3.1.1 Context

The particular project I recount here is a stylised amalgamation of projects I have encountered while in the employ of multilateral donors, bilateral donors, implementing actors, and universities. The project takes place in Country, a stylised small country on a coast of sub-Saharan Africa (reflecting a mix of my experiences across West and East Africa).Footnote ³ My position in this project is as an employee of the ‘Development Agency’, or ‘DA’, itself a synthesis of my work with the World Bank, the UN, DfID, and other such organisations. I work for the DA’s rule of law department. Counterparts include:

• a government agency (the National Agricultural Agency or ‘NAA’);

• a bilateral donor who was one of the other major aid providers to Country (the Other Donor, or ‘OD’);

• a mid-sized agricultural multinational with an early-stage large agricultural concession in Country (the Agricultural Concessionaire, or ‘AC’); and

• a national NGO with a focus on development and human rights and experience as an implementer of development research and projects (‘the NGO’).

The thematic substance of the project is agricultural reform, specifically the provision of market access and capital to smallholder farmers. Participants in project implementation span donors, researchers, project staff, national administrators, NGOs, local communities, and companies. My reflections on and representation of the project’s substance – the legal frameworks, the financial state and macroeconomic importance of the sector, and so on – are faithful to the specific country contexts in which I have done this work. My reflections on and representation of the project process are similarly faithful.

Throughout the chapter, Jackie, Greg, and Ted provide commentary on my accounts of the project in comment bubbles in the margins of the text – a common modality of asserting one’s knowledge in development work, with an aesthetic that is literally marginal. At the same time, it invokes the ‘humanist legal tradition’, in which the law as such emerges from the conversations, contestations, and accretions of marginal engagement with the text and which works to interrupt social-scientific studies of law through an engagement with law’s forms and open-ended textures – in line with one of the themes of this book.Footnote ⁴

Jackie, now working for the DA, oversees the project as a whole from the DA’s headquarters as part of the portfolio of projects she manages. Greg, by now working for a smaller European bilateral donor with an interest in Country, observes the project from a distance through our chats about it as well as information refracted through his colleagues working on other issues in Country. In Country, I work with Ted, a Country national with a doctorate from the USA who has joined our team to lead on-the-ground implementation of the work.

Through the three different accounts of the project – as well as the comments by Greg, Jackie, and Ted – the chapter as a whole concretely represents how ignorance about the rule of law destabilises a set of stable positions: the participants are each other’s’ experts and laypeople, insiders and outsiders (to the project), subjects and objects. At the same time, it shows how each account structures and limits such representations, since the form of each one affords the project a particular type of coherence (thence my understanding of the accounts as genre pieces). In doing so, I show how each account overlays experts’ willingness to be ignorant onto the contours of expert authority. Thus, in the organisational sociological account, I focus on the expert social structures that make ignorance of the rule of law stick: epistemic communities, bureaucratic constraints, and so on. In the critical discourse analysis, I focus on the discursive conditions that offer hidden closure to open-ended ignorance about the rule of law – and which inflect experts’ ignorance with ideology and hidden aspirations of world-making. Finally, the ethnographic account of expert practices is not so wedded to an underlying order that makes ignorance analytically possible. Instead, it focuses on the routines and socio-material networks in which experts are entangled, such that the project slowly coheres. As such, in this account, experts’ ignorance meets real practices.

For each account, I introduce the genre, write a generic account of the project, and offer a conclusion about the project from within the genre. At the end of the chapter, I reflect on and contrast the genres. I explore how each form regulates the content of projects, their protagonists, their spatial and temporal boundaries, their function, and how they couple with and shape local political economies. I argue that, in doing so, these different genres give shape to expert ignorance and thus might not fully capture its workings and effects.

3.2.1 The Project

A few years ago, a colleague at the DA asked me to participate in a project designed to support smallholder farmers in Country. The project, then in the planning phase, was designed to provide loans totalling tens of millions of dollars to local smallholder farming cooperatives. Agriculture contributed over 50 per cent of Country’s GDP and had done so for the past decade; it employed over two-thirds of the workforce and was dominated by the production of staple crops. Cash crops constituted less than a quarter of agricultural output. Country’s government had decided, following broad multi-stakeholder consultation (funded by the World Bank and some bilateral donors), to make commercial agriculture the main engine for socio-economic growth. The project was intended to support the scaling-up of smallholder farming in a sustainable and socially sensitive way; it would be linked to market access, technology improvement, and value chain integration projects run by other donors and the government.

My colleague, an agricultural economist, went on to explain that some cooperatives had already been constituted as part of smaller projects by other donors, including the OD. Others were in the process of being set up (although he did not know by whom or in what form). My colleague explained that he had visited some of these cooperatives in the field; he knew that these financial flows would boost productivity and increase scale when paired with another component of the project to build roads to local markets. The cooperatives were thus a key component of the project – allocators of the capital inputs that should spur growth. Choosing them, then, was an extremely important moment of decision within the project cycle. The problem, as he put it, was ‘politics’ – the risk of local elite capture of the funds, inappropriate or inefficient selection of beneficiaries, and land conflicts between cooperatives and non-coop farmers, to name a few.

My colleague claimed that for the project to be successful, the choice of cooperatives would have to be locally contextual, locally embedded, fair, and legitimate (although he did not specify what that meant). As a result, he figured that the cooperatives should be chosen by as-yet-unconstituted local multi-stakeholder committees. Would I (and my rule of law team back at DA headquarters) support the project by helping to design the multi-stakeholder committees to mitigate capture and contain conflict? In particular, he hoped that we could incorporate social accountability and grievance redress mechanisms into the functioning of the local multi-stakeholder committees – so that they could monitor cooperatives, hear peoples’ grievances about them, and hold them accountable. He was asking me because one of my old teammates in the rule of law reform unit had worked with him in the past and ‘opened his eyes to the fundamental importance’ of law and accountability. He would allocate project funds to support our work.Footnote ⁹

I discussed this with my rule of law colleagues upon returning to HQ. We didn’t really know what he meant by multi-stakeholder, social accountability, and grievance redress. But this sounded like a good opportunity to demonstrate how we could add value to big projects – and secure a bit of a reputational boost for our team across the DA (and perhaps a financial one, if he would pay for our work). We had long been working on the ‘community’ dimensions of natural resource extraction in the region – including some limited work in Country itself. The use of natural resources as a driver of development – oil, mining, forests, agriculture, and the like – changed the nature and distribution of land and rights over it (from ownership rights, to usufruct rights, to the political power of Chiefs that emanate from their symbolic stewardship of land use). We had long-standing concerns about the local effects across the region of struggles emanating from the changing valence of land – and how those effects might accumulate and intersect with national political and developmental trajectories, undermining development objectives, transforming governance, and potentially increasing the risk of violent conflict. As a result, we believed that our role was to grapple with social and legal accountability and dispute resolution institutions that might (fail to) manage the evolving political struggles over land and farming and the consequent patterns of social, economic, and political marginalisation.

We also knew that there was a clear space for this sort of work in Country. An instrument called an ‘Agricultural Development Agreement’ (ADA) was enshrined in a section of broader agricultural legislation passed a few years earlier (the Agriculture Act, or ‘AA’). The AA was funded and driven by a group of bilateral and multilateral donors including the OD and the DA – a standard piece of ‘technical assistance’ aimed at modernising the agricultural laws (the precursor legislation dated from British colonial times and had been updated in the 1970s) and facilitating inward investment in the sector. Country had been marked by sporadic – and frequently, but not exclusively, ethnic – civil conflict, in particular a protracted and brutal civil conflict in the fertile north that had ceased five years earlier. Its government was keen to show that it was open for business and had determined that agriculture would be a key sector to attract foreign investors. The AA had been drafted by a Western consulting company, which had copied some of the language from similar legislation in countries such as Sierra Leone, Nigeria, Australia, and Canada. Not simply an investment law, it contained various provisions for the governance of the social, environmental, and labour impacts of agricultural investment, of which ADAs were one.Footnote ¹⁰

The AA specifically required that agricultural companies establish ADAs with the main communities whose land played host to the investment (to be determined in the first instance by an agreement between the local government and the company); companies would have to contribute a minimum of 0.1 per cent of annual revenue to a community development fund, whose governing board and scope of activities would be governed by the provisions of the ADA. The general thrust of the ADA provisions was that the participatory governance of agreements was an end in itself (although not stated explicitly, in contrast to ad hoc or local elite-captured corporate redistribution of rents at the community level). The law mandated that the fund be governed by a local multi-stakeholder group, which would act ‘transparently’ and in the interests of the community. The group was required to reserve positions for the local government and community representatives.

The legislation also gave the National Agricultural Agency – an executive offshoot of Country’s agriculture ministry – the power to supervise the implementation of ADAs. Subsequent regulations also mandated that the governance board be gender-sensitive, suggested – but did not require – that funds be distributed to communities on the basis of small project proposals that they would prepare and submit to the board, and listed a number of things the fund’s money could not be spent on (like private vehicles). Beyond that, the AA provided little detail. For example, it did not specify the process to determine who the ‘main’ community (or ‘MC’) was. Should the discussions between the local government and the company involve community representatives of some sort? What should the involvement of local Chiefs (key traditional power-holders) be? And what role would landowners play? In general, Country’s government had not yet worked out how to implement the ADA provision but had convened a taskforce (funded by the OD) to do so. The group had begun to meet and was hammering out a document that, depending on who in the taskforce you asked, contained ‘model’ ADA provisions, or a ‘framework’, or ‘principles’ for ADAs. The first question they needed to tackle – but had yet to do so – was how to identify the MC, who would then go on to negotiate the substance of the ADA.

Because of the broader scope of our work, we had seen similar provisions in natural resource governance legislation across the subcontinent. Indeed, there was a loosely ordered global community of natural resource governance professionals and policymakers.Footnote ¹¹ Many among them were interested in the effectiveness of such instruments across the world in promoting socially responsible resource investment, sustaining ‘good’ governance in the sector (usually taken to mean a combination of transparency of resource flows, accountability for their expenditure, and procedures to mitigate the social and environmental impacts of sector activities), and reducing conflict risk around investments.

The causal ideas about the benefits of local development agreements held by this community emerged from experiences in the Global North, particularly Australia and Canada, which have developed a significant body of policy experience in and academic analysis of their implementation.Footnote ¹² According to these ideas, agreements are ‘a means of resolving disputes, delivering government programmes, or establishing common understandings …’ between main communities and companies.Footnote ¹³ They do so by ‘i) [addressing] the adverse effects of commercial mining activities on local communities and their environments, and ii) [ensuring] that [communities] receive benefits from the development of mineral resources’.Footnote ¹⁴ These, in turn, are the products of four new governance ideas this community holds about the functioning of agreements. They ‘respond flexibly to local conditions’, ‘achieve lower regulatory costs by stimulating collective action’, ‘reduce transaction costs associated with fragmented service delivery’, and ‘increase legitimacy through increased participation in decision making’.Footnote ¹⁵

As a result, agreements had become a key part of the natural resource governance policy toolkit. The World Bank, the UN, and private and public donors of many other stripes commended them and have pushed for their implementation.Footnote ¹⁶ The enforceable legal form of these local development agreements is, in the view of this global community, key; at the same time, scholars and policymakers clearly recognise the importance of the gap between the law on the books and the law in action. Indeed, this implementation gap has become the central object of study by this global community. As a result, they have produced a wealth of loosely social-scientific case studies to accumulate knowledge about the validity of their normative and causal ideas.Footnote ¹⁷

We had contributed to the discussions among this group by participating in global conferences and writing papers and blog posts. We did so having adopted a marginal perspective, claiming to integrate insights into the debate over local development agreements from the experience of our other community: rule of law reformers. Specifically, we drew on insights from a group of reformers who believed like us that rule of law reform was a type of policy change that provided the basis to produce other policy changes – it was both a distinct means of social organisation and a cross-cutting good that shaped the procedures and allocations of rights through which development occurred. This community was geographically diverse, including practitioners from various donors, state actors, and NGO members in Country. The community was avowedly mixed-methods in its assessment of how and why the rule of law mattered and comprised lawyers and social scientists. Community members shared a fundamental belief in producing legal change that emphasised bottom-up and contextual perspectives on what the law should do rather than approaching the rule of law as a policy object to be deductively designed and centrally implemented.

ADAs, then, were a good hook to engage with my economist colleague’s concerns about ‘politics’. They were expressly concerned with the social changes wrought by agricultural reform. They were a pre-existing instrument that was not overly prescriptive and had not yet been implemented; they might thus function to fix the ‘political’ challenges our economist colleague had mentioned. The participatory governance mechanism might help in setting up a process to pinpoint beneficiaries, disburse funds, and reveal and mitigate land conflicts. At the same time, domestic ADA implementation in Country furthered and contributed to a broader global agenda that meant we would not just reactively be dealing with the specific political problems of this project – indeed, we could make use of the comparative experiences and legitimacy of the global natural resource governance people in turning a component of a project into a globally relevant experience.

Our institutional setting, however, was still oriented around projects as ‘privileged particles’. It would not be easy to turn a component of the project into something semi-autonomous and globally relevant. On my return to headquarters, my team discussed our approach. We decided we should begin with the problems our economist colleague was trying to solve. Did we know what sort of land conflicts might arise and what local elite capture of funds really looked like? Between us we had only spent about six months in Country (in contrast to the economist, who had lived there for many years); we would have to conduct some research. This, of course, might be challenging in an institution whose core function was to develop and implement projects.

We decided in the end that our fundamental aim would be to find someone intelligent, credible, and entrepreneurial who we could install on the ground.Footnote ¹⁸ That person could build relationships within the project team – so we would be team players within the context of the DA. They could also network with members of the DA’s country team in Country to see if we might be able to support the DA’s work in Country more broadly (and make its projects more sensitive to the local governance issues we cared about). At the same time, that person could use some of the agriculture project’s funds to support (and also try to steer) the implementation process for ADAs: as it stood, the NAA was being funded by the OD to implement ADAs. For the OD, implementation meant getting an ADA on the books – they simply wanted the NAA to design a model draft agreement and then have a handful of agricultural investors sign versions of it. Finally, that person could also network with other donors in Country, like the World Bank and the Swiss government, who might be interested in the process and who generally have a greater propensity to fund research that would support both that person’s job and potentially a couple more local hires.

We hired a Country local named Ted shortly afterwards – a former long-term consultant with the British government with a PhD from an American university, who knew the sector and was also a strong field researcher.Footnote ¹⁹ He began participating in the ADA taskforce, bringing with him the promise of some project money for ADA implementation. The taskforce was considering how to set up a multi-stakeholder process to determine the identity of the ‘main’ community, which included working out and engaging the broader universe of communities affected by agricultural concessions, from which the main communities would be drawn. This also involved working out who the key local powerbrokers might be on whom the success of ADA implementation might rest (e.g., how important were the local Chiefs, and in what ways might they be engaged without risking wholesale capture of the process?).

I then travelled to Country for a three-week trip. Ted had organised an initial workshop in the NAA’s offices in the capital for the members of the ADA taskforce. This included NAA officials, the OD, the NGO, and the AC. After the workshop, I would travel ‘outland’ (the local term for the further-flung non-urban regions) to observe the elections of community representatives who might negotiate the terms of the ADA. I would then spend a couple of weeks in the field with Ted doing some scoping research around a few potential agricultural concessions to give me a basic sense of some of the ADA-implementation challenges as well as the challenges we might face in executing a research programme (and thus some insight into the quality of data we might generate).

We developed the agenda for the workshop in collaboration with the taskforce participants. The OD, NAA, AC, and NGOs wanted to focus on global best practices on community engagement, the lessons from which would flow through the identification of main communities to the establishment of governance boards to the implementation of projects. My team had developed a set of briefing documents for the taskforce, circulated in advance, that set out comparative practical examples from other experiences of local development agreements (mainly from Australia and Canada). At the same time, the documents stressed that these experiences were not prescriptions and that the best approach would be contextual; in particular, they raised the importance of the context of each concession when identifying the specific beneficiary or ‘main’ communities and the exact amount or percentage of an ADA.Footnote ²⁰

At the workshop, the participants asked Ted and me, as ‘technical experts’ on ADAs, how they might best structure ongoing community engagement. We referred back to the briefing notes: lessons from rule of law reform suggested that community engagement was socially contextual and could take the form of consultations, public forums, dialogues, village-by-village negotiations, and so on. However, a common set of process norms should be put in place to ensure the meaningful participation of marginalised and vulnerable groups. Lessons from natural resource governance reform suggested the importance of quickly putting in place a community engagement process to make communities feel like they had a voice with respect to the project. The challenge would be to work out the right balance between flexibility or responsiveness, and the urgency of providing a determinate structure for dialogue, in the particular context of each potential concession. The AC’s concession would be an important testing ground.

The other participants provided their views. The main NGO representative talked about the NGO’s wealth of experience engaging with communities and helping communities assert their rights against companies and the government. They argued forcefully that they would be best placed to run and manage community engagement as they had the scope (they had chapters in every local district in the country) and depth (they had built a significant community trust over the years) to ensure that engagement would be meaningful. The representative from the OD agreed. She asserted that the taskforce should fund the NGO to conduct outreach to communities and help them establish representative committees so that companies could negotiate and sign ADAs with each one. The AC said they would defer to the NAA, as they held firm to the view that it was the NAA’s responsibility in law to manage community consultations.

Ted and I expressed some concern. From a natural resource governance perspective, did we know which communities were MCs? And from a rule of law perspective, what would the process and threshold be to determine the answer to that question (and could they be derived from the context of each concession)? Both the NAA and the OD referred immediately to the law but with differing interpretations. The NAA asserted that the ‘main community’ meant the people impacted by the concession. The OD said that ‘main’ referred to geographic proximity to the boundaries of the concession.

By contrast, the NGO representatives argued that we should not get too hung up on the law. Another said that we should focus instead on the ‘participatory method – where everybody is informed and involved’. The law didn’t capture the key local actors anyway – the local councils, traditional Chiefs, local landowners’ associations, smallholder farmers’ groups, and so on. Moreover, the NGO had a long history of working with these actors to come up with locally sustainable decision-making processes. They pointed to discussions that had already begun around the AC’s concession area. The discussions focused on who the main community would be for the ADA. A local Chief, whose chiefdom was closest to the concession, had asserted that his chiefdom was the only main community. However, his chiefdom included half – but not all – of a major town in the area. The other half of the town belonged to another chiefdom. The NGO had convened an informal stakeholder meeting between the AC, the two Chiefs, and the head of the town council to begin discussions over how to divide up the main community (although they had not yet developed a standard to determine ‘main’). This, they said, would continue, an example of the ‘pragmatism that must come in [implementation] as well’. Involving communities and local powerbrokers in the identification of ADA beneficiaries would guard against local elite capture and at the same time against those who would ‘take the law and ride it’ by ‘pick[ing the choice of main communities] and tak[ing] it to court’.

All the members of the taskforce agreed that the taskforce should build on the NGO’s existing work and proceed with community engagement. We should bracket definitional questions and not let them inhibit us; the engagement process would lead to some sort of consensus (whether practical or legal) on the identity of the community. And the multi-stakeholder committees should be designed in such a way that their membership and constitution could be revised as different communities came forward to stake a claim for inclusion and as the impacts of the concession evolved. For the OD in particular, the most important thing to do was to get something signed and then keep revisiting the definitional questions throughout implementation.Footnote ²¹

The taskforce members were also not concerned by the risk of conflict between communities fighting to be defined as ADA beneficiaries, nor were they altogether concerned about the cost of the community engagement process with respect to the actual value of the ADAs. Ted asked the taskforce members what they estimated the value of the ADAs might be per year so that we could contextualise how much would be expended in implementation (and how much might be left for development projects). No one was able to answer his question, although the NAA and NGO representatives asked if we might provide technical input on the global best practice on the amount that concessionaires should put into the local development fund – and whether 0.1 per cent of revenue was too low.

I subsequently met for a drink with our main counterpart at the NAA to discuss how to relate the ADA implementation work to our agriculture project, such that the project’s funds could be used to support the implementation of ADAs. We began with what he wanted the ADAs to achieve. He spoke of their ‘beautiful potential’ to ensure that companies would remain committed to supporting the people it affected most, through good and bad economic times, and through a process driven by those people. He rejected a vision of local development in which ad hoc corporate social responsibility investments by concessionaires or the desires of captured local elites like some Chiefs dictated the benefits local communities received. That vision, he said, was the contemporary status quo: it was disordered and gave life to below-the-surface conflicts (e.g., between Chiefs and local councils). I asked him how to move from the status quo to realising the potential of ADAs; he said that ADAs needed to be implemented in such a way as to get the ‘buy-in’ of communities so they would have ‘ownership’ and wouldn’t become ‘disenchanted with the potential’ of the ADA. I asked him for more detail about his vision of implementation; he did not provide any, instead amicably reiterating his belief in the potential of ADAs.

A few days later, I travelled to the town that the NGO had discussed. Each chiefdom had decided to elect community representatives to negotiate the terms of the ADA as and when that process began. I attended the elections, held in the respective Chief’s courthouse. At both, a small group of community members was in attendance. The Chiefs provided transportation money for the select few. Ted and I were seated at the top of the hall next to the Chief and NAA representatives. We were both introduced as ‘white men’ (in the local language) here to observe the elections to the local ADA committee. Voting proceeded without controversy with almost all of the representatives being unanimously confirmed. We returned to the capital that evening.Footnote ²²

3.2.2 Analysis

This account offers insights as to how the social organisation of expertise structured and limited the horizons of implementation, including bureaucracies, epistemic communities, social hierarchies, and organisational interests. Within these structures, the project should be understood as a series of actions and interactions by experts with the power to shape the form and substance of the project. As the project was decentralised, participating experts were not just technical actors at the global or national level; locals could lay claim to expertise based on their embedded, emplaced, or tacit knowledge about what the community struggled with and needed.

To begin with, the DA chose to engage with the project and had an interest in sustaining it owing to the instrumental role it would play in the larger, longer-term agricultural industrialisation project. Some of the ADA’s form thus took shape as a result of both broader bureaucratic incentives within the DA (rewarding larger projects; the importance of cross-departmental work; the availability of larger pots of funding in non-governance departments) and interpersonal networks within the DA. Similarly, the OD and NAA clearly manifested their own set of bureaucratic incentives in pushing the ADA taskforce to prioritise the signing of the ADA despite concerns over its contextual appropriateness.

Next, the project took shape at the nexus of the natural resource governance and global rule of law epistemic communities. The OD and NAA were clearly influenced more by the former than the latter; the DA team attempted to introduce more of the sociological and contextual approach of the latter into the project. This conflict was staged in a series of material sites: reports, the workshop, and the identification meeting. The conflict manifested itself through instantiating different interpretive approaches to the meaning of the AA, in particular how to identify the MC (with impact requiring a more nuanced, socially oriented policy analysis than proximity as the MC identification criterion).

Similarly, the project took shape around the social markers of professional power. The DA team (‘white men’ to their local interlocutors, despite being African and South Asian males) were given the platform to provide ‘best practice’ input as a result of their institutional position and Ted’s qualifications. Indeed, in spite of their efforts to talk about contextual lessons, they continued to be treated as global purveyors of universal knowledge (with respect to the percentage of revenues that should have gone into the ADA). At the same time, the DA team sought to weave itself into the project informally by establishing informal relationships with key brokers in the process, specifically the NAA representative.

The project also took shape through the conflicts between the background agendas or interests of the participants, whether the DA’s desire to incorporate the multi-stakeholder groups into a different project, the NGO’s desire for ADA implementation to build on and expand its existing work, or the Chief’s attempts to stack the room in the selection meeting. Finally, the project took shape through the strategies available to participants to manage and resolve conflicts within and between their different modes of organisation. For example, the NGO representative used deferral (of MC identification) to manage what was simultaneously an interpretive conflict between the NAA and OD (impact versus proximity); and an interest-based conflict between the DA and OD (to slow down or speed up implementation).

In sum, the project was socially structured through the organisation of its expertise. At one level, the project was thus a function of the conditions that structured those social structures, both internal (like the bureaucratic incentives in the DA) and external (like racialised hierarchies of development knowledge). At the same time, the project was dynamic – as it concretised, it came to shape the social structures that produced it. Thus, Ted and myself were ensconced as providers of global best practices for the project even as we sought to embed the project in its local context.

3.3.1 Context

The stakes of agriculture governance in Country are high: Country is one of sub-Sahara’s larger palm oil and cocoa producers. Country is also ripe for a rapid expansion and industrialisation of cash crop agriculture, given that agriculture has constituted over 50 per cent of national GDP over the last decade, but cash crops constituted less than a quarter of agricultural output over the same period. The government has promoted an agriculture-driven growth programme since the end of the civil conflict, liberalising foreign investment laws and seeking foreign investors in the sector. The AC purchased a large concession, which the government hopes will be successful enough to function as proof of concept for other multinationals. As a result, social expectations for near-term agricultural job creation are high, even as around half the population still lives in poverty and lacks the skills necessary for many of those putative jobs.

At the same time, social and communal tensions are high in a context marked by memories of sporadic civil conflict, especially the long-lasting northern conflict. The northern conflict, in particular, was rooted in a set of socio-economic local struggles over the control of land, agriculture, and agricultural labour. These struggles began and ended in the particular institution of the traditional Chief or ruler. The Chiefs’ authority was reified and constitutionalised in the mid-nineteenth century during British indirect rule as a counterweight to an increasingly assertive domestic urban mercantile elite. At this time, Chiefs claimed – and the British formalised – authority over their Chiefdoms on the basis of their ‘guardianship’ of Chiefdom land (including land actually owned by others). While there were ethno-regional variations, Chiefs could in effect regulate or even veto investment or other activities that touched their land – whether agriculture, natural resource extraction, or construction – through their position as tribunals of first instance for land disputes,Footnote ²⁶ their ability to bureaucratically obstruct investment by refusing to provide their assent where required, and their capacity to rabble-rouse.

Furthermore, the political authority of the Chiefs was not simply a cultural phenomenon but a socio-economic one as well. Prior to the British abolition of slavery in Country, Chiefs relied on slave labour to tend their land-holdings. Following abolition, they leveraged their power of assent to marriages of women within the Chiefdom to indenture young men to years of free labour in exchange for the ability to marry. This latter tradition in particular, and the concomitant frustration of young men’s ability to achieve their ideals of masculinity, led to a broad-based revolt in the north against Chiefs and their political patrons in Country’s ruling party headquarters, a revolt that eventually took on an ethno-regionalist (and sometimes separatist) bent, as well as a natural resource rentier dimension (with illegal mining and logging funding warring groups). After quashing the revolt (and widespread atrocities and retribution on both sides), a fragile Chiefly authority has returned. Chiefs have managed to promote themselves to the government and international donors as sources of solutions to a problem that they helped cause: as locally legitimate authorities, ripe to be trusted with a post-conflict agenda of administrative decentralisation (sponsored by donors as a means of producing accountable governance to disaffected northern citizens). As putative local administrators and continuing ‘guardians’ of the land, they hold a great deal of power when it comes to distributing benefits from agricultural investments.

The government, while supporting Chiefly authority insofar as it helps stimulate aid revenues into Country as donors fund decentralisation programmes, remains wary of their local power. Country’s political economy is dominated by major ethno-regionalist political parties, with the largest forming rotating alliances with one or two of the smaller ones to dominate the political scene, having held power for most of the post-independence era. Resources and power travel through party networks. The relationships between local party bigwigs and Chiefs are important and fragile nodes in those networks, especially given a Chief’s capacity to rabble-rouse. Large-scale inward investment focused on local land thus places pressure on those nodes, simultaneously making them more important relative to other bits of the party system and more volatile as the stakes of the party–Chief relationship increase. The only exception to this pattern is in the post-conflict north, where the ruling regime’s central apparatus informally bargains directly with Chiefs to ensure investment and maintain physical control.

This political legacy and contemporary political economy give rise to three discursive frames of natural resource governance in Country:

1. (1) government and donor-driven development discourses of growth on the basis of industrialised private agriculture;

2. (2) donor-driven post-conflict political economy or state-building discourses based on administrative decentralisation, which are intended to reduce conflict risk; and

3. (3) a prevalent property-based local administrative or governance discourse, in which traditional Chieftaincy plays a central role in managing the pressures and resources that come with inward investment and intensive land use.

The confluence of these three discourses is not uncommon in natural resource-rich environments, from Papua New Guinea to Sierra Leone.Footnote ²⁷ They underpin arguments in favour of local development agreements for land, agricultural, and natural resource extraction. The discourses share an orientation towards the concession, rather than the state, as the object of policy. As a result, the success or failure of policy in this mode is contingent on the company–community(–state), rather than company–state, relations. This means that policy must be contextualised to the specific social and political dynamics of the community and concession area.

I argue that, although the text of ADAs and related documents in Country on their face require context specificity and responsiveness to local needs, they actually attempt to discipline the politics of natural resource governance. ADAs, like other natural resource local development agreements, provide a framework for company–community relations and offer some promise of immediate resource redistribution from the company to the community. The framework is not completely reconfigurable; it is limited by its form. Furthermore, the redistribution of resources enabled by the framework has embedded in it a view of what constitutes matters of public and private concern. Specifically, communities around concessions must articulate their own preferences privately; companies’ social obligations are to be met through private negotiation, and the state’s governance, conflict, and development roles are supervisory at best. In doing so, ADAs facilitate inward investment, produce a pliant participatory public, and support a legal imaginary of the state as a regulator of the private sphere.

3.3.2 Ambiguity in the Law

The legislative and regulatory framework for ADAs in Country is extremely ambiguous.Footnote ²⁸ Indeed, according to officials at the OD, DA, and other donors who funded the drafting, and officials at Country’s agriculture ministry, this is supposed to be one of its strengths: an ADA is supposed to be adapted to the specific needs of the particular company–community relationship around the concession. Part XV of the Act sets out the main legislative provisions giving rise to and regulating ADAs; pursuant to other provisions in the Act, the NAA has the responsibility to implement ADAs. A set of regulations required by the Act were adopted a short while later (the Agricultural Administration Regulations, or ‘AA Regs’); scattered throughout the regulations are some further provisions about ADAs. Both the AA and AA Regs were funded by donors and drafted by international consultants.

The AA and AA Regs provide some guidance as to the form and process of ADAs but little as to their purpose and substance. The holder of a large-scale agricultural concession (defined in the Act as larger than 200 hectares) is required to have and implement an ADA. Per the AA Part XV(i), ADAs should ‘promote development, enhance the welfare and the quality of life of inhabitants, and recognise and respect the rights, customs, and traditions of local communities’. Part XV gives no further guidance as to the purpose of this provision. The AA and AA Regs are similarly unclear with respect to the amount and use of ADA funds. The AA provides that at least 0.1 per cent of gross revenue should be spent on each ADA per annum (Part XV(v)), a figure which several potential concessionaires in Country themselves have suggested is too low to both avoid conflict and foster local development. It also provides that the ADA funds should be governed and spent through a local multi-stakeholder process (giving the example of a ‘board’: AA Part XV(v)). The Act and Regs are silent on whether other funding streams between companies and communities (e.g., private trusts or charitable foundations) should be included in the ADAs. These would be a means of increasing development funds but through channels with different standards and mechanisms of governance to ADAs. Finally, Part XV(iii–iv) of the AA list things that may be included and must not be included, respectively (e.g., apprenticeships for community members may be included; the purchase of passenger cars must not), but the guidance is extremely limited.

Crucial to both process and substance, the terms of the ADA are to be directly negotiated between the concessionaire and the ‘main community’ (‘MC’; AA Part XV(ii)). Most of the ambiguities in the law and regulations are meant to be resolved through this negotiation and spelled out in the final document. There is thus much at stake with respect to who is in and out of the MC. The AA Part XV(ii) provides the following definition:

At the same time, AA Part XV(i) provides that the ADA ‘shall assist in the development of communities affected by [the concessionaire’s] operations’ (emphasis added).

Clearly, the concessionaire and local government have the initial authority to determine who is eligible to be part of the MC. Moreover, in requiring that the local government be one of the two initial parties, the legislation suggests that ADAs should be designed on the basis of local context rather than a central mandate. At the same time, the Act leaves open significant questions. Specifically, to what extent should MC determination be a matter of pure discretion and agreement on the part of the concessionaire and local government, and to what extent should that discretion be fettered by principles and processes? As for those principles, is ‘affected[ness]’ the most relevant criterion to determine MC membership, or (given the reference to ‘twenty kilometres’ in the Act) proximity to the concession (whereby proximity may exclude those who are far away but affected by infrastructure, transport, and processing activities)? And what is the nature of a ‘community of persons’ as a single unit – might it be administrative, such as a village, town, Chiefdom, some other kinship grouping, and so on, or might it be a generic descriptor of a collectivity? As for processes, should other participants be included in the determination process (e.g., the NAA or traditional Chief as the arbiter of the fairness of the process)?

The Act and Regs thus present a challenge to implementers of ADAs. They must work out the scope, content, and purpose of each ADA according to the concession’s local context; however, they are provided with no clear means of determining exactly who and what are ‘local’. These are no mere lacunae in the law, to be filled either through policy pragmatism or principles of statutory construction. They are products of deliberate vagueness in drafting (along with poor drafting) designed to produce ADAs that can be context-responsive. In doing so, these lacunae produce a first-order set of policy questions that must be answered clearly before the legislation can be implemented. To take a simple example, if an MC is a specific administrative unit, like a village, we may see hundreds of potential MCs for one concession. Even if we assume that many MCs can be a signatory to one ADA (to avoid a scenario in which a company has to sign hundreds of ADAs, each worth 0.1 per cent of revenue), must each MC be represented on the ADA’s governance board? Moreover, the fashion in which such questions are answered will shape the implementation of the Act. For example, if the initial determination of the MC is taken to be contingent and open to challenge or revision, the ADA itself may be understood more as a policy framework than as a private contractual arrangement between two predefined parties.

The Act’s ambiguity is a product of ideas about Country’s decentralisation and industrialisation, ideas which emerge from Country’s current policy environment and which are part of the global natural resource governance toolkit. As I argue in the next section, interpretive limits on this ambiguity can be derived from the same sources in a way that reproduces those ideas in the name of local contextualisation.

3.3.3 Discourses of Conflict, Development, and Governance in ADAs

ADAs are stablemates of several types of natural resource company–community agreements. Deriving from experiences in the Global North, particularly Australia and Canada, these agreements are new governance-inflected instruments through which global policymakers seek to frame the company–community relationship (through redistribution and the discipline of deliberation)Footnote ²⁹ to minimise conflict, improve governance, and promote local development around the concession.

As a result, ADAs are embedded in a set of discourses about their purpose and implementation, covering conflict, development, and governance. These discourses suggest some assumed content to ADAs along with the limits of how they might be implemented. These discourses clearly overlap but draw on different ends and discursive resources. I treat them separately here.

3.3.4 Analysis: Re-Contextualising the Contextualisation of ADAs in Country

On their face, the AA and AA Regs leave much of the form and content of ADAs to implementation, from the identity of MCs, to the nature and multi-stakeholder composition of its governance board, to the content of the very idea of development that ADAs will enable. However, in exploring and connecting the discourses of conflict prevention, local governance, and local development that frame ADAs, I have argued that ADAs contain clear ideas about the form and content of implementation:

• ADAs, while creatures of legislation and regulation, draw on private legal forms. These forms presume that the parties have private preferences and bargain over them to produce an agreement. The set of preferences with which the ADA is concerned is not a vision of development but rather a vision of company–community communication and resource distribution that avoids violent conflict and spoilers.

• With respect to ADAs, intra-communal power dynamics, especially the relationship between the Chief and communities, are a private matter, neither subject to public regulation nor a matter of public concern.

• Preferences and intra-communal contests can become a matter of public concern if they escalate to the level of conflict. However, the AA and AA Regs recognise the process of MC identification as a key site of conflict – meaning that contests over the decision to sell a concession, policy preferences, or representation have to be translated into contests over MC eligibility.

• The process of determining the MC is also the moment where the state’s regulatory role is most clear; the law provides that the local government negotiates with the company to identify the MC. As a result, this specific issue can be treated as an ongoing policy concern, subject to renegotiation or reinterpretation. Other matters, such as the adequacy of community representation, are the object of unclear but light-touch state supervision, as communities are supposed to use the deliberative platform of the ADA to regulate their own collective action.

Placing these ideas in Country’s specific context immediately raises the following concerns, emerging out of the political economy of land and Chieftaincy:

• As the history of Country demonstrates, local power structures are attuned to quashing incipient resistance to them. An emphasis on conflict avoidance, while well-meaning, may result in legitimate contests and conflicts – such as strikes and blockades – being suppressed by local leaders who claim the right to represent the community in the ADA. In other words, the participatory and collective action dimensions of the ADA may be a means of producing a pliant or quiescent public whose Chiefs will not allow them to interrupt agricultural operations.

• The executive state is produced not as a development planner but as a light-touch regulator of a limited set of engagements between company and community as well as a provider of violence to enforce bargains and suppress violence by others. In doing so, the state is supposed to ensure the continuity of business operations and facilitate inward investment while leaving economic and social matters to bilateral negotiation between company and community – or more properly, company and Chief.

• Chiefs hold a quasi-public role as protectors of the land. In containing no provision for the participation of Chiefs in the ADA process but rather leaving that to implementation, policymakers have left open the possibility that Chiefs will instrumentalise how the ADA has drawn the public/private divide. For example, Chiefs may exercise their public power in the ADA’s private space to influence the community’s choice of a representative to the ADA governing board.

• In leaving the question of Chiefly participation ADA implementation, the ADA process takes no clear position on the politics of Chiefly involvement in the causes of Country’s civil conflict and their proper position in the post-conflict political settlement. Instead, it keeps that question open and defers its resolution.

ADAs, then, have particular politics. They are identity politics in the sense that they do not interrogate intra-community political dynamics, seeing them instead as a function of local collective action based on cooperation and not coercion. They are power political in the sense that they produce the company–community relationship in deliberative terms rather than as a power struggle (and thus power disparities as a matter of unequal speech). Finally, they are politically instrumental in the sense that they enable the bracketing of certain types of communities’ political claims in the name of conflict avoidance and good governance. They thus frame the state’s role as a light-touch regulator of deliberative interaction rather than as an articulator or enforcer of political claims. These politics serve to discipline the identities of both the state and the community in favour of continuing business activity, thereby serving the interests of capital and the Chiefs (as key translators of financial capital into local political contexts).

3.4.1 Boarding

This is an account of a trip. More precisely, it is an account of several trips, each nested within, and reaching out far beyond, the others. The purpose of writing about the trip is not a solipsistic exercise in autoethnography, nor is it an extended paean to travel writing. Rather, I seek to analyse a specific quality about the trip that, I will argue, makes it both a valuable metaphor and a synecdoche for a particular type of development work.Footnote ⁵³ That quality is fluidity or movement – that is, how the ‘trip’ becomes shorthand for the fragmentation and reorganisation of space and time endemic to development work, and in doing so, the identities of the trip-taker and those she encounters.

The particular type of development work I am concerned with is a set of contextually minded governance or institutional reform projects, the ones that are concerned with claiming and producing their own contextual embeddedness even as they seek to reorganise power in a polity. The trip is a vehicle for describing – as well as a means for doing – those projects that are concerned about the conditions and limits of their own ability to travel from place to place.

A trip is, trivially, travel. How does a trip cohere and distinguish itself from everyday movement? I focus in this chapter on the materiel – objects, artefacts, techniques – of the trip that give its fluidity shape, direct and channel it, and create eddies or spatio-temporal moments of experience. Given the trip’s central role in doing development work, this materiel is a key component of making development projects cohere as well as a metaphor for how an agreement over a specific level of contextuality is reached or stabilised such that a project activity can occur. In contrast to others’ accounts of projects as particles (bounded by and invested with money, work, and time), through my dual use of the ‘trip’ I show how ‘projects’ are enacted as cognisably project-like while still remaining multiple, misshapen, and overflowing.Footnote ⁵⁴ The project is, in that sense, trippy – an accumulation of movements, anchored in materiel, that produce moments of space-time, a hazy totality emerging from disjointed and uncanny experience.

Given that the question of contextualisation is central to the project I am describing, my account of the project rests on how the actors and objects that constitute the project contextualise themselves and each other. My account of the trip is not immune to this dynamic: the form of narration is a strategy of implementation, of an attempt to place an order on a trip while keeping a sense of its fluidity.

3.4.2 On-BoardFootnote ⁵⁵

A small, upholstered screen intrudes into my peripheral vision, then the hard membrane of a laptop screen flattens my fingers and a keyboard presses into my stomach. I lift my knees to the folded tray table and attempt to keep typing. I give up and save the document. I place the laptop in the seatback pocket and clamber out of liveried claustrophobia into the relative freedom of the aisle. The cabin offers up a chiaroscuro landscape. Scattered flecks illuminate white and black chins; next to me, those lights resolve into screens full of memoranda, research reports and briefing notes. They shed enough light to make out the contours of the scenes to either side of them – blankets draped over heads, fingers lazily assaulting the back of the headrest in front. I walk up to the business class cabin. It contains a more diverse set of crowns – Chinese, West African, white, South Asian. They otherwise look the same as their compatriots behind the iron curtain: some embrace the trappings of boredom, others wield laptops. Their screens display yet more bulleted lists, pie charts, executive summaries. I try to spy the logos at the top of the documents. I recognise a mining company, the World Bank, and Oxfam. A loud warble emanates from a nose in their midst, at the same frequency as the hum of the engines. A screen at the front of the cabin shows a collection of pixels (a large, dark bat?)Footnote ⁵⁶ that hovers above the legend: ‘Time to Destination: 2:56’.

I return to my cocoon. I am drafting the agenda for our workshop with the ADA taskforce in a few days’ time, focusing on community engagement and identifying the MC: ‘Community engagement workshop draft agenda v7_TC_TK_DD.docx’. We drafted Version 1 at the behest of the NAA, who then circulated it to the taskforce members, including donors, NGOs, and representatives from the AC. Versions 2 to 6 reflected their priorities: a kaleidoscope of Track Changes with coloured strike-throughs, underlines, and comment bubbles. For the sake of clarity, Ted (‘TK’) had just accepted all the changes up to and including Version 6 (creating ‘v7’), inserted his own Track Changes (‘TC’), and emailed it over for my thoughts (‘DD’) (Figure 3.1).Footnote ⁵⁷

Figure 3.1 Community engagement workshop draft agenda v7_TC_TK_DD.docx

Source: Author

The other taskforce members, including the NAA, had initially asked that we – along with the OD – present them with examples of ‘best practice’ in community engagement from other parts of the world. The OD was quite happy with this language. However, we had pushed back against the idea of ‘best practice’ for something as complex as structuring dialogue and resource redistribution between companies and communities: what purchase would our templates and analytics have when power was structured so differently from place to place and from moment to moment? Ted rejigged the schedule to spend more time on lessons from Country itself; we had also proposed the alternative language of ‘lessons learned’ from other places. The NAA quickly adopted – or appropriated – our terminology, along with variations on the word ‘context’. Whenever they changed the agenda to reflect this language, they added a comment bubble proposing study trips for themselves and the NGO representatives to learn these lessons – for example, to Ghana and Canada.

We had hoped to use this workshop to emphasise to participants the contextual complexity of community engagement and set up an agreement among the taskforce members to treat the ADA as a fluid process rather than a fixed, signed contract. The questions and bulleted sub-questions in the draft agenda gave an analytic framing to the proposed discussion; however, we had tried to ask them at a level of generality that meant we would probably spend time talking about the contextual differences between all the different potential concession areas (similarly, removing the word ‘Chiefs’ promised to focus our discussion on local power-holders more generally rather than participants’ old and well-trodden views of Chiefly conduct).Footnote ⁵⁸ Nor had anyone pushed back against the substance of the workshop being on MC identification and local representation – the two most contentious and locally politicised aspects of the ADA.

In the end, we hoped that we could find ways to help the taskforce conduct and incorporate long-term qualitative research of local conditions around the concession into the implementation process of the ADA. In fact, Ted had already prepared a draft workshop outcome document for the workshop; he had included these research activities in his provisional list of next steps.Footnote ⁵⁹ Ted’s draft document influenced our edits to the agenda. We tried to make the agenda and the plan match up; for example, by suggesting in the agenda that we have a discussion about monitoring and evaluation, I was hoping to initiate discussions on the relationship between ongoing research and ADA implementation. At the same time, the outcome document remained interim. We were concerned that we did not know how much the ADAs would actually be worth and so did not want to propose an implementation process whose costs far outstripped the value of the ADA itself to main communities.

‘Time to Destination: 1:31’. The cabin fills with the smell of bread and hot salt. I cross-reference the agenda with the draft outcome document again. I close my laptop, finally finishing and emailing the document when I have a wi-fi connection at the hotel that night.

3.4.3 Boardroom

The National Agricultural Agency’s offices are on the top three floors of a five-story building in the commercial district of Capital. The first two floors house a commercial bank. The building is squatter than its neighbours but hard to miss, its façade a dirtying mustard yellow punctuated by rows of circular cabin windows. The NAA’s executive boardroom sits behind several of those windows. It is bounded by a whitewashed floor and walls of black-flecked beige plastic, the prima essentia of anonymous cubicles. In the middle resides an expanse of walnut and black leather, surrounded by matching deep recliner chairs. At one end sits a vast flat-screen TV; its virtual meeting space is dark, all the participants for this meeting coming in person.

Ted and I arrive a few minutes early, clambering out of our shared Nissan taxi with our laptop bags wedged under our arms. We weave through the white Toyotas 4×4s surrounding us and climb up the stairs to the third floor. The air conditioning is running, but the boardroom is empty. Schwartzman argues that meetings give an organisation ‘a form for making itself visible and apparent to its members’, thereby ‘provid[ing] individuals with a place for making sense of what it is that they are doing and saying … and what their relationships are to each other in this context’.Footnote ⁶⁰ In other words, meetings are ‘the organization or community writ small’.Footnote ⁶¹ If this is the case, our relationship with the other members of the taskforce is not looking great.

This is, of course, part of the dance of bureaucratic status.Footnote ⁶² As we gaze out of a porthole, more Toyota pick-ups park alongside the front wall of the building. We recognise the vehicles as those of the other taskforce members, even if their inhabitants remain obscured behind peeling window tints. Eventually, a little more than half an hour after our allotted start time, white doors swing open. At that same moment, the door to the boardroom opens, and Yahya, our director-level counterpart at the NAA, steps in. He wipes sweat from his thin brow with a handkerchief. He greets us with a warm smile and deep handshake, welcoming me back to Country. He has missed me!

The others arrive moments later. Seats are taken, laptops and paper notebooks are discharged from bags. Despite the fact that we know each other well and greet each other as old friends, business cards are passed around. Each recipient recognises the giving of each one, makes some show of inspecting it and places it in his notebook. Yahya then calls loudly for a secretary seated outside the door and demands that she print and circulate copies of the agenda immediately. By 9:45 or so, the copies arrive.

Yahya apologises for the late start of the meeting. He welcomes us, invoking the name of the President, Minister, and Director of the NAA in turn. As is customary in Country, he requests that we bow our heads in prayer before beginning any business. He thanks God for His guidance and asks Him to steer our conversation today so we can help His children ‘outland’ [out in the rural areas, where the concessions are found]. What we discuss and decide today will be His will.

Kennedy suggests that expertise is a ‘terrain of struggle’Footnote ⁶³ precisely because it deals in strategy, not faith; shades of grey, not clear totalities; (antinomian) reason, not charisma: as an expert, ‘[y]ou cannot say God has authorised your victory …’Footnote ⁶⁴ Recent anthropological accounts of bureaucracies and meetings have similarly reasserted the legacy of the Weberian frame of bureaucratic rationality, counterposing it to charismatic authority.Footnote ⁶⁵ They even pinpoint routinised moments of exception in meetings that express and contain matters that go beyond an organisation’s bureaucratic rationality – for example, ‘Any Other Business’ – and thus act as a bulwark against charismatic authority.Footnote ⁶⁶

Yet Kirsch, studying meetings in Protestant and evangelical churches in sub-Saharan Africa, has pointed out how such meetings in fact comfortably syncretise bureaucratic and charismatic authority – specifically divine will and bureaucratic routine – through preparatory fasting, the interjection of casual prayer or religious metaphor, breaking into song, and so on.Footnote ⁶⁷ This is part of the meeting’s power. The meeting might ordinarily be understood as an attempt to order uncertainty and specify a sequence of actions that structure the present and the future.Footnote ⁶⁸ However, Yahya’s prayer does not seek to manage uncertainty and ignorance; rather, it embraces and expresses them. The taskforce participants are asked to be passive vessels of faith even as we sit down to the hard struggle of argument.

Who are these passive vessels of faith? I know most of the faces: government officials, AC officials, OD officers and NGO representatives. Still, a tour de table; we introduce ourselves, our institution and our official position. One person is new – a middle-aged Caucasian woman sat next to me. I recognise her name: she is a well-known anthropologist of Country and its civil war. She participates as part of the delegation from the OD. Her role, she mentions, is to provide contextual insight to the OD on issues of land and the political role of the Chiefs. She is in the meeting ‘to observe’. I know her work well; I wonder if the workshop itself is part of her fieldwork.Footnote ⁶⁹ I cannot resist glancing over at her notebook on the desk to get a sense of what she is writing. I see a list of names and institutional positions. Would we end up alongside the young rural Country men whose economic lives she has so diligently chronicled?Footnote ⁷⁰

We begin the initial discussion on lessons from various community engagement processes in Country. The NAA speaks about its aspirations for, rather than experience in, ‘context-specific’ community engagement (referring to those study trips again). Emmanuel, the flamboyant main representative of the NGO cohort, then takes the floor. Who knows the context better than his organisation? he asks. It has a long history of ‘going down to communities outland’ and helping them speak with companies. He is concerned that the DA is side-tracking the taskforce by talking about research and local power dynamics. The NGO knows that no one in potential concession-area communities has a major problem with local Chiefs. Emmanuel tells us that he has just come back from outland and has seen it himself; as he does so, he pulls out his mobile phone and dabs at the screen. A video begins to play. He brandishes the phone; the video is too small for anyone to see. He talks over its faint sounds. There is no need for all the research and reports, he tells us: the NGO has videos. It has been videotaping testimonials of community members (including women and youth), Chiefs, and landholders in the potential concession areas.

He reports that this very video shows a Chief saying that he doesn’t want to be in charge of the ADA – proof that we have nothing to worry about. All that community members are concerned with are the imminent impacts of large industrial agriculture – ‘This is what we hear whenever we go down to the community’. The NGO, Emmanuel says, has been trying to tell them that they should not be concerned, that the ADA will bring investment into their communities; for that to happen, the NGO has stressed that communities have to ‘get the [concession] space open’. (By this, he means that communities should show investors that the area is conflict-free, meaning their investment can proceed without the threat of protests or violence.)

Ted expresses concern. How can we realistically say that no one wants to control the process, and that everyone should avoid conflict, if we don’t even know who the MCs are or how much they might get from the ADAs? Yahya immediately interjects: ‘Of course we know who the MCs are. They are the ones most impacted by the concession’. Betty, the OD representative, immediately disputes this: ‘As a [Country] lawyer’, she knows that the law does not refer to impact. She opens her laptop; its internet dongle lights up. The meeting pauses as everyone proceeds to open theirs. I scroll through my hard drive until I find the text of the Act. In the very first word, it proclaims its status – ‘ACT’.

Black and white, clearly organised, with a table of contents, it is reassuring. As I press ‘Ctrl+F’ and search the document for ‘Part XV’, Betty reads out Part XV(ii) of the Act.

She then turns her laptop around on the table to face the other participants: ‘See?’ Yahya reads out the exact same provision from his screen, emphasising the word ‘community’ a bit more. This seems to be probative of his position. The AC representatives point out that Yahya’s interpretation matters most to them. Others cut in, reading out the same provision again and again, over the top of each other, with mildly different emphasis. I join in. The Act becomes noise; its sheet music is notebook pages with circles, radii and ‘20km’ scribbled on them.

Emmanuel interrupts. We need not worry about setting out rules just yet to avoid conflict or capture. ‘You have very clever people in the community. People take the law and ride it … They go to court and want to clarify this [the exclusion of some communities from the MC] … This will come in implementation … for some people will pick this and take it to court’. To that end, there is a ‘pragmatism [i.e., flexibility] that must come in [implementation] as well’ – something the NGO has a track record in. He says we must follow a ‘participatory method – where everybody is informed and involved’ – to identify the MC and its representatives.

This would require immediate sensitisation meetings at the community level so all community members could be made aware of the forthcoming ADA, its content, their right to lobby to be an MC, and their right to choose their representatives to the ADA governing body. This would also require specifically sensitising Chiefs about ADAs – after all, nothing can really get done in their Chieftaincy without their say-so, especially on issues related to land. The next step, he suggested, would be for the NGO and the NAA to draw up and circulate a sensitisation plan and budget as soon as possible. Betty continues to point to her laptop screen, saying that nowhere does the law provide for Chiefly involvement or community sensitisation. Emmanuel retorts that the law does not prohibit it, either. The AC demands that we get have an official discussion on how to interpret the law at some point in the near future, with government lawyers, AC lawyers, and others participating.

We move on without agreement. Ted and I discuss our briefing notes, which participants pull up on their laptops. Our research suggests the importance of regular and long-term local dialogue regulated by some process norms to ensure that the marginalised and vulnerable are not left out. However, we point out that such a process could be very costly to implement, especially if it involves ongoing research and the proposed local sensitisation. We have not seen any revenue projections for the potential concessions but presume the NAA have them. What will the actual value of the ADAs be each year?

The Act becomes noise again – the participants proceed to read from their screens the provision specifying that companies place a minimum of 0.1 per cent of annual revenue into the ADA. The AC reminds everyone that they intend to provide more than that – the exact percentage depends on the circumstances of the concession. We repeat that we are wondering about a dollar value but to no avail.

Yahya and Emmanuel instead ask us how 0.1 per cent of revenue compares to international best practice for local development agreements. It is our turn to return to our screens. Our briefing notes show that the context of each concession varies dramatically; the amount should vary by impact, the robustness of the local economy, and the local power dynamic. The best practice is unhelpful in such a complicated environment. I eventually come up with a range of about 1–3 per cent and a promise of more research on the matter. We move on again.

Later, in another Nissan taxi, Ted expresses surprise that I gave them any figure at all. Using Track Changes, I subsequently add a request for more research on the percentages in the draft outcome document.

3.4.4 Back-Room

WeCountry Lager is Country’s bestselling, domestically produced, beer. It is a sour, thick, pinkish-brown suspension, brewed from a malted mix that is 60 per cent sorghum and 40 per cent maize.Footnote ⁷¹ Its texture is close to that of gruel,Footnote ⁷² and for good reason. As with the indigenous beer in many other countries on the subcontinent, it contains enough carbohydrates to provide between roughly 1–5 per cent (per Food and Agriculture Organization estimates) or 6–12 per cent (per more targeted studies) of total national caloric intake.Footnote ⁷³ It is consumed rather than drunk: indeed, in the dialects in the north of Country, one asks whether one may ‘eat’ a glass of beer.

WeCountry Breweries Inc. began life as a small northern brewer during British colonial times. It grew through a series of mergers into the main national brewer and was nationalised at independence. Losses mounted, and it was privatised during structural adjustment. Having solicited investment from Diageo (the global drinks conglomerate) in the late 1990s, it embarked on efforts to expand regionally. It purchased large-scale agricultural concessions and modernised its supply chains. Its efforts failed. Sorghum-based beers spoil rapidly (within one to five days), since they are ‘consumed while [they are] still fermenting … The resulting beer is thus microbiologically unstable, i.e. infected at varying levels with yeasts and bacteria’.Footnote ⁷⁴ WeCountry Breweries attempted to develop pasteurisation techniques as the basis for its expansion strategy; however, pasteurising malted sorghum produced a reaction that changed the taste and texture of the beer, making it viscous, stringy, and flat.Footnote ⁷⁵ Their supply chains were also unable to keep up with the short spoliation period. WeCountry Breweries declared their attempts to balance indigenous taste with modern efficiency a failure, until such time as scientific research developed an effective way to pasteurise sorghum-based beers. They scaled back production, reduced agricultural capacity, and left their concessions; many young men who had moved into new Chiefdoms in the hope of employment found themselves with no jobs, no income, and enduring obligations to Chiefs or creditors that they could not pay off or could discharge through labour.

WeCountry remains a bestselling beer in Country. Its labels display a hand holding a mug of cold beer – an image of the future ideal currently bottled up. More often than not, the labels fall off; robust adhesives are expensive. Bar floors are littered with labels and bar tops with naked bottles. Right now, the label on my beer comes off in my hands as I sip from a bottle. Yahya’s has already fallen off.

We are meeting for an ‘informal drink’ (at my request) a few days after the workshop and before my trip outland. There might be some funds at the DA to support ADA implementation – from my colleague’s agriculture project or from some technical assistance funds from our rule of law department. However, at the workshop, the OD, NAA, and NGO were pushing for an elaborate ADA-implementation process without weighing the costs and benefits. So I want to get a sense of Yahya’s ambitions for ADA implementation and the direction in which he imagines implementation going.

He is, he says, no fool. He does not want ADA implementation to ‘be a cost centre’. As ‘a professional’, he has to make the most effective use of resources. He will look bad to the donors and the head of the NAA otherwise. But he is passionate about the ADA. As ‘the law’ [legal instrument], it ‘chains’ companies [they are bound to their obligations]. It is at moments of shock and reversal in particular that communities need protecting – for example, if grain prices plummet. I take another swig of my beer.

He ‘rejects local development where CSR [corporate social responsibility] money’ is paid in an ad hoc fashion to quell those who shout loudest or to capture local elites. The ADA has ‘beautiful potential’ to bring about benefits to communities that they truly want and need. At the same time, as it stands, potential concession areas are politically charged, complex, full of conflicts, and rent-seeking. It is imperative that community people do not become ‘disenchanted with the law’ [the ADA], seeing it as a vehicle to further the interests of corporate and local elites. For him, this is the importance of the ADA taskforce’s task: not to implement the ADA but to ensure the ‘buy-in’ of community members. I ask him how buy-in will happen and how it will be sustained such that the current levels of cynicism and conflict might transform into the realisation of ‘beautiful potential’. What, in effect, will implementation look like? He repeats that all we need to do is ensure buy-in, for Country’s ‘future’.

I cannot help thinking that this feels nothing like ‘project time’, which is supposed to be regimented and ordered by project plans, terms of reference, logical frameworks, and the like.Footnote ⁷⁶ Instead, I hear that there will be a modern, socially responsible industrial agriculture sector; there is a concession riven by competing land claims, a traditional political economy of Chieftaincy and a legacy of conflict. There is no imaginary of the terrain in between.

WeCountry Breweries tried to use science to chart a ‘food’-producing path between tradition and modernity, between an emplaced now and an expansionary future. Science failed it, contributing to the conflict and mistrust that constitutes the contemporary now in potential concession areas. Despite the requests for ‘best practice’ percentages and practices, Yahya seemed to have no desire to call on technocratic ‘science’ to work out the course between the present and the future – nor did my team have to be so worried about pushing back against that view of our role and advocating for contextualised, locally driven solutions. There is a void between the ADA as a vessel of the present and a fantasy of the future – they are two scenes, juxtaposed. A neatly legally-packaged future and a concession-based present; a label in one hand and a bottle in the other.

3.4.5 Back-and-Forth

The driver of our battered Nissan taxi presses his right foot down to the floor. The car trundles to a halt next to two bright white Toyota pick-up trucks. I see Betty’s outline through one of the tints. A little further ahead stands the Chief’s courthouse, a concrete canopy resting on four pillars, bounded by a waist-high wall but otherwise open to the elements. Emmanuel and Yahya are already there, greeting people. Emmanuel is also arranging the speakers and microphone. We are here to observe the elections of some local community representatives who might negotiate the terms of an ADA. Emmanuel mentioned that this community had been ‘sensitised’ by his NGO to the impending ADA-implementation process and had decided to try to organise themselves.

Like two cheerful explorers, Ted and I are keen to get out of the car and look around. We enjoy the intrepid feeling of being out in the field; indeed, Ted has made his bread as a field researcher in Country for many years. We are, of course, well aware that we might be on the receiving end of a well-choreographed kabuki of research responses, as with many foreign visitors whose research pretensions are in short-term trips to rural sub-Saharan Africa.Footnote ⁷⁷ Yet we figure we’ll see what we can glean from observing local Chiefs and subjects in action.

We walk around the courthouse, pausing by a small metal sign at its entrance that displays a battered European flag and proclaims that it was built thanks to European Union funding as well as the generosity of the AC. It is hot, and soon our shirt collars are a province of dirt and sweat. We retreat to the concrete shade of the courthouse. Some locals offer us plastic chairs; we decline and perch on the low wall in the back corner of the building where other villagers are lounging. Hip-hop blares out of their tinny phone speakers. Ted and I chat with them about their livelihoods, their Chief, their attitudes towards the concession. I practice my smattering of local dialect; they are nonplussed. Together, we watch a group of what appear to be locals, congregated in the middle of the courthouse. They are a mix of men and women, young and old; some bear what I assume are scars from the civil conflict.

Emmanuel walks across the courthouse to chat with us. We ask him about the group. He tells us they are from the surrounding villages; the Chief has paid for their transport to the meeting. Emmanuel escorts us over to the far side of the courthouse to meet a man wearing leopard-print loafers. He is the Chief. We shake hands, and he bids us all sit on the plastic chairs next to him. Emmanuel proceeds to facilitate a discussion between us, asking questions and translating when necessary. At the same time, the group in the middle of the courthouse suddenly leaves the building to the left. As we continue talking, I can just about see them form a crowd, their backs facing us. Following some jostling, they return. As they do, Emmanuel pauses the conversation and walks over to Yahya. A member of the group joins them and hands Yahya a piece of paper. They confer, and Emmanuel and Yahya join us again next to the Chief.

The Chief grasps the microphone and says ‘Hello’ repeatedly until everyone is seated. He formally welcomes all those who are present and passes the microphone to Yahya, who introduces the four of us at the front to the community. Eventually: ‘The two white men sitting next to the Chief are here to observe’. I smile awkwardly. ‘They are with the DA’. Everyone applauds. Yahya then proceeds to read names from the piece of paper; as he does, people stand and are confirmed by acclaim.

On the way back to the cars, I ask Emmanuel what had happened just before the meeting. People ‘didn’t want to air their dirty laundry in public’ or ‘let the meeting get too hot’, he explains, so they had a ‘pre-meeting’ to decide who would be nominated. I couldn’t help but notice that our ‘white’ gaze created a space that was beyond the Chief. I ask Emmanuel who the man with the list of names was. One of their local activists, he replies.

3.4.6 Trip Report

What does the trip reveal about the substance of the ADA-implementation project? My account of the trip shows the project as an accumulation of struggles over the contingent concretisation of the project and distribution of its resources. On its face, the ADA-implementation project consists of bilateral and multilateral donor support and technical assistance to a government agency, influencing a multi-stakeholder process to govern a community-driven development programme. It should thus be straightforward and mundane; one might imagine the reified project emerging out of the bureaucratic contests between the objectives and process preferences of different stakeholders as well as the ideational orientations of their broader epistemic communities.Footnote ⁷⁸

In fact, my team imagined and attempted to push back against this type of ADA implementation. We sought to counter what we presumed would be an overdetermined implementation process by invoking the importance of the context of implementation and then setting up structures to incorporate that context into the project (such as iteratively incorporating field research into an adaptive ADA-implementation process). However, it turned out that no one involved in implementation was committed to overdetermined best practices. From the openness of the legislation to the fluid composition of the ADA working group, everything remained fluid, and the ADA emerged as a nested series of provisional determinations. Who is the MC? Perhaps this Chiefdom. Should the Chiefs be involved? Maybe. What is the amount of the ADA? We don’t know yet. And so on. Notably, it turned out that the provisional nature of the ADA persisted well beyond this trip and through a crash in the price of cocoa and palm oil – that had it been expected would have changed the AC’s tolerance for ADA implementation. And yet, something did happen in the courthouse – a decision was taken, with people nominated (although for what purposes remains unclear).

What can be gleaned from this trip in terms of how decisions happen and their governance effects? At a general level, those ways will differ from those we assume to be part of ordinary bureaucratic work – the work of rationalising, developing processes, meeting, and making lists, which mark bureaucracies and whose form, function, instrumentalisation, and hybridisation are the stuff of much ethnographic inquiry into sub-Saharan administration. The relevant techniques in rule of law reform are negative: denying assertions about who should participate in the ADA by referring to the many lacunae in the Act; or rejecting the validity of any knowledge about what should be the relevant royalty rate. These are used in the service of collapsing someone else’s pragmatic provisional suggestion and establishing your own.

The techniques are also material-political: leveraging the materiality of surroundings to give greater weight to your provisional suggestion. Those who can pull the text of the Act up on their screen in the workshop can most effectively contest and recontextualise its meaning. Those who can arrange a side meeting by producing a bodily barrier between themselves and other authorities can isolate and concretise a moment of decision. Yet this should not be overemphasised. To take a trivial example, the decision about community representatives is not a direct product of the financial influence that donors brought to bear on the ADA process, even though the DA and OD sought to leverage their finances to shape implementation down to the level of local governance. Nor is it directly influenced by the AC, even though their sign bears watch over the Chief’s courthouse. This sort of open-ended bureaucratic work thus does not take place in what we might imagine governance reform to be: a legal or institutional framework for political struggles over policy and implementation. The ADA is a series of deferrals of those struggles as well as of the legal or institutional framework in which they take place.

How can one narrate these struggles – the recursive, reflexive, and antagonistic relationships between policy and implementation or acting and doing? I begin with the trip itself: it is at once a means of writing about the project and a mode of constituting it. For example, my trip to the selection meeting was undertaken as a means of getting out to the field, observing local political dynamics, and linking the global, national, and local political endeavours involved in implementing the ADA. It produced the opposite effect – enabling the creation of a space invisible to global, national, and local actors (including the Chief).

More formally, the language of place appears throughout: ‘terrain’, ‘emplaced’, ‘situated’, and so on. The scenes also have a beginning and end. All the scenes are thus bounded physically and spatio-temporally, yet they are described through a narrative that seeks to surpass those bounds. The end of the first scene, for example, projects the closure of the scene forward to the hotel that evening – yet another place and time. Movement between and beyond the scenes is further sustained through the ordering of subjects and objects within the narrative. Subjects become objects, and vice versa: Ted independently comments on my interpretation of his outcome document; I imagine the anthropologist and myself mutually narrating each other. That relationship is one of struggles between subjecthood/objecthood as well as contextualisation/decontextualisation. The draft agenda is clearly interim, invoking its blank past and its eventual, clean future; at the same time, it sits within the text as an image, which appears as an uneditable object with clear visual boundaries.

To contextualise and concretise the project, actors must contextualise and concretise their role within it: as knowers of local spaces (like Emmanuel); guardians of the process (Yahya); funders and thus priority setters (Betty); contextualisers who determine the relevant frame and map the landscape upon which the debate will be held (Ted and myself). Practices make reformers’ ignorance authoritative – both the material practices of reformers themselves as well as the formal practices of writing about them. In an effort to describe the politics of the antagonistic and mutually constitutive movement of reformers between acting and doing, or subjecthood and objecthood, the scholar must both analyse and visually enact that movement as hazy – for example, Betty appears through a car’s window tint, while people appear partially lit by their laptop screens. Our strategies of writing about projects – of breaking them into objects and moments and stitching them together again – are strategies of implementation. We are already a part of the anthropologist’s story, and she a part of ours.