In recent years, neo-institutional sociologists, political scientists and geographers have engaged in a lively set of theoretical debates about how policy ideas move from one place to another. This paper seeks to engage with claims about global norm diffusion or policy transfer by studying policy discourses on agricultural development in the East African country of Tanzania. Using documents produced by international donors and research institutions, the Tanzanian government, and national and transnational civil society organizations; transcripts of parliamentary debates; and over 30 interviews with policy actors in Tanzania, we identify and compare three discourses that are currently circulating on African agricultural development policy: a global discourse, a dominant national discourse, and a subordinate national discourse. Based on an analysis of these discourses’ similarities and differences—and of the policy coalitions that are promoting them—we advance arguments about (a) the role of national contexts and historical legacies in shaping the diffusion of a global discourse; (b) power dynamics and political contention within the state itself; and (c) the transnational networks of both dominant and subordinate discourse coalitions.
Over the last two decades, sociologists, political scientists, and geographers have advanced a series of claims about how policy ideas are constructed and travel around the globe. In the 1990s, world polity theorists in sociology (Boli and Thomas 1999; Meyer et al. 1997) theorized that global norms “diffuse down” to various nation-states through the creation of a world culture that is spread by international non-governmental organizations. The result of this amorphous cultural process, according to these scholars, is policy isomorphism across different national contexts. Political science scholarship on “policy transfer,” along with some of the more recent work by neo-institutional sociologists, implicitly challenges world polity theory's central claim of “sameness” by focusing on particular cases of transfer and showing how domestic conditions mediate the process of reception.1 An important branch of this literature points to the role that domestic politics and institutions play in shaping the way that global norms, discourses, and ideas are accepted, altered or rejected at the national level (Hay 2000; Ozkan 2013). Finally, critical geographers interested in the movement of policy models across space focus on the actors, actor networks, and nodal sites where these models are developed and exchanged. Rejecting the notion that global culture flows amorphously from “the global” to “the local,” as some studies suggest, these scholars emphasize the roles specific actors and interests play in constructing and transferring policy ideas. They also highlight the fact that policy movement does not follow a unidirectional pathway, from more powerful countries to less powerful ones (McCann and Ward 2012; Peck and Theodore 2015).
We attempt to contribute to this debate about policy transfer by analyzing the production and circulation of policy discourses on agricultural development using the case of Tanzania. Policy discourse on agricultural development in Tanzania offers an excellent case through which to examine claims about whether and how policy models move, for two reasons. First, since the early 2000s, a group of large and well-coordinated international institutions, including the Rockefeller and Gates Foundations, USAID, the British Department for International Development (DfID), the World Bank, and other organizations, have generated and spread a powerful new discourse on agricultural development in Africa, with the goal of transforming African agriculture as a means of addressing poverty and food insecurity on the continent. With roots in the economic theories of U.S.- and European-based agricultural economists, international development experts, and international business consultants, this discourse can be considered “global” in that it carries the weight of global institutions of power. Second, as a relatively poor nation that is heavily dependent on these same international institutions for foreign aid, has abandoned its path of state socialism, and has bought into the idea that it can stimulate its economy through agriculture, Tanzania is precisely the sort of country where one would expect policy transfer to occur. With pressures from donors to take their advice, and that advice being copiously given, it would be reasonable to expect that Tanzania's policy approach would closely mirror the global model.
We have chosen to focus on policy discourses rather than on policy implementation because we are interested in the broad models being advanced as well as the actors who are advancing them. Drawing on previous studies (Hajer 1995), we use the term policy discourse to refer to the ideas, concepts, categories, and framings contained in particular sets of policy proposals and embraced by identifiable policy coalitions. We analyze these more or less coherent sets of ideas or representations of reality as “necessarily bound up with relations of power, agenda setting, inclusion and exclusion, selective attention and neglect” (Fischer and Forester 1993). We also follow the lead of scholars who emphasize development policy discourses as a form of argumentation, highlighting differences between the broad arguments informing each discourse of agricultural policy. In this view, “policy making is a constant discursive struggle over the criteria of social classification, the boundaries of problem categories, the intersubjective interpretation of common experiences, the conceptual framing of problems, and the definition of ideas that guide the ways people create the shared meanings which motivate them to act” (1–2). The discourses analyzed below are each based on particular claims-making stories about agricultural development that inform the contentious politics of agricultural policymaking in Tanzania. Our analysis documents the central descriptive and normative claims of each discourse, and we institutionally and politically contextualize them by identifying what Maarten Hajer (1995) labels discourse coalitions. In other words, our focus is on the agenda-setting dimension of power and the formulation of policy proposals, that is how problems are constructed via various claims made in policy debates, rather than on the implementation or evaluation of policies.
Our research compares the global discourse of agricultural development with two domestic discourses on agricultural development in Tanzania in the post–structural adjustment period, that is from the 2000s onward. By identifying the discourse coalitions and narratives underpinning these three discourses, and the similarities and differences between them regarding smallholder farmers, land rights, foreign investments, the role of the state, and genetically modified organisms (GMOs), we contribute to the aforementioned debates concerning policy transfer by answering the following questions: To what extent is the global discourse of agricultural development incorporated into the ideational framework adopted by the Tanzanian state, and what does this suggest about the movement of policy ideas from the global to the national level? What groups comprise the different discourse coalitions supporting these models of development, and what do these coalitions reveal about the contemporary character of the Tanzanian state? How, if at all, do domestic politics and distinctive historical legacies mediate the reception, revision, or rejection of the global discourse?
The data for our analysis come from published government documents and policy papers, transcripts of parliamentary debates (United Republic of Tanzania, various years), newspaper articles, NGO publications, and secondary sources. We analyze these data via a very close, interpretative reading of these documents, rather than relying on a quantitative form of content analysis. We also conducted three dozen interviews in 2013 and 2014 with a range of key informants involved in policymaking and/or discourse construction. Our interviewees included government officials, members of parliament (MPs) from the ruling and opposition parties, leaders of civil society organizations (e.g. smallholder farmer associations), and foreign aid officials.
The rest of this paper proceeds as follows. We begin with a discussion of the theoretical scholarship on processes of “national” policymaking, focusing on the literatures introduced briefly above. Subsequently, we describe the global discourse on agricultural development and compare it with the two domestic policy discourses that we found circulating in Tanzania after the early 2000s. One of these domestic policy discourses can be considered dominant at the national level, and the other represents an alternative (and subordinate) discourse. We compare the similarities and differences among these discourses and utilize these comparisons to advance three central arguments. First, we argue that what might seem to be a fairly straightforward story of the diffusion and adoption of a global discourse is actually more complex than it first appears. When looked at from afar, the dominant domestic discourse appears to be a product of the diffusion of the prevailing global policy consensus on agricultural development. Closer inspection reveals that domestic actors—that is Tanzanian state technocrats and domestic entrepreneurs—played a central role in constructing some of the major national policy visions regarding agriculture while rejecting one of the central tenets of the global discourse: the emphasis on smallholder agriculture as the optimal route to reducing food insecurity and poverty. As members of a modernizing bureaucratic elite that sees itself as the vanguard of change, these actors pursued their own interests and ideas rather than simply parroting all of the elements of the global discourse. Since the struggle for independence from colonial rule, well-educated and technically competent Tanzanian state bureaucrats have regarded themselves as a vanguard of modernity (Schneider 2007).
Second, the fact that there are two competing domestic discourses—each rooted in coalitions that include state officials—indicates that the state is not a unified entity or cohesive actor. Indeed, our findings highlight a central division within the state between the two main opposition parties in parliament (CHADEMA and CUF), on the one hand, and bureaucratic officials associated with the ruling party (CCM) in the executive branch, on the other.2 The former, we argue, were more attuned to electoral pressures emanating from a largely rural electorate, making them more open to the criticisms of the dominant discourse and to joining the subordinate discourse coalition. This subordinate discourse was rooted in a distinctive national history and context of decades of state socialism that had treated land as a public community resource rather than a commodity and propagated a political culture based on values of national self-reliance and equality. These values continue to appeal to many of Tanzania's poor farmers, who harbor fears of losing their land to foreign investors, becoming dependent on foreign seed companies, and ingesting genetically modified food pushed by foreign corporations. Government bureaucrats are more concerned with technical rationality, problem-solving expertise, and securing foreign aid and investments as well as maintaining their legitimacy among donors and other development professionals. This led them to embrace ideas that were relatively in line with (although not identical to) the global discourse of agricultural development.
Finally, our research suggests that not only did ideas move through the global networks connecting foreign donors and international organizations to state bureaucrats promulgating the dominant national discourse, but parallel networks shuttled other ideas among opposition leaders in the Tanzanian parliament, domestic civil society organizations, and transnational activist groups who constituted an alternative discourse coalition. In other words, the transnational movement of ideas occurs within multiple networks at the same time, not just in those emphasized by the existing literature, which focuses on the transfer of dominant policy models. We now discuss the theoretical debates that animate our interest in how policy models circulate.
World polity neo-institutionalism (WPN) offers an influential account of how nation-states around the globe adopt similar policies despite very different socio-historical contexts, documenting, for example, the implementation of common public policies against female genital cutting (Boyle 2002) and educational curricula (Ramirez and Meyer 2002), environmental protection (Frank, Hironaka, and Schofer 2000), and “women and gender” as development assistance priorities (Swiss 2012). According to these scholars, the global diffusion of Western “norms of rationalization, universalism, belief in progress, and individualism” shapes the public policies of nation-states around the world (Schofer et al. 2012). Global cultural models provide external legitimacy for the states that embrace them and project an appearance of universal agreement (Meyer et al. 1997). As a cultural theory, WPN is decidedly not actor-centric; rather, the effects of world culture are seen as highly diffuse, almost like the air we breathe (Schofer et al. 2012). Change may happen in a variety of ways and take a range of forms. Schofer and Hironaka (2005) suggest a “dynamic imagery, in which actors at all levels of the social system may take cues from exogenous norms and cultural models.” Reflecting their focus on the diffusion of Western norms and institutional forms around the world, WPN's focus is on sameness, rather than difference (Schofer et al. 2012).
Critics of WPN theory have pointed to the need to recognize tensions and contradictions between different elements of world culture that limit global convergence/isomorphism, such as tension between the norm of progress and that of justice (Finnemore 1996). However, a revised WPN theory that acknowledges contradictions between different global norms still ignores power relations in the global system, including the power of donors to pressure domestic actors to adopt particular policies. It also marginalizes national political contention arising from the contradiction between local and global norms. Importantly, WPN ignores differences in domestic actors’ interests and how struggles among domestic actors shape discourses of public policy. Amitav Acharya (2004) argues that variation in the acceptance of global norms can be explained by the differential ability of local actors to reconstruct foreign norms so as to ensure that they are congruent with the agents’ prior norms and beliefs. These criticisms suggest that the WPN literature does not engage in any ‘deep study’ of norms to see what actually diffuses and what forms these presumably similar models or ideas take in situ.
The political science scholarship on “policy transfer” overcomes these limitations by exploring the role of national institutions and institutional legacies as significant shapers of the process of reception. In interrogating the intersection of the global and local, this research moves beyond a focus on Western transnational actors and norms, international issue constructions, and consensus-based diffusion mechanisms by incorporating national actors and issue constructions. It also highlights conflicts that often accompany the process of diffusion as national actors and institutions struggle over, reject, or modify global norms and policy prescriptions. Colin Hay (2000: 509, 512) importantly recognizes variation in “national capitalisms” and the “contingent, political and frequently coercive nature of the convergence process,” noting that the particular institutional and cultural characteristics of a state will shape how it responds to external pressures for change. He argues that such institutional and cultural features “serve to channel, filter and mediate common challenges in highly specific ways, often reinforcing the distinctiveness … of distinct national … capitalisms” (512).
Political scientist Peter Newell's (2008) work comparing the “domestication” of global biotechnology policy in India and China offers additional insights into the means by which global norms and political commitments get interpreted and implemented at the national level. Newell draws attention to the fact that there are different, and often contending, interests within the state that need to be taken into account when analyzing the process of policy transfer. His work also highlights the political alliances that may form between state actors and non-state actors, both domestic and transnational. He demonstrates that both of these contending interests and alliances can help explain policy inconsistencies and tensions, as well as variations in the way that global policies translate into domestic policies in distinct national contexts. Finally, Newell suggests that there has been a blurring of the domestic and the global, as reflected in the existence of “globalising state bureaucrats,” or domestic officials who are deeply embedded in transnational policy networks, and “localized global actors,” such as transnational corporations operating at a national level (117).
Finally, the “policy assemblages, mutations and mobilities” literature in geography (or “policy mobilities,” for short—McCann and Ward 2012; Peck and Theodore 2015) takes some of these insights even farther. This literature recognizes that policy models morph as they travel from one place to another. Indeed, the central focus of policy mobilities research is the process by which policies move and mutate and the mechanisms involved, rather than the “fact” of their movement (see e.g. Peck and Theodore 2015, ch. 1). Policy mobilities scholars insist on grounding their analyses in concrete social, political, and institutional contexts, and eschew the high level of abstraction characteristic of world polity approaches, in which these processes are not spelled out or subjected to detailed, empirical analysis.3
The methodological approach of the policy mobilities and policy transfer literatures involves focusing deeply on specific cases to understand how new policy models are adopted (or rejected), translated, and modified, as they move across space. These approaches reveal what highly abstract and macro theories of policy diffusion cannot, namely, the way policy discourses change in particular national circumstances, as well as why they change and how such change is accomplished. Whereas the WPN scholarship points to broad and diffuse global pressures resulting in the spread of particular institutional forms, the policy mobilities literature concerns itself with identifying specific forces for and agents of change. The policy translation and policy mobilities literatures embrace complexity, toward the goal of illuminating concrete and impactful processes of change. In the process, they show how both “global models” and domestic circumstances matter. We now turn to our case study of policy discourses in Tanzania, beginning with the global agro-industrial discourse on agricultural development.
THE GLOBAL DISCOURSE
A new global discourse on African agricultural development emerged in the early 2000s from groups concerned about the growth in African poverty and hunger during the 1990s.4 This new discourse is broadly neoliberal in character, but does not take the market-fundamentalist form that characterized the “Washington consensus” of the 1990s. Its foundational assumption is a market-based economy in which smallholder farmers are at the center of the agricultural development strategy.5 It also contends that agricultural productivity in Africa must be enhanced through smallholders’ adoption of modern technologies and inputs; that agricultural growth should be private sector–led and market-driven; and that the state's role is to create a conducive environment for private investors. If these conditions are met, African agriculture will not only be able to feed the continent, but will also represent a motor for economic growth for African countries.
This discourse had its roots in several different policy communities, which became increasingly connected over time to form a discourse coalition. These included (a) the Rockefeller and Gates Foundations and the Alliance for a Green Revolution in Africa, which collectively argued for the need to raise African agricultural productivity through a science-and-technology-oriented approach; (b) the community surrounding UN Secretary-General Kofi Annan, who used the UN's Millennium Development Goals to call for a “uniquely African green revolution” to address the region's persistent poverty; (c) the international aid donor community, comprising USAID, DfID, and the World Bank, among others; and (d) African officials working for the Comprehensive African Agricultural Development Program.6 In 2007–08, the world food price crisis drew enormous attention to the global food security issue and helped bring these four groups together, buttressing an emergent global discourse with a storyline of smallholder entrepreneurship, gender equity, private-sector investment, and agro-industry. The following discussion is based on a close reading of reports by the institutions involved in generating and circulating this discourse. Central among them are the World Bank's 2008 World Development Report, Agriculture for Development; various USAID documents; papers written by agricultural economists in U.S. land grant universities and the Washington, DC–based International Food Policy Research Institute; and materials from the Rockefeller and Gates Foundations. While not every actor involved in constructing the new discourse emphasizes the same elements, they broadly agree on the tenets identified here.
Smallholder farmers—particularly women7—play a central role in the global discourse, which rejects the view that smallholders are inefficient producers who should leave the land to make way for larger, more productive enterprises that can take advantage of economies of scale. Indeed, smallholder farmers are seen as “micro-entrepreneurs” who are capable of lifting themselves out of poverty if they can be capacitated to produce more agricultural surplus and connected to new markets (Oslo Conference 2006; World Bank 2008). The former can be achieved by helping them access commercially produced “improved” seeds (rather than farmer-saved seeds) and inorganic fertilizer, both of which will raise their productivity. New market linkages will come through private investment in regional and international value chains that have emerged because of globalization and the rise of new middle classes in Africa and elsewhere (Gates Foundation 2011; USAID Feed the Future 2013). The goal is to shift smallholder farmers from subsistence to commercial farming (Fan et al. 2013). The global discourse contends that property rights should be made clear and unambiguous in order to incentivize farmer investments in land improvements, and so that land can be transferred to more productive uses when it is not being utilized efficiently (World Bank 2008: 138–141). The donor community thus supports land titling, on the assumption that land titles will make it easier for farmers to obtain the credit they need to purchase inputs and to sell their land if they decide to do so. Furthermore, rapid population growth in many countries has created such great land fragmentation that it is no longer viable for some smallholders to farm, even for subsistence (World Bank 2008). As a result, some land parcels will need to be aggregated to create an economically viable level of production, a process that can be facilitated through land rentals or sales. For farmers whose plots are too small to be commercially viable, the discourse envisions work in new agro-processing industries and other rural labor markets or migration as more promising “pathways out of poverty” (Alliance for a Green Revolution in Africa 2011: 28; Grow Africa Secretariat 2013; World Bank 2008: 72–96).
Firms that develop and sell inputs, produce farm equipment, or work in food processing and distribution are also key actors in the global discourse. Support for private-sector investment figures prominently among the donor community, and the global agro-industrial discourse explicitly suggests that the only economically sustainable and cost-effective way for the agricultural sector to grow is through private-sector investment. The private sector is believed to have the know-how, the experience, and the right economic incentives to engage in what is seen by this discourse coalition as a business activity. The World Bank, USAID, the Alliance for a Green Revolution in Africa, and other proponents of the global discourse thus widely promote private investment in African agriculture and advocate “public–private partnerships” as an important means to achieve it.
While the global discourse is not explicitly pro–foreign investment, many of the private-sector partners to which the advocates of this discourse refer are assumed to be foreign investors. This is obvious when it comes to the many U.S. agribusiness firms that became active members of this discourse coalition after the 2008 world food price crisis.8 It is also evident in the position taken by the Grow Africa Partnership, which was co-founded by the African Union, the New Partnership for Africa's Development, and the World Economic Forum in 2011 to broker agricultural investments in Africa.9 In short, the global discourse is not overly concerned with the origins of new agricultural investments; the goal is simply to encourage any private investors who will advance the objective of agricultural development.
The role of the state in the global discourse is to encourage private-sector investment by creating an enabling business environment. This means cutting red tape, regulations, and taxes that discourage companies from investing in the agricultural sector. The discourse also urges governments to establish clear and stable rules for doing business, protect property rights and promote land titling and the commodification of land, and liberalize trade by eliminating trade barriers. Because the task of developing a competitive, commercial agricultural sector is so challenging, the state and the private sector need to become development partners. The proponents of this discourse see governments and private actors as sharing the same goals and having common interests. States can support private-sector investment by establishing “growth corridors” (clusters of agribusinesses, infrastructure, service providers, and farms that generate economies of scale); helping agricultural investors gain access to land; and facilitating partnerships between public agricultural research organizations and private seed companies. Finally, the state should invest in public goods that the private sector has no incentive to develop, for example, roads that increase physical access to markets and other public infrastructure. These new state roles do not involve direct government intervention in markets to regulate prices for smallholder farmers or challenge market oligopolies; they do, however, insist that African governments develop their institutional capacity, eschew political favoritism and rent-seeking, and adopt the “good governance” norms followed by states in the global North.
Finally, the global discourse promotes a wide variety of technologies that are seen as important for increasing agricultural output, namely, improved seeds, fertilizers, agrochemicals, and new information technologies such as cellphone apps for farmers. Proponents of this discourse are vocal about the need for governments to support research and development for high-yield varieties of seeds on the grounds that Africa has to increase its agricultural productivity if it is going to feed itself. Transgenic crops are seen as offering particularly promising solutions to pest problems, drought conditions, climate change, and plant diseases.
THE DOMINANT DOMESTIC DISCOURSE
In this section, we analyze the dominant national discourse on agricultural development in Tanzania. This discourse espouses a development strategy reliant on private investment in commercial farming and agro-enterprise, contract farming in the form of an anchor farm (and agribusiness) that works with smallholder contract producers, and partnerships between domestic and foreign capital as well as between the state and the private sector. This formulation, we argue, is broadly consistent with the global discourse on agricultural development based on an embrace of the private sector and emphasis on the use of modern agricultural inputs, global value chains and commercial markets. But it differs from the global discourse in that it privileges large and medium-size enterprises, de-emphasizes smallholder producers as a source of dynamism and change, and openly encourages foreign capital as business partners in new agro-industrial activities. The difference reflects a change in the political economy of Tanzania, in which there is a growing black Tanzanian business elite that is openly claiming space and political legitimacy. Indeed, it is this business elite that produced Agriculture First, which former President Kikwete and the prime minister's office embraced as their official model of agricultural development. In other words, this discourse did not simply get transferred from outside, but was in part produced by an increasingly vocal and powerful set of domestic interests, members of an emerging national bourgeoisie seeking opportunities for foreign partnerships in a global economy.
We characterize this discourse as dominant because it was effectively institutionalized by a powerful discourse coalition at the national level, where it informs both policies and institutional arrangements as the most credible and acceptable approach to agricultural development. The ruling party, CCM, which has dominated both the executive and legislative branches of government for five decades, embraces this discourse in its agricultural policy platforms and proposals. Three main groups of actors have been engaged in constructing this discourse: technocrats in the executive branch of the Tanzanian government, especially the Ministry of Agriculture and the prime minister's and president's offices; private-sector actors associated with the Tanzanian National Business Council, who in 2008–09 formulated an agricultural strategy called Kilimo Kwanza, or Agriculture First; and a group of domestic and international actors associated with an organization called the Southern Agricultural Growth Corridor of Tanzania (SAGCOT), which seeks to put the Agriculture First framework into action (Cooksey 2013a). This discourse is reflected in various policy documents that the government and private sector have produced, and in public statements made by government officials, the prime minister, the president, and the private sector. Among the documents we analyzed are the Ministry of Agriculture's Agricultural Development Strategy Plan 2006–2011; a “ten pillar” Agriculture First framework that the Tanzanian private sector produced in 2008–09; the Blueprint for the Southern Agricultural Corridor of Tanzania; and the Ministry of Agriculture's Tanzanian Agriculture and Food Security Investment Plan (TAFSIP) and its Big Results Now program. Some analysts (Cooksey 2013a) have highlighted sharp differences between these various policy documents and visions, characterizing TAFSIP as prioritizing smallholders and state intervention, in contrast to Agriculture First and its implementation in the SAGCOT, which emphasizes large-scale investments, global value chains, and the private sector. Our analysis suggests that although these seemingly incompatible policy visions have been a source of political contention within the government and ruling party, they have been reconciled over time, giving way to a discourse that promotes “modern” agriculture based on agro-industrial inputs and large-scale private investments, while also claiming to further the interests of smallholder farmers.
The dominant domestic discourse asserts that Tanzania has tremendous agricultural potential but is not living up to it, a claim that also informed socialist-era development planning. Rather than being the breadbasket of the region and a major food exporter to other countries in the world—which proponents of this discourse suggest it could be—Tanzanian agriculture is seen as suffering from a lack of modern technology and severe underinvestment. Among the factors hindering the sector are “low productivity, overdependence on rain-fed agriculture; inadequate agriculture support services; poor infrastructure; weak agro-industries; low quality of agricultural produce; inadequate participation of the country's private sector in agriculture; environmental degradation and crop pests and diseases” (MAFC 2013: xii). Other factors viewed as constraining include gender relations, weak producer organizations, and “insecure property rights to land, which limit its use as collateral for credit and limited involvement of the private sector” (5). In short, the dominant national narrative about Tanzanian agriculture is that the country has been caught in a subsistence-agriculture trap and must transition to a modern, commercial agricultural system.
According to this discourse, modernizing the sector will require major investments in infrastructure (irrigation, roads, and ports) and in mechanization (e.g. tractors), so Tanzania's farmers can become more productive, achieve scale economies, and become internationally competitive. This will promote the sector's growth. Developing the agricultural sector will also necessitate the increased use of commercially produced hybrid and GMO seeds, fertilizers, agrochemicals, and pesticides, as well as enhanced research and extension services (MAFC 2013: 6; Tanzania National Business Council 2009, esp. Pillar 7). The dominant national discourse is that the state, in conjunction with its development partners, should make infrastructural investments (United Republic of Tanzania 2011; author interviews with MAFC officials). But it is private investors, both domestic and foreign, who are seen as the primary engines of growth for Tanzania's agricultural economy (MAFC 2013: xi–xii). It is the state's job to incentivize these actors by making policy reforms that are friendly to them, including making large tracts of land available for commercial farms, even if it requires removing villagers currently occupying these lands. For instance, Agriculture First calls for land to be sold and investors’ land rights secured; furthermore, land banks should be created for commercial production and investment by documenting all “underutilized” land for agricultural investments (Tanzania National Business Council 2009: 10). This is meant to lay grounds for easy and equitable access to village land (Mbunda 2013: 15). This claim regarding underutilized land also appears in the TAFSIP, which notes that “Tanzania still has large areas of arable land that are not used for crop production, but could be developed for commercial farming. This form of extensive agriculture is rather capital intensive and will require substantial private sector participation, including possibly foreign direct investment” (United Republic of Tanzania 2011: 36). In short, although the dominant national discourse on land issues is very similar to the global discourse, it privileges large-scale commercial producers, rather than entrepreneurial smallholders, as the central dynamic force in agricultural development.
As suggested by the last quotation, the dominant national discourse recognizes the need for foreign investment in order to develop the agricultural sector. Agriculture First proposes a paradigm shift from state-led to market-led agricultural development that will stimulate investment in the sector (Cooksey 2013b: 12; Tanzania National Business Council 2009). As part of this paradigm shift, the government is focusing on the SAGCOT project to mobilize large-scale investments in agriculture, global value chains, and private-sector participation via contract farming (SAGCOT 2011: 15, 2015: 4). The state should play an enabling role by promoting large-scale, socially responsible foreign investments and encouraging joint ventures between foreign investors and domestic capitalists (Cooksey 2013a; SAGCOT 2011: 7).
The dominant discourse points toward a difference in the government's assessment of the kinds of farmers who are capable of transforming the sector. Rather than adopting the global discourse's focus on smallholder farmers, government technocrats and private-sector actors are increasingly setting their sights on large and medium-scale producers as the instigators of change. This is especially evident in the more recent documents on agricultural policy. For example, the government's National Agricultural Plan, published in October 2013, notes that “a transformation approach toward the creation of farming systems based on more intensive and permanent use of crop land areas, by efficiently run and planned farms of economic size, is crucial in order to improve productivity. The transformation in production technology and productivity necessary to achieve high rural incomes and growth will take place in the smallholder sector given time and the correct incentives” (MAFC 2013: 12, emphasis ours). In 2013, a Big Results Now report on agriculture portrays smallholders as obstacles to, rather than opportunities for, agricultural development, reporting that “the dominance of small-scale farms is one of the key obstacles in the way of higher agricultural productivity in the country.” The report continues: “Lacking scale, smallholders forego advantages enjoyed by commercial farmers such as lower operational costs, better access to farming inputs and greater farming expertise” (Tanzania Development Vision 2025 2013: 6).
While smallholder farmers do feature in the dominant national discourse, they are seen as incapable of leading a transformation and dependent on the state for improvements in their well-being. From this perspective, most smallholders are too poor to buy the modern inputs that could raise their farms’ productivity, and they often do not have adequate knowledge of them (MAFC 2013: 8).10 The dominant discourse (and the government's actual agricultural strategies) envisions two potential futures for smallholders. One involves contract farming, in which farmers continue to focus on agricultural production, but do so through a hub-and-outgrower scheme in which smallholders are connected to large “anchor” farms owned by agribusinesses engaged in supplying national, regional, and global value chains. In this scenario, farmers who live in the vicinity of these large-scale corporate farms will be able to access inputs, extension services, value-adding facilities, and markets (SAGCOT 2011: 9), and will thus be converted from subsistence farmers to successful commercial farmers. In this view, small and large-scale farmers have compatible rather than antagonistic interests and will both benefit from contract farming. The other future, which is more subdued in the global discourse, involves an increase in the demand for rural labor associated with the expected increase in Tanzania's agro-enterprises, thus drawing smallholder farmers off the land into non-farm employment.
In sum, the dominant discourse is an amalgam of ideas, some consistent with the prevailing global discourse on African agricultural development and others which are distinct from it. Despite a difference concerning the privileging of large-scale commercial farming in the former, the two discourses coincide in many ways, including the notions that agricultural development must be private sector–led; that farmers should adopt modern inputs and orient themselves toward commercial markets (value chains); and that the state should encourage private-sector investment, rather than directly intervene in markets, by creating an attractive investment climate through land, trade, and tax policy and by investing in public infrastructure. Whether these ideas come from “inside” or “outside” is impossible to determine, as many of the state and private-sector actors advocating these ideas are closely connected to transnational policy networks. Many have worked or studied abroad, meet regularly with donors, move in international circles, or are part of an internationalized business class. Furthermore, as Graham Harrison (2004) argues, many officials working in African government positions have imbibed neoliberal ideas for decades and are often more neoliberal than those in the international financial institutions that “structurally adjusted” African economies in the 1980s and 1990s; he refers to this phenomenon as a “post-conditionality state.” At the same time, the dominant national discourse clearly has some strong domestic elements, ideas that distinguish it from the global discourse. Prime among them is the idea that Tanzania should encourage large and medium-scale industrial agriculture and agriculturalists. Small-scale farmers are not absent from this discourse, but they are certainly not prioritized the way they are in the global discourse. This is, we argue, due to the nature of the discourse coalition, in which an emergent domestic bourgeoisie and modernizing state administrators play a central role.
THE SUBORDINATE DOMESTIC DISCOURSE
Alongside the global and dominant national discourses of agricultural development, there exists an alternative domestic discourse elaborated and propagated primarily by national and transnational civil society organizations (CSOs) and elected opposition party MPs. As elected officials, these MPs speak to the grievances of rural voters, whose interests, they argue, are not being adequately taken into account by non-elected bureaucrats in the administrative branch of the state. This discourse emphasizes a commitment to improving conditions for smallholder farmers. It challenges claims and assumptions in the dominant domestic and global discourses regarding the commodification of land, the proper role of the state, the need for large-scale foreign investments and agro-chemicals, and the benefits of genetically modified seeds. The discourse questions claims about the shared interests of smallholder farmers and large-scale investors, highlighted in the global discourse on contract farming and foreign investment, suggesting instead that these groups have antagonistic class interests and a neo-colonial relationship. The discourse also challenges the global discourse's advocacy of a “free market” in land made possible by land titling, arguing instead for the treatment of land as a community resource, not a commodity, and for policies to counter land-grabbing by large-scale foreign investors. It also envisions a very different relationship between the state and the market, advocating the use of state power to support informal and local markets (rather than export markets/global value chains), to guarantee fair prices for smallholder farmers, and to promote organic rather than agro-chemical inputs. Finally, in contrast to the other two discourses, it urges the government to prohibit the use of genetically modified seeds, which are viewed as part of an unsustainable agro-industrial model.
We label this a subordinate discourse because the major actors in this discourse coalition, namely, opposition political parties and civil society organizations, lack the political clout of those promoting the policies favored by the dominant discourse coalition. The label “subordinate” is not meant to imply that this discourse coalition has been entirely ineffective in influencing public policies. Although it has limited traction in government agricultural policymaking, it has managed to force the government to modify or hold back on a number of policies affecting rural areas and has mobilized vocal opposition to new policies. For example, in 2009 the government suspended new biofuels investments and halted land acquisitions by foreign biofuel companies until tighter regulations were drawn up; and in 2010 new mining legislation required a government stake in all future mining projects, set aside areas for small miners, and increased taxes for foreign multinational mining companies (Aminzade 2013: 292–295). Criticism of land-grabbing led to passage of a new policy at the end of 2012 limiting the amount of land that single large-scale investors were allowed to lease for agricultural use (Kiishweko 2012). In addition, the government faced intense opposition in parliament in 2013 when it proposed legislation to abandon the 2009 strict liability laws that treated anyone importing, storing, or using GMOs as legally liable for any claims of harm, injury or loss (including environmental degradation and a loss of biodiversity) and to replace it with a new system that placed the burden of legal proof on those claiming harm or loss (Schmickle 2013). These policies and policy debates reveal both the influence of the subordinate discourse coalition and the way in which an oppositional discourse can help alter elements of the dominant discourse in response to criticisms arising in the contentious politics of agricultural development.
This section makes use of CSO documents and websites, as well as interviews with opposition party MPs, to explore key dimensions of this subordinate national discourse. The discourse is elaborated by a number of domestic and transnational CSO networks, including the Tanzania Land Alliance, the Tanzania Alliance for Biodiversity, and the Alliance for Food Sovereignty in Africa. The Tanzania Land Alliance was created in 2010 to coordinate the activities of Tanzanian land rights organizations, with prominent member organizations including the Land Rights Research and Resources Institute (HAKIARDHI) and Mtandao wa Vikundi vya Wakulima Tanzania (Network of Farmers’ Groups in Tanzania, MVIWATA), a national network of small farmers. A second important CSO network is the Tanzania Alliance for Biodiversity (TABIO), formed in 2011. Although most of the 19 farming and environmental organizations belonging to TABIO are domestically based civil society organizations, such as MVIWATA and Sustainable Agriculture Tanzania, many members are European-based, including transnational CSOs such as Envirocare and Action Aid International, private-sector members such as Biolands, and the European aid organization Swissaid (TABIO 2016c). TABIO belongs to a pan-African transnational network of farmer organizations, the Alliance for Food Sovereignty in Africa, founded in South Africa in 2011. The presence of these transnational organizations in this discourse coalition suggests the existence of global networks that are parallel to the dominant ones and which transfer alternative ideas.
These national and transnational umbrella organizations and their members have helped develop and disseminate an alternative discourse that has found its way into Tanzanian parliamentary debates and the policy proposals of opposition party leaders in parliament. An Oxfam (2012: 26) report on Tanzania noted the overall lack of inputs from smallholder farmers in the process of policymaking and implementation, but noted that “barriers to voice and participation are slowly being broken down, and in more recent times some smallholder farmers’ organizations and civil society organizations (CSOs) (such as MVIWATA and ANSAF in Tanzania), as well as some members of parliament, are increasingly echoing the voices of smallholder farmers in the corridors of power where policies are made and reviewed.” The following sections document key elements of this subordinate domestic discourse.
A central claim of this discourse, rooted in the legacy of state socialist policies that regarded land as a public good, is that the treatment of land as a commodity rather than as a community resource has led to land-grabbing due to class inequalities of power. A 2011 report by HAKIARDHI (Mwami and Kamata 2011: 11) on land-grabbing attributes land conflicts to “the expansion and deepening of the interests and needs of international capitalist accumulation which, under neoliberalism, are realizable through the transformation of everything into a commodity.” The report characterizes market-based land reforms as an effort to “commodify land and in doing so usher in a whole process of expropriating the peasants, and in turn strengthen private property rights of the agrarian capitalists (large commercial farmers).” MP Tundu Lissu of the CHADEMA Party echoed these criticisms:
Market-driven solutions are the kind of solutions that will render millions of people landless. They are called market solutions because all they look at is the economic side of things. Even then, the so-called market solutions are heavily dependent on the state. … The problem is not necessarily illegal activity. It may have more to do with the legal arrangements that have been in place the past two decades which have allowed the proliferation of land-grabbing. … It may be done legally, but it is land-grabbing all the same because the relationship between the investor and land owner is such … an asymmetrical relationship. (Author interview, May 15, 2014)
CSO leaders point to the large number of land conflicts between investors and smallholders as evidence of land-grabbing. Yefred Myenzi, executive director of the Tanzanian Land Rights and Resources Institute, stated at the end of 2012: “Land conflicts pitting poor villagers against powerful investors now number more than 1,000 reported incidents. On average, there are five land disputes daily in the country and three of these involve powerful investors” (Kiishweko 2012).
The subordinate discourse also criticizes the erosion of the commons under the guise of utilizing supposedly empty land for large-scale agriculture. Critics question this claim, noting that much of this land is already in use for various purposes, including communal activities, national parks, game and forest reserves, and grazing by pastoralists, and that much of it is unsuitable for agriculture due to poor soils, inadequate water supplies, or the danger of soil erosion. Local populations, observes an Oxfam report (2012: 31), “often have legal claim only to the land that is in ‘productive use’, which undermines local claims to rangelands, hunting-gathering grounds, sacred sites, and land reserves set aside for future generations—lands that are often traditionally managed as common property resources.”
The subordinate discourse echoes socialist-era notions of national self-reliance, warning of the threat posed by foreign investors to national interests and to national self-determination. Whereas the dominant discourse regards large-scale foreign investors as key partners in any effort to modernize agriculture, proponents of the subordinate discourse warn of the land dispossession that will inevitably accompany large-scale corporate farming. “If you want to invest in large-scale agriculture,” stated CHADEMA MP Tundu Lissu, “you are going to have to create a ‘rural proletariat,’ if you want to use the old terms. You cannot have one and retain the other. … Where are they going to take all these millions of peasants once we make the American Midwest out of Tanzania?” (author interview, May 15, 2014). Some CSOs have rejected the claims of foreign investors that they have received consent from landowners for the purchase of their land, viewing the active consent of the governed as an example of the exercise of hegemonic power. “Land is alienated to foreign companies and local elites, and peasants are promised development and employment opportunities,” stated a HAKIARDHI report on land-grabbing. “This is a consent building mechanism and makes dispossession appear as if it was consented by the dispossessed themselves. But where this is not easy, … the state intervenes on behalf of investors. … The state has turned itself into an agent of foreign interests against the interests of the popular masses” (Mwami and Kamata 2011: 12, 33)
Proponents of an alternative approach to agricultural development suggest that the state should protect the interests of vulnerable smallholders by closely monitoring and regulating large land deals, breaking up marketing cartels, promoting informal and local markets, guaranteeing fair prices for crops, and encouraging natural fertilizers and plant-based insecticides (TABIO 2016c). The Tanzania Gender Networking Programme (2010: 2) policy brief on Agriculture First questioned who would benefit from the government's promotion of a “green revolution” for Tanzania and warned that an initiative based heavily on the use of petrochemical fertilizers, insecticides, and GM seeds was not only hazardous to the environment and human health but would also “lock producers into dependency on the products of corporate agribusiness.” The alternative, it argued, was a government “commitment to support—through financing, training and provision of services—peasants and small scale farmers, the majority of whom are women.”
Whereas the executive branch has made it a priority to ease the way for large-scale investors in agriculture, oppositional voices within parliament have called for greater state oversight and regulation of those investors. “The state sees its role as being to assist the investor to succeed to get the land,” noted Tundu Lissu of CHADEMA. “Parliamentary oversight is very important, particularly if it comes in prior to the allocation of these large tracts of land. It shouldn’t come after the fact” (author interview, May 15, 2014). He also called for greater state intervention in the marketplace to limit the power of cartels, which have the ability to force small-scale farmers to sell their crops at low prices.
Critics of the dominant discourse question whether government efforts to encourage commercial investors to launch large-scale projects, including the outgrower schemes in SAGCOT's plans, can really benefit most smallholders. Oxfam (2012:6), for example, acknowledges that public policies “can support more and fairer connections between commercial investors and small-scale producers—such as oversight of contracts, incentives for producer organization, and oversight of trading relations between producers and buyers.” However, it notes that “these value chain linkages are more likely to work best for a particular segment of rural societies. This segment involves only the top 2–10 percent of small-scale producers, primarily men, who have the assets and access to capital, organization, information, and infrastructure to ‘step up’ to formal and coordinated markets.” The alternative, suggest the authors, is to
strengthen the organization and market power of small-scale producers (including channeling support for informal groups, given women's predominance in these organizations), encourage preferential public/food security procurement, provide a greater diversity of market outlets, make standards and certification work better for small farmers, break up cartels and monopolies, improve price stability and producer share of market value, and/or provide market preferences for small- and family-scale producers, for instance in export quotas or public procurement, or through protecting small-farm sectors from import competition. (7)
This approach envisions an activist state that fosters “a diversity of market outlets—from street vendors to local wet markets to national wholesale markets and terminal markets—as a public good” (53). While in agreement with the global discourse on the need to prioritize smallholder farmers, calls for the government to break up cartels, stabilize prices, impose export quotas, and protect small farmers from foreign imports challenge the global discourse assertion that states should facilitate rather than directly intervene in markets.
Opposition party MPs also expressed reservations about contract farming arrangements that made small farmers highly dependent on foreign investors and of government policies to attract multinational corporations. MP Magdelena Sakaaya (CUFS) spoke of rice farmers who had lost land to outside investors, which they then had to rent at a high price; of the inability of smallholders to access water, which was being diverted to the investor's larger farm; and of sugarcane farmers receiving low prices for crops deemed to be of low quality. “That is why I am discouraging the system of bringing in the big companies,” she stated. “Because they are not coming to help the smallholder. They are coming to get business” (author interview, May 13, 2014). MP Halima Mdee (CHADEMA) expressed confidence that smallholder farmers could dramatically increase production if provided proper support, including modern equipment, training, and irrigation support, and was also wary of foreign multinationals. “Monsanto is one of the companies that people or small-scale farmers have been complaining about all over the world,” he stated. “Yes, we need investors but what kind of investor are we welcoming? Are they worth it? … The challenge we are facing is investors who want to come [but] they don’t want partnerships. They want to create a master-and-servant relationship. We can’t allow that” (author interview, May 16, 2014). MP Rose Kamil Sukum (CHADEMA) criticized the government for failing to ensure that sugarcane farmers received fair prices for their crops, complaining that “big companies … decide on when and how much to buy the sugar cane from the farmers around the industrial area. Under such an exploitative system the one who suffers is the farmer. The opposition camp demands that the government find a solution to this domination in order to help the farmers” (United Republic of Tanzania, April 22, 2013, 11th meeting, 10th section, 125).
In sharp contrast to the global discourse, the agro-ecology vision elaborated in the subordinate discourse advocates state policies that promote biodiversity and enable farmers to avoid dependency on commercial inputs, and it is highly critical of policies that subsidize and promote industrial agriculture. “Government policy,” argues TABIO (2016b), “is focused on facilitating the seed companies to take over seed supply to the nation, with new laws and policies that support and protect the seed companies, while marginalizing small farmers’ seed-saving practice.” The core principles of agro-ecology, according to TABIO (2016a), include integrating crops, animals, fish, and farming communities to minimize the need for expensive external inputs and encouraging the use of local and often freely available resources so that “farmers are less dependent on access to capital or credit.” TABIO (2016a) is also opposed to GMOs, on the grounds that they “pose unnecessary risks to human health, destroy biodiversity, lead to increased costs for farmers, increase corporate control of the food chain, and fail to combat global hunger.”
The Alliance for Food Sovereignty in Africa (2014) makes a case for the need for public policies that reject the industrial agriculture model of the dominant discourse, and promotes an agro-ecological vision:
Industrial agriculture is a dead end. It claims to have raised yields in places but it has done so at great cost, with extensive soil damage, huge biodiversity loss and negative impacts on nutrition, food contamination and groundwater pollution. Agroecology is the antithesis of conventional, corporate-driven, monoculture-based farming systems. … Conventional agriculture seeks to simplify … eliminates biodiversity … de-skills farmers … offers one-size-fits-all fixes like GMOs, chemical fertilizers and pesticides … pollutes and degrades.
Whereas government policies to promote industrial agriculture emphasize the need to introduce tractors and power tillers, advocates for smallholder farmers suggest a more appropriate technology. “Smallholder farmers don’t need massive tractors, they need a bridging technology,” argued ANSAF leader Audax Rukonge (author interview, May 20, 2014). “For me, the emphasis would have been on the use of plows. We are the second-largest livestock-keeping country in Africa. Why can’t you use animals? … How do you give a tractor to someone who never has owned a bicycle?”
Critics of the dominant discourse argue in favor of farm-saved seeds and warn of the dangers of GM seeds. Whereas government officials who buy into the dominant discourse typically push hybrid or GM seeds, proponents of the alternative discourse emphasize seed saving and view GM seeds as part of unsustainable industrial agriculture model. This model, they argue, poses risks to health, destroys biodiversity, increases corporate control over the food chain, and fosters dependency on corporations. TABIO (2016b) has responded to proponents of the dominant discourse who portray opponents as anti-science by arguing that it is not opposed to the adoption of modern technologies, such as gene mapping and marker-assisted selection; however, it asserts, “The new Green Revolution proposed for Africa … appears destined to repeat the tragic record left by the fertilizer dependent miracle seeds in Latin America and Asia by increasing dependency on foreign inputs and patent-protected plant varieties which poor farmers cannot afford.” Elizabeth Mpofuthe, vice-chair of the Eastern and Southern Africa Small Farmers Forum, urged policymakers to “ensure that big private sector agrochemical companies like Monsanto, DuPont, Syngenta and others, do not colonize the seeds industry and stop farmers from saving and sharing because by doing so they will be infringing their patents on seeds” (Kiishweko 2013).
We have explored the claims made by three related social science literatures—world policy/neo-institutionalism, policy transfer, and policy mobilities—about how policy ideas move from one context to another. Using a case of policy discourses about agricultural development, we documented and compared the discourse circulating in the “global” realm (dominated, not surprisingly, by G8 actors) with the policy discourses circulating in Tanzania. The presence of not one but two domestic discourses of agricultural development in Tanzania, one which incorporates central elements of the global discourse and another which challenges many of its key claims, supports the view that policy discourses are not directly or unproblematically transferred from the global to the national. As the policy transfer and policy mobilities perspectives argue, their reception, revision, and rejection at the national level are heavily mediated by domestic politics––in our case, by a national political culture shaped by the historical legacies of a state socialist era that refused to treat land as a commodity and fostered values of equality and national self-reliance, and by the emergence in the 1990s of a growing civil society and of opposition political parties following liberal democratic political reforms that guaranteed freedom of association and ended single-party rule.
The claims of the subordinate discourse are rooted in norms of equality and national self-reliance that were central to Tanzania's socialist project and continue to resonate among the country's predominantly rural population. Despite the abandonment of socialist-era policies, notes Aili Tripp (1999: 48), “social equality continued to be a part of the political rhetoric and to a considerable extent a political culture of egalitarianism has taken hold as part of a national ethic.” Even ardent proponents of neoliberalism, such as former Finance Minister Edwin Mtei (2009: 4), embraced ideas that were (in his own words) “clearly egalitarian.” While distancing himself from “decades of futile doctrinaire socialist experimentation,” he warned of “the yawning gap between the new filthy, mega-rich, in league with avaricious foreigners masquerading as development partners, and the masses of the poor.” The importance of the norm of national self-reliance was asserted by “founding father” of the nation and central architect of the socialist experiment, former President Julius Nyerere (2000: 21), who, in response to the sale of the nation's public assets to foreigners, noted: “Throw away all our ideas about socialism. … Embrace capitalism, fine! But you have to be self-reliant.”
These domestic norms helped shape policy debates over issues of land-grabbing, foreign investments, and the protection of smallholder farmers from the predatory pricing of oligopolies, strengthening the appeal of the subordinate discourse. The influence of the subordinate discourse coalition on public policy did not stem from numerical strength in either the legislative or executive branches of government but from distinctive national characteristics, including rules within parliament that enabled opposition party leaders to chair important oversight committees investigating the executive branch and allowed individual MPs to introduce private motions, rather than simply debate motions proposed by the government. Opposition party leaders used their positions within the legislature, whose meetings were, until recently, publicly televised and widely viewed, to denounce government policies that failed to protect small-scale farmers from dispossession by foreign multinational corporations.
The fact that both the dominant and subordinate domestic discourses of agricultural development mobilize high-level state actors as part of their discourse coalition underscores the point that the state is not a unified actor. WPN scholars often marginalize national political contention by equating government technocrats with the state, thus failing to explore the role of other actors, such as elected legislators, and their differential links to foreign actors and discourses. They tend to treat the state as a unitary and cohesive actor represented by the voices of the executive branch and appointed officials, rather than as a contested terrain. Our identification of two domestic discourse coalitions of agricultural development reveals a division within the state between dissident elected legislative officials and high-ranking civil servants. The need for elected officials, including the president, to mobilize voter support and secure legitimacy among a majority of the predominantly rural electorate helps account for divisions within the ruling party and the state. It also helps account for the ability of the opposition to force the government to reverse or moderate its positions on land and foreign agricultural investments and to not push harder to introduce GMOs.
A domestic discourse that was quite similar to the global discourse did achieve dominance in Tanzania in the post-2000 period, however. This dominance resulted in part from the huge resources that donors provided to their allies in the executive branch and in part from the relative weakness of civil society organizations and of opposition political parties, both of which were prohibited by a single-party authoritarian state until the final decade of the twentieth century. Another contributing factor was the presence of an emergent Tanzanian bourgeoisie that is eager to forge partnerships with foreign multinationals and shares many of the central precepts of the global discourse on agricultural development. An important reason that this bourgeoisie shared these precepts is not because it simply absorbed these ideas from outside, but rather because it viewed many of them (e.g., reliance on the private sector, public–private partnerships) as being in its own interests. It thus advanced them independently, and as an organized group, to government ministries.
Despite their many similarities, there are important ways in which this dominant domestic discourse also diverges from the global discourse. The most significant has to do with the centrality of small-scale farmers to the storyline of how to address the shortcomings of Tanzania's agricultural sector. While a focus on improving the productivity of smallholder African farmers lies at the heart of the global discourse, the dominant discourse in Tanzania, particularly in recent years, has moved toward envisioning larger-scale farmers and agro-industrial enterprises as the most promising route to agricultural development. Smallholder farmers are not absent from this discourse, but in policy visions like Big Results Now they clearly lack the central role attributed to them in the global discourse. We have argued that this difference is in large part a product of the emergence of a national bourgeoisie in search of foreign investment partners and of the ideology of modernity that has long informed high-level, and in many cases foreign-educated, state administrators, who view Western science and technology as part of a broader ideology of development and modernity.11 Senior civil servants of the twenty-first century continue to regard themselves as a development vanguard, embracing Western science and technology as the solution to their country's agricultural problems. It is this that helps explain their enthusiasm for large-scale agriculture and agro-industry and why they see it as the way forward.
Our final point has to do with the fuzzy boundary between the global and the local and the problem of thinking in terms of “external” or “internal” factors shaping discourses. In post-conditionality states like Tanzania, policy discourses are not imported from the outside but are simultaneously inside and outside. Notably, foreign actors played an important role in the elaboration and dissemination of both the dominant and the subordinate discourses circulating in Tanzania. This is evident in the role played by foreign donors and foundations in pushing the government to incorporate the tenets of the global discourse into national policymaking. It is also evident in the role played by foreign civil society organizations in shaping certain elements of the subordinate national discourse. National actors were the central players in elaborating elements of the subordinate discourse concerning land rights, foreign investments, and the need for government intervention against predatory pricing by market oligopolies. At the same time, transnational (and predominantly European) actors played a central role in formulating the anti-GMO and agro-ecology elements of the subordinate domestic discourse. The blurring of the domestic and the global, highlighted in the policy transfer literature, is evident in the policymaking imprint of globalizing state bureaucrats and localized global actors. However, as the policy mobilities literature argues, national political actors and circumstances play a critical role in the global movement and reception of policy discourses.