‘Teach a Man to Fish’: The Doctrine of Sustainability and Its Effects on Three Strata of Malawian Society

Abstract

This paper analyzes the social impacts of the commitment to “sustainability” in donor-funded AIDS programs. Using survey, interview, and ethnographic data from rural Malawi, we examine how efforts to mobilize and empower local communities affect three strata of Malawian society: the villagers these programs are meant to help, the insecure local elites whose efforts directly link programs to their intended beneficiaries, and, more briefly, national elites who implement AIDS policies and programs. We describe indirect effects of sustainability on the experiences, identities, and aspirations of Malawians—effects that are much broader and deeper than the direct impacts of funding.

1. Introduction

Humanitarian assistance for HIV prevention and AIDS mitigation in sub-Saharan Africa has unintended yet profound consequences far beyond the realm of AIDS prevention. Current efforts by foreign donors to mobilize and empower village communities to solve their own AIDS-related problems in a sustainable way have paradoxical effects, failing to provide the sustainable programs donors hope for, but nonetheless profoundly affecting the aspirations and identities of local actors. Just as teaching a man to fish—rather than simply giving him fish—should provide him with a source of food for life, so also should communities, once empowered, on their own turn the tide of the AIDS epidemic and mitigate its effects. Or so the sustainability doctrine presumes. We describe the actual consequences of these efforts on three strata of Malawian society: the villagers whom these policies and programs are meant to help, the elites in the capital who staff the NGOs charged with implementing AIDS policies and programs, and an intermediate group that has received much less attention: the interstitial elites living in district capitals and trading centers, whose efforts directly link policies and programs to their intended beneficiaries.

We are certainly not the first to critique the development enterprise. James Ferguson's (1990) work on the myth of “development” in Lesotho stands as the outstanding exemplar: Mansuri and Rao (2004), Jackson (2005), Englund (2006), Easterly (2006), and Li (2007) are among the more recent critiques that focus on the failure of development projects to confront underlying inequalities in power. We agree with these critiques, but our project here differs in two key respects. First, we demonstrate how the cumulative weight of AIDS projects, over extended periods, indirectly but powerfully shapes what we, following Foucault, think of as local subjectivities. Second, an unusually rich combination of data permits us to link the programs emanating from western donors and their Malawian partners in the capital, Lilongwe, with a detailed examination of the reception of these programs in the villages. Unlike most of the critical scholars who have come before us, many of whom are anthropologists working alone or in small teams, we can draw on the survey, interview, and ethnographic data that the Malawi Research Group (MRG) has been collecting in three districts of rural Malawi for almost ten years. We also draw on qualitative interviews with NGO staff conducted by several members of the MRG, including the authors, in 2005 and 2006.¹

In Malawi, each stratum invested the donor programs with a distinctive social imaginary. In the capital, NGO staff—capable, energetic Malawian professionals—worked in partnership with international experts to devise policies and programs that, they expected, would transform backward, poor and dependent village communities by “offering people concrete opportunities to develop their sense of internal and external control and self-efficacy” (Malawi Bridge Project, 2004). The villagers were no less busy, engaging in a variety of contingent and opportunistic strategies to make ends meet: subsistence farming, growing crops for sale, petty trading such as selling tomatoes in the market or making mats from palm fronds, and sometimes migrating for temporary work. They sought basic income security, so that they would no longer have to seek hither and yon for the resources to sustain their families. The third group, the interstitial elites—those still based in villages or small trading centers, but with enough education and gumption to nurse the hope of finding a job in the formal sector rather than fishing or farming—hoped that they might be able to parlay their years of volunteering with NGOs into a salaried position that would confirm their status among the educated and urban rather than the backward and rural.

Those who are supposed to be the most empowered by sustainable projects, the villagers and the interstitial elites, learned lessons learned from these projects that undermined the projects' manifest purposes. Rather than developing “their sense of internal and external control and self-efficacy,” donors' insistence that CBO [Community Based Organization] projects be sustainable without further outside aid created a capricious, irrationalizing environment that reinforced a contingent, opportunistic orientation among recipients. In essence, the donors encouraged hunting and gathering in a terrain of AIDS NGO projects (Bird-David, 1983; 1990).

We begin by first briefly summarizing the theoretical and empirical work that has influenced our approach and by describing our data. We then turn to explicating the key elements of the global ideology of sustainability. Finally, we demonstrate how sustainability is practiced and examine its effects. We find little evidence of sustained or even potentially sustainable community mobilization in the villages; however, we do find that these programs sustain the fortunes of the national elites and the dreams of the aspiring elites in practical as well as theoretically interesting ways. Although our data refer to humanitarian assistance to combat the AIDS epidemic, our results are likely to be generalizable to other efforts to “teach a man to fish.”

(a) Background

A large academic literature interprets successive fashions in foreign aid (e.g. Clark, 1991; Mohan & Stokke 2000; Smillie, 2001; Mosse & Lewis, 2005; Lancaster, 2006). Another evaluates the expected effects of foreign aid strategies and typically documents failures either at the global level using statistical data (e.g. Stiglitz, 2002; Easterly, 2006) or at the local level through case studies (e.g. Bebbington, 2005; Igoe & Kelsall, 2005). Descriptions and analyses of the unexpected social effects of foreign aid, however, are rarer (for exceptions see Justice, 1986; Cooper, 1998; Englund, 2006; Li, 2007).

We draw on several bodies of theory that are not often brought together in the existing literature on foreign aid. Theories of the new institutionalism, especially those that that focus on cultural rules that constitute new realities (Meyer, 1983; 1987; 1990; Finnemore, 1996; Jepperson & Meyer, 2000; Frank & Meyer, 2002) led us to look for effects of funding strategies beyond those directly intended by donors. So, for example, when Home Based Care becomes an official activity of CBOs that have to be registered with the National AIDS Commission, it may alter the meaning of village practices for caring for the sick or orphans (Meyer & Rowan, 1977; DiMaggio & Powell, 1983; for related arguments see Nguyen, 2004; Marsland, 2006). Institutional theory also focuses attention on the “hidden curriculum” of the AIDS enterprise—not what it teaches through explicit didactic efforts to disseminate information or values, but what its institutional rules and practices enact as new realities on the ground (Dreeben, 1968; Thomas, 1994; Garland, 1999; Englund, 2006; see also Heimer, 2007). We have also been influenced by the Foucault-inspired literature on new kinds of subjects/human possibilities (cf. Foucault, 1983; 1988) that are both imagined and enacted through the interactions of organizations with their clients and potential beneficiaries. Much of the Foucauldian literature focuses on institutionalized categories that require new identities and forms of subjectivity (Nguyen, 2004; 2005). We are struck by how much NGOs influence not only those who directly become beneficiaries or employees, but those whose imaginations the NGOs beguile. A third important literature is that on the principal-agent problem, which suggests how difficult it is for organizations to control processes at a distance, to monitor and evaluate what their agents on the ground are actually doing (Adams, 1996; Gauri & Vawda 2003; Abraham & Plateau, 2004; Erickson & Bearman, 2005)

(b) Data

Rather than following a particular project (see, for example, Bornstein, 2003; Campbell, 2003; Igoe & Kelsall, 2005; Mosse, 2005; Englund, 2006), our approach is geographically-based, focusing on three rural districts in Malawi, in the Southern, Central and Northern regions. This allows us to see the cumulative effects of multiple projects to understand how they influence not only the intended beneficiaries but also the wider society. We focus on these three districts because we know them well, both through time spent there and through the remarkably rich data collected since 1997 by members of the Malawi Research Group (MRG). ²

Our interest in the ideology of sustainability was stimulated by the many months we spent in rural trading centers while participating in MRG fieldwork. The inexpensive motels that accommodated us also often accommodated a succession of training workshops, thus providing us with opportunities to chat with trainers and with the eager volunteers being trained to mobilize their communities to combat AIDS. We were fascinated and puzzled when trainees from a youth group told us, for example, that they had learned “decision making” or, as one young man earnestly explained, that semen and urine both came from the penis.

These experiences led to new data collection for this paper. In a pilot project in 2005 we tried to establish how many youth clubs were active in the MRG southern site, where we happened to be staying. Almost invariably, if we talked with the school headmaster we were told that the club was active, but if we talked with others we found that there had been an AIDS youth club, but it existed only briefly, having become defunct when, after training a few peer educators at a workshop, the NGO failed to provide promised resources and members became “disillusioned” and drifted away in search of more rewarding activities.³ In the 2006 round of the MRG's longitudinal survey we included questions on the activities of CBOs in respondents' villages and supplemented these with qualitative interviews with villagers and NGO staff. We found little evidence of sustained or sustainable community mobilization but much to suggest the powerful if unanticipated social effects of donor commitment to the ideology of sustainability.

2. The Doctrine of Sustainability

To make sense of the paradoxical effects of AIDS funding, we first examine the principle of “sustainability” and its underlying ideals. The term “sustainable development” originated in the search for ecologically sustainable forms of economic development.⁴ But the ideal of sustainability has increasingly become a conscious policy of donors in a more comprehensive sense: donors seek projects that (donors believe) recipients will be able and willing to sustain after the donor—and the donor's funding—departs.⁵

(a) Sustainability as an ideal

In the AIDS world, sustainability is a kind of mantra, even when it is manifestly unrealistic (see Kremer & Miguel, 2007).⁶ A flavor of this doctrine can be found in World Vision's description of its local Area Development Programmes or ADPs. Each description of a local ADP, whether in Malawi, Zambia, or Tanzania, ends with an invocation of the community's participatory governance and hopes for its eventual autonomy. The following is from Chata, one of the ADPs in Malawi:

Sustainable development

World Vision and the people of Chata envisage that by 2010 the community's capacity to access resources for ongoing development will be at a level that no longer requires World Vision assistance. We will keep you updated on their progress. (25 August 2005; accessed 3/20/2007: http://www.worldvision.org.nz/wherewework/profiles/p_Malawi_Chata.asp)

A report on a World Vision ADP in Zambia expresses an identical sentiment: the community plans to establish a trust, which “will facilitate development activities in Nakonde after World Vision has left the area” (25 August 2005 [accessed 3/20/2007] http://www.worldvision.org.nz/wherewework/profiles/p_Zambia_Nakonde.asp). Even where the term sustainability never appears, its underlying logic is evident in donors' funding priorities. Thus, for example, Harri Englund (2006, p. 99-122ff) describes how the major human rights organization in Malawi refused to provide material help or concrete benefits, which are dismissed as creating “dependency.” Instead, donors insist on providing education and training, so that people will know what their rights are and what kinds of “freedom” they should aspire to. Even more succinct was the national executive director of a faith-based youth organization in Malawi, who said of HIV/AIDS, “If a person is self-reliant then the chances of getting the disease is low” (Interview by A. Esacove and J. Watkins 2006).

The sustainability doctrine incorporates a set of ideals about the way the world should work that derive from participatory approaches to foreign aid that have been in fashion for several decades (see Mansuri & Rao, 2004). One ideal is local “ownership”: the project should “belong to” the local community (sometimes the local “stakeholders” [Campbell, 2003]), both in the sense that it should arise out of their needs and interests, rather than those of the donors, and that the locals should take responsibility for it. The “local community” is understood as having consensual needs and interests, expressed in democratically chosen leadership.⁷ Related to the ideal of local “ownership” is an insistence on local community buy-in. The community should show its commitment by contributing some of the resources.

Community mobilization ideologies invoke basic assumptions about the virtues of community participation (Stiefel & Wolfe, 1994). One assumption is that communities have better information about what they need than do national or international technocrats (Fisher, 1997; Bratton, 1989; Kelsall, 2001; Webb, 2004; Mansuri & Rao 2004). Second, local communities are presumed to draw on local social capital, so that the NGO can harness the “generalized reciprocity” of village life to provide Home Based Care for the ill and dying, to care for Orphans and Vulnerable Children, or to create youth groups to teach AIDS-prevention to other young people.

The doctrine of sustainability adds to these participatory approaches a profoundly moral invocation to the poor: that they should become self-reliant, mobilizing their own energies and resources to solve their own problems. This is the essence of “teaching a man to fish.” What does this mean in practice? That the goal is to fund only projects that will be sustained after the donor's funding ceases. Donors' views of what might make a program or activity sustainable vary. Activities might be sustainable because they generate their own revenues, as micro-lending projects are expected to do; because they can attract new donors who will keep the project going after the initial seed money runs out; or because, like “teaching a man to fish,” they provide something self-renewing, that once transmitted will sustain itself without further donor inputs.

In the donor imaginary, sustainable projects will lead to profound social transformation by imbuing the members of democratic and self-reliant communities with a “rational, planned sense of personal and social responsibility associated with the virtues of modernity”⁸ and empowering them to take control of their own futures. The key contrast is thus between “sustainability,” with its connotation of autonomy, self-reliance, and rationality, and “dependence,” with its taint of helplessness, passivity, and irresponsibility.

(b) What can be funded ‘sustainably’?

Only a few of the many things local people might want, or donors might wish to fund, meet the criteria of an idealized sustainability.

(i) Training and the “Workshop Mentality”

In Malawi as elsewhere Sub-Saharan Africa, the supply of “training” has created a huge demand. In an incisive analysis of family planning programs in Nigeria, Daniel Jordan Smith (2003) has described the “workshop mentality,” arguing that training and workshops provide the ideal intersection of donor and recipient interests. Donors can believe they are doing something self-renewing by providing training, while workshop facilitators can build their patronage ties by providing access to the per diems, travel allowances, and opportunities for networking that workshops provide. But we argue that the predominance of “training” as a core donor-sponsored activity also arises from the constraints of the sustainability doctrine. If donors are supposed to help, but without funding substantive programs that could breed dependency, then training and workshops are the ideal donor-funded activity: experts will teach people skills, or better yet teach them to teach skills, which will provide all with the capacity to provide for their own needs.

(ii) Reliance on volunteers

A second way donors try to make their interventions in local communities ‘sustainable’ is to rely on volunteer labor, both from local communities and from the interstitial elites. The expectation that these village and small-town based participants—and only they, not the elites in the capital—contribute their labor for free may strike outsiders as strange; where else but in the impoverished Third World would anyone expect people to work without pay? Relying on volunteers is, of course, most obviously a way to save money. But even if turnover among volunteers and the continual need for new training make volunteers as costly as paid staff, the donors' belief that they can rely on volunteers makes their projects qualify on paper as sustainable. The donors provide start-up funds to organize people and train them, but then the volunteers will carry on the good work—distributing family planning in Kenya (Kaler & Watkins, 2001), providing home-based care for the sick, serving as youth peer-educators, or caring for orphans.⁹ Thus money can be spent on training and workshops to help people learn how to care for or educate others, but aside from the elites who plan how the donor project is to be implemented and those fortunate enough to become “trainers” or “trainers of trainers,” no one need be paid.

(iii) Income-generating projects

When resources other than those associated with training are provided to community-based organizations, these are typically for income-generation projects that are to make the project sustainable without further input from the donors.

3. The Social Consequences of Sustainabilty

Here we examine how the doctrine of sustainability works in practice by documenting its direct and indirect effects on three strata of Malawian society. Direct effects are everything from being a recipient of resources or training to being a volunteer, staff member or executive of an implementing organization. The indirect effects of sustainability are changes in the experiences, identities, and aspirations of Malawians more generally—changes that are, we argue, much broader and deeper than the direct effects of funding.¹⁰

(a) Villagers

In a country where per capita income is approximately US$170 (about 25,000 kwacha) per year (and less in the rural areas), the newspapers regularly announce donations to the cause of AIDS prevention and mitigation: “Canada grants K700 million for HIV/Aids” (Nyoni, 2002), “UNFPA, BLM Launch K437.5m Youth Project” (Times Reporter, 2006); “MSF assists orphans with items worth K130,000” (Phalula, 2000), “Bush pledges $500m … to help fight HIV/Aids” (Reuters, 2002); “GAIA launches K42.6m project for orphans, Aids projects” (Somanje, 2008).¹¹ Many rural villagers have access to these announcements through newspapers and radio, but they can also see signs of donor wealth in the 4×4's with NGO logos zipping along the roads, or stories of a friend of a friend who got 825 kwacha/day at a three-day NGO workshop. The donors have fantasies about community mobilization, volunteerism, democracy, empowerment, and sustainability—but those in the village have their own fantasies. We begin by asking what villagers want from donors, then describe the procedures villagers have to follow to seek funding, and then the outcomes of their requests.

Villagers have long lived in an insecure world. Despite years of foreign aid and technical advisors, the main economic resource for subsistence and cash in the rural areas is agriculture (there is virtually no formal labor market in rural Malawi). But rainfall is capricious, as are government subsidies of fertilizer for the depleted land and international prices for cash crops. Economic insecurity trumps AIDS, as shown in a national survey in several countries of sub-Saharan that asked respondents to name the most pressing problems their governments should address: 24% in Botswana mentioned AIDS, 10% in South Africa and Namibia, but only 2% in Malawi (Whiteside et al, 2004; see also de Waal, 2006)—this despite the fact that Malawi has the 8^th highest HIV prevalence in the world, that 65 % of Malawians report having relatives and neighbors who have died of AIDS, and that many Malawians expect that they are either now HIV positive or will be in the future (Anglewicz, 2007).¹²

A moving example of worries about “materiality” comes from a conversation in a women's group. This group is just what donors have in mind when they talk about communities solving their own problems: it was formed spontaneously and has never applied for or received donor funding (although Save the Children did give one of the women some soccer balls for the children in her village). This particular meeting occurred in the “hunger season,” the months when food stocks are low; the report came from one of the members, in written form.¹³

On the 15^th November, it was the meeting day for the Vutitsani women club where I am treasurer, and we normally meet at my house. At 2:00 p.m as usual my fellow women came for our meeting and the topic that we had on that day was about hunger and business…. The one who proposed that topic was Mrs. Nchinga…. She is now single and she usually sells the firewood to help her for anything which she needs in her life. Now the time has changed. It is now the time of hunger because many people have no food. Many people are buying some maize and the problem is that food is now expensive because they have seen that many people are lacking food. Since we have no farms, estates, and companies for people to work, many people especially women are depending on selling the firewood. Since there are many people who are selling the firewood for buying food, the business is not going on well because people are selling their firewood at a low price. What they need is to sell enough so that they should manage to buy food for that day. If one is misfortunate it happens that you can stay at the market for three days without selling your firewood and because of that, things are becoming tough and worse. What can a person do in order to save his/her life there?

Points were coming out from other women who were there. One of the women who answered first was our chairman Miss Everson. She answered that she thinks that it is better to change the business and try another one because if she just stick on that business which she is failing to sell or win then she can die of hunger. But Miss Chaweya answered that there is no other business which can help a person in these days, firewood was good because it does not need any capital but all other business needs a capital to start with therefore it is better to have a [sexual] partner who can be helping you during the time that you have nothing at your house. But Miss Danken answered, she said that Miss Chaweya was cheating [deceiving, fooling] her friends. These days are very dangerous. They are not the days which one can depend on having a sexual partner because of the Aids disease. [Alice 031005]

How then do NGOs try to meet the needs of rural Malawians, including those, like the women above, who are threatened by AIDS?

(i) Funding sustainable community-based organizations

Much of the donor funding for AIDS is channeled through the National AIDS Commission (NAC).¹⁴ Malawi, along with Zambia and Uganda, is among the privileged countries to have received more than $200 million in the first 5 rounds of Global Fund grants, including $35,000,000US in 2004, and $22,880,000 July-Sept 2005 from the Global Fund and a set of large bilateral donors (Weinreb, 2006, p. 5; NAC 2005a, p. vi). Much of this funding is for national-level efforts, especially the provision of HIV testing and antiretroviral therapy, but also condom purchase and distribution, and posters and radio programs encouraging abstinence and faithfulness.¹⁵

NAC set aside $10 million for community mobilization and empowerment, primarily in the areas of HIV prevention and AIDS mitigation. The guiding assumption was that the funds would enable existing but struggling community groups to better care for the orphans and the chronically ill, as well as to conduct prevention activities. In the event, and not surprisingly, the call for proposals stimulated the formation of new CBOs. The community mobilization project also funded five Malawi branches of international non-governmental organizations (such as World Vision and Save the Children) that had offices in the local districts; the district-level offices of these Umbrella Organizations were to play a role in reviewing proposals and in monitoring and evaluating the funded CBOs.¹⁶

Below we examine not only the direct consequences of the process through which CBOs are funded, but also the effects of the doctrine of sustainability on the resources people can access and the procedures they must follow to do so. Guided by institutional theory, we also look for evidence that the procedures that NAC stipulated taught a hidden curriculum—what villagers learned indirectly from the new realities NAC's institutional rules and practices enacted.

We can get a broad overview of NAC's grant making by looking at the NAC Project Tracking Data Base (n.d. [2007]).¹⁷ From March 2004 to June 2007, NAC recorded contacts with 1,248 CBOs, Faith Based Organizations, NGOs, and Private and Public Sector Organizations, of which some 800 had applied for funding.¹⁸ Applications from CBOs in each district went to the district-level offices of the Umbrella Organizations for review.

In 2006, we investigated how the proposals from CBOs for NAC funding were produced in two districts that are part of the MRG's research areas: Mchinji in the central region, which had 51 AIDS CBOs funded by June 2006, with 200 approved but waiting for funding,¹⁹ and Balaka in the southern region, which by June 2006 had 37 funded (Save the Children, 2006). For Balaka we also have a data set compiled by Kim Yi Dionne in 2006 recording details on 180 CBO proposals submitted to the local Umbrella Organization, Save the Children, including the 37 that were funded, almost 100 rejected proposals, and 44 waiting for review. We also interviewed district and national officials,²⁰ and we supplement these data with interviews with CBO committee members and village residents conducted by MRG survey supervisors (university graduates) and interviewers (secondary school graduates from the district).²¹

On paper, NAC's procedures for distributing money to local groups fit perfectly with donor ideals of community mobilization and empowerment. Existing community-based organizations are to decide themselves how NAC resources could help them implement their AIDS-related activities in a way that will be sustainable and then write a proposal. There are details, however, which in the world of Lilongwe are reasonable, but which turn out to be more problematic in the villages. For example, the CBO needs to provide evidence that it already exists (a bank account, a site visit by Umbrella Organization representatives); the members of the CBO are to be elected democratically by the village members; the proposal should follow a standard format (aims, target groups, budget, work plan); and the proposal needs to be in English. Significantly, because of NAC's commitment to the doctrine of sustainability, resources could be requested only for a limited number of activities: training (primarily in home based care or in mobilizing village youth for prevention activities) and seed money for income generating projects to support the committee's activities into the future. The proposals were submitted to the local Umbrella Organization for review by the District Assembly (with the guidance and staff support of the Umbrella Organization), and, ultimately, reviewed by NAC, which made the final funding decision.²²

The actual process was a fascinating demonstration of community ingenuity. First, if there were no existing CBO, which was often the case, one could be formed in response to the NAC offer of resources. The opportunity was too important to be left to village democracy; typically a village notable—often the chief, sometimes a village influential who had contacts in the District Assembly or with the local Umbrella Organization—formed a committee.²³ Since proposal writing is not a familiar village activity, nor are most villagers able to read and write in English, the committee usually included the few village elites (e.g. a schoolteacher, a businessman with extra-local connections). Other villagers may have ratified the choice at a village meeting, but our impression is that this did not happen often, or if it did, ratification was pro forma, as it generally is for other committees mandated by the government (Paz Soldan, 2003). If an existing or a newly formed committee did not have a bank account, they scrambled to open one, either with the contribution of a wealthy patron or by engaging in enough agricultural piece-work to satisfy the requirement.

The Umbrella Organizations reviewing the proposals quickly discovered that even a committee of village influentials did not know how to write a proper proposal. A review of rejected CBO proposals and the rationales for rejecting them from the office of Save the Children, the Umbrella Organization in Balaka, shows that the most common comment was some version of “Need to be enlightened on proposal writing” or “The group has to be assisted in developing a good proposal.” There are also many comments about the need for a “clear work plan,” that the proposal was “not in line with the National Strategic Framework,” or other application formalities. Despite the assumption that local communities knew best what they needed, there was substantial slippage between what was initially requested and what could be approved. Proposals that requested support for activities that were not considered by NAC to be AIDS-related were rejected, as were proposals that were deemed to ask for real resources; “No hand outs relief without means of sustainability” (noted in rejecting a proposal).

The actual proposals are poignant in their detailed aspirations. One [Tiyende Home Based Care], handwritten in pen on lined paper, under “Metho[do]logy and Plan” says “project will be better and fine in twelve <12> months time. We will have seminar to gain more knowledge before the project take place.” It requests funding to train 21 volunteers, 3 facilitators, and 4 “trustees” [possibly “well wishers” who contributed to get the CBO's bank account established] in “business management and CBO management,” with 500 kwacha (about $4.00) a day for the volunteers and trustees, money for two sodas and two buns a day per person, and “stationery,” including a pen and two notebooks for each participant, flip chart paper, and a roll of cellophane tape. The applicants carefully follow the outline they have been given, but supply their own interpretations. For example, under the heading “What Are the Coping Mechanisms of <PLWA> and Their Families,” a proposal elaborated that if “a wife or a husband has got no real love, she or he starts to chase [away] his or her husband” and that “People with HIV/AIDS get worried because it is a killer disease.” The committee members plan to help people with AIDS by “drawing water for them,” “firewood collection,” and offering advice and entertainment.

Once the committee members learned the limits of what could be requested for income-generating projects, they scaled down their aspirations: funds were requested to buy and raise chickens so the eggs could be sold for money to help the elderly and disabled, those living with AIDS, and the youth, or eight treadle pumps were requested for a youth group to irrigate crops that would be sold to support their drama group in distributing HIV prevention messages. The authors dutifully list items under the heading “Project Sustainability,” and “What Will Be Done If Project Has Stopped,” though their ideas here seem mainly wistful. One proposal said, “We will continue informing communities about our economic status”; “There must be transparency and accountability”; and “There should be unity amongst people and our group.” Despite these efforts, however, this aspiring home-based-care group, like most others, had its proposal turned down.

Once they submitted their proposal, the CBO committees waited for the resources to come. Eventually, a handful of CBOs received word that their proposals had been funded. These were usually larger CBOs that had previously received support from other donors. Then there was another wait for the money to be disbursed. The call for proposals was issued in 2003; the first money to be distributed to organizations (including large NGOs and INGOs) was announced in a press release on 1 December 2004 (National AIDS Commission, 2004). Even then, NAC distributed the funds in tranches, such that one tranche had to be accounted for (in the proper form) before another was disbursed. Moreover, even when the accounting was approved, if NAC's budget was tight, subsequent disbursements were delayed. And then, if somewhere in the process there was failure in reporting or managing funds, the whole grants process could be suspended, as happened in several districts in early 2007, due to “mismanagement of funds.”²⁴

While the $10 million that NAC allocated to community mobilization is a substantial sum—and must have seemed fantastically vast to those villagers who knew how much donors were spending on AIDS—the resources given to any single CBO were small. In Mchinji, the MDICP district in the central region, a total of 31 CBO proposals had been approved as of June 30, 2005 (NAC, 2005b), and only 5 in the smaller administrative area where we did our survey—all of them proposing to work in several villages—for an average of approximately US$10,000.²⁵ Some of the money went for the chickens or goats or treadle pumps, but typically the largest budget item was for a training workshop, usually attended by two or three members of a CBO who were then expected to return and train the other members of the CBO. Participation was prized, for it came with per diems to cover transport and at least one meal. The per diems were in the range of US$4.00 to $8.00, large enough so that participants could return home with money in their pockets (interview with a youth representative to the Mchinji District AIDS Coordinating Committee, Mchinji, 2006).

With such limited and erratic resources, it is not surprising that we found little persuasive evidence of sustained CBO activity in the villages. Since we had learned from the 2005 pilot study of youth groups that it was necessary to talk not only with members of the CBO committee but also with village residents who were not members, during the 2006 MRG survey round we asked the six survey supervisors and some of the 45 interviewers who were in the field in Mchinji to take advantage of the endless waiting characteristic of field work to chat with people they met and ask about any CBOs in their village involved in HIV prevention or AIDS mitigation (e.g. home-based care of the sick or orphans). This produced 500 informal interviews scattered across 57 villages.

The most striking result was that in 38 of the 57 villages in which the informal interviews were conducted, multiple respondents in the same village disagreed as to whether there were any CBOs in their village (Browning, 2006). These results were replicated in the survey data, where respondents were distributed across all 69 villages in the MRG sample in Michinji. Survey respondents were asked “Outside of churches and family, is there any group in your village that is assisting orphans?” Here again, in the vast majority of villages (58 of the 69), villagers disagreed about whether a CBO existed or not. In only 4 villages did 100% of the survey respondents agree that there was an orphan group and in another 6 villages a clear majority (70% or more) said there was an orphan group. In one village 100% of the respondents said there was no such group. The results were similarly mixed for the home-based care groups. We conclude that in the majority of villages there were no currently active CBOs. Since the villages are small and residents are keenly interested in who gets resources from the outside and who doesn't, it is highly unlikely that residents were unaware of active community groups (interview with a youth member of the District AIDS Coordinating Committee, Mchinji, 2007). The likely reason for the contradictions in the data is that various AIDS groups had been formed but had not been sustained. A similar question on groups assisting with home-based care gives the impression that there were even fewer of those groups than orphan groups, and in response to a question about groups assisting with economic development, a clear majority of almost every village said there was no such group. The results were similar in the other two MRG survey sites.

The systematic survey data and the informal interviews show that whatever any individual CBO did or did not do and whether or not any CBO was funded by NAC, few “communities” had been mobilized, and that many, perhaps most, CBO activities were not sustained (see also Peters, Walker, & Kambewa, 2008, pp. 22-4).²⁶ Given the meagerness and the types of resources provided under the doctrine of sustainability, this conclusion should not be surprising.

(ii) Sustainability and the creation of “modern” subjects

Although the Global Fund resources channeled through NAC do not appear to have generated sustainable community mobilization, this is not to say that these efforts had no lasting consequences. Villagers learned a great deal as they used their ingenuity to satisfy the expectations of donors while pursuing their own agendas and using donor resources in ways that made sense to them.

Members of the CBO committees who responded to the call for proposals (and perhaps those with whom they talked) learned that donor resources came with conditions. To access the funds they had to jump through very specific hoops, such as organizing a gender-balanced committee to “represent” the community; if funded, they had to fill out quarterly reports for “monitoring and evaluation.” They also learned that even though they might be village elites, they were not competent to write a proposal; they had to identify someone who could write in English and knew the jargon of community mobilization or both.²⁷ As Mansuri and Rao (2004, p. 23) conclude in a review of the literature on “participatory development,” often “elites are the only ones who can communicate with outsiders, read project documents, keep accounts and records, and write proposals.” Nor could they get the resources in the form they might want, such as food to sustain them through the hunger period or micro-credit to support entrepreneurial activities that might be more successful than selling firewood.

Second, committee members learned that as villagers they were not worth being paid: they were expected to work as volunteers. But being a volunteer when others are paid is considered by many to be foolish. The Chairman of a village health committee—composed of volunteers—in Mphepho Village told an interviewer, “The people at the hospital have money for these things and they do not think about us. They told us at the beginning that the assistance to set up these village committees in the rural areas came from Canada. We receive no money [therefore] many people mock us that we are just working for the hospital and they do not take us serious.”²⁸ And even the groups that actually had been struggling to care for orphans and the sick found that the material resources that were on offer were quite limited. It is unclear whether the policy-makers at NAC actually thought that the meager resources provided were sufficient to sustain the CBO activities indefinitely; it seems unlikely, however, that those who knew anything about the struggles of poor villagers to support their families could believe this.

For those villagers who were not on the CBO committee but had heard about it, the lessons learned were also multiple. They learned that there are sometimes resources out there for those who know how to access them. Villagers also observed that, as in the past, village elites dominate the process, thus contributing to envy and mistrust. This was especially the case where the chief, who is expected to redistribute resources that come to the village (Collier, 2004), permitted the committee to monopolize the resources or even kept them for himself. Certainly there is the perception that this occurs: in the informal interviews in both Mchinji and Balaka, we heard complaints of “bad chiefs” who couldn't or didn't redistribute resources that came to the village from donors or government. One village received three cows to produce milk for orphans. But when the respondent was asked if the chief benefited more than others, he said, “What do you expect? His children milk the cows, do you think they go without?”

There is also an important question about whether the all the “training” for which the CBOs are encouraged to apply (as opposed to the material benefits which villagers clearly very much want) really tell villagers anything they do not already know. Often, those who were “sensitized” about the epidemic —for example, through the radio or health talks at the clinics, were told by youth groups what the MDICP survey and qualitative data show they had known at least since 1998: that AIDS is spread by sex and that it can be prevented by abstinence before marriage, faithfulness in marriage, and/or condom use (Watkins, 2004). Those who were taught about caring for “Orphans and Other Vulnerable Children” and “Home-Based Care” may have found the teachings especially ironic, given that these are ordinary family responsibilities.

We have seen, then, that the actual process of community mobilization and empowerment was far from that envisioned by the donors (see Swidler, 2006; forthcoming). Committees were mobilized, but the process was instrumental rather than democratic. Communities were not empowered to ask for what they needed, but rather limited to asking only for resources that donors deemed sustainable. We also learned, however, that neither the resources nor the projects were sustained. Bicycles broke, but no funds could be requested for repairing them; the maize crop a youth group planned to irrigate with its treadle pumps was eaten by hippos from the nearby river, and the pumps were given away; when the small amounts of resources provided to those CBOs were spent, the projects ended.

Despite the ideal of sustainable development, which is supposed not only to improve lives but to engender autonomy, empowerment, a sense of internal and external control—and ultimately the ability of local communities to take responsibility for their own welfare—much of what we found was an opaque, thoroughly capricious process, one which sent many messages to villagers about how disempowered they really were. It is unlikely that the trickle of funding, or even the much wider mobilization to seek funding, could have greatly altered villagers' lives. Indeed, the unpredictable resources donors sometimes provide tend to get incorporated into the same ways of coping with life that villagers usually depend on.²⁹ Just as the weather is unpredictable, the price of tobacco or cotton rises and falls, and other resources come and go, so also do the resources provided by “sustainable” NGO projects sometimes appear and just as unpredictably depart, or, more frequently, are promised but never materialize.

This is a world a modern American consulting firm, or someone trying to pitch scripts for TV shows in Hollywood might recognize. But it is far from the coherent, modern, rational self-reliance the sustainability ideology intends to produce. Indeed, it engenders something closer to hunting and gathering, a kind of unpredictable supplement to the marginal subsistence agriculture that sustains daily life (Bird-David, 1983; 1990).

(b) Aspiring elites

We now turn to the implications of the doctrine of sustainability for a particularly interesting social stratum, those hoping to cross the boundary from unemployed villager to a salaried job. We refer to this group as “aspiring elites” or “interstitial elites.” The difficulty of finding the right term arises because this category is genuinely ambiguous: its boundaries are fluid, and its members' status is often precarious. But that ambiguity is at the core of the experience of these aspiring elites as they struggle to escape the subsistence farming and trading of the village and to reach toward the imagined status and wealth of a job in the formal economy. This often-overlooked stratum provides many of the volunteers who populate the workshops, and even more, the social imaginary that animates NGO strategies. Understanding the situation of this group illuminates otherwise-opaque aspects of the practices that sustain the current embrace between NGOs and locals: the reliance on “training” and on “volunteers.” Through the lens of this stratum's experience we learn further lessons about what teaching a man to fish really imparts in practice.

The aspiring elites are secondary school graduates in a country where only a minority achieve this status; almost all can speak and write English, which also distinguishes them from most in the rural areas. Understandably, they—and their relatives and neighbors—expect them to have a “bright future.” But the defining feature of these interstitial elites is that despite their education and their aspiration to fulfill their families' hopes and escape the agricultural labor and petty trading that define village life (Englund, 2006, pp. 87-95), their lives are precarious. Since jobs in the formal sector are scarce, many are unemployed or inconsistently employed. By village standards, they are an educated elite; by the standards of the elites who work for NGOs in the capital, they are not sufficiently educated to be hired at a high salary to implement donor projects. These aspiring but underemployed elites are precisely the stratum most likely to become the volunteers in donor programs.

AIDS programs often measure their success by the numbers of volunteers trained. Malawi NAC's Quarterly Service Coverage Report (2006a), for example, reported that from January-March 2006, 25,000 volunteers were trained (Table 16); in the previous quarter, the figure was 20,000. We provide two examples of what training—and volunteering—means to those who become volunteers.

Our first example is a young man who currently appears to be on the way to achieving the goal of many youth volunteers: a sustainable career. After graduating from high school in 2002, “I didn't go anywhere,” he told us, He did, however, join a youth group: “It was not my wish to be a volunteer. I wanted money to go to school. We have many people who have done the high school, but for a job you need more education.” When we first talked with him in 2006 he had been a volunteer for four years and was the youth representative on the District AIDS Coordinating Committee. He had also been a founder of a Youth Alive, a group modeled on a program he and others had heard about in Zambia. With the help of a vehicle provided by a parish priest and fuel paid for by raising money from “well-wishers,” his group visited the Zambian group and founded their own branch, which flourished. Youth Alive eventually drew attention from a large international NGO, Family Health International, which wanted the group to help design programs for a UNICEF-funded project. But, he said, “We found that they just wanted to use us. At the time we had no donors so that was useful. But FHI only had money for training, so that was the end.” Youth Alive is “now gone down.” When we returned in 2007, after five years of volunteering he finally had a paid job with a NGO and was hoping to save enough to get another credential.

The second example shows how donors' beliefs about sustainability conflict directly with the aspirations of volunteers. It comes from an evaluation of a Ministry of Gender, Youth and Community Services (2000) project to train peer educators. The initiative was apparently well funded, since the introduction acknowledges the technical and financial support of an impressive list of donors, including major international NGOs and bilateral agencies. For the evaluation, monitors from the Ministry conducted focus groups with youth committee members, who identified “weaknesses” and “constraints” in their programs: “Lack of individual financial gains due to the voluntary work”; “Lack of means of transport which makes it difficult to conduct other outreach activities”; “Lack of capital to start sustainable and reliable income generating activities”; “Lack of stationery and other basic office facilities”; “Lack of training”; “Lack of IEC [information, education and communication] and recreational materials”; and “High turnover.” Despite these plaints from the young volunteers, the monitors make only two recommendations that involve resources: 1) to provide IEC and recreational materials, and 2) training. There is–not surprisingly—no mention of the youths' wish to be paid something, nor do the monitors recommend resources for IGAs. Rather, “Youth clubs should be encouraged to maximize use of locally available resources to implement their activities,” and “Sustainable and reliable IGAs should be encouraged and carried out in youth club activities to facilitate smooth implementation of club activities.” These examples suggest that income and jobs are the real priorities of the interstitial elites who become volunteers. But we shall see that the training NGOs provide can also fulfill other aspirations.

(i) Credentials, training, and aspirations

While the funders' image of trained volunteers are village residents who then go on to care for or educate their fellows, the enthusiastic recruits to training—those who actually attend workshops and found volunteer organizations—are very often aspiring elites who seek not to live in the village, but to leave it (see Bebbington, 2005). Their passionate aspirations fulfill the donors' faith that there is a group wishing to be “trained” and the donors' commitment—echoed by those very aspirants—to the ideal of the NGO volunteer.

At one end of the interstitial-elite range are those secondary school graduates who aspire to find paid employment, ideally an office job in the city with a regular salary. Those who have patrons may be able to do so, but most remain in their village farming or, at best, engage in petty trading in a large trading center or small city while waiting for an opportunity to place a foot on a career ladder.³⁰ Many perceive that the best available route to a steady job is to volunteer, often for several years and for several organizations. At the other end of the interstitial-elite range—but just as essential to its identity and the hopes it inspires—are those with apparently identical backgrounds and skills who do manage to make it in the formal economy. Some of these find a government position, which, compared to employment with an NGO, is relatively poorly paid. Others find a job with the district branch of an international NGO, perhaps beginning as a driver but then advancing to a low-level white-collar position. NGO jobs are relatively well paid—and moreover come with perks such as internet connections and white Land Rovers—but they are insecure: the growing number of NGOs increasingly compete for donor support, scrambling for short-term contracts (Cooley & Ron, 2002) and firing and rehiring the reserve army of interstitial elites as successive humanitarian crises capture international attention.

The very insecurity that the interstitial elites face—both the intermittently employed and the still aspiring—increases their eagerness for more training. For the aspiring elites, training workshops satisfy a variety of needs. The most obvious is cash: the per diems and travel allowances may be the only source of cash that permits young people to contribute to their family's livelihood and, if the workshop funding is generous, to buy a new shirt for themselves (see also Smith, 2003). A second, and to us unexpected, advantage of workshops is that at the end participants are given what they consider to be a valuable credential. As one participant in a workshop in rural Balaka district explained, “Life is very competitive here in Malawi, with the next credential maybe I could get a job in the city.” We have been struck by the hunger for credentials, for certificates, apparently regardless of the topic or degree of expertise that they certify. A third advantage of workshops is that they permit participants to make contacts that may, someday, be useful in obtaining a job. In Malawi, patron-client relations are foundational. In an insecure world, everyone needs as many and as well-placed patrons as possible (Barnes, 1986; Chabal & Daloz, 1999; Smith, 2003; 2006; Peters & Kambewa, 2007; Swidler and Watkins, 2007). At a workshop, one can expand one's networks through contacts with the trainers or the trainers of trainers, with other aspiring elites who may know about available jobs, or even with foreigners staying at the same motel who may potentially offer resources.

Despite our initial cynicism about the money spent on per diems and travel allowances, certificates and contacts, for the participants these do not exhaust the appeal of workshops. Youth (usually young men and women in their twenties or early thirties) loved them. For one thing, they were fun: away from home, trainees could pool small amounts of their per diems to hire a boom box and dance the night away, or sing hymns into the early hours of the morning. More importantly, the youth were deeply convinced that through the workshop they had become “enlightened,” a word that has religious overtones (Englund, 2006). At first we were disbelieving: we read the training manuals composed in Lilongwe or South Africa, which largely disseminated information that our project surveys and qualitative data showed was already well-known to young adults in rural Malawi. But we became believers after a series of interviews.

Our enlightenment began when we asked a young man of 22 what he had learned from his four days in a workshop on “Life Skills.” He answered “decision-making.” We joked, “Surely you knew about decision-making before you came here!” No, he insisted, he learned it here. After probing, he agreed that of course he had made decisions in the past, but he didn't know it was called “decision-making.” ³¹ Similarly, a female participant, age 24, explained that she learned “gender roles.” In his excellent ethnography of human rights organizations in Malawi, Harri Englund (2006) considers the acquisition of NGO jargon to be important because it permits the volunteers to sustain their identity as different from village folk and to show potential patrons in the NGO community that they are cosmopolitan. We agree that this is an important consequence of the workshops supported by the ideology of sustainability. But we also find that workshop learning has a more profound significance.

It is hard to doubt the evident passion expressed by a young woman volunteer with Youth Against AIDS when we talked with her at the Mpaweni Motel, at a dusty, rural crossroads, in 2006. The NGO began as a youth club in 2002. “At first we started with nothing, I could beg my parents could you please give me some money, 100k [about US$1.00] to contribute to my group. We were walking on foot for 20-25 km to go to some villages to give the messages.” Late in 2005 the NGO received US$32,000 from NAC, which helped them to get a bus to expand their activities.

How can we be quiet, it's us the young people who are infected…. We train them so that they can resist pressures from their friends, so they can negotiate. We teach them assertiveness, self esteem, decision making, communication skills, so that they can know that their friends are getting the messages we are giving them, they can learn how best to deliver the information we can give them. Information on relationships, how can they have better relationships with parents, friends, friends of opposite sexes…. [N]ot all of them went to school. So we have to train them how their bodies work, each and every part of their bodies, so they don't have pregnancy.” [Interview with Dusila, 23 June 2005]

Far from alienating these young volunteers, the sometimes arcane, elaborately formalized lessons about anatomy and leadership skills fulfilled deep aspirations for identity and status, preserving symbolically the sense of self that the pragmatic demands of their everyday lives continually threatened to overwhelm.

(ii) Social mobility and identity

For the interstitial elites, the hopes sustained by training and volunteering offered something more precious (at least in the short term) than material rewards: a modern, educated identity (see Johnson-Hanks, 2002; 2006). One of our most remarkable interviews was with two young men who had been part of a peer-education AIDS training, enthusiastically debating whether using condoms could be reconciled with the Bible's prohibition on spilling one's seed. The young men told of having applied through the local World Vision office for funding for their youth AIDS club, returning some months later to inquire about their application, and being shouted at and told to go away. We expected they would have been angry. Their response was that, no, they weren't angry. They said, “for us, World Vision is a ‘glittering castle.’” Their story was that the glittering castle (the mirage) WV held out is what led them to form their group and to pursue their important mission. Obviously they had managed to get funding to finance the meeting at the motel where we met them. But the real point was that World Vision and the other funding mirages had allowed them to pursue the “trainings” that kept alive their sense of themselves as modern—in their clean, pressed western garb, discussing decision-making, the physiology of sex, and the morality of condoms—in the absence, but still in the hope, of jobs in the formal economy.

Many “AIDS clubs” are formed by secondary school students. After they graduate, they can use the club to stay together and affirm the educated modern identity they have worked so hard to attain. Together, they keep alive the hope for—and the networks through which they might hear of—possible jobs, perhaps first as workshop trainers, then trainers of trainers, then even as paid staff. But in meeting in seminar-like settings and conducting earnest discussions of important topics like reproductive anatomy or how to conduct a youth workshop, they also enact and thus reaffirm a central identity.

(iii) The hidden curriculum

What these aspiring elites really learn, then, goes far beyond the formal content of their “training.” On the one hand, they display enormous commitment to the forms of rationality and self-development the modern sector seems to make possible (Nguyen, 2004; 2005; Englund, 2006). They become invested in the sense of dignity that comes from formal education and credentials, in ways that are consequential for the aspirations of AIDS NGOs themselves. For these aspiring elites, whose families have sacrificed to pay school fees and allow their bright youngsters to succeed in school, the dignity that comes with that education is of great value, as is the hope that volunteering might eventually lead to paid employment. Indeed, one of the major things these precarious elites learn is their distance from the “backward” villagers whose ignorance it is their mission to correct.

The logic of sustainability reinforces the interstitial elites' commitment to esoteric knowledge. If funders won't finance substantive projects (VCT, nutrition supplements, paid health-care workers, paid teachers or counselors) on an on-going basis because they wouldn't be sustainable, then “training” is one of the only fundable activities. And what is all that training to consist of? Since the training is in some ways an end in itself for both donors and those trained, the content of the training becomes elaborate formalizations of what would otherwise be common sense.

This tendency toward ornate embellishment of what might seem obvious or trivial matters is exacerbated by the economic logic of the volunteer/training system. If the only way volunteers can legitimately receive material compensation in a sustainable program is through the per diems and transportation money they receive for training, then training needs to come often to keep volunteers “motivated” and there need to be “refresher” courses. Taining and workshops provide the occasions when like-minded aspiring elites can gather, soaking up authoritative wisdom and reinforcing modern identities.³² So sustainability both requires and reinforces not only the link between volunteering and the modest rewards of attending workshops and participating in training, but the elaboration of complex truths which are taught and reinforced in those workshops. Belief in these complex truths in turn strengthens volunteers' commitment to identities that the trainings reinforce, as donors hold out hopes of funding, or even jobs, which all too rarely materialize.

(c) National elites

Although the aims and ideologies of foreign aid come from abroad, the donors depend on members of the national elite to actually implement projects. The national elites are middlemen, brokers strategically placed at the intersection of international and national networks (see Evans, 1979). Such middlemen are especially critical for projects—like educational, human rights, and AIDS interventions—that seek to change the human materials, rather than simply the physical infrastructure, of other societies.³³

In some ways the university-educated Malawians who staff the international and national non-governmental organizations, as well as civil servants and academics who supplement their incomes by consulting for international agencies, lead lives that resemble those of their professional and government cohorts in the Global North. But the actual texture of their lives is radically different. The clearest difference is simply the number of commitments African elites juggle, and the constant disruptions to their work. In Lilongwe, at NAC and other ministries, we met many capable, effective professionals. But the fact that we met them at all suggests the problem. In the AIDS world such professionals are continually expected to stop their work to meet and greet visitors, adding to daily calendars that are already filled with meetings and routine office tasks.

These professional responsibilities are not, however, the national elites' only tasks. They also deal with the demands of their personal lives. Aside from ordinary domestic responsibilities, members of the national elites typically maintain a whole separate sphere of responsibilities to their families, which often means their home village, clan, or lineage group. Since they are educated and wealthy, relatives send them orphans to raise, who arrive on their doorsteps hungry or ill, and they expect these cosmopolitan kin to solve all sorts of family problems, from a young person who is going astray, to a baby who needs medicine, to a cousin who has been turned away by the hospital offering antiretroviral treatment.

There are three crucial differences between the experiences of the national elites and the other two strata. First, they are largely exempt from the doctrine of sustainability. Unlike the villagers and the aspiring elites, they are paid by the donors. By the same token, however, their livelihoods depend on foreign funding. They perceive, perhaps rightly, that this means that they have to dance to the donors' tune (Horning, 2008).

Second, the national elites' income supports a lifestyle that distinguishes them markedly from the villagers and the aspiring elites. The continual array of seminars, dissemination workshops and international conferences demanded by donors has, over the decades, inducted participants into a common style of life, which includes transportation in donor vehicles and expense-account meals, and the expectation that they will be able to provide their families with first-world health care and their children with first-class educations. As long as their income is sustained—as they have reason to believe it will be—so also is their lifestyle. At the least, they do not expect that they or their children will ever live in a village.

A third, and related, distinction from the interstitial elites is the national elites' confidence that their identity as cosmopolitan is secure. Their education, the cultural capital they have accumulated (signaled, for example, by the fluency with which they speak the jargon of the international aid community) and, no less important, the potential patrons and potential clients they have developed will continue to support an identity that is distinct from that of the other two strata, an identity that permits them to enlighten international visitors about their nation's “traditional customs” without compromising their own claims to modernity (see Collier, 1997).

Despite the national elites' fundamental confidence that they will never return to the village, they too live in an insecure environment. The donors' commitment to sustainability means that projects are short, or at least meant to be short; additionally, the explosive growth in the numbers of NGOs since the mid-80s has meant more competition for donor funds (Cooley & Ron, 2002; Heimer, 2007).³⁴ Thus, no matter how much an individual may be personally committed to family planning or to AIDS prevention, he or she must always be prepared to find another project. This means they must scramble to accumulate resources to permit them to move from one project to another.

Through their own experiences and observing or hearing about the experiences of others, elites have also learned a hidden curriculum: the skills necessary to adjust to an environment of “sustainable”—if in practice unsustained—projects. They also learn that it is not only their performance in a particular job that will provide for their future, but even more that their network of contacts, their patronage ties, and their good political relations with other members of the elite really determine their futures. Thus, much like the interstitial elites or the commercial elites of 19^th century France described by Balzac, they are in constant Brownian motion.

Donors think that through “capacity building” NGOs will become more professional, more organized, more modern. But the doctrine of sustainability has taught another curriculum, with opposite effects. Elites' need to accumulate resources to make their incomes secure leads to lives that are highly irrationalizing: multitasking, fragmented, fluid, opportunistic, as they constantly scan the environment for new opportunities and try to juggle multiple responsibilities. The critical resource that allows the elites to remain buoyantly suspended above the froth of projects and donor interests is their network of contacts, being on good terms with those in other ministries, in academia, and in other NGOs, and being visible to potential patrons and employers. One cannot hole up in the office working on a single task, just as one cannot count on a unified, linear career track. This is a straight leap into “post-modern” fragmented lives and hybrid identities, as one jets off to international conferences and comes home with generous per diems that allow one to buy land on which to grow enough maize to feed one's many dependents. Donors' emphasis on sustainability means that no one focus is sustainable; even members of the national elite must jump from one unsustainable task to another.

4. Conclusions: Sustaining the Unsustainable

In examining the effects of AIDS NGOs in Malawi, we have confirmed some of what is already known: that projects for community mobilization frequently fail to achieve their goals and that resources are sometimes diverted from their intended beneficiaries. But we have also pointed to the importance of the culture and the social imaginary of the donors as it interacts with the aspirations and interests of three strata of Malawian society.

By taking advantage of unusually broad and deep evidence from three regions in rural Malawi, we have shown first how erratic, patchy, and uncertain are the effects of AIDS funding for Malawian villagers. While global donors aspire to mobilize and empower the desperately poor rural Malawians, their efforts rarely trickle down to the villages, and when they do, they arrive in forms that tantalize with the promise of added resources but that can rarely stimulate the sort of sustained collective enterprise donors imagine. Instead, village influentials strategize about ways to camouflage what they really need—support for the elderly, the poor, children, and the ill; agricultural inputs like fertilizer or breeding hens; blankets and school uniforms—in ways that meet donors' arcane criteria. What villagers are allowed to ask for, under the guise of empowerment and sustainability, are paltry sums for training or small amounts for what are supposed to be self-sustaining income generating activities. Most of villagers' laboriously-produced proposals are turned down because they fail to meet stringent donor criteria for the form of proposals (with precise budgets and worksheets) or the activities donors want to fund. Even when projects are funded, funding periods are surprisingly brief—usually well under a year—and projects can be cut off abruptly if the required reporting is not done correctly, or if any of the links in the funding chain breaks. It is no wonder that the villagers we interviewed were uncertain about whether CBOs were operating in their villages.

We focused especially on the ideology of ‘sustainability’ because this aspect of the donors' social imaginary creates such bitter ironies for the stratum of interstitial elites, those eager to escape the village through participation in AIDS organizations. The “glittering castle” of training, certificates and the possibility of a paying job lead these young men and women to embrace opportunities to volunteer in donor-sponsored projects. Some of them volunteer for years, always hoping that the NAC grant they had submitted years earlier would finally come through, or that the workshops and meetings they attended and the certificates they earned might someday lead to a paying job.

The sustainability agenda has yet other unintended effects. Most generally, in contrast to its promise of autonomy, empowerment, self-reliance, and a coherent, rational modernity, the actual practices dictated by the sustainability doctrine have created nearly its opposite for each stratum of Malawian society. To meet the stringent criteria of sustainability, villagers must cast their quite pragmatic concerns in the often risible language of “gender sensitization” and “decision-making.” Furthermore, the unpredictable trickle of actual resources does nothing to mitigate the insecurity of rural Malawians' everyday lives. For national elites, who benefit enormously from the ways of life made possible by donor humanitarianism, the actual texture of life is more post-modern than modern, as they must break up their work days and their sense of purpose to blow with the winds of institutional fads and fashions. And, finally, for the aspiring elites, whose imaginations, we have argued, are most profoundly shaped by the promises NGOs make but rarely deliver, the clash between donors' dreams and local realities creates the most powerful effects. The interstitial elites really do aspire to the expert knowledge, the self-improvement, and the social impact the NGOs seem to promise. But because the donors are so committed to sustainability, they offer these intelligent, educated locals only years of insecure work as “volunteers,” punctuated by occasional access to “training” in knowledge and skills that are often irrelevant. The NGOs inspire aspirations to a rationalized modernity and then deflect them into an opportunistic, anxiety-provoking chase after an ever-receding mirage.

It is hard to say precisely what constructive recommendations follow from the perspective we have offered here, but we do have several suggestions. First, the ideal of sustainability is a convenient self-delusion for funders and they would do much better if they could systematically and rigorously determine what projects are effective and then sustain them by paying local workers to actually do good—provide health care, sell discounted seeds and fertilizers, treat STIs, provide ARVs, supply children with books and school uniforms, or care for the ill and elderly (Kremer & Miguel, 2007). Second, since few of the approaches to AIDS prevention currently in vogue have shown any measurable effect (Potts et al, 2008), we encourage funding that responds to Malawians' desire to take care of the vulnerable in their communities, provide for their children's futures, and build economic security, independent of the issue of HIV and AIDS. Indeed, reading the proposals that Malawian villagers submitted in their usually vain attempts to gain access to AIDS funding convinces us that villagers do know what they want, but little of it is training in how to prevent, mitigate, or treat AIDS. The first two they already know how to do as well as the experts who try to advise them (Watkins, 2004), and treating AIDS has to be done through the health-care system.

Finally, we suggest that donors consider the “hidden curriculum” their procedures teach. Requirements for elaborate proposals, bank accounts, and monitoring and evaluation might better be replaced by simple procedures that would funnel more resources to villagers and less to monitors. Such resources would create continuing projects that both villagers and employees (perhaps the brighter, more successful of the villagers' children) might rely upon. Rather than projecting a social imaginary that they find morally gratifying, donors and NGOs might provide opportunities that could sustain the realistic aspirations of those they claim to help.