'high modernism': The past is an impediment, a history that must be transcended; the present is the platform for launching plans for a better future.

This ahistorical tendency operates at the level of the ideas and concepts that are used within development policy, and within the broader contexts into which project interventions are introduced. At a micro-level, it also obscures the individual life histories that can tell us a great deal about the people and practices involved in development processes, and the factors that shape people's room for manoeuvre within these broader institutions and structures.

History, Anthropology and Development

There has long been work undertaken by anthropologists that has tried to contextualize development projects within history and politics, drawing attention to the ways in which project documents may over-focus on the present at the expense of wider context, politics and history.² For example, a classic study of this kind was Barnett's (1977) analysis of the Gezira land-leasing scheme in Sudan that was introduced by the British in the 1920s. The project aimed to control local labour and secure cotton exports, taking a largely top-down approach that failed to engage with historical factors that helped determine the incentives of producers. This tradition of the anthropological project critique has remained a small but popular theme within the anthropology of development. For example, more recently, Wrangham (2004, p. 100) has shown how a Department for International Development (DFID) rural development project document in Mozambique contained just 17 references to 'a non-project related past' in over 132 detailed pages. This, she argues, reflected a project design that paid insufficient attention to three important sets of historical factors: the significance of the still-recent civil war, the ways in which local communities were likely to perceive outsiders given their earlier encounters, and finally the 'long-standing diversity of rural livelihoods'. The result was that ignoring history led to the agricultural component of ZADP being designed with inappropriate aims and objectives, and with an apolitical grasp of the reasons for rural poverty ... (p. 124)

Drawing on the work of Isaacman (1997) on the production of 'historical amnesia', Wrangham's study shows the ways in which dominant project staff became overly focused on imagined views of the future, which then served to obscure important historical and political continuities with the past.³

Another highly relevant field is that of the excavation of the archaeology of ideas about development, which has become a lively area of scholarship and debate within development studies. The landmark work of Cowen and Shenton (1995, p. 34, 1996) traced the Enlightenment roots of the concept of development through to the present day. The authors argued that the Saint Simonian tradition of the 1820s served to remove ideas about development from the realm of history to become 'the means whereby the present epoch might be transformed into another superior order through the actions of those who were entrusted with the future of society'. Despite this interest, there are many who remain critical of the inability of development theorists and practitioners to look back beyond the idea that development began after Second World War and engage further with colonial histories (Kothari, 2005). In a recent review of the current state of the discipline, Corbridge (2007, p. 202) suggests that 'there seems to be a reluctance for people within the discipline of development studies to examine the history and present condition of the forms of knowledge to which they may be committed (knowingly or otherwise)'.

Perhaps it is not so surprising then that history is downplayed or ignored within development agencies. In addition to these wider ahistorical tendencies, there are sets of additional pressures and certain distinctive organizational characteristics of development agencies that may also contribute to the problem. For example, the use of expatriate staff by Western agencies often involves relatively short-time frames which are based around 2 or 3 year postings, making it difficult for individual knowledge and experience to be documented, contained or distilled. At the same time, the bureaucratic logic of an office or department may create powerful incentives for a new appointee to a particular position to show their effectiveness by deliberately down-playing what went on before they arrived in post, as a way to demonstrate their own particular 'added value'. The result may be the unnecessary development of new initiatives, terms and approaches in a process that further contributes to the suppression of past experiences and restricted learning.

Such practices form part of the wider infusion of ideologies of managerialism into the organized worlds of development (Roberts et al, 2005). These bring a tendency to discourage engagement with the very ideas of past and present through a relentless emphasis on novelty and change.⁴ In so doing, the world of development practice becomes further decontextualized as Cornwall and Brock (2005, p. 13) show in their analysis of the ways in which development 'buzzwords' such as empowerment, participation and poverty reduction each function to help maintain what they term 'the imagined, decontextualised world of the consensus narrative'.

In the 1990s, critical work by anthropologists influenced by Foucault's work on power and discourse has also taken a more historical approach to the analysis of development policy and practice. For example, Ferguson (1990) famously described the way the World Bank's livestock project in Lesotho constructed a version of agrarian relations that downplayed and distorted important elements of the past, thereby depoliticizing ideas about development. Hobart (1993) concerned himself with histories of the construction of knowledge about development, and the consequent and equally constructed growth of ignorance as a counterpart outcome. Escobar's (1995) analysis, which has been extremely influential, was concerned with the growth of the ideology of 'developmentalism' through which Western interests after the Second World War constructed the idea of the third world as something to be known and acted upon.

Although post-structuralist anthropological accounts such as those written by Ferguson and Escobar have been highly influential and instructive, they have attracted criticism for taking a monolithic view of a single dominant history of development. As Grillo (1997, p. 20) pointed out in a critique of what he calls 'the myth of development' and its unitary 'all-powerful' character (along with that of the state), such accounts have been somewhat limited in their historical vision: Like most myths it is based on poor or partial history, betraying a lack of knowledge of both colonialism and decentralisation ... [and] ill-informed about the history of government ... (pp. 20–21)

Being in touch with the past does not imply that for those working in development looking back is more important than looking forward, or that 'there is nothing new under the sun'. Instead, the point is that a stronger engagement with historical perspectives can provide a much clearer view of the political field in which development thinking is situated, and the constraints that this implies.

What follows from this contextualization is then the possibility of providing the development practitioner with a firmer grasp of what is and is not possible, and the importance of longer-term thinking in relation to policy. This contextualization has two distinct but inter-related dimensions: firstly, the need to locate international development activities more fully within historical and political contexts, and to argue more forcefully for the reflection of such a perspective in the design of activities and policies; and secondly, the need for development practitioners to operate with as clear as possible an awareness of the histories of the ideas, concepts and experiences that they themselves are engaged in deploying through their work.

Anthropology and Development Policy

The relationship between anthropology and the ideas and practices of development dates from the earliest days of the discipline during the colonial period (Gardner and Lewis, 1996). It has taken various forms since that time. It became common in the 1980s to make a distinction between the 'development anthropologists' who worked broadly within the agendas of the development agencies in their applied research or advocacy of particular policies, and those 'anthropologists of development' who set themselves the challenges of analysing the meanings and effects of the ideas and practices of development itself, often taking up a highly critical stance that questioned its impacts, values and purposes (Grillo and Rew, 1985). Through this dual approach, anthropologists generally adopted either a position of sympathetic involvement with development practice, or the more disengaged stance of critique of or hostility to development.

Unpacking these positions further, Lewis (2005b) suggested that these stances can be more accurately characterized as three different – but sometimes inter-related – roles: anthropologists as 'engaged activists', as 'reluctant participants', or as 'antagonistic observers'. For 'engaged activists', the role of an applied anthropology of development – as with an applied anthropology of other fields such as business or media – is seen as relatively unproblematic. The field of applied anthropology, defined as the use of anthropological methods and ideas in practical or policy contexts, has involved collaboration with activists, policy makers and professionals. Anthropology came to be seen as a tool which potentially provided the means to understand, and therefore to some extent control, people's behaviour, either as beneficiaries, employees or customers. For example, in the Office of Indian Affairs in the United States anthropologists have contributed their research findings to policy makers on many issues such as local customs, dispute settlement and land rights. The gradual professionalization of the development industry from the 1970s onwards led to a growth of opportunities for anthropologists to work within development agencies as staff or consultants, just as anthropologists also took up jobs within fields as diverse as community work and corporate personnel departments. In this role, anthropologists often worked to interpret local realities for administrators and planners (Gardner and Lewis, 1996).

The second role, that of the 'reluctant participants' sometimes overlapped with the first, but resulted from somewhat different pressures. This was the contraction of academic resources in the United Kingdom and other countries during the 1980s, along with a growth of opportunities to work as contracted policy advisers and consultants. Some anthropologists seeking to work within academic settings were drawn into more applied consultancy work within the development field – often as a condition of their employment – in order to generate more resources for their cash-strapped university departments. Others left academia for more lucrative employment working full-time for development agencies.

Finally, there has always been a high-profile group of 'antagonistic observers' within the anthropology community. At one level, such work has flowed seamlessly from many anthropologists' longstanding concerns with the social and cultural effects of economic change in less developed areas. In this type of work, development agencies are frequently seen as the modernizing enemies of local communities and cultures, a view that has also been echoed during the 1990s among the so-called 'post-development' school of thinkers. More recently there are anthropologists, such as Ferguson (1990), who have selected the ideas, processes and institutions of development as part of their field of study, but have taken a distanced critical approach rather than one that attempts to engage directly with development agencies themselves.

Within this framework, there are different individual motivations for forms of anthropological engagement: those who engage with development instrumentally, seeking to play a normative role from a range of different positions within development agencies, such as Michael Cernea's work on resettlement within the World Bank; those who engage with development ideologically through forms of participation that stress localism, such as Robert Chambers; and those whose motivation has, as in the case of Escobar and Ferguson's work, been primarily critical and deconstructive (Mosse and Lewis, 2006, p. 2).

It is now more important than ever to dissolve and move beyond such distinctions for the future. The inter-mingling of the three stances outlined here, and the illusory nature of the belief that one can separate anthropological work 'on' and 'in' development, requires us to travel beyond the dualist position that distinguishes between applied and non-applied categories. As Harrison and Crewe (1998) suggest, boundaries between development anthropologists and anthropologists of development no longer hold firm under criticism of their artificiality. Binary thinking obscures the positioning of all research work within a powerful set of organizing ideas about development: what Ferguson (1990) termed the dominant 'interpretive grid' of development discourse.

The 1990s brought a strong revival of interest in development and anthropology, and today there are signs of a shift into forms of engagement that move beyond these simple distinctions towards ones which can reveal more of the organizations and agencies of development, along with a deeper analysis of the ways in which ideas about development have come to play a central role in our lives. As these old boundaries have broken down, new and more nuanced perspectives on development are emerging. For example, Olivier de Sardan's (2005) work draws attention to the ways in which the study of increasingly decentralized and localized practices by development actors reflect their broader roles within national and international political economy. Mosse and Lewis's (2006) explore the ways in which the ethnography of micro-level processes of aid 'brokerage' can illuminate the construction of development meanings and representations alongside functional development roles. Anthropological understandings of development can therefore produce new insights and analyses within the shifting landscapes of development ideas, institutions and activities.

Engaging with Development Histories

In the following section we consider some areas of ideas and practices which illustrate the value of an anthropologically 're-historicised' approach to development. Quarles Van Ufford et al (2003) suggest three main ways in which the idea of development has been conceived. It has been characterized as 'hope' in that it carries ideas about shaping a better future, as 'administration' in that it has since the 1950s amassed a range of agencies and technologies designed to produce it, and as 'critical understanding' because it constitutes a site of knowledge about the world. Yet despite these multiple relatively open meanings, it has gradually narrowed so that development has increasingly come to be defined simply as a management problem: The new creed of market and result-orientation has emptied itself of the earlier historical awareness of development as an initiative in global responsibility first expressed in the agenda of global development which emerged towards the end of the Second World War. (p. 8)

Development agencies have increasingly replaced the wider, more open-ended goals and aspirations of development with a stronger focus on results and 'manageability'. This is one of the areas of 'disjuncture' identified by Lewis and Mosse (2006) that forms part of a growing set of tensions and contradictions around development ideas and aims, and development practices. A more explicitly historical anthropological perspective can therefore challenge this increasingly impoverished view of development. It can do this at two inter-related levels: (i) in relation to the histories and contexts in which development interventions are constructed, and (ii) in relation to the historical trajectories of ideas, organizations and people within policy processes.

In the first section below, I will look at the way in which an understanding of the history of rapidly shifting development knowledge and ideas is important, using the example of ideas about non-governmental organizations (NGOs) and civil society. In the second, I move on to consider development history at the level of people and practices, and summarize briefly some recent research on the individual 'life histories' of managers and activists who cross between civil society and state during their careers.

Knowledge and ideas: NGOs and civil society

The rapid and largely non-reflexive growth of interest in ideas about 'civil society' and 'NGOs' among development agencies during the 1990s is one illustration of the rapidly changing fashions within development thinking (Lewis, 2005a). Although civil society was a concept several centuries old, and various types of NGOs had long been active in many parts of the world, neither subject had had a profile within development policy or development studies during the 1980s and earlier.⁵ In their different ways, both of these ideas then quickly found their way into development discourse, but without much reflection on history and background taking place. An appreciation of the historical context of the rise to prominence of these ideas brings a more nuanced understanding to each.

NGOs were the first of these two ideas to make a rapid entry into mainstream development policy and research. By the end of the 1980s, a World Bank publication announced that 'NGOs have been heralded as new agents with the capacity and commitment to make up for the shortcomings of the state and the market in reducing poverty', and they were seen as offering elements of 'comparative advantage' in relation to government and business that included cost-effectiveness, administrative flexibility and an ability to work closeness to the poor (Paul, 1991, p. 1). Sogge (1996, p. 1) remarked 'After decades of quiet and respectable middle-class existences, private development agencies have come up in the world'. By the early 1990s, NGOs had become central to the emergence of the 'good governance' agenda and their roles also underpinned new thinking in relation to promoting the flexible delivery of services by private non-state actors (Lewis, 2005a).

NGOs at this time were presented largely as – to take a cue from Eric Wolf (1982) – organizations 'without a history'. Yet there is a long history to the NGO phenomenon which far predates their rise to prominence within development. Charnovitz (1997, p. 185) argues that this history has remained largely hidden, and is critical of the ahistoricity of both NGO researchers and supporters: Although some observers seem to perceive NGO involvement as a late-twentieth-century phenomenon, in fact it has occurred for over 200 years. Advocates of a more extensive role for NGOs weaken their cause by neglecting this history because it shows a long time custom of governmental interaction with NGOs in the making of international policy.

Charnovitz traces NGO 'emergence' from the late eighteenth century up to the period before the First World War, through to a period of 'empowerment' in which NGOs have played greater international and developmental roles since the 1992 United Nations Environment conference at Rio. There had been national level issue-based organizations during the eighteenth century that focused on issues such as the abolition of the slave trade and support to the peace movement. By the early twentieth century, NGOs had become more active internationally on issues as varied as transportation, intellectual property rights, narcotics control, public health, agriculture and environmental protection. After Second World War, the United Nations Charter had provided for NGO involvement in UN activities, with an increased 'intensification' of NGO roles through the 1960s and 1970s culminating in the 1992 Rio conference and the Beijing women's conference in 1995. These processes also brought into sharper focus the existence of communities of Southern NGOs which had emerged not simply as an outcome of international aid but also, as in Bangladesh, a complex set of other factors such as the post-1971 national reconstruction effort, local innovation around poverty reduction within the new country and the disillusionment with mainstream political processes (Lewis, 1997).⁶ Rather than the sudden emergence implied by Paul (1991), there was already a long history attributable to NGOs.

The relatively sudden appearance of NGOs within development policy from the late 1980s onwards can be understood in the context of newly ascendant neo-liberalism, both at the level of ideological recruitment and the more practical level of creating new opportunities for applied researchers (Lewis, 2005a). A convergence of priorities among donors, activists and researchers served to elevate these ideas into a position of development orthodoxy. Policies of privatization, market liberalization and administrative reform had come to represent the dominant solutions to development problems (Schech and Haggis, 2000). A crisis of development theory in the 1980s had helped contribute to a loss of confidence that development could be produced by the state, and this coincided with the rise of neo-liberal analysis which saw state intervention as problem rather than solution. Neo-liberals had come to dominate the international financial institutions and many governments, influencing significant sections of the development industry. This led to greater levels of funding for NGOs, particularly those engaged in service delivery.

The rise of the 'good governance' agenda in the 1990s tempered more extreme approaches towards privatization in favour of a view that saw potential synergies existing between state, market and the non-governmental sector. This period saw an expansion of funding of NGO activities beyond service delivery to include advocacy. But it also began to lead donors away from an emphasis on NGOs towards the newly discovered discourse of 'civil society'. Although there have been differences among donors in the ways this was defined, civil society brought a broader interest in supporting grassroots membership organizations, business associations, 'faith-based groups' and sometimes even trade union. By the end of the 1990s, in donors circles these organizations were now 'in', and NGOs were old news.⁷ Many international donors have now shifted towards a broader focus on working with governments more directly through budget support and with a broader notion of 'civil society' actors engaged in governance and services.

The resurgence of interest in the concept of 'civil society' can also be analysed as an ideological act of 're-remembering' – or perhaps differently remembering – rather than an engagement with an idea that was qualitatively new (Comaroff and Comaroff, 2000). But the concept of civil society, as gradually became clear to development agencies, was one with a long and complex genealogy that created problems for those who seized too unreflectively upon its promised role in development. This became clear, for example, as development agencies soon came to realize that civil society did not only contain organizations that were pursuing progressive aims, but also a range of uncivil exclusionary groups and business fronts (Bebbington et al, 2008). A more historical approach also led some observers to reflect on the question of whether the concept's essentially Western origins allowed for an application to non-Western contexts (Lewis, 2002).

At the same time, a diversity of often non-compatible strands of thinking about civil society made it difficult for development agencies, government policy makers and activists to communicate easily with each other (Lewis, 2002). It was a version of civil society that was derived from the work of Alexis de Tocqueville that found most fertile ground in development agencies interested in 'good governance' and in ways of supporting civil society. De Tocqueville's account of nineteenth century associationalism in the United States stressed volunteerism and associational life as bulwarks against state domination, and as counterweights forces through which citizens could help keep government accountable. By contrast, Gramsci had argued that civil society constituted an arena, separate from state and market, in which ideological hegemony is contested, implying that civil society contained a wide range of different organizations and ideologies which both challenged and upheld the existing order. Gramscian ideas about civil society had been influential in the context of resistance to totalitarian regimes in Eastern Europe and Latin America from the 1970s onwards, and were also linked with ideas about transformative potential of 'social movements' (Escobar and Alvarez, 1992; Howell and Pearce, 2001).

A historically informed view of both NGOs and changing ideas about civil society can therefore serve to clarify concepts, assess more realistically their potential value to development practice, and draw attention to the importance of embedding them more fully within contextual analysis. More importantly, the shifting fashions of a largely ahistorical engagement by funders with NGOs has direct implications for organizations and their partners at the grassroots, as funding opportunities come and go. For example, at its start in 2001, DFID's Human Rights and Governance funding initiative in Bangladesh was a flagship programme in the form of a 'local fund' designed to seek out and support small local civil society partners doing innovative work (Beall, 2005). But by 2005, the programme looked increasingly out of place among the multi-donor budget support programmes which now engaged most of the energies of DFID staff, and its employees and local partners had to increasingly fight for attention and resources among a set of new economic target-related priorities. It was also under pressure to begin to act as a cost-effective delivery point for DFID funding to its other civil society partners, potentially drawing the programme – now an independent local trust – away from its innovative roots. For the small grassroots partner organizations working closely with marginalized communities around the country, these changing priorities have the potential to impact negatively on day-to-day work and relationships.⁸

People and practices: 'boundary-crossing' between state and civil society

A more historically informed approach to development policy can also be useful in understanding the changing roles of particular kinds of professionals and activists within development institutions and policy processes. Recent work on the experiences and motivations of individuals who have crossed between the civil society or 'third sector' and the government or public sector during their lives and careers takes a historical approach to understanding policy (Lewis, 2008a).⁹ The work drew upon ethnographic 'life history' interviews collected across contrasting country settings – including Philippines and Bangladesh – in order to document and compare individual experiences of operating on either side of the boundary between government and civil society. The study analysed the reasons for these boundary shifts in relation to concepts of power and innovation, and explored their implications for understanding the boundaries, both conceptual and tangible, that separate government and civil society in different institutional contexts and at different times.

A 'three sector' model of state, market and civil society has remained central to the development models and policy agendas that underpin 'good governance'. Although policy makers have found it analytically convenient to separate these sectors, the realities are of course far more complex. NGOs and states tend to be linked through explicit and implicit sets of personal relationships, resource flows and transactions. At particular historical junctures, such links have become more apparent, such as when there is an exchange of personnel between the government and the civil society or 'third' sector. For example, in the Philippines, many NGO activists and professionals joined the democratic Aquino government in 1986 in order to pursue the implementation of agrarian reform agendas that had previously been ignored under the authoritarian Marcos regime, which had imposed Martial Law in 1972. After years of opposing the regime, the section of the NGO community that had formed among activists of the left became divided between those who saw value in entering the new government to seek reformist spaces on issues such as land reform from within, and those who favoured a continuation of the strategy of campaigning and mobilization from outside. Other civil society activists sought election to government motivated by new ideas about non-traditional forms of politics, owing to the limitations of mainstream political parties and elites.

The research has therefore been able to illustrate the ways individual life histories can help reveal more of the ways in which broader policy history is constructed. In particular, it has generated insights that tell us more about the tensions that exist between a rigidly conceived three sector policy model and the more unstable boundary between government and civil society that exists in practice.

The findings point to two sets of issues relevant to our argument here. First, they suggest a range of 'hidden' linkages between the two sectors, such as informal political, patron – client or kin relationships between key players. For example, an elite family may advance its interests by having representatives within government, business and third sector organizations, raising new forms of 'elite circulation' that need to be analysed (Lewis, 2008a). Second, the findings highlight the importance of comparative historical analysis in understanding the considerable differences that exist between state and civil society relationships in different contexts. In the Philippines, for example, efforts by policy reformers to work within government structures to bring about change have generated some progressive gains in the agrarian reform, social welfare and housing sectors, but they have also exposed the limitations of unrealistic civil society strategies that aim to secure long-term change from within. In Bangladesh, where the boundary-crossing is more likely to take place from government into civil society and other development agencies, the direction of travel instead reflects the strong role played by foreign aid and the large NGO sector. Here, the life histories speak of the decline of civil service capacity and notions of 'public service', and of narratives of 'escape' by public officials into better paid and more personally rewarding work outside government (Lewis, 2008a).

Cross-over activity of this kind takes on different forms depending on context, and reflects broader historical changes in resource incentives and political opportunities more widely – such as shifts in policy priorities among governments and international agencies. For example, more 'flexible' organizational structures are increasingly favoured under neo-liberal economic and social policy frameworks. The new public management which seeks to combine elements of markets and public administration in its model of the mixed economy of welfare and 'the purchaser – provider split' in service delivery. Another outcome may be the co-option of previously independent activists from civil society within government structures.¹⁰ Government and civil society boundaries are therefore in a constant state of being constructed and unmade, as individuals make purposive shifts as leaders, organizers, activists and managers between the different sectors. Such shifts suggest the need to analyse the importance of the roles and experiences of these 'sector brokers' (cf. Lewis and Mosse, 2006) in relation to changing configurations of power within governmental and non-governmental activities.

In more practical terms, the life histories also highlight the importance of personal and professional histories as a means of gaining insight into the realities of how policy fashions play out on the ground. For example, the inadequacies of securing policy change through the route of NGO advocacy is discussed within the narratives of individuals who cross over and see the reality and complexity of policy change processes within government. As more personal experience is accumulated, NGO advocacy models can be revealed as naïve or ill-informed, but may then be strengthened and improved as certain individuals gain better knowledge and return to civil society after a period working inside government. Research on the boundary-crossers also tells us more about how inter-sectoral relationships operate within project settings, and this may further inform and strengthen the complex practices of partnership that frequently seem to under-perform in various ways (cf. Lewis, 1998).

The use of the life history method therefore makes possible a deeper understanding of the ways in which peoples' individual histories and experiences both contribute to, and are embedded within, wider patterns of change and transformation (Lewis, 2008b). These individuals are constituted into knowledge communities in ways that cannot simply be analysed in terms of ideological hegemony, but need also to take account of agency and contingency at the level of individual actors and their everyday practices. A focus on the histories of individuals within policy processes draws attention to the importance of individual agency within processes of institutional change but it also highlights the ways in which it is historically framed and formed by structures.

Conclusion

The lack of historical perspective within development agencies stems partly from the pressures of development work in which activities remain powerfully (and understandably) focused on the promise of generating future change, but it is also part of a broader problem of ideologically constructed managerialism. An anthropological view of development is one that takes account of its multiple histories and its capacity to embody diverse ideas about hope, administration and critical understanding rather than simply a set of technical and managerial processes. Within such an approach, it also becomes possible to build an anthropological engagement with development that goes beyond the three roles described (problematic forms of co-option, reluctant involvement or simple antagonism) towards a more unified form of critical engagement.

Following from this, the case for raising the profile of history fits with recent anthropological arguments made by Eyben (2006) for researchers and policy makers to pay more attention to the 'relational' aspects of development work – within and between organizations and individuals working in development. A historical perspective recognizes and contributes to ways of working that are sensitive to issues of power, politics and learning that such a 'relationships perspective' requires. The tendency of development thinking to be characterized by a perpetual present of changing buzzwords and fashions does, of course, imply some relationship with the past – but it is a past that is only invoked in a superficial way to justify the present, and only rarely to challenge it.

The need to pay more attention to history within process of institutional change has also been emphasized by cultural critics of public management such as Hood (1998, p. 14), who stresses the value of searching for 'broader patterns of recurrence beneath apparent novelty': Limiting discussion to the very recent past and neglecting the major historical traditions of thought in public management can narrow debate and criticism, by implying there is no alternative to whatever modernity is held to mean ... Historical knowledge is a good antidote to naïve acceptance of novelty claims and history in this sense is potentially subversive. (p. 17)

A historical perspective therefore serves as a counterweight to the changing fashions that characterize the world of development agencies and ideas. Not only do these constantly shifting priorities work against longer-term relationship building, but they also, as Wrangham (2004, p. 264) points out, have a bearing on practices on the ground: 'fads and fashions at the international level have a real effect on poor people'.

This paper has not discussed different types of historical approach that might prove most valuable or relevant to development work, and space does not allow for detailed consideration here. However, the work of Norman Long (2001, p. 62) is particularly useful in clarifying the value of history in terms of the interplay between agency and structure, drawing as he does on Kosik's development of the idea of 'praxis': History never relates in a uniform or unilinear to the present and the future ... their relation is essentially dialectical, both elements of the possible and the real. That is, history always contains more than one possibility, where the present is the realisation of only one of these; and the same holds for the interrelations between the present and the future.

In this way, a historical perspective can sensitize us to alternative states of being and ways of acting, and thereby challenge managerialist 'one best way' type of thinking. Anthropological work that can move beyond longstanding 'critical versus applied' tensions can contribute usefully to this re-historicized approach. It can bring a historical perspective that is both critical and constructive, and one which seeks to combine theoretical insight with practical application. In so doing, such work will perhaps provide more nuanced insights into the 'black box' of development intervention, and open up more historically informed forms of development policy and practice.

Notes

¹ Mosse (2005) is a recent and notable exception.

² Anthropologists themselves have from time to time come under heavy criticism for their lack of a sufficiently historical approach. For example, during anthropology's period of post-modern critical reflection in the 1980s, Marcus and Fischer (1986) questioned the tendency of anthropologists to focus on an ahistorical or exotic 'other' and instead argued for work on the ideas and institutions of anthropologists' own contexts, emphasizing the need to study power and history locally and globally.

³ This is not to say that such historical perspectives were not available to the project or argued for by certain project staff, but that they were in practice given low priority.

⁴ Hood (1998) has pointed out that the 1990s practice of 'hot desking' that was common among private sector companies such as Xerox and presented as the latest in up-to-date management thinking were not so dissimilar to ideas that had been advocated by Jeremy Bentham to improve the performance of public officials two centuries earlier!

⁵ When I was a postgraduate development studies student in the early 1980s in the Unite Kingdom, neither NGOs nor civil society were ever mentioned during teaching or within the set readings.

⁶ Different sets of NGO-related histories are also recoverable from other parts of the world: in Latin America, the growth of 'liberation theology' in the 1960s, the political ideas of radical Brazilian educator Paolo Freire, from peasant movements seeking improved rights to land and against authoritarianism, from the influence of Christian missionaries, the reformist middle classes and Gandhian ideas in India (Lewis, 2007). These histories help reveal more of the 'indigenous' sources and forces that also helped promote the rise to prominence of NGOs in the South.

⁷ The current interest in 'faith-based organisations' in place of NGOs is a particularly good example of the fads and fashions of the development industry, as such organizations are a longstanding feature of most societies that were long ignored by development agencies.

⁸ This observation is based on earlier work as an adviser to the initiative between 2002 and 2005, and continuing informal contact during regular visits to Bangladesh in the subsequent period.

⁹ The research project was funded by the UK Economic Social Science Research Council as part of its Non-Governmental Public Action Programme (Grant RES-155-25-0064). Twenty detailed life history interviews were collected in each of three contrasting country contexts: the Philippines, where many NGO activists have crossed into successive post-Marcos democratic governments to work on agrarian reform, social welfare and other issues; Bangladesh, where there is little movement from civil society into government, but considerable movement in the other direction and extensive informal linkages among key individuals between NGOs and government; and the United Kingdom, where there has been an intensification of movement between the two sectors in the past decade since the 1997 New Labour government came to power as a result of both purposeful exchange in the form of secondments, and increased flexibility and mobility in the labour market.

¹⁰ For example, the rise of 'audit culture' has been analysed in neo-Foucauldian terms as part of a shift to neo-liberal forms of governance which depends in large part on the role of individual agency in which 'individuals, as active agents, are co-opted into regimes of power' (Shore and Wright, 2001, p. 760).

References

1. Barnett, T. (1977) The Gezira Scheme: An Illusion of Development. London: Frank Cass.
2. Beall, J. (2005) Local Funds for Local Governance. London: ITDG Publications.
3. Bebbington, A., Hickey, S. and Mitlin, D. (2008) Introduction: Can NGOs Make a Difference? The Challenge of Development Alternatives. In: A. Bebbington, S. Hickey and D. Mitlin (eds.) Can NGOs Make a Difference? The Challenge of Development Alternatives, Chapter 1, London: Zed Books, pp. 3–37.
4. Charnovitz, S. (1997) Two centuries of participation: NGOs and international governance. Michigan Journal of International Law 18(2): 183–286.
5. Comaroff, J.L. and Comaroff, J. (2000) Civil Society and the Critical Imagination in Africa: Critical Perspectives, Chicago: University of Chicago Press.
6. Corbridge, S. (2007) The (im)possibility of development studies. Economy and Society 36(2): 179–211. | Article |
7. Cornwall, A. and Brock, K. (2005) Beyond Buzzwords: 'poverty reduction', 'participation' and 'empowerment' in Development Policy. Overarching Concerns Programme Paper 10, UNRISD, Geneva.
8. Cowen, M. and Shenton, R. (1995) The Invention of Development. In: J. Crush (ed.) Power of Development, Chapter 1, London: Routledge.
9. Cowen, M. and Shenton, R. (1996) Doctrines of Development. London: Routledge.
10. Escobar, A. (1995) Encountering Development: The Making and Unmaking of the Third World. Princeton, NJ: Princeton University Press.
11. Escobar, A. and Alvarez, S.E. (eds.) (1992) The Making of Social Movements in Latin America: Identity, Strategy and Democracy, Boulder: Westview Press.
12. Eyben, R. (ed.) (2006) Relationships for Aid. London: Earthscan.
13. Ferguson, J. (1990) The Anti-Politics Machine: 'Development', Depoliticisation and Bureaucratic Power in Lesotho. Cambridge: Cambridge University Press.
14. Gardner, K. and Lewis, D. (1996) Anthropology, Development and the Post-modern Challenge. London: Pluto Press.
15. Grillo, R. (1997) Discourses of Development: The View from Anthropology. In: R. Grillo and J. Stirrat (eds.) Discourses of Development: Anthropological Perspectives, Chapter 1, Oxford: Berg.
16. Grillo, R. and Rew, A. (eds.) (1985) Social Anthropology and Development Policy. London: Tavistock.
17. Harrison, E. and Crewe, E. (1998) Whose Development? An Ethnography of Aid. London: Zed Books.
18. Hobart, M. (1993) Introduction: The Growth of Ignorance?, An Anthropological Critique of Development. Chapter 1, London: Routledge, pp. 1–30.
19. Hood, C. (1998) The Art of the State: Culture, Rhetoric and Public Management. Oxford: OUP.
20. Howell, J. and Pearce, J. (2001) Civil Society and Development: A Critical Exploration. Boulder, CO: Lynne Rienner.
21. Isaacman, A. (1997) Historical amnesia or the logic of capital accumulation: Cotton production in colonial and postcolonial Mozambique. Environment and Planning D: Society and Space 15: 757–790. | Article |
22. Kothari, U. (2005) From colonialism to development: Oral histories, life geographies and travelling cultures. Antipode 37(3): 49–60. | Article |
23. Lewis, D. (1997) NGOs and the state in Bangladesh: Donors, development and the discourse of partnership. The Annals of the American Academy of Political and Social Science 554 (November): 33–45. | Article |
24. Lewis, D. (1998) Partnership as Process: Building an Institutional Ethnography of an Inter-Agency Aquaculture Project in Bangladesh. In: D. Mosse, J. Farrington and A. Rew (eds.) Development as Process: Concepts and Methods for Working with Complexity. London: Routledge, pp. 99–114.
25. Lewis, D. (2002) Civil society in African contexts: Reflections on the 'usefulness' of a concept. Development and Change 33(4): 569–586. | Article |
26. Lewis, D. (2005a) Individuals, Organisations and Public Action: Trajectories of the 'non-governmental' in Development Studies. In: U. Kothari (ed.) A Radical History of Development Studies. London: Zed Books.
27. Lewis, D. (2005b) Anthropology and Development: The Uneasy Relationship. In: J. Carrier (ed.) A Handbook of Economic Anthropology, Chapter 29, London: Edward Elgar, pp. 472–486.
28. Lewis, D. (2007) The Management of Non-Governmental Development Organisations, 2nd edn, London: Routledge.
29. Lewis, D. (2008a) Crossing the boundaries between 'third sector' and state: Life-work histories from Philippines, Bangladesh and the UK. Third World Quarterly 29: 1. | Article |
30. Lewis, D. (2008b) Using life histories in social policy research: The case of third sector/public sector boundary crossing. Journal of Social Policy 37(4): 559–578 | Article |
31. Lewis, D. and Mosse, D. (2006) Encountering order and disjuncture: Contemporary anthropological perspectives on the organisation of development. Oxford Development Studies 34(1): 1–14. | Article |
32. Long, N. (2001) Development Sociology: Actor Perspectives. London: Routledge.
33. Marcus, G. and Fischer, M. (1986) Anthropology as Cultural Critique. Chicago: University of Chicago Press.
34. Mosse, D. (2005) Cultivating Development: An Ethnography of Aid Policy and Practice. London: Pluto Press.
35. Mosse, D. and Lewis, D. (eds.) (2006) Introduction. In: D. Lewis and D. Mosse (eds.) Development Brokers and Translators: The Ethnography of Aid and Agencies. Bloomfield, CT: Kumarian.
36. Olivier de Sardan, J.-P. (2005) Anthropology and Development: Understanding Contemporary Social Change. London: Zed Books.
37. Paul, S. (1991) Non-governmental Organizations in the World Bank: An Overview. In S. Paul and A. Israel (eds.) Nongovernmental Organizations and the World Bank: Cooperation for Development, Washington DC: World Bank, pp. 1–19.
38. Quarles Van Ufford, P., Giri, A.K. and Mosse, D. (2003) Interventions in Development: Towards a Moral Understanding of our Experiences and an Agenda for the Future. In: P. Quarles Van Ufford and A.K. Giri (eds.) A Moral Critique of Development: In Search of Global Responsibilities, Chapter 1, London: Routledge.
39. Roberts, S.M., Jones, J.P. and Frohling, O. (2005) NGOs and the globalization of managerialism. World Development 33(11): 1845–1864. | Article |
40. Schech, S. and Haggis, J. (2000) Culture and Development: A Critical Introduction. Oxford: Blackwell.
41. Scott, J.C. (1998) Seeing Like A State: How Certain Schemes to Improve the Human Condition Have Failed. New Haven: Yale University Press.
42. Shore, C. and Wright, S. (eds.) (1997) Anthropology of Policy: Critical Perspectives on Governance and Power. London: Routledge.
43. Shore, C. and Wright, S. (2001) Reply to Maguire. Journal of the Royal Anthropological Institute 7(4): 759–763. | Article |
44. Sogge, D. (1996) Northern lights. In: D. Sogge, K. Biekart and J. Saxby (eds.) Compassion and Calculation: The Business of Private Foreign Aid, Chapter 5, London: Pluto Press, pp. 144–178.
45. Wolf, E. (1982) Europe and the People Without History. Berkeley: University of California Press.
46. Wrangham, R. (2004) Negotiating meaning and practice in the Zambezia Agricultural Development Project, Mozambique. Unpublished PhD thesis, London School of Economics.