Multiple accountabilities: development cooperation, transparency, and the politics of unknowing in Tanzania’s health sector

Abstract

Accountability and transparency are considered best practices within development cooperation frameworks characteristic of global health practice today. In this article, I ask: How do accountability and transparency work, and for whom? I develop three main arguments. Drawing on Geissler’s concept ‘unknowing,’ I first demonstrate that global health actors are aware, yet strategically obscure, the instabilities and problematics of data and indicators in Tanzania. Second, I suggest that multiple and contradictory forms of accountability are pursued by global health actors, while this multiplicity is often unspoken in order to render accountability frameworks legitimate to sustain the existing development cooperation system. Third, I argue that foreign and Tanzanian actors within the health sector perpetuate accountability and development cooperation frameworks which are neither cooperative, nor accountable to citizens and purported beneficiaries of aid, because doing so allows actors to pursue interests often unrelated to formal policy goals.

Keywords:

• Tanzania
• accountability
• transparency
• global health
• ethnographic fieldwork

Acknowledgements

This research could not have been completed without the support of Dr. Faustin Njau. The project was approved by the University of Florida’s Institutional Review Board, and by the Tanzanian Commission for Science and Technology. I am grateful to Susan Erikson, Amanda Logan, Peter Locke, guest editors and anonymous reviewers for insightful feedback. Remaining shortcomings are my own.

Notes

1. The turn toward evidence-based metrics emerged in the early 1990s to increase aid accountability and evaluate disease burdens. This was exemplified in the World Bank’s invention of the DALY (Disability Adjusted Life Year) metric to quantify disease burden and cost-effectiveness simultaneously (Adams, 2016). James (2010) outlines the history of the shift of the American government toward results – and performance-based financing and its effects on USAID.

2. Aspects of the Paris Declaration (mutual accountability, sharing data, government ownership) were foreshadowed in Tanzania’s Helleiner Report (Helleiner, Killick, Lipumba, Ndulu, & Svendsen, 1995), which outlined ways to reduce tensions between donors and the Tanzanian government.

3. I employ the real names of public figures, as their attendance and statements are public record.

4. Established in 2004, DPG Health is a multi-stakeholder initiative promoting ‘aid effectiveness’ in Tanzania’s health sector, in accordance with transnational policies.

5. A digital copy of this speech is available within the minutes of the Ministry of Health and Social Welfare, Tanzania’s (2008) 9th Joint Annual Health Sector Review 89th-10th October 2008, Annex 2, pp. 29–33.

6. There is a rich literature contemplating the notion of ‘publics’ (see for instance Langwick, 2015; Warner, 2002). Warner (2002) argues that the public is not a particular group of people, but rather a form of address to strangers, who may or may not take up the possibility of becoming an audience. By ‘public writ large,’ I mean that the information is made available through media sources or websites for anyone with an interest in becoming an audience, and who has access to an internet connection or media sources through which to do so. Even then, whether or not an audience will interpret the information as intended is never assured.

7. As this paper relates primarily to actors working in the health sector, I utilize the term ‘global health actors’ from this point forward. However, development actors outside of health operate with similar frameworks and best practices, so much of what is outlined here would be similar in other development practice contexts.

8. As of 2015, some bilateral donors funded efforts to digitize data collection through a series of ‘apps’, but due to frequent electrical outages, the paper system was maintained, thereby duplicating data collection burdens. See also Erikson (2016) and Green (2015).

9. For a thorough review of scholarly engagement with ‘not knowing’, see Geissler (2013).

10. Analise Riles (2000) notes a similar form of unknowing during negotiations of an intergovernmental agreement in Fiji. When negotiations about language and meaning were complete, the document was thought of as ‘concrete’ – able to move from local to national to global level and produce other documents. Yet she footnotes, ‘the entire game of the negotiation consisted in lulling oneself and others into an apprehension of the concreteness of language even as one knew that meanings must fail to hold’ (fn. 14, pp. 200–201).

11. The identity of participants who were not public figures, such as ‘John’, have been anonymized.

12. John worked directly for the Ministry of Health to improve governance capacity. His salary was paid by a bilateral agency, but the agency had no input into his work at the Ministry; he considered himself part of the Ministry’s staff rather than a donor employee.