ARTICLE Re-Forming the Post-Apartheid State? Citizenship and Rural Development in Contemporary South Africa William Munro Introduction In as much as they involve the promise of social transformation and the consolidation of new socio-political regimes, political transitions are messy and contested processes. As the electoral resurgence of communists in eastern Europe and Russia demonstrates, incoming governments confront the formidable political challenges of rapidly re-asserting and re-orienting the political authority of the state to signal the distinction between the new order and the ancien regime, while also taking account of the entrenched legacies of pre-existing institutions. Transitional states tend to be weakly rooted in society. Thus, under often volatile political conditions, governments must entrench state apparatuses that are able to manage society, constrain social conflicts, direct or promote economic development, and generate a degree of sustainable social allegiance to a 'new' socio-political order that is measurably different from the 'old' draconian system. For post-transition political leaders, these tasks require a 'hegemonic project' through which they not only secure the incumbency of the new government, but also legitimate the state as the over-riding locus of political authority in society.1 Such a hegemonic project underpins what one might call the politics of citizenship and state construction, which focus on establishing the Rules of the Game that regulate state-society relations in the post-transition era. Such 'rules' typically include appeals to the normative postulates of bourgeois democracy: an impartial state apparatus, mechanisms of accountability, juridical and legal structures that underwrite citizenship rights, and inclusive institutions of political competition. They also specify the access to civic resources that people get by virtue of being citizens, and thereby define structures of social authority, political allegiance and the provision of public goods. In this sense, they provide an institutional framework for what one might, paraphrasing EP Thompson, call a 'moral economy of the state', according to which popular cultural conceptions of need and expectation (both between citizens and between citizens and the state) are modelled along the lines of an inclusive national community. In periods TRANSFORMATION 30 (1996) ARTICLE MUNRO of transition, these 'rules' are fluid, and post-transition governments are under tremendous pressure to rapidly design and construct effective institutions of development and governance that can secure the over-riding social authority of the state. The aim of this paper is to examine some ramifications of this political challenge as it confronts the post-apartheid state in South Africa. The paper sets out to show, firstly, that a post-transition reconstruction of social authority is a development problem; that is to say, one cannot think about appropriate development strategies separately from hegemonic strategies, and vice versa. The paper then applies this argument to South Africa, drawing on occasional examples from KwaZulu-Natal, and suggests that the importance of rural development strategies in state construction is generally underestimated in development thinking. The politics of citizenship can be conceptualised in terms of different types of rights that accrue to citizens as members of a political community. These include civic rights, which are rights of access to the law; political rights, which are rights of participation; and social rights, which are rights of basic social provision - the 'moral economy' rights of expectation and need. In.the context of a post-transition hegemonic project, the extension of political rights does not present a problem. But in post-colonial agrarian settings, the other kinds of rights do. In this light, the paper examines the South African approach to the relationship between development and hegemony as it is addressed in the general framework of the Reconstruction and Development Programme (RDP), and highlights three sources of tension in the RDP as a vehicle for state construction and hegemony. One is the problem of 'urban bias' in development thinking, which relegates rural communities to the fringes of macro-development strategies. This problem generates a tension between the (reconstruction of inclusive national citizenship and the macro-accumulation strategy. The second issue is the design of appropriate institutions for constructing rural hegemony. These institutions must both extend state power in rural communities and incorporate those communities politically. Consequently, they rest on a persistent tension between development as empowerment of the state and development as empowerment of the people. The third issue is the political difficulty of inculcating expensive social rights and defining the state's development responsibilities in a context of scarce resources. Finally, the paper argues that, in the light of these constraints, a greater rather than a smaller state presence in rural communities is required. TRANSFORMATION 30 (1996) MUNRO ART1CLE Hegemony, Development and the South African State On the ground, the political dynamics of constructing hegemony are complex, and depend both upon the structures of political authority (or domination) and upon the economic class structure. On the one hand, the hegemonic projects that states pursue aim to contain and suppress social conflicts, including class conflicts at the level of the social formation as well as localised conflicts over the quality of life - distribution of food, jobs, etc. Thus, they aim to build a 'national consensus' around a common sense of citizenship; On the other hand, such projects generate conflicts and negotiations between 'stakeholders' (state agents, parties, dominant classes, community associations, citizens) over the control of social life and over what state interventions in social life are to be socially understood as legitimate and appropriate. A core consideration in these conflicts is how revenues are to be extracted and how the common weal is to be redistributed across social groups. At the level of policy discourses, development plans and political institutions, these dynamics centre on several politically-charged development questions: how is economic development to be promoted and managed? how is social security to be secured and distributed? how is the supply of services to be arranged? what is to count as a public good? The answers to these questions define the location of the state in the social order, as well as the quality of citizenship for local populations.2 These challenges are particularly acute in Third World agrarian contexts where marginalised and frequently oppressed rural populations live on the fringes of macro-development strategies.3 African economies, where the authority of the state has never reached very deeply into rural communities, provide the most obvious example. But the persistence of agrarian social movements in countries with much stronger industrial bases, such as Mexico and India, vividly demonstrates the profound importance of constructing state hegemony among rural populations, at least to the extent that social stability is desirable. The demands of agrarian social movements are generally both about their access to material resources and about their citizenship status within the polity - in short, about both development and the nation-state. The point was eloquently made by a Zapatista leader in Chiapas: Why does the government [refuse to put] national politics on the agenda for negotiation? Are the indigenous people of Chiapas 'Mexicans' only for the purpose of being exploited? Do they have no right to speak out on national politics? Does the nation claim Chiapas's petroleum, its raw materials, its labor - in effect all of Chiapas's life blood - except Chiaponecans' opinion regarding the future? What sort of citizens does the government take indigenous TRANSFORMATION 30 (1996) M U N R O ARTICLE Chiapanecans to be? Are they 'citizens in formation?' Does the government still treat them as little children, as 'adults in formation?'4 This statement emphasises that nationality is a serious conduit of the relationship between a state and its citizens; states must negotiate the contours of citizenship with their populations. It also implies that in contexts where the presence of the state in rural society is weak or limited, such negotiations are channeled both through local institutions of governance and through local structures of service delivery. In general, however, the main interest of Third World governments towards marginalised rural populations is to keep them socially stable and politically quiescent at the lowest possible cost. To do so, they can (and often do) deploy the coercive apparatuses of the state. But this is a costly and often risky strategy that few post-colonial governments have the political will or capacity to carry out systematically.5 More generally, governments recognise the importance of containing social struggle and resistance within the ambit of state authority, and they deploy more hegemonic social management strategies to stabilise rural populations and to overcome centrifugal political forces. Rural development policies of managed land reform, populist agrarian strategies that promote self-sufficient peasant producers, and 'community development' or 'basic needs' policies that promote 'self-reliance' among rural communities, play a central role in such stabilising strategies. As Doug Porter (1995:64) puts it, 'throughout the post-war period, diverse metaphors have been introduced which promote the impression of radical change without threatening the basic project of controlled and orderly manipulation of change'. In the context of developing states, therefore, rural development is never only a set of objectives or processes for promoting economic growth or improving the livelihoods of citizens. It is also a political process whereby states set out to manage state-society relations. A growing body of literature has argued that the role of development plans and discourses in facilitating these political tasks is to de-politicise public policies by casting them in a technocratic, politically 'neutral' language of development which privileges ideas of technical expertise and economic efficiency, and which promotes a generalised conception of social improvement (Ferguson, 1992; Escobar, 1995; Chatterjee, 1994; Crush, 1995). In effect, 'development' policies and plans provide a non-political discourse through which post-transition governments set out to legitimate their efforts to re-structure politics. In doing so, they also try to specify the parameters of the 'new' (post-transition) state's legitimate and appropriate role in shaping the social order. In part, such strategies aim to secure the state as the final arbiter of TRANSFORMATION 30 (1996) MUNRO ARTICLE social and political authority by establishing state control over the distribution of public goods (thereby undermining the suasive capacities of alternative authority sources), and also by controlling the social definition of what is to count as a public good (thereby delimiting social expectations of state responsibilities and material outlays for social welfare). Thus they place 'development' at the epicentre of post- transition state construction. In short, they forge the intimate links between the rural development policies and the hegemonic projects of insecure states. It is clear that these general conditions also prevail in contemporary South Africa, and it is no surprise that the post-transition government has cast its objectives to re-shape the socio-political order in the language of development.6 In the wake of the first democratic elections, the ANC-dominated government confronts the massive task of re-linking the old bantustans and their impoverished populations with the central state and the national political economy by increasing both the quality of economic opportunity and the quality of citizenship for rural people. At a minimum, this requires a new impetus for rural development and service delivery, as well as new institutions for local governance. State-sponsored rural development agencies lack legitimacy and capacity, especially where apartheid administrative structures are strongly entrenched (Macintosh, 1990; May, 1993:23-24; Kotze et al, 1987; Spiegel, 1991). Local government agencies are underfunded, uncoordinated and inefficient, creating space for fluid local politics which invites clientelism and deeply complicates government approaches to state construction. In this respect, South Africa faces the same structural development challenges as other post-colonial regimes (Munro, 1996). But this also means that post-transition state construction demands a drastic overhaul of the organising principles of the South African state. Under apartheid the bantustans performed the function of controlling a massive marginalised population. David Kaplan (1980) has characterised the form of state under which this was achieved as a racially exclusive bourgeois democracy. This state form, driven by the demands of capitalist accumulation at the economic core, was defined by the inclusion of all whites within the state through universal democratic representation, and the exclusion of all blacks through the strengthening of tribal juridical, political and ideological structures in the bantustans. On Kaplan's account, the sustained stability of this state form was made possible by the ability of tribal control to sustain both non-capitalist production relations and a continuing attachment of Africans to individual plots of land in the reserves. This power structure prevented the eruption of localised social conflicts associated with rural class formation as well as the emergence of TRANSFORMATION 30 (1996) ARTICLE MUNRO anti-state agrarian social movements. Mahmood Mamdani (1996) has elaborated this idea into a more general model of a 'bifurcated state' that rests on two distinct power principles: a principle of civil power grounded in a legitimating language of rights, and a principle of customary power grounded in a legitimating language of tradition and community. The first principle is institutionally sustained by the tenets of statutory law, and is compatible with market relations; the latter is sustained by customary law and regulates non-market relations in land, family affairs and community affairs. In Mamdani's view, this state form, exemplified in South Africa, is the quintessential form of organisation of African states more generally. It rests on a form of 'decentralised despotism' which separates the urban from the rural as well as one ethnic group from another. On Mamdani's account, the most important imperative and the most intractable difficulty of post-colonial African regimes is to dismantle and replace this overarching power structure by 'detribalising' social power. It is one that the post-transition government in South Africa faces most starkly today. Taken together, these arguments represent power structures in the bantustans as driven by centralising logics of accumulation and domination. These structures were designed to feed those logics while maintaining the quiescence of rural populations by defining in stark institutional - and to some degree ideological - terms, the structures of allegiance and the meaning of citizenship for different sectors of the population. One of the central implications of Kaplan's and Mamdani's arguments is that, under these different forms of social authority, different populations experience different structural and ideological relationships with the state, and therefore have different expectations of citizenship. Nevertheless, the 'bifurcation' of the state was never thorough-going. Migrant workers moved to and fro between these different realms and used the resources generated in one realm to negotiate constraints in the other. Bantustan governments had one foot in the camp of traditional power andthe other in the camp of modern political institutions, with the result that local bureaucrats and local traditional authorities were sometimes at loggerheads. In some instances, rural citizens could seek legal redress in either the magistrate's court or the traditional court, and sometimes play one off against the other. The poverty of resources available to bantustan governments ensured that the central state intervened repeatedly, though inconsistently and unpredictably, in the social lives of rural citizens. State-sponsored rural development initiatives, such as the 'betterment' programmes, provided important channels through which the central state tried to manage rural social relations at the local level without disrupting bifurcated structures of social authority. TRANSFORMATION 30 (1996) MUNRO ARTICLE Thus this state form depended on a radical splintering of accountability between legitimating principles of rights, traditions, and technocratic rationality. For rural citizens, it represented a kind of 'shell game' state in which responsibility for social conditions was constantly dis-placed in order to sustain authoritarian control. For them, therefore, neither form of social authority was reliable, and both tended to undermine local moral economies (see, for instance, Spiegel, 1991). The relationship between the public, private and community realms of social life remained indeterminate and uncertain even under the 'decentralised despotism' of tribal authorities. The defining feature of citizenship for rural people was not so much centralised or decentralised despotism but generalised uncertainty. It is some measure of the success of this state form that the social movement that finally negotiated the end of apartheid does not have strong rural roots. But the current political transition does of course have serious ramifications for this power structure. The transition has not significantly shifted the national economic structure to which, on Kaplan's account, the bifurcated state form was functional. But it has thrown into disarray all the mechanisms that sustained the definitions of citizenship on which that state form rested. To the extent that those mechanisms maintained social stability by precluding anti-state rural social movements, the new government (like other Third World governments) has an interest in replacing them. It might adopt a new authoritarian strategy of reconstructing those mechanisms, in the way that other African states have. But this is an uninspiring model. It is likely to be economically undesirable, difficult to sustain, and, as Mamdani (1996:293) warns, may undermine the 'urban civil power' of the state. Moreover, the national democratic struggle has placed tremendous popular pressures on the state, both for political inclusion and for material provision. In short, the post-transition government faces urgent structural and political imperatives to create a new state hegemony, not only among urban populations where the institutional and procedural elements of a democratic order are already in place, but more particularly among rural populations where the tenets of inclusive citizenship are very shallowly rooted. The challenge of post-apartheid state construction, therefore, is to secure institutions that can expunge the generalised uncertainty that has characterised the quality of citizenship for rural citizens, without resorting to contradictory strategies of social control that aim to maintain quiescent rural populations on the fringes of the national polity. Thus, the state politics of political transition in South Africa place a premium not only on renewed development initiatives, but also on re-ordering the structures of social power in order to secure the overarching social authority of the state. Elsewhere, I have suggested that such TRANSFORMATION 30 (1996) ARTICLE MUNRO a project involves re-moralising political forms of local authority so as to place the state's institutions at the centre of the community - creating, in a sense, an 'imagined community' in which the universality of the state can be secured and it becomes part of the common-sense reality of village life (Munro, 1996). Since rural populations are largely poor, and social security is a central element in the make-up of political communities, the reconstruction of citizenship - the forging of a 'new social partnership' - must take account of local or domestic institutions that are designed to mitigate risk and insecurity. This requires not only that differences in lines of governance be expunged, but that structures and channels for the provision of public goods be clearly established. But once one recognises that in the context of political transition and a weak, skewed economy, development and hegemony cannot be thought about separately, the difficulties of state construction loom large. In South Africa, as in other parts of post-colonial Africa, the history of sustained (though also transformed) 'tribalism' has confounded these imperatives. Indeed, this history suggests, against the current of development theory and transition politics, that the key to consolidating the post-apartheid state lies in the countryside. The complexity of political competition in the transition, especially as highlighted in the recent local government elections, make this an extraordinarily difficult task.7 It is through this historical and sociological prism that one might best consider the post-apartheid government's general framework for development as it has been articulated in the RDP. State Construction, Rural Development and the RDP The RDP draws upon an implicit recognition that state construction, nation- building and development are inextricably linked processes. It lays stress on a 'people-driven process' to forge a new concept and quality of national citizenship, understood as equal membership in a new national political community (GNU, 1994:6-7). It sets out to establish a 'new social compact' or "new social partnership' which will break down the adversarial relationship between state and society that developed under apartheid, will eradicate perceptions of the state's overweening role in society, and will strengthen the legitimacy of state institutions. A central objective of the RDP is to 'build the capacity* of community-based organisations to fill the void left by a retreating state. The thrust of this approach is to bring governance closer to the people, mainly by expanding the role of local authorities in delivering basic services and spreading the responsibility for the distribution of public resources across community-based organisations (CBOs), non-government organisations (NGOs), as well as state institutions. Behind it all is a concept of 'empowerment,' TRANSFORMATION 30 (1996) MUNRO ARTICLE by which is understood the increased capacity of citizens to take control of their own destiny.8 As a general framework for development, the RDP is an innovative sociological initiative. It represents the fullest working out, at the level of policy and state-making, of the 1980s conventional wisdom that development is best pursued by unfettered markets, an active civil society and a small, non-interventionist and facilitatory state.9 On this account, development proceeds best by expanding the participation of the citizenry in planning and managing the structures that affect their lives by strengthening the institutions of civil society. Development theorists stress the importance of 'development from below' if effective and stable development institutions are to be secured at the local level, and the quality of rural life improved. This development approach places a premium on developmental collective action by citizens at the level of local 'communities', and therefore on the design of local institutions that will facilitate such collective action. But the initiative also invokes serious sociological tensions that must be recognised if it is to be a model for development, especially rural development. Firstly, there is the danger that a 'people-driven' approach to re-ordering relations between state and society will be overwhelmed by a market-driven approach to growth and development. Many of the strongest institutions in civil society are market-related. It is not sufficient to simply roll back the state, because the promotion of economic growth and the reconstruction of social authority are both essential, but distinct, post-transition tasks. Markets cannot underwrite hegemony in weak, externally oriented economies with large poor rural populations. As Philip Raikes (1988:79, 84) has noted, markets distribute asymmetrically because they 'push goods towards money not need'. Among agrarian populations, especially in varied or fragile ecologies which are common in Africa, market-driven rural development tends to be highly inequitable, both according to region and according to class or gender. It frequently produces conditions in which significant agrarian wealth exists alongside chronic hunger. Such conditions, which drive poor communities back onto their own resources or into localised survival networks, undermine the construction of state hegemony because they undermine the capacity of the state to underwrite social rights or to become the final arbiter of social provision. They necessitate a more socially active state in promoting development initiatives where the distribution of public goods is at stake. Secondly, there is a tension between development as a political process for entrenching and legitimating a new political order, and a participatory 'people-driven' process of development 'from below'. One of the main TRANSFORMATION 30 (1996) ARTICLE MUNRO objectives of participatory people-driven development is to strenthen the resources of citizens against the managerial tendencies and extractive capabilities of the state, which theorists increasingly interpret as driven by essentially authoritarian impulses (for instance Porter, 1995). This approach assumes that the state's managerial tendencies make it ineffective, untrustworthy or inimical to sustainable development initiatives. But this is a questionable assumption, not least because there is no compelling reason to believe that institutions in civil society have a more coherent commitment to development, or to the redistribution of resources, than state agencies. In government rhetoric, demands for development and resources should be voiced by local 'communities', a term which implies a commonality of interests and allegiances. In KwaZulu-Natal, many rural communities are highly politicised, and indeed defined by their allegiance to a particular party or to a particular traditional chief, which are themselves often closely identified. Insofar as a commonality of local identity and allegiance rests upon the social authority of 'customary power', it is precisely the nature of community identities that is at stake in restructuring the organising principles of the state. In addition, communities on the ground, especially in agrarian contexts, comprise localised webs of social relations that are subject to, and sometimes rest upon, severe conflicts. Local organisations often emerge at the level of the community, but are group-based rather than community-based (Uphoff, 1993). Thus the key category of development from below cannot be taken at face value although the politics of partnership rests upon the establishment of effective community institutions. The new social partnership, therefore, requires development institutions which can not only engage the state in aquest for resources, but which can also underwrite the 'community' itself. In addition, the context of political transition has an important effect on the definition of a 'new social partnership'. During political transitions, as we have seen, neither the state nor societal interests are clearly dominant; the entrenchment and legitimation of any new political order necessitates the strengthening of a new state form. In effect, the empowerment of citizens - both as individuals and as communities - in a post-transition context is intimately tied to the empowerment of the state. The development language of 'social partnership' seems quite appropriate in this situation. But it cannot be allowed to sweep aside a concept of 'social contract' according to which citizens hold the state to account for meeting the particular social needs and expectations that underwrite an inclusive (national) conception of citizenship. In these circumstances, current development theory notwithstanding, the driving concern of development institution-building should not be to reduce the state and inhibit 10 TRANSFORMATION 30 (1996) MUNRO ARTICLE its managerial tendencies but to secure and institutionalise its distributive capabilities. The post-transition project of re-orienting the political order according to more inclusive principles generates complex tensions between market, state and civil society. Most broadly, the central conundrum of rural state construction under the RDP is to put in place a set of local development institutions which can resolve the tensions between the demands of state hegemony, social management and rural empowerment within the national political economy. The success of the RDP depends importantly on the extent to which the state is able to do so. But it faces serious structural constraints. Urban Bias and Rural Development as a Residual Category The defining logic of development theory over the past half-century has been its pre-occupation with industry-led economic growth within national economies. Its yardstick has been the expansion of a domestic manufacturing sector through the transfer of human and fixed capital into industrial production. In this sense, dominant conceptions of development have traditionally been evolutionist, and driven by an urbanising teleology.10 The effect of this bias is that rural development is a residual development category that deals principally with populations which are at the fringes of national and international political economies, and tend to be politically and economically marginal. Macro-accumulation strategies predicated on such an urban bias sharply delimit state commitments to rural populations, and rural development aims mainly to stabilise these populations. In many developing countries, development planners have sought consciously to extract surpluses from agriculture in order to build up industry, often through the manipulation of exchange rates and food prices. Such strategies are premissed on the existence of a viable small-scale rural production sector. In apartheid South Africa, however, development strategies actively inhibited the consolidation of such a sector; extraction from the countryside focussed principally on the extraction of cheap labour through forms of primitive accumulation based on race. Overcoming this legacy presents the post-transition government with complex and inter-related problems of structuring redistribution, managing population mobility and 'de-tribalising* social authority in the countryside. At the same time, the South African government's macro-economic strategy which was released in June 1996 reflects the overwhelming urban bias in conventional development theory. Built around the dramatic need for labour absorption and for industry-led growth in the national TRANSFORMATION 30 (1996) 11 M U N R O AR-HCLE economy, it is remarkably quiet about the role of rural citizens in national development. The nature of transition politics, including the urban support bases of the main political parties, the importance of the labour movement, as well as the industrial core of the economy, created pressures towards what Mike Morris (1993; 1991) has called a '50 percent solution' - a growth-oriented macro-development strategy in which the boundaries of inclusion and exclusion in the social order (broadly determining access to a range of social benefits such as health care, education, unemployment insurance, infrastructural services, housing, etc) shift from being racially defined to being defined by class, and national redistribution through state action is constrained. In this view, there is a danger that South African society will not be transformed into a new national community but will be re-cast into a class-based 'Two Nation' society. This danger seems to be enhanced by COSATU's predilection for a high-wage, high-skill industrialisation strategy (MERG, 1993; Nattrass, 1994). It is difficult, today, to imagine a macro-development strategy that is not growth-oriented, market-led and based on the manufacturing sector. But Morris' formulation usefully highlights the dual character of the development problem. On the one hand, it is a problem of redistribution, not only of material resources but also of the civic benefits that accrue to people by virtue of citizenship, such as education, health provision, and some form of social security. This problem thus arises out of the structures and strategies of national accumulation. On the other hand, it is a problem of hegemony, ie a problem of re-ordering state-society relations and restructuring social authority. Morris' account shows that there is a deep-seated political tension between a political hegemonic project which invokes an inclusive unitary national community, and a macro-accumulation strategy that incorporates economically 'those blacks at the upper end of the class ladder' (including organised labour) and leaves 'the rest of the black population living on the periphery of the mass consumption economy' to fend for themselves (1991:57). Any national development strategy needs to reconcile these two dimensions of state-building. The urban bias of development strategies shows clearly that the problem of 'Two Nations' goes beyond class. Given the structuring of citizenship and social authority under the 'bifurcated' apartheid state, it has a profound spatial dimension that engages the rural-urban split. Neither development theory nor development politics have ever addressed this dimension of development adequately. The broadest objective of rural development initiatives is to improve the quality of life of rural populations, either by alleviating poverty or by promoting economic growth in rural communities. In general, proponents of 12 TRANSFORMATION 30 (1996) MUNRO ARTICLE rural development are concerned with the question of what rural economic organisation should look like in conditions where one or more of the factors of production - land, labour, capital and expertise/knowledge - are in short supply. Since the communities whose welfare they seek to advance are generally agrarian communities, the core issue is agricultural productivity. In South Africa, this issue is enormously complicated because there is no clearly identifiable and commercially viable small-scale agricultural sector that can be readily incorporated into national or international markets through pricing, marketing or input policies." Furthermore, the social and income structure of rural households is so complicated (and sketchily understood) that clear distinctions between the urban and rural sectors, as well as between household and inter-household relationships, and between agricultural and non-agricultural strategies, are often unsustainable. Under these conditions, it is extremely difficult to define the relationship between rural poverty alleviation and economic growth strategies at the macro-level. A more locally determined, piecemeal approach to rural economic strategies is likely to be more fruitful. The government's macro-economic development strategy does not provide a framework for such an approach; it does not, in fact, address the issue of small-scale rural production at all. As a legacy of the apartheid economic structure, rural markets are poorly developed. While a recent poverty study by Data Research Africa contends that an income transfer of R9.6 bn per annum to rural people would be necessary to secure all with a subsistence income, the macro-development strategy offers no overarching theoretical, conceptual or institutional machinery to do so. Even the role of poverty reduction is not systematically addressed. The macro-economic strategy relies upon land reform to address the problem of rural poverty, and 'emergent farmers' to address the problem of economic growth. The Rural Development Strategy (GNU, 1995) calls for 'entrepreneurialism, employment and empowerment* but offers no development strategy, agrarian or otherwise, to provide a framework; the relationship between distribution and production in the rural areas remains unclear. In rather sweeping terms, these issues are assigned to the RDP. In effect, the macro-economic strategy follows a modernist industrial paradigm in which the conditions of socio-economic citizenship for marginalised rural populations - which are social justice issues - are tacked on. In one important sense, the relationship between rural development and inclusive citizenship rests on the question of how rural populations can hold the state to account. The RDP aims specifically to tackle the tensions between macro-accumulation strategy and hegemonic imperative by presenting development strategies as the outcome of consultative deliberation. The TRANSFORMATION 30 (1996) 13 M U N R O ARTICLE government set up the National Economic, Development and Labour Council as a corporatist mechanism for consultation, coordination and negotiation by key stakeholders, including labour, business, civics and the government. But this corporatist structure is overwhelmingly urban. Organised interest groups are much weaker in the countryside. There is no organised sectoral representation for small-scale agrarian producers at the national level. The government has insisted that it will not sell short rural communities where capacity is poor, or where local institutions are not integrated into local government structures. It requires government ministries to match their urban initiatives with rural initiatives. But in the absence of a development framework that clearly addresses rural populations, as well as the emphasis in the macro-economic strategy on job creation, it remains critically unclear how 'rural development" as a planning category fits into 'national development'. Certainly, proponents of rural development who demand the transfer of resources from the national state or from urban constituencies are not in a strong bargaining position. At the level of national planning, the government seems content to leave rural initiatives to weak on-the-ground structures, either state or private. It seems likely that the structural role of the bantustans within the national political economy will not change significantly in the medium-term: they will remain 'holding areas' for populations on the fringes of the formal economy. In effect, a racially exclusive bourgeois democracy runs the risk of becoming a spatially defined bourgeois democracy. This spatial constraint on national development is perhaps intractable at the level of macro-economic strategy. But it is potentially undermining at the level of nation-building. To the extent that the national political economy inhibits the transfer ofresourcesto the countryside, it constrains the resolution of the national question which engages not only the centralising logics of accumulation and domination but also the structures of social authority. The dangers for state construction of a spatially structured 50 percent solution are well illuminated in contemporary Mexico: the realm of 'urban civil power' is characterised by extensive and deep-seated corruption and several rural states are in incipient revolt. It is a model to avoid. If 'urban bias' in the national development strategy delineates the limits of state commitment to rural populations, state construction nevertheless requires that state authority be secured in the countryside. If there is one thing that post-colonial history elsewhere in the Third World tells us about state construction, it is that a 50 percent solution is not a good solution and may be no solution at all. Thus, it is a central imperative of post-transition development not only to maintain rural stability but also to construct rural !4 TRANSFORMATION 30 (1996) MUNRO ARTICLE hegemony by re-orienting the organising principles of social power in the countryside. Rural Institutions and the Construction of Hegemony The construction of state hegemony in the countryside has two institutional components. One component is the ensemble of local governance structures that the government puts in place. It is through these institutions that political participation and competition are organised at the primary level, and it is through them that the provision of public goods is channeled. The other component is the legal regime that the government puts in place. It is through its legal and juridical institutions that the state sustains public authority and exercises the power of sanction. In the context of political transition, and of a 'bifurcated' state, both these dimensions are highly contested. In the first case, local government structures are subject to intense inter-party conflicts. Thus, although they are the core conduit of state-community relations, they blur the distinction between state and party, and are susceptible to the entrenchment of local clientelist networks which can undermine the authority of state-based institutions. As a result, they are often politically volatile and unstable.12 In the second case, different forms of social authority, often sustained by competing tenets of statutory and customary law, are at loggerheads. As we have noted, such competitions weaken the legitimacy of all contenders, tend to promote local authoritarianism, and inject greater levels of uncertainty into the livelihoods of rural citizens. The construction of state hegemony, which involves both a revision of the organising principles of the state and a re-orientation of local concepts of citizenship, requires that both these components of state construction be de-politicised. In rural South Africa, as in other countries where the presence of the state in the countryside is weak or erratic, this is a difficult task. Two broad models exist. One is the one-party state model which once enjoyed vogue in post-colonial Africa. On this model, the party is brought within the state and the distinction between state power and party power expunged. However, this approach tends to destroy nationalist coalitions, push popular classes out of the polity, and promote ethnic or regional resistance movements. The other approach stresses the distinction between state and party, and stresses either the state's technocratic efficiency or its accountability and transparency. The latter model is clearly preferable for the construction of state hegemony, but in the heightened political atmosphere of transition it is very difficult to pursue, especially where party lines follow urban-rural splits closely, as they do in KwaZulu-Natal. TRANSFORMATION 30 (1996) 15 ARTICLE MUNRO In agrarian contexts, the control of land and the manipulation of property regimes generally plays a central role in the attempts by governments to put the stamp of state authority on rural society. The main reason for this is that land is the linchpin of social relations in agrarian societies. It establishes the crucial link between relations of production and relations of social production. For centralising states, it is important to codify an effective land management system in accordance with state-based institutions of law and sanction.13 Some scholars view adjustment of property regimes as the key to managing social change and promoting modernising development; as Robert Bates (1989:28) puts it, 'to alter property rights is to redefine social relationships'. In this sense, the 'land question' and the 'national question' are intimately connected. But, for the same reason, control of land is a particularly delicate and potentially volatile focus for state-society relations. Adjusting property rights is by no means a straightforward political process. The state's hegemonic efforts may conflict with local contingencies arising out of social change and/or political transition. On the ground, property regimes tend to be flexible and dynamic - both the subject and the source of intense political conflict.14 Comparative research has shown that, in the processes of social change associated with colonial and post-colonial development, property regimes and the forms of social authority linked with them have frequently fragmented rather than transformed (see for instance McKenzie, 1994; Munro, 1995; Shipton, 1988). Such processes have complicated the hegemonic projects of post-transition governments: while the land question is a central part of state construction, it is also a source of potential conflict in restructuring social authority. How, then, can a post-transition government re-orient property rights, and other social rights linked to them, without undermining its own objectives? This powerful dilemma lies at the heart of both land reform and state construction initiatives in post-transition South Africa.1' The government's land policy addresses the problem directly by proposing a 'rights-based' approach to tenure reform (GNU, 1996:43). It recognises the flexibility of property regimes and argues that group rights and individual rights are not at odds. Thus, it sets out to extend security of tenure under diverse forms of tenure, and to 'bring the law in line with the actual practices and realities which exist on the ground' (GNU, 1996:43-44). People must be able to select their own form of tenure with the state assisting communities 'to develop various models of communal or group tenure, to allocate land rights among community members, and to manage common property resources ... in ways that reflect local preferences' (GNU, 1996:45-46). 'These arguments focus attention sharply on the rural community. They shift the notion of real rights to property towards a notion of rights in the 16 TRANSFORMATION 30 (1996) MUNRO ARTICLE community, a notion captured in the idea of a Community Ownership Trust (Claassens, 1994). Thus, the new conundrum of property is to institutionalise flexibility at the level of the 'community'. In the current development vision, this objective can be achieved through the 'new social partnership' which is to be operationalised and cemented by the RDP. It is an innovative, but extraordinarily challenging, approach, for it rests on a double move. On the one hand, it appeals to existing (land-based) community identities to secure a new property regime. On the other hand, it seeks to re-orient community identities away from the social authority of tradition towards the social authority of the state. This task confronts several powerful constraints. In the first place, the resources and capacity of the government for managing land reform and land development are very limited. Secondly, there is considerable fluidity in lines of authority over land. The key institutional role of traditional local authorities promises to make the institutionalisation of an new 'social partnership' a difficult and messy process in which the overriding authority of the state is by no means central.17 Thirdly, the parameters and resilience of community identities cannot be taken for granted, nor can their propensity to underwrite a new state-based regime of social and legal rights. In short, the land question highlights, but does not necessarily resolve, the tension between social control and social incorporation that bedevils any effort at state construction and nation-building. It does so, paradoxically, by bringing into relief the importance of defining 'communities' if the conceptual distance between 'national development' and 'rural development' is to be narrowed. It seems apparent, on the face of it, that the structures of local government should provide the natural conduit for 'people-driven' development based on state-community alliances. Local government institutions can underwrite local communities and provide the formal structures for popular participation. To be sure, local government is regarded as the 'hands and feet of the RDP', and is expected to extend local control, manage local economic development, redistribute public resources, and provide access to services such as sanitation, water, transport, electricity, primary health care and housing. But this is a severe challenge. As the Rural Development Strategy (GNU, 1995:49) notes, 'Rural people have long been the worst educated, least organised, and therefore least able to demand assistance through formal or informal structures. Yet their ability to take charge of local government and to contribute to decision-making will be critical to the effectiveness of rural local government'. Nevertheless, the government insists, under the philosophy of demand-driven development, that communities themselves should provide the initiative and impetus for these activities." TRANSFORMATION 30 (1996) 17 ARTICLE MUNRO The government acknowledges that there is more to capacity-building than training community members or transmitting expertise. It includes broader forms of socio-political empowerment, which cannot be achieved without the establishment of effective local government institutions. But it has found it politically and administratively difficult to design such institutions in rural areas. This is partly the result of the urban bias of government interests, both in terms of its development vision and its political base.18 But it is also a function of the traditional authorities' lobby and the slow process of land reform. Especially in KwaZulu-Natal, where the Inkatha Freedom Party dominates rural politics largely through its close alliance with traditional authorities, this situation sustains the rural-urban split in social authority and inhibits the re-integration of a bifurcated power structure. The local government structure currently being implemented makes no clear provision for primary-level local government institutions in rural areas, though it does so in urban areas. Under the current structure, rural decision-making is to be located at the Regional Council (RC) level because poor rural areas lack the capacity to implement local government functions independently (GNU, 1995:11). The RCs are territorially large - there are seven in KwaZulu-Natal - and unwieldy, comprising anything between 200 and 400 representatives, including non-elected traditional authorities, who comprise 20 percent of the representatives.20 Rural representatives are selected by party-list proportional representation, with the result that representatives do not necessarily speak for particular areas or community constituents. Thus, local government institutions in rural areas will draw their authority partly from democracy and partly from tradition, and their local accountability will be mediated by the principles of tradition and party competition. This set-up is almost certain to compromise local state construction. As Barnes and Morris (1996:19) point out, the system will work at a distance from rural communities, which makes it difficult 'to envisage how a demand driven process like the RDP is to be co-ordinated via a government dominated institutional environment that exists at a removed level from the people who are supposed to be driving the whole process'. The system will continue to be application-based but local government legislation prohibits the exercise of some primary local powers, such as water supply, electricity supply and sewerage purification, at the primary local level. The large size of RCs militates against effective decision-making without extensive delegation. But within RCs, urban citizens, who comprise the bulk of ratepayers and are organised at the municipal level through local councils, are in a much stronger position than rural communities to lobby for resources. In rural areas, services are still channeled through the apartheid-era regional authority system, a statutory non-representative tribal 18 TRANSFORMATION 30 (1996) MUNRO ARTICLE structure that represented black rural areas on the Joint Services Boards in the 1980s. In addition, the RCs encompass rural areas of great ecological, social and economic variety. As we have already noted, such variety demands close attention to localised development imperatives. But the local government structure does not make provision for such flexibility, especially given the absence of direct rural representation. There is as yet no model for the devolution of responsibility, accountability or input to the local level. In the highly-charged political atmosphere of KwaZulu-Natal, it is likely that those areas with greater resources, organisation or party patronage will determine development priorities within RCs. For weaker communities, local government will mean far less, either in terms of participation or in terms of service delivery. In effect, the government has abdicated a significant degree of influence at the local level, either to traditional leaders or to non-governmental interests. It has established a demand-driven system with no systematic framework for the articulation and evaluation of demands. The structure offers at best a very piecemeal approach to the tasks of equitable rural development, 'empowerment', or capacity-building. More broadly, it has left substantially intact spatially-differentiated definitions of citizenship according to which urban citizens elect their local representatives directly and claim their rights directly against the state, while rural citizens must rely on more distant party-list representation and claim their rights largely through the entrenched power of traditional authorities. The RDP, the land reform initiative and the local government structure are all integral components of the broader hegemonic project of re-orienting community identities and rooting state presence in the countryside. Yet there are serious disjunctures between them. According to the land reform initiative, property regimes are to be determined at the community level, but according to the local government structure, development decisions are to be made at the district level. According to the RDP, decisions are to be made generally within civil society, and the state's role is to respond to the demands of civil society organisations. As Julian Baskin (1994:8-9) has indicated, the 'social compact' demands not only acountability of the state but also the accountability of civic leaders. While access to land depends on community membership, land is allocated by the local authority, and local government decisions are taken elsewhere, it is difficult to see how the accountability of local civic leaders will be systematically secured. Local struggles over social authority are likely to intensify, and the role of the local state in forging a new national 'social partnership' remain tenuous. Under these conditions, development and service delivery are likely to become politicised rapidly, and the meaning of a 'people-driven process' ever murkier. TRANSFORMATION 30 (1996) 19 ARTICLE MUNRO This has, indeed, been the outcome in other developing countries where similar models have been adopted (Bryceson, 1988; Helmsing, 1991). Governments have been unable to transform the dominant structures of social authority, and they have tended to fragment. The local state has tended to become mired in local networks of clientelist relations. Quite frequently the result has been prolonged social volatility and a progressive weakening of both the local state and local economic structures. There is a very real danger that in South Africa (especially in KwaZulu-Natal where rural-urban distinctions are powerfully echoed in party politics) a similar outcome will occur. In effect, the privatisation of development in the context of a hegemonically weak state is a risky business, whether it occurs through the withdrawal of state resources and a reliance on civil society or markets, or through the channeling of state resources via patronage and clientelist networks that undercut the capacity of state-based institutions. As the decline of states elsewhere in Africa has shown, one danger is that popular empowerment becomes empowerment against the state rather than within the state.21 Where development and hegemony are inextricably linked projects, and the resource-base for both is small, it is crucial that a clear social understanding of the responsibilities of the state for development, and for the welfare of its citizens, be established. State Responsibilities and Popular Empowerment Political transitions, as well as processes of state construction, involve negotiations between societal interests and the state, and between political parties, over the precise parameters of state responsibilities in development. Governments are reluctant to define these parameters clearly, especially at times of limited resources, because it reduces their flexibility in responding to economic and political shifts. The conventional wisdom of liberalisation, vigorously promoted by the World Bank and the IMF, encourages a drastic curtailing of state responsibilities. At the same time, governments are determined to locate the state at the centre of political authority and to secure it as the final arbiter of social provision and public orde'r. This tension has created a dilemma for development theorists which remains unresolved: just what should the role of the state be in development? The same tension is reflected in the RDP. On the one hand, the government sees the expansion of services and amenities to the population as a task of government that is critical to consolidating a new political order.22 It has stressed the importance of taking back 'as part of its normal operations' the roles of planning, education, policy development and support that were taken over by 20 TRANSFORMATION 30 (1996) MUNRO ARTICLE anti-apartheid organisations such as local forums, community trusts, and NGOs during the 1980s. On the other hand, given its limited resources as well as its commitment to people-driven development and an active civil society, it has been reluctant to play a pre-eminent role. The RDP insists that development will be delivered according to the initiative and responsibility that specific communities take to secure it, and maintains (GNU, 1994:50) that 'Organisations of civil society should continue to have the choice of access to alternative sources of services ... l23 In short, the government has declined to define the state's development responsibilities with any precision. This has important ramifications for state construction. The tension between state activism and state retreat places a premium on institutions for negotiating what is to count as a public good, who is responsible for the provision of services, and who has ultimate sanction over their distribution. It makes negotiations over appropriate state responsibilities very sensitive, especially as contending political parties jockey for position. In particular, conflicts tend to arise between NGOs as providers of services and the state as the focus of allegiance. NGOs offer particular advantages for service delivery. They can move money, expertise and technical capacity quickly and sometimes more efficiently than the state, especially given the fragmented character of state institutions inherited from the apartheid era. Nonetheless, NGOs are also in competition among each other, and under pressure to show results. Where one NGO comes to dominant the local development terrain, it may succumb to the same authoritarian tendencies often imputed to state agencies, without the same mechanisms of accountability. As Alan Fowler (1988) points out, the advantages of NGOs over the state are only potential advantages and are limited to particular activities. Citizens are left to deal with resultant uncertainties. It is thus by no means clear that the accountability of the state will be improved by reducing its responsibilities, or that provision of public goods will become any less capricious when taken out of state hands. On the ground, conflicts between the state and NGOs hamper the ability of state agents to secure state authority. This is particularly the case, as in the case of South Africa, where the state lacks development resources and capacity, especially at the lower management levels. It wants both to rely on, and control, development initiatives of non-public agencies. For instance, the 1995 draft Non-Profit Organisations Bill sought to establish a compulsory register of NGOs and a statutory NGO council to provide government oversight; NGOs, moreover, would be enticed to register through tax incentives. Unsure of how properly to negotiate its relations with NGOs on the ground, the government has hedged its bets in defining the particular social responsibilities of the state. TRANSFORMATION 30 (1996) 21 ARTICLE MUNRO This queasiness has been especially clearly manifested in the difficulties that the RDP office experienced in disbursing funds for development projects and programmes. It is not simply the result of a laudable new pre-occupation with control and transparency, but also a determination to place the responsibility for development firmly on citizens themselves and curtail the role of the state. Most egregiously, the Rural Development Strategy (GNU, 1995:55) declares that 'rural people who wish to obtain funding assistance for capacity building, service delivery, or infrastructural development must learn the importance of obtaining and using statistical information about themselves in their applications for funding'. Such an unrealistic expectation places the demand-driven process beyond the reach of many rural communities, or it places their development trajectory in the control of organisations who can generate such information whether their interests coincide with those of community members or not. This makes some rural communities vulnerable to environmentally destructive development initiatives promoted by powerful enterprises. These conditions are currently creating a fragmented development and delivery system on the ground, which is driven very much by local capacity to organise. In some cases, anti-apartheid organisations are becoming entrenched as service deliverers. In some cases, large corporations have insisted on routing their social responsibility development programmes through local traditional authorities who are at loggerheads with local development committees. At the local government level, the lines of development funding and provision are tangled between central government competencies, provincial government competencies, and the non-public sector. Not only does this create a problem of co-ordination in which the role of local government institutions is murky, but in KwaZulu-Natal, where ANC/IFP tensions break down along national/provincial as well as urban/rural lines, development funding is likely to become rapidly politicised, and the separation between state and party expunged at the local level. As the government has tried to be as flexible as possible in defining the state's position in the "new social partnership', capacity-building and empowerment have moved slowly and haltingly. This process has not advanced the hegemonic project embodied in the RDP: the more the government hedges its bets, the less coherent its development project is. This ambivalence about the developmental role of the state is not insignificant. In a post-transition context where the state is weakly rooted in rural society the construction of hegemony depends on a clear demarcation of the state's development responsibilities, as well as a clear demonstration of the state's effectiveness in meeting those responsibilities. It is partly for this reason that the state/market and state/civil society face-offs that lay at the core of 1980s 22 TRANSFORMATION 30 (1996) MUNRO ARTICLE conventional development wisdom proved hopelessly inappropriate. At stake are the social rights - the 'moral economy' rights of expectation and need - that are an integral component of citizenship. Structural adjustment programmes elsewhere have shown that development without the state is a risky business because the social costs of economic development have to be politically mediated, and this depends partly on state authority (Duncan and Howell, 1992). Within the RDP rhetoric of popular empowerment, there is a subtle but important philosophical disjuncture: the right of rural communities to demand public goods has been blended into their responsibility to do so. This shift undermines the state's hegemonic project inasmuch as it fails to re-orient 'bifurcated' social authority structures and community identities at the local level, or to underwrite any meaningful new 'social partnership'. Conclusion Clearly, rural development under the RDP faces massive insititutional and structural constraints, especially in the absence of a viable overarching agrarian development strategy. The aim of this paper is not simply to paint a morose picture of these constraints. Nor is it to follow the currently fashionable argument that development is an essentially authoritarian process in which states extend their control over populations, and therefore to be treated with suspicion. Rather, the aim is to take account of the essentially political nature of development and to note that, for structural and historical reasons, rural development cannot be thought about separately from the 'national question'. Both are integral components of the post-transition process of re-ordering the principles of state power and social authority. Indeed, given the 'bifurcated' and authoritarian structure of social power shored up under the apartheid state, rural development plays a larger role in the national question than is generally conceded. Thus, the inevitable tensions in development processes between popular empowerment and state empowerment can only be properly addressed by recognising the imperatives of state that drive the post-transition politics of rural citizenship. As a result, the design, implementation and financing of development projects cannot be thought about as a technocratic process separate from social negotiations over the supply of public goods and legitimate state interventions in social life. The crucial issue of post-transition state construction is how such negotiations can be institutionalised. Such a hegemonic project has both institutional and ideological components which bring development, service delivery, empowerment, and the new social partnership together into an integrated political package. As I have suggested elsewhere, if the village is to be brought into the state, the state must be brought into the village (Munro, 1996). TRANSFORMATION 30 (1996) 23 ARTICLE MUNRO In this light, the RDP represents a sophisticated response to a highly complex political and development problem, which the government has tackled in a commendably open and public fashion. But in shifting between state empowerment, consolidation of party power, and demand-driven development, the politics of partnership shifts uneasily between the state's willingness to promote specific projects and its reluctance to define its own development responsibilities clearly. Consequently, it is by no means clear that the radical splintering of social authority and power principles that was central to the apartheid state form is being overcome. But such a fragmentation of authority inhibits both development and state construction in the post-apartheid period. It limits the capacity of state agencies to regulate social relations and property rights, and to manage rural populations; it promotes patronage politics as party elites try to maintain reliable political allies in the countryside; it casts shadows of uncertainty over local citizens' contingency plans for survival in a difficult environment. Under these circumstances, building the capacity of communities through a people-driven development process does not mean taking the state out of development and allowing civil society free rein. Nor does it mean 'empowering' rural citizens to get by on the fringes of the polity by investing their own resources in the common weal. In the context of political transition, where social authority in the countryside is highly fractured, building the capacity of communities is inextricable from building the capacity of the state. In KwaZulu-Natal, as Aldington and Lund (1995:574) note, 'All rural development - whether arising organically from local efforts or encouraged and facilitated by outside sources, state or private - at present suffers from the local absence of the state on a permanent basis. The seriousness of this for enabling the broad programmes to succeed is constantly underestimated'. Development theorists (Rapley, 1994; Bromley, 1995) increasingly acknowledge that state effectiveness rather than state size or state activism is the critical factor for successful development. Constructing an effective state is not necessarily an authoritarian process. If we take seriously the idea that legitimacy is a critical aspect of state capacity, and that the social authority of the state is a critical aspect of legitimacy, the relationship between the provision of public goods, a culture of politics (and political community) and democratic institutions provides a critical nexus for development processes. As a generalised approach to state-building, questions of 'participation' and 'capacity-building' have to be politically negotiated both at the level of local politics and at the state/society interface. But for these negotiations to effectively promote a national hegemonic project it is essential that the role of the rural population in the national 24 TRANSFORMATION 30 (1996) MUNRO ARTICLE development strategy be more clearly defined, that state-based institutions be rapidly secured at the community level, and that the state's role in securing both local governance and local conceptions of expectation and need be clearly defined. Notes The research and writing of this article was made possible by a Jill Nattrass Visiting Fellowship at the Centre for Social and Development Studies, University of Natal (Durban). Without implicating anyone else in the argument and interpretations presented here 1 would like to thank the CSDS staff for providing generous research assistance and intellectual community. I would also like to thank Sandy Johnston for useful comments on an earlier draft. 1. The distinction between these two dimensions of legitimation is necessarily imprecise. It varies with the relationship between the state apparatus and the ruling party. In developing countries, this relationship has been particularly close because the spoils of party control over state power have been so great, and the relationship between party largesse and public goods has been so strong. Nevertheless, the analytical distinction between these two legitimation projects is important inasmuch as it highlights the different sociological dynamics of party power and state power in the politics of political transition and development. It is the latter, which is frequently neglected in analyses of development politics, that focusses most centrally on the politics of citizenship. 2. The current dismantling of the welfare state in the USA and parts of Europe provides a particularly pertinent example of such conflicts. See, for instance, Keane (1988:7-13). There is no space here for a detailed conceptual discussion of 'stateness' or state tasks. States undertake a variety of political tasks including social control. For a more extensive theoretical discussion, see Munro (1996). 3. These challenges can be traced in the overlapping literatures of rural sociology and rural development. 4. Quoted in Collier (1994:15); see also Barry (1995). On Africa, see Munro (1996). On the new farmers' movements in India, see Brass (ed) (1995). 5. There are of course notable exceptions, such as China during the Great Leap Forward and Guatemala today. 6. For a short discussion, see Tapscott (1995). It is notable that the National Party has cast its own post-ethnic recruitment efforts in terms of a critique of the RDP, and an alternative poverty-reducing programme (as yet undefined). 7. Recently, IFP leader Mangosuthu Buthelezi suggested that local elections are usually about 'bread-and- butter' issues, but that the KwaZulu-Natal elections were more fraught because they were also about constitutional questions. In fact, local elections in any context where the state is weakly rooted in society are about both these issues, insofar as constitutional matters underpin local meanings of citizenship and public goods. 8. As the government's White Paper on the RDP (GNU, 1994:48) declared: 'The empowerment of institutions of civil society is a fundamental aim of the Government's approach to building national consensus. Through this process the Government aims to draw on the TRANSFORMATION 30 (1996) 25 ARTICLE MUNRO creative energy of communities. To facilitate effective involvement, the Government will introduce programmes that will enhance the capacity of community organisations'. 9. In development circles, the growing importance of this vision was reflected in the creation of the Development Bank of Southern Africa in 1980 to reduce the role of the state and to develop the bantustan areas economically on a 'participatory' basis. It is noteworthy that the World Bank regards South Africa as a role model for its own project on African Management in the 1990s. 10. Thus developmentalism draws on an essentially industrial and modernist ideology. While this ideology has been challenged from a variety of perspectives - postmodern, feminist, environmentalist, etc - it has not (yet?) been substantially displaced from the core of development theory. 11. One obvious exception is the small-scale sugar farming sector in KwaZulu-Natal and Mpumalanga. 12. In many post-colonial African regimes, the solution to this difficulty was to eliminate party competition, and to bring the party within the state. This strategy had the effect of smashing nationalist alliances and removing rural constituents from national politics. It thereby contributed markedly to the on-going hegemonic crises of these regimes. 13. It is notable that a large number of African governments restructured land tenure arrangements shortly after independence in order to place the state at the centre of authority and control over land. 14. For an excellent discussion, see Berry (1993). Berry shows that the customs on which land tenure arrangements were based were ambiguous and contested even before the imposition of colonial rule, and that colonial administrations did not succeed either in putting their stamp on such contests or in 'freezing' customary legal regimes. 15. This dilemma is particularly acute in areas, such as KwaZulu-Natal, where land played a central role in local power struggles associated with the collapse of apartheid. The passage of the Ingonyama Land Trust Act by the outgoing Nationalist Government and the Isipakhanyiswa Act by the KwaZulu Government has intensified conflicts over control of land in the province. The constitutional status of both these acts remains uncertain. 16. These ideas were elaborated by Mr Glen Thomas, Director: Restitution, in the Department of Land Affairs, at a public discussion of the Green Paper on Land Policy in Pietermaritzburg, March 15, 1996. 17. On this issue, see for instance the prolonged and very sensitive negotiations between Mandela, Buthelezi, Zwclithini and the amaKhosi over holding an imbizo. 18. As the Rural Development Strategy (GNU, 1995:54) puts it, rural communities 'must mandate their local and district councils to demand their fair share of funding for capacity building.' 19. Notably, it was not until mid-1995, under great pressure from rural advocacy groups, that the government amended the Transitional Local Government Act to make special provision for rural areas. 20. In Regional Council 7, for instance, 25 percent of the councillors are traditional chiefs. The structure aljo makes allowance for a small proportion of 'Special Representatives' representing levy-payers and women. Women's representatives are selected by party list, so that their interests do not necessarily follow gender issues. Currently, all chiefs in KwaZulu-Natal have ex officio membership on the RCs. The large size of the RCs represents an attempt to balance out the role of the traditional authorities in decision-making 26 TRANSFORMATION 30 (1996) MUNRO ART|CLE until the precise role of chiefs is constitutionally determined. It is highly unlikely that the chiefs will be excluded from the structures in the final dispensation. 21. As Bayart has argued, politics in post-colonial Africa is driven largely by a 'principle of escape', whereby subordinate political actors try to escape the reach of the state by whatever techniques are available to them. 22. The White Paper on the RDP (GNU, 1994:7) argues that the link between state construction, nation building and development is 'an infrastructure! programme that will provide access to modern and effective services such as electricity, water, telecommunications, transport, health, education, and training for all people. This programme wiU both meet basic needs and open up previously suppressed economic and human potential in urban and rural areas.' 23. The White Paper (GNU, 1994:53) declares: 'Communities must be prepared to commit themselves to performance contracts in terms of which they undertake to participate in the planning, management and protection of development programmes'. It continues: 'For the RDP to be "people-driven," there must be a vibrant civil society. However, there is a tendency at present to believe that "the Government will deliver on its own". This has the potential to disempower the organisations of civil society ... organisations and communities should themselves set up projects and expect the Government to co-operate, not to finance'. REFERENCES Ardington, E and F Lund (1995) 'Pensions and development: social security as complementary to programmes of reconstruction and development', in Development Southern Africa, 12(4). Barnes, J and M Morris (1996) 'KwaZulu-Natal's rural institutional environment: its impact on local service delivery', LAPC Working Paper 49, August. Barry, T (1995) Zapata 's Revenge: free trade and the farm crisis in Mexico, Boston: South End Press. Baskin, J (1994) 'Unpacking the "social compact'", paper presented at workshop: Can Urban-based community development make a major contribution to development and production? University of Durban-Westville, August 29. Bates, R (1989) Beyond the Miracle of the Market: the political economy of agrarian change in Kenya, Cambridge: Cambridge University Press. Berry, S (1993) No Condition is Permanent: the social dynamics of agrarian change in sub-Saharan Africa, Madison WI: University of Wisconsin Press. Brass, T (ed) (1995) New Farmers' Movements in India, London: Frank Cass. Bromley, S (1995) 'Making sense of structural adjustment', in Review of African Political Economy, 65. Bryceson, D (1988) "Peasant cash-cropping vs. food self-sufficiency in Tanzania: a historical perspective', in IDS Bulletin, 19(2). Chatterjee, P (1994) 'Development planning and the Indian state', in T Byres (ed) The State and Development Planning in India, Delhi: Oxford University Press. Collier, G (1994) Basta! Land and the Zapatista Rebellion in Chiapas, Oakland CA: Institute for Food Development Policy. Crush, J (ed) (1995) Power of Development, London: Routledge. TRANSFORMATION 30 (1996) 27 ARTICLE MUNRO Duncan, A and J Howell (eds) (1992) Structural Adjustment and the African Farmer, London: James Currey. Escobar, A (1995) Encountering Development: the making and unmaking of the Third World, Princeton: Princeton University Press. Ferguson, J (1990) The Anti-Politics Machine: development, depoMcisation and bureaucratic power in Lesotho, Cambridge: Cambridge University Press. Fowler, A (1988) 'Non-governmental organisations in Africa: achieving comparative advantage in relief and micro-development', IDS Discussion Paper, 249. GNU (South African Government) (1994) White Paper in the Reconstruction and Development Programme, September. GNU (1995) The Rural Development Strategy of the Government of National Unity: a discussion document, September. GNU (1996) Green Paper on Land Policy, March. Helmsing, A et al (1991) Limits to Decentralisation in Zimbabwe, Harare: University of Zimbabwe. Kaplan, D (1980) 'The South African state: the origins of a racially exclusive democracy', in Insurgent Sociologist, 10(2). Keane, J (1988) Civil Society and the State, London: Verso. Kotze, Detal(1987) Rural Development Administration in South Africa, Pretoria: Africa Institute. Macintosh, A (1990) 'Rethinking chieftaincy and the future of rural local government', in Transformation, 13. Mackenzie, F (1994) 'A piece of land never shrinks', in T Bassett and D Crummey (eds) Land in African Agrarian Systems, Madison: University of Wisconsin Press. Mamdani, M (1996) Citizen and Subject: contemporary Africa and the legacy of late colonialism, Princeton: Princeton University Press. May, J (1993) Transforming Public Expenditures on Development: agriculural development. Centre for Social and Development Studies, University of Natal, Durban. MERG (1993) Making Democracy Work: aframeworkfor macroeconomic policy in South Africa, Cape Town: Centre for Development Studies. Morris, M (1991) 'State, capital and growth: the political economy of the national question', in S Gelb (ed) South Africa's Economic Crisis, Cape Town: David Philip. Morris, M (1993) 'Who's in? Who's out?', in Work in Progress, 87. Munro, W (1995) 'Building the post-colonial state: villagization and resource management in Zimbabwe', in Politics and Society, 23(1). Munro, W (1996) 'Power, peasants and political development: reconsidering state construction in Africa', in Comparative Studies in Society and History, 38(1). Nattrass, N (1994) 'South Africa: the economic restructuring agenda, a critique of the MERG report', in Third World Quarterly, 15(2). Porter, D (1995) 'Scenes from childhood: the homesickness of development discourses', in Crush (ed). Raikes, P (1988) Modernising Hunger: famine, food surplus andfarm policy in the EEC and Africa, London: James Currey. 28 TRANSFORMATION 30 (1996) MUNRO ART|CLE Rapley, J (1994) 'New directions in the political economy of development1, in Review ofAfrican Political Economy, 62. Shipton, P (1988) 'The Kenyan land tenure reform: misunderstandings in the public creation of private property', in R Downs and S Reyna (eds) Land and Society in Contemporary Africa, Hanover NH: University Press of New England. Spiegel, A (1991) 'Village politics and administrative practices in Matatiele during the early-1980s', paper presented to the conference on political transition and economic development in the Transkei, Rhodes University; November. Tapscott, C (1995) 'Changing discourses of development in South Africa', in Crush (ed). Uphoff, N (1993) 'Grassroots organisations and NGOs in rural development: opportunities with diminishing states and expanding markets', in World Development, 21(4). TRANSFORMATION 30 (1996) 29