Development Discourse and the "Conduct of Conduct:" Promoting Good Governance in Tanzania

Following the failure of Structural Adjustment Programmes (SAPs) in the 1980s, and the liberal triumphalism caused by the end of the Cold War, development discourse underwent a significant transformation. Key to the new development paradigm was the promotion of “good governance,” stressing the role of governments in providing a sound legal and institutional framework for economic growth and development, and conversely locating the failures of past development programmes in the “poor governance” of aid recipient states (World Bank, 1992: 9). According to the good governance agenda, the state should at once be strong enough to uphold the “institutional, regulatory and legal frameworks for development,” and sufficiently weak to allow for the participation of civil society to ensure accountability and efficiency in policy-making (Williams, 2010: 408). This essay critically examines the extent to which donors have successfully promoted good governance in Sub-Saharan Africa since 1989. Drawing on a Foucauldian conception of discourse, power and governmentality, the good governance agenda is seen as constituting a grid of intelligibility legitimising a liberal project of societal transformation via Western intervention in developing countries.

These interventions operationalise a liberal governmentality, characterised by the attempt to produce self-governing societies through the conduct of conduct – “conduire des conduites” (Foucault, 1994: 237) – that is, through productive power relations embodied in particular technologies and techniques of liberal governance (Death, 2013: 764; Gordon, 1994). This framework is utilised to illuminate both change and continuity in development policies in Tanzania, as well as the very real effects that the good governance agenda has had despite perceived failures according to its own criteria of success. Although continuities in aid conditionality are evident, the good governance agenda should not be dismissed as a mere rhetorical ploy to obscure a relation of outright domination. The governmentality approach uncovers the ways in which the agenda has promoted new forms of subjectivity in Tanzania, thus explaining how complex hierarchies of power between a multitude of actors shape variable outcomes on the ground.

The simultaneity of the emergence of the good governance agenda with the breakdown of the Cold War is not coincidental. The standoff between the United States and the Soviet Union had previously “provided the conditions of existence for development discourse” as both superpowers viewed the peripheral Third World as an important site of ideological and geopolitical contestation (Abrahamsen, 2000: 32). With the fall of Communism in the East, African states lost their bargaining position which had previously allowed them to play off the superpowers, and Western development policy was able, perhaps required, to more strictly adhere to a liberal agenda given the sudden absence of the Cold War threat as a justification for Western involvement across the globe (Ibid.: 35). Also, development agencies such as the International Monetary Fund (IMF) and World Bank faced criticisms following the failure of SAPs in the 1980s, when many Sub-Saharan African countries experienced negative economic growth and declining social conditions despite extensive international donor involvement (World Bank, 1989).

Thus, the good governance agenda allowed for locating the failure of SAPs not in the economic framework itself, but rather the “poor governance” of African states (IMF, 1997: 8). This new agenda ostensibly represents a break from past mistakes, towards a more sustainable path of economic growth via development “partnerships.” Main tenets of such partnerships include democratisation and local ownership of the development process, in which external donors provide a supporting role while the developing countries themselves take the “driver’s seat” (Wolfensohn, quoted in Mercer, 2003: 744). Attempts to promote good governance were not limited to strengthening the regulatory apparatus of central governments, but included a strong focus on the promotion of “civil society” to ensure accountability and sustainability in the policy-making process (Williams & Young, 2012: 10; UNDP, 2010). The rationality underpinning this liberal developmental model is outlined below, as are the discrepancies between stated goals of civil society promotion and the concrete outcomes of these attempts in Tanzania.

Despite such pronouncements of a new development paradigm, the good governance agenda has been criticised as a mere rhetorical device, obscuring continuities in relations of domination between external donors exerting instrumental control over the policy-making processes of developing countries (Crawford, 2003). Indeed, for some, states in the developing world have no choice but to adapt to the vagaries of the global economy as capitalism continues its relentless expansion across the world, driven by transnational organisations such as the United Nations, World Trade Organization, and the International Financial Institutions (IFIs). These institutions are characterised by hierarchical relations in which weak states “have very little say over the direction of these institutions or even their de facto participation in them” (Ó Tuathail et al., 1997: 14). Such criticisms raise important questions concerning apparent continuity in aid conditionality and the extent to which development “partnerships” really are equal.

However, in arguing that the good governance agenda serves as a ploy to veil external control over African states, these approaches gloss over “how this control is effected,” as well as why this agenda is called upon to address developmental concerns despite its apparent failures (Ferguson, 1994: 13 –original emphasis). In his interrogation of the maintenance of the prison system despite frequent proclamations of its failure, Foucault asserted that “perhaps one should reverse the problem and ask oneself what is served by the failure of the prison; what is the use of these different phenomena that are continually being criticised” (1991a: 272). A similar line of inquiry is required for a critical examination of the good governance agenda. A Foucauldian approach to discourse, power and governmentality is better able to capture the way in which the good governance discourse legitimises the operationalisation of a liberal form of power.

The good governance agenda must be placed within the broader context of development discourse. Discourse here is understood as a manifestation of power relations, forming a grid of intelligibility or regime of knowledge – “régime du savoir”– through which to understand, categorise and arrange the world in a particular way, thereby legitimating certain representations and forms of knowledge as “truths” and “expertise,” while simultaneously marginalising alternative conceptions (Foucault, 1982: 781; 1997). Power, then, is not conceptualised in wholly negative terms of domination, constraint, and control, as often represented by accounts of development as a tool for imperialist or class-based domination (Ferguson, 1994: 13). Rather, power “induces pleasure, forms knowledge, produces discourse [and therefore] needs to be considered as a productive network which runs through the whole social body, much more than as a negative instance whose function is repression” (Foucault, 1984: 61).

Development discourse constitutes a grid of intelligibility by which to represent and understand the “underdeveloped” world. It does not simply describe empirical realities on the ground, but rather constitutes the “Third World” as a particular object of knowledge, to be understood and acted upon on the basis of the structure of knowledge created around it (Ferguson, 1994: xiv-xv). The image this discourse constructs of Africa is often formulated in negative terms, that is, with regard to “what it is not …its lack of development, the absence of ‘technical knowledge’, ‘scientific advances’, prosperity, progress, and so on” (Abrahamsen, 2000: 17–original emphasis). This is not to say that depictions of instability, poverty, spread of disease, etc. are wholly imaginative with no basis in reality (Ferguson, 2006: 8-12), but rather to begin to critically interrogate the forms of knowledge, expertise and practice which are legitimised by this grid of intelligibility in order to show that “things are not as obvious as people believe … To do criticism is to make harder those acts which are now too easy” (Foucault, 2002: 456).

The negative imagery through which development discourse constitutes the “Third World” or “Africa” as an object of knowledge is based on a dichotomisation between such “underdeveloped” regions and the more developed, stable and progressive “West.” Thus, this imagery constructs a teleological and hierarchical dichotomisation, not unlike earlier colonial discourse of “the Orient” (Said, 1991), which justifies intervening in the “underdeveloped” societies to bring about the economic, social and political progress already enjoyed in the West. Such interventions are justified both by the rhetorical deployment of development discourse, and more broadly as this grid of intelligibility constitutes the “underdeveloped” world as a “field of intervention of power,” with particular “problems and possible solutions” to be addressed by “development experts” employed by a plethora of development institutions (Escobar, 1984: 387-388). Development and poverty, then, become depoliticised as technical rather than political issues, allowing major development actors such as the IMF claiming not to “interfere in the domestic or foreign politics of any member” even while extensively determining their economic policies, with significant sociopolitical effects (IMF, 1997: 4).

Although the good governance agenda must be situated within development discourse more broadly conceived, the shift from earlier forms of explicit aid conditionality towards ostensibly equal and uncoerced “partnerships” and “locally owned” development programmes is not inconsequential. In fact, while this discursive shift did indeed provide a powerful legitimation for continuity in the promotion of a neoliberal economic policy platform by locating responsibility for past failures in the “poor governance” of recipient states, it also resulted in significant change in how power relations embodied in development discourse were manifested in practice. No longer operationalised via direct aid conditionality, potentially subject to accusations of neocolonialism, development discourse began to stress democratisation and local ownership, conceptualising donors and development institutions as “partners” providing technical advice and support in the implementation of home-grown development policies (World Bank, 1998).

Dismissing this reconceptualisation of development as mere rhetoric overlooks the agency accorded to local actors in shaping social, political and economic development (Mercer, 2003: 742). This shortcoming is rectified by conceptualising the good governance agenda as reflecting a liberal governmentality (Foucault, 1991b). This framework illuminates how power is operationalised not only via coercion and constraint, but also through the “conduct of conduct of (at least partially) free and multiple subjectivities through specific techniques and technologies, within particular fields of visibility” (Death, 2013: 764). Relations of power, therefore, are not manifested only in direct domination, but in all forms of knowledge, regulations, incentive structures, value systems etc. which “not only constrain actors, but also constitute them,” thus reflecting “[t]he use of freedom as a formula for rule,” although this is not to say that coercive and disciplinary power disappear from view entirely (Abrahamsen, 2004: 1459; Foucault, 1991b: 102).

Examining good governance reform in Tanzania reveals continuity in donors’ economic rationality obscured by the new development discourse, while also uncovering the multiple and productive ways in which a liberal governmentality is operationalised to conduct the conduct of multiple subjectivities on the ground. While the outcomes of good governance reform in Tanzania cannot straightforwardly be described as “successful” according to the agenda’s own criteria of success, the governmentality approach is able to illuminate how state capacity for governance has been strengthened and some groups have indeed been “empowered,” although not always in a manner conducive to democratisation or socioeconomic development.

A liberal governmentality based on the conduct of conduct is already evident in the very form of the developmental partnership between international donors and Tanzania. With the move away from explicit conditionality, one of the preferred forms of development partnerships became the Poverty Reduction Strategy Papers (PRSPs) in the context of the Heavily Indebted Poor Country Initiative (HIPC). The PRSPs should ostensibly be drafted up by recipient states themselves, in consultation with both international donors and subnational interest groups in order to ensure sustainability, accountability, and local ownership of development initiatives (Abrahamsen, 2004: 1456). However, the IFIs retain veto power over proposed PRSPs, and in order to qualify for the HIPC recipient countries must possess a track record of compliance with neoliberal macroeconomic reform, in accordance to the IMF’s Poverty Reduction and Growth Facility (PRGF) and the World Bank’s Programmatic Structural Adjustment Credit (PSAC-1) (Mercer, 2003: 745, 751).

While the continuities between earlier SAPs and the PRSP process are evident, the PRSP and HIPC initiatives nevertheless reflect a significant shift towards a liberal governmentality, as recipient states are incentivised to transform themselves in order to qualify for funding, and to be categorised as “good adjusters” (Harrison, 2001: 659-660). Instead of rhetorically veiling a form of coercive power, then, the good governance agenda involves “rationalised attempts to conduct the conduct of free subjects, within particular fields of visibility and mobilising particular forms of knowledge and practical techniques” (Death, 2014: 71).

The development programmes adopted within Tanzania further reflect a liberal governmentality. Comprehensive programmes such as the National Strategy for Growth and the Reduction of Poverty (NSGRP), commonly known by its Swahili acronym MKUKUTA (Mpango wa Pili wa Kukuza Uchumi na Kuondoa Umaskini Tanzania, succeeded by MKUKUTA II), embody administrative, market-based and democratic rationalities (Death, 2014: 75-78). Firstly, these initiatives aim to strengthen the state’s capacity for the “technical, depoliticised practice of governance” by introducing new administrative techniques and technologies for monitoring, surveillance and management, such as the Platinum computer system for tracking government expenditures (Kelsall, 2002: 601). Second, a market-based rationality is reflected in the importance accorded to markets and the private sector in producing sustainable productivity growth (Death, 2014: 76).Continued on Next Page »