Power Relations and Poverty Reduction

Author(s): David Mosse, School of Oriental and African Studies, University of London
Publication date: Tuesday 08 February 2005

This article is a chapter of the book Power, Rights, and Poverty: Concepts and Connections, (edited by Ruth Alsop) which is the result of a working meeting sponsored by DFID and the World Bank, March 23–24, 2004.

For many reasons, the concept of power has attracted relatively little attention in policy analysis among international development agencies. And yet, power has been central to the frameworks that inform academic social science; emphasized, perhaps, to a fault. [ 1] Nonetheless, power, disempowerment, and empowerment are increasingly part of the analysis of poverty and its alleviation, and inequality in power relations is taken to explain important constraints to poverty reduction measures.
Power inequalities inhere in interpersonal relations and in the community. They are part of the dynamics among beneficiaries, development agencies, and the state, and can be found in the hierarchies of organizations as well as in interagency and donor-client relations. The effects of power relations on poverty reduction are many. The interests of national elites and the electoral concerns of those in power affect the state’s policy choices, sector priorities, and programs, with important consequences for the poor. Equally, well-intentioned sector reform programs can run aground where they challenge vested interests, and democratic reforms often have limited or unpredictable effect on power relations. And where local elites are well placed to capture benefi ts and reservations (in education, employment, or for elections), or to manipulate the administrative system upon which the poor depend for their livelihood and for access to anti-poverty schemes, formal processes of decentralization may do little to reduce informal forces of domination (Jenkins 2002).
Power is pervasive; it is not just structure, but also the electrical current of society. In this paper I want to set out some ways of thinking about power, and then look at some different approaches to empowerment within development. In various ways these approaches attempt to address the effects of power inequalities on the achievement of poverty reduction goals.

Conceptualizing Power
Concepts of power vary widely, ranging from Weber’s (1964) pluralistic notion of the command of force to Foucault’s (1980) discourses of truth and knowledge. Indeed, thinking about power relations, empowerment, and poverty requires a broad framework. For present purposes, it may help to draw six preliminary distinctions.
First, formal and legitimate forms of power, such as government councils, the police, and so on, can be distinguished from informal, dispersed, or, in Foucault’s terms, “capillary” power. Of course, formal power can be distinguished in terms of the different sources from which it is derived (or legitimized); for example, the power of bureaucratic authority versus the power of popular approval (Weber’s rational legal versus charismatic authority). When thinking about empowerment, however, it is as important to think about the relationship between such formal authority and informal relations of power. It is well known that the operation of formal structures is underpinned by informal relations, and it is the hope of reform agendas that formal processes (policy, decentralization, structures of representation) will have an impact on unequal, informal power relations (for example, democratic decentralization on inter-caste relations in India).
Second, analyses of power can be distinguished in terms of whether they emphasize the modes of domination from “the top,” or the everyday exercise of power at “the bottom.” Of course, power can be analyzed at many different levels.
Third, sometimes power is conceptualized as infi nitely expanding and augmented by economic growth; but at other times as finite, as a scarce resource like land, water, or state resources over which groups compete (Cheater 1999). The latter is a zero-sum view of power that Moore and Putzel (1999) characterize as “interest group economism.” This distinction is related to that between the promotion of power to do various things, and competition for power over things or people (Nelson and Wright 1994).
Fourth, there is a distinction between actor-oriented and structural views of power. In actor-oriented or transactionalist (Weberian) views, power is a non-economic resource that individuals seek to maximize, rather as they might maximize economic returns. Power is subject to rational choice, and ways of strategically maximizing power can be modeled (for example, Bailey 1969, in Gledhill 1994). Other actor-focused analysts are more interested in the goals, interests, and unequal effects of power plays, such as the uneven accumulation of political capital, or profit, in the form of symbolic or cultural capital, prestige, honor, or popularity (Bourdieu 1977). Political capital allows certain groups in society privileged access to public resources, whether public works contracts or jobs within an administrative service (cf. Wade 1982). But political capital is also necessary for the poor, whose rights and assets have to be negotiated and defended politically (Baumann 2000). In economistic formulations, political capital is regarded as an asset that links individuals or groups to the power structure (ibid).
A problem with viewing power in terms of individual or group strategies is that it does not explain the systematic nature of social behavior (Gledhill 1994). Symbolic and political capital are closely related to the accumulation of economic capital and the reproduction of class structures, but it is also a matter of the effects of these on individual behavior. People’s behavior is conditioned by durable dispositions (cognitive and behavioral) derived from historical forces and “tend[s] to reproduce the regularities immanent in the objective conditions,” which Bourdieu terms habitus (1977, 78, in Gledhill 1994).[ 2] This is a useful concept, not least because it indicates the complex subjective and objective demands of empowerment. For example, those working to empower socially subordinated groups like the Dalits in India explain that their work involves changing assumed meanings and entrenched habits by initiating a process of re-socialization, in which individuals learn through practice to modify the distinctions and schemes that produced their disempowerment, as well as to achieve economic independence (Arun 2004). There are parallels with work on gender relations. It is significant that activists often see confl ict, even violence, as necessary to disrupt old meanings, and education as necessary to prouce new ones. Of course, unequal power relations are naturalized as habits and classifi cations to different degrees, so opportunities for change and reactions to it vary in different cultural contexts. The violent retaliation to efforts to change unequal relations between upper castes and Dalits (evident in atrocities against Dalits), or to change gender relations (evident in links between domestic violence and micro-finance initiatives)[ 3] demonstrates a strongly embedded habitus. Opposing actor-focused positions is the idea of power as structure, which has a strong tradition in social science. Of course, such structures are not visible; rather, they are ideas about the distribution or balance of power in a given society (Leach 1964). The idea of a power structure usually makes sense only in relation to other things such as the distribution of land, wealth, or other assets. The idea of a power structure also depends upon the cultural construction of power, which varies from society to society. For example, the relationship between religious, political, and economic power or status is not constant. Much scholarly ink has been spilt on this issue.[ 4] The idea of a power structure that defines individuals and groups by their position in the socioeconomic order can off er a rather static view. Indeed, generally speaking, structural theories that explain the reproduction of power relations are not so good at explaining change in power structures (Gledhill 1994). But power structures and the institutions through which they are expressed are indeed profoundly changed by historical circumstances.[ 5] For example, introducing landed property rights in many colonial settings where power derived from the control over people rather than over assets (or from redistribution rather than production) had profound implications for the structure of power relations at local and regional levels. Similarly, the shift to a market economy and a new power of money in relation to institutions and office in post-Soviet societies also had transforming eff ects on power relations. Th e world over, structures of power have been changed in regionally and locally specific ways by universal franchise and electoral politics that give a new power in numbers, and which can also unleash populist violence turned on “market-dominant minorities” (Chua 2003).
In a fifth conception of power, unequal power relations are more or less concealed from those whose lives are shaped by them. Some theorists stress the idea of a hegemony of the powerful, the domination or oppression of minds and aspirations, and the production of a “false consciousness” among subordinate groups who appear to be in consensus with systems that oppress them. Other theorists emphasize the agency and everyday resistance of the poor, such as Scott’s (1985) many “weapons of the weak,” as well as the more overt forms of rebellion or conflict.
Finally, power can be understood as political representation. In this case, power does no concern only people’s actions and relations, but also the language, classifications, and organizations through which they are represented as interests and groups within political systems (Gledhill 1994). The point here is that power relations in society are always shaped by wider political systems. The power that people have (as individuals and groups) depends upon the capacity of others (for example, labor union leaders and party workers) to impose social classifications upon them and then speak on their behalf. It is the process of classifi cation that “turns the group from a collection of individuals to a political force” (ibid. 139).[ 6] In this view, political parties or organizations do not reflect any naturally occurring classes, castes, ethnicities, and the like, but rather manufacture these categories through the process of determining who gets political representation. The party precedes the class struggle. Further, the political system is a professionalized field in which political capital is held in the hands of a few (ibid).

Empowerment and the Poor
Next, I want to identify some of the different approaches to empowerment as a means to poverty reduction, which draw on these different ideas of power.

Capacity Building vs. Struggle
Voluntaristic approaches to empowerment emphasize training, awareness raising, and capacity building for individuals and groups. Power here is the power to achieve ends. It is an infinitely expanding resource, but its conception is often limited to “having a place, a voice, [and] being represented within administrative or managerial systems” (James 1999, 14). Empowering organizations make the most of their staff potential (ibid). However, critics suggest that this sort of empowerment may be linked to restructuring, downsizing, cost-cutting, and flattening management structures. In this sense, it can be antithetical to acquiring political power through collective bargaining or union action (ibid), and may even strengthen the power of managers, bosses, and owners. In contrast to power to is a view of empowerment as struggle for power over resources (or other people), often within a zero-sum game in which the rich and the poor, managers and workers, are opponents. The “interest group economism” (Moore and Putzel 1999) that promotes this view of empowerment is in turn criticized for its failure to see the coalitions and mutual interests between rich and poor, state and citizens, that can sustain pro-poor agendas (see below).

Anti-Poverty Programs vs. Decentralized Democracy
There are many NGO and state-run community-driven development interventions that have an empowerment agenda as part of specific programs. Such projects aim to empower from the bottom up through participative planning, technology development, and the promotion of self-help groups or users’ associations for improved management of resources such as water, forests, grazing land, finance, public utilities, and the like. The potential of such programs to enhance poor people’s power to achieve their ends is rarely in question. Real needs and interests are addressed through resource user associations, which are more accessible and inclusive than elite-dominated systems of local government. The need for poor people to form associations to contend with the power of the rich, or as a means to deal with injustice, formed the core of NGO strategies in South Asia from the early 1980s. However, the capacity of such interventions to overcome rather than reproduce wider unequal power relations is also questioned. Development programs and their user groups can operate in ways that limit poor people’s potential to enhance their political capabilities or to sustain political organization, and may actually demobilize existing organizations (Moore and Putzel 1999). Significantly, such a critical position is taken both by those who favor operating within state systems and those who favor working against the state; by those who prefer to work through formal politics as well as by those who operate through extra-political mass action. Let me illustrate the point.
Numerous rural development initiatives undertaken in India in the 1990s, including DFID’s rainfed farming projects and the GOI’s watershed development programs, aimed to empower the poor, especially through mobilizing the grassroots for the management of key livelihood resources. However, these initiatives have been subject to criticism in terms of their capacity to address power relations.
For one thing, the village-level associations they promote tend to be dominate by the more affluent and powerful members of society, especially since they are avenues to the important material and political resources of outside agencies. Kumar and Corbridge (2002) argue that the effect of such programs is not only to concentrate local power, but also to weaken existing institutions of collective action (grain banks, reciprocal labor) that offer some livelihood security to the poorest of the poor. Moreover, NGOs and other implementing agencies tend to develop clientalist relationships with their villager beneficiaries, who, as individuals or self-help groups, are willing recipients of efficiently delivered programs (Mosse 2004a). So development interventions, even those with explicit agendas of participation, community driven development, or empowerment, tend to affirm power structures; they tend to be inherently conservative, reconstituting rather than challenging relations of power, authority, and patronage at every level—in target villages (between landlords and labor), in the project teams, or within the corporate organization, donor, and beyond (cf. Harriss 2002).
These unequal power relations can shape the very instruments that are intended to be empowering. Thus, in “participatory planning” outsider expert perspectives win over local knowledge, needs and plans being determined with reference to outsider agendas, and local organizations develop as dependent client bodies seeking patronage. The concern with strengthening demand occurs in the context of supply. This means that despite the ideals of participation, “people become empowered not in themselves, but through relationships with outsiders; and not through the validation of their existing knowledge and actions, but by seeking out and acknowledging the superiority of modern technology and lifestyles, and by aligning themselves with dominant cultural forms” (Mosse 2004a, 218; cf. Fiedrich 2002). Project structures entail a degree of uncertainty, arbitrariness, inequality, and patronage, which does not provide an environment in which collective action by the poor is encouraged. Moreover, these development programs are invariably implemented through non-state Project Implementing Agencies; they establish authorities against which rights cannot be asserted (cf. Moore and Putzel 1999, 16).
The empowerment credentials of Indian state anti-poverty programs that aim to transfer resources management from state bureaucracies to user communities are hardly better. User groups in India, are often regarded as undemocratic, unaccountable, and easily controlled or manipulated by the departmental bureaucracies that promote them, and as unlikely to be sustained or to foster wider forms of organization among the poor. Typically, the powers and rights that poor people acquire through them are heavily circumscribed, especially by lower level bureaucrats (Manor 2002). This is not to say that transferring resource management does not have effects on power relations (e.g., by making new demands on offi cials), merely that their effects are not systematically in favor of the poor. Precisely because they threaten powerful interests in the bureaucracy, user groups may be controlled, resisted, or resented. In south Indian irrigation, for example, there is an unconstructive stalemate. While irrigation officers have little to gain and much to lose from increased accountability to water users associations (WUAs) under participatory irrigation management—that is, their capacity to benefit personally from increasing risk and rent seeking through privileged knowledge is reduced—farmers organizations themselves have no formal means of holding officials accountable. “And although WUAs have little authority and no legal right to manage irrigation systems, their existence, and the links which they are able to forge with politicians and senior levels in the bureaucracy, still signal a de-motivating loss of power to officials within state level irrigation bureaucracies” (Mosse 2003a, 291). Many studies suggest that the evidence on the long-term sustainability of user groups, especially with the removal of external incentives, is equivocal— even in the case of groups supported intensively by NGOs, some for as many as seven to ten years (Saxena 2001). Of course there are other studies that show how new user groups can be routes to empowerment; for instance, challenging existing social exclusions of lower castes from resources. Associations may also preserve indigenous institutional arrangements (as well as erode them). One thing is clear: without longitudinal studies it is impossible to predict the longerterm effects of new associations.
Now, a view that commands growing support among policymakers is that the poor will be empowered not through central anti-poverty programs, but through the process of political decentralization that devolves resources and decision making to elected local councils, such as Panchyati raj institutions in India. Indeed, the strongest critics of centrally controlled programs and their users’ associations and self-help groups (SHGs) in India are often the greatest supporters of political decentralization and Panchayati raj as the best route to empowerment of the poor (Manor 2002). They argue that the indiscriminate promotion of well-resourced user committees and other parallel bodies undermines the processes of democratic decentralization and diverts resources from elected bodies, which is detrimental to the long-term interests of the poor. Problems include the overlapping of jurisdiction, confusion and the usurpation of roles and functions, a fragmentation of popular participation, and the balkanizing of accountability to diverse programs, departments, and donors (ibid). However, some research suggests that democratic decentralization itself may fail to change unequal power relations in favor of the poor, while bureaucratically managed program delivery may actually be more effective at enhancing their political capabilities.

The Politics of Governance and Poverty Reduction: the Case of Madhya Pradesh and Andhra Pradesh
Different state governments in India have emphasized to different degrees the promotion of anti-poverty programs and user groups and democratic decentralization. There is an interesting comparison between the state of Madhya Pradesh (MP), which officially launched a radical program of political decentralization, and Andhra Pradesh (AP) which devolved powers to a lesser extent and adopted a populist approach to poverty reduction through line departments and parallel bodies, such as local user groups and SHGs. Both are DFID partner states. A recent study (Johnson, Deshingkar, and Start 2003) reaches two interesting conclusions.
The first conclusion is that political decentralization in MP has been less effective in addressing the needs of the poor than expected because of “the failure to challenge [the] well-entrenched power of the village chiefs, the Sarpanches” (Johnson, Deshingkar, and Start 2003, vi). The government’s attempt to challenge the power of these authorities by empowering village councils (gram sabhas) and putting power back into the ruling party machinery through District Planning Committees failed. Gram sabhas remain controlled by existing leaders, they have not increased political competition, and they are poor in resources and ineffective. In other words, the functioning of formal democratic structures (for example, reserved constituencies of women or low castes) has been substantially undermined by informal power relations.
The second conclusion is that, despite a more limited process of political decentralization in AP (even some hostility to Panchayati raj), the government’s populist approach to development (its janmabhoomi program of community development through watershed rehabilitation, joint forest management, and women’s credit) has empowered the poor in more effective ways. Contributing factors include the AP government’s need to secure political support among its key constituencies (women, backward castes, agricultural laborers) especially in face of painful reforms (Johnson, Deshingkar, and Start 2003,11); program delivery arrangements, which brought public officials close to the people even while weakening the panchayats; and the creation of incentives for political participation through populist schemes delivered to the poor, which ensured better attendance and participation in AP gram sabhas, even though they were bureaucratically controlled (ibid).[ 7] The negative finding (decentralization empowers local elites to capture resources from the poor) is consistent with a stream of recent research emphasizing that the existence of a strong center that is able and willing to resist the power of local elites (to earmark funds, support strong local staffing, and so on) is a necessary precondition for decentralization (e.g., Tendler 1997).[ 8] It also speaks to the more general theme that democracy and poverty reduction are not necessarily mutually supportive. The positive finding (populist, central programs can empower) draws attention to the importance of wider political systems and electoral strategies for enhancing the political capabilities of the poor.

Political Representation and Poverty Reduction
In both states, poverty reduction programs and decentralization were shaped by political interests of the government in power, as well as by state-wide caste/class structures (historical continuity of upper caste/class dominance in MP, and historical challenge to landowning dominant castes in AP). But while political strategies worked in favor of the poor in AP, they were less able to do so in MP.
Despite limited political decentralization, the government of Andhra Pradesh put eff orts into central anti-poverty programs (such as subsidized rice, credit for women, watershed rehabilitation) that worked for the poor because their interests had become part of the government’s electoral strategy; that is to mobilize necessary electoral support from women, low castes, and laborers in face of unpopular reforms. In other words, the wider political system (the nature of political constituencies and party competition) enabled the poor in Andhra Pradesh to develop a political capacity that they did not have in Madhya Pradesh. Put another way, poor and unorganized people do not have a chance for political representation unless their interests can become a weapon in the struggles of the professional political fi eld (Bourdieu 1991, 188; in Gledhill 1994, 139). The politicization of poverty is necessary for the empowerment of the poor. Making poverty a public, moral, and political issue is often the basis upon which the poor gain leverage by making power work to their advantage through enrolling elite interests, through pro-poor coalitions, and from competition between elite groups (Moore and Putzel 1999). This view of political representation argues against both interest group economism’s zerosum view of structurally opposed interests dividing up the power cake (ibid), and voluntaristic approaches to empowerment through capacity building.
Rights-based approaches to poverty reduction also depend upon effective politicization. Rights awareness and legal aid may be important first steps, but if the rights concerned do not become part of a political agenda, and if those whose rights need to be protected do not comprise a political constituency, the outcomes are unlikely to be very positive. Inter-state adivasi (tribal) migrant laborers and construction site workers in India are a case in point since they fail to become a constituency for political parties, line departments, or labor unions, and remain subject to appalling exploitation despite the existence of progressive Indian labor laws (Mosse 2004b).[ 9]
One difficulty, as Sandra Pepera (2002) notes, may be that “most countries do not defi ne themselves by their poverty, so we cannot assume that poverty reduction is at the heart of politics in developing countries.” It is political systems—not policy—that determine the interests and identities that shape the democratic process. Political systems have their own logic, which may or may not enhance the political capabilities of the poor. Since votes are rarely cast simply on the basis of economic interests, the development of political capability among the poor depends upon the adoption and manipulation of identities that allow effective representation. These are often caste, religious, language, or ethnic identities.[ 10] The problem is that while political systems determine the identities around which people gain empowerment, interests framed in “communal” terms can become self-limiting and dangerous, especially when—as is common—they turn to violence and conflict among the poor themselves.[ 11]
There is a wider dilemma, often glossed as “political manipulation.” Empowerment depends upon political representation, but political capacity is gained at the cost of conceding power to a political system with its own logic about maximizing votes, retaining power and coalitions, disseminating ideology, and the like, which further concentrates political capital (Bourdieu 1991, Gledhill 1994). Of course, this does not preclude strategic manipulation by the poor from below. But in India and elsewhere, disillusionment with party politics has fuelled a variety of extra-political social movements with the aim of increasing the capacity of the poor to organize around their interests as forest users, fishermen, indigenous people, and the like. Sometimes these organizations are linked to international NGOs.

Governance, Sector Reform, and Politicization: Contradictions
Moore and Putzel (1999) rightly point out that development agencies need a far greater understanding of the different political systems and scenarios in which they intervene (politically, inevitably), in order to identify opportunities to strengthen the political capabilities of the poor. Developing an operational political analysis is complex and difficult, and it will certainly involve revisiting some existing policy concepts (community-driven development, participation, and social capital, to name just three). The good news, perhaps, is that in the end, effective government—a goal of most donors—is the best condition for developing such political capabilities (ibid). Eff ective government means, among other things, a center able to defend the interests and rights of the poor; institutionalized populist policies that enable poor people’s organizations to access resources or claim rights (including labor rights and land reform); and competition for the votes of the poor through relatively stable party lineups. [ 12]
The bad news is that once again, development is most likely to occur where it is already half accomplished (ibid). The poor are rarely well served by weak governments that mobilize their votes in the short term through patronage and violent identity politics. Moreover, the politicization of development programs can often work against the poor. The sectors and schemes that attract most political attention and priority for funding are often those high-profile programs that offer most opportunities to redistribute resources to political supporters so as to generate political capital cashed as votes. Comparing different types of poverty reduction schemes in India, for example, Farrington (2002) draws a contrast between the National Housing Scheme and the National Old Age Pension Scheme. The former (involving large lump sum payments) is high profile and subject to political interference and corruption, while the latter performs well in terms of poverty-reducing impacts but lacks “political champions” and so is never expanded.
Politicization of this kind acts as a powerful break on reform. The power sector, for instance, is a rich source of election funds. Here, as Sumir Lal points out, political interference is the result of—not the cause of—utilities mismanagement and the “redistribution” of public funds (helped by Indian election funding rules). The problems here are how to depoliticize the power sector, how to contend with internal resistance, and how to rectify the subservient relationship of sector managers to political leadership (Lal 2003). The key point is that poverty reduction is always political, but the way in which interventions are politicized has a signifi cant bearing on the interests of the poor.

Depoliticize or Politicize? Contradictions of Sector Reform
A focus on power relations brings out some contradictions in current policy frameworks, for instance, between the demands of public sector reform (corporatization) and rights-based approaches and good governance (democratization). Sector reform attempts to order public institutions in prescribed terms (autonomous from the political and elected civic bodies). At the same time, striving for democratic rights and entitlements can involve violating these institutional norms (Coelho 2003, citing Partha Chatterjee). Karen Coelho (2003) illustrates the point with a case of urban water supply in Chennai (Madras). Her ethnography shows that the engineers and frontline staff who put public sector reform and corporatization into practice tend to equate the citizenry with private paying consumers. In these terms, slumdwelling poor people are further marginalized as non-paying, non-deserving, difficult customers (i.e., with few legitimate claims to water rights). On the one hand, the engineers regard group rights-based claims to water and politics more generally as a disorderly nuisance, a rabble; on the other, the poor do not present their claims as individual consumers filling out complaints forms, but collectively, publicly, and confi dently in the currency of rights as members of political society.

Donor-Client Power Relations
One final point, which can only be mentioned, is that any poverty reduction strategy that boasts an empowerment component must take into account donor-client relations and the political interdependences of donor and countrylevel coalitions. As the sharp conditionality of 1980s structural adjustment lending is replaced by debt-relief initiatives linked to pro-poor policy reform, PRSPs, Comprehensive Development Frameworks, sector-wide approaches (SWAps) and the like, and as aid relationships are re-framed in the language of partnership or local ownership, these goals need careful conceptual and operational scrutiny.

[ top]

Notes
1. Sahlins (1999, 404) takes to task the “afterologists” who end up knowing everything only “functionally, as devices of power . . . not substantially or structurally.” [ back]
2. Here, Bourdieu offers a theory which addresses an issue that is left unresolved in social capital studies (especially in the weak modeling of causal mechanisms in econometric analyses of the relation between social capital and poverty indicators), namely, what constitutes a social influence on individual behavior (Durlauf 2002). [ back]
3. Micro-finance init iatives can also have the opposite effect of reducing domestic violence, for example in adivasi western India, by morally delegitimizing existing forms of social capital mediated by alcohol (see Mosse 2004a, 216–7).[ back]

4. See, for example, the debates on secular power and religious status in India opened up by Louis Dumont’s work (1980).[ back]

5. My own work in south India (Mosse 2003a) shows that it is not just inequality or the concentration of power and authority that is relevant for social outcomes, but also how power is articulated: whether for example through public institutions (such as temples and water systems) or through more diffuse private networks of patronage, alliance, and personal obligation. [ back]

6. “The fact that the working classes are widely deemed to exist is based on their political representation by political and trade union apparatuses and party officials ‘who have a vital interest in believing that this class exists and in spreading this belief among those who consider themselves part of it as well as those who are excluded from it’” (Bourdieu 1991, 250; quoted in Gledhill 1994, 139). [ back]

7. The research was undertaken before the ruling of the 2004 elections. The fact that the AP government dramatically lost power suggests, perhaps, the limitations of this political strategy, and the growing unpopularity of the government’s reform agenda. [ back]

8. The Indian state of West Bengal is often cited as a case that exemplifies effective decentralization based on a strong center providing long-term support. [ back]

9. Artisanal fishing communities along the Indian coast are another group of the poor who are unable to acquire the political capacity to protect their interests (against mechanized operations) within mainstream politics, largely because current boundaries preclude coastal constituencies.[ back]

10. My work in Tamil Nadu shows that even local struggles over resources are increasingly expressed in such terms. When wealthy farmers illegally divert common water sources to their dry land cash crops, poor farmers organize and petition the district authorities as Dalits, alerting the officials to “imminent communal riot.” They do so because they understand that this is the language that will secure state backing for local disputes, that will ensure that the next morning four jeeps will arrive in the village with local officials and police to resolve the conflict. Women also use the politics of identity tactically to secure necessary access to resources and services. [ back]

11. Growing awareness of this fact may have played its part in the turn against the Hindu nationalist party, the BJP, in India’s 2004 national elections. [ back]

12. The emphasis on governance and stability is crucial, and Chua (2003) offers a timely warning about the dangers of rapid democratization in the context of free-market growth bringing disproportionate wealth to market dominant ethnic minorities. [ back]

[ top]

References

Arun, C. Joe. 2004. “From outcaste to caste: The use of symbols and myths in the construction of identity; A study of confl ict between the Paraiyars and the Vanniyars in Tamil Nadu, south India.” PhD diss., University of Oxford.

Bailey, F. G. (Frederick George). 1969. Stratagems and spoils: A social anthropology ofpolitics. Oxford: Blackwell.

Baumann, Pari. 2000. Sustainable livelihoods and political capital: Arguments and evidence from decentralization and natural resource management in India. ODI Working Paper 136. London: Overseas Development Institute.

Bourdieu, Pierre. 1977. Outline of a theory of practice. Cambridge: Cambridge University Press.

———. 1991. Language and symbolic power. Ed. John B. Thompson, trans. Gino Raymond and Matthew Adamson. Cambridge: Polity Press.

Cheater, Angela. 1999. Power in the postmodern era. In The anthropology of power:Empowerment and disempowerment in changing structures, ed. A. Cheater. London: Routledge.

Chua, Amy. 2003. World on fire: How exporting free market democracy breeds ethnichatred and global instability. London: Arrow Books.

Coelho, Karen. 2003. Unstating “the public”: An ethnography of reform in an urban public sector utility in south India. Paper presented at the EIDOS Workshop on Order and Disjuncture: the Organisation of Aid and Development, London.

Dumont, Louis. 1980. Homo hierarchicus: The caste system and its implications. Complete revised edition. Chicago: Chicago University Press.

Durlauf, Steven N. 2002. Th e Empirics of social capital: Some skeptical thoughts. 66 Power, Rights, and Poverty: Concepts and Connections. Roundtable Paper. Washington, DC: World Bank, Social Development Department.

Farrington, John. 2002. Poverty reduction prospects in India: What Difference does state politics make? Presentation for ODI/DESTIN conference, Putting Politics Back into Development: Are We Getting There? London, May 22.

Fiedrich, Marc. 2002. “Domesticating Modernity: Understanding Women’s Aspirations in Participatory Literacy Programmes in Uganda.” PhD diss., University of Sussex.

Foucault, Michel. 1980. Power/knowledge: Selected interviews and other writings, 1972-1977. Ed. Colin Gordon. New York: Pantheon Books.

Gledhill, John. 1994 (2000). Power and its disguises: Anthropological perspectives onpolitics. London: Pluto.

Harriss, John. 2002. Civil society: Universal concept or donor fad? Presentation for ODI/DESTIN conference Putting Politics Back into Development: Are We Getting There? London, May 22.

James, Wendy. 1999. Empowering Ambiguities. In Anthropology of power: Empowerment and disempowerment in changing structures. Ed. Angela Cheater. London: Routledge.

Johnson, Craig, Priya Deshingkar, and Daniel Start. 2003. Grounding the state: Poverty, inequality and the politics of governance in India’s panchayats. ODI Working Paper 226. London: Overseas Development Institute.

Jenkins, Rob. 2002. Domestic politics and the WTO. Presentation for ODI/DESTIN conference Putting Politics Back into Development: Are We Getting There? London, May 22.

Kumar, Sanjay and Stuart Corbridge. 2002. Programmed to fail? Development projects and the politics of participation. Journal of Development Studies 29 (2): 73–103.

Lal, Sumir. 2003. Political factors aff ecting power sector reform in India: An internaldiscussion note. New Delhi: World Bank.

Leach, E.R (Edmund Ronald). 1964. Political systems of Highland Burma: A study of Kachin social structure. Cambridge: Cambridge University Press.

Manor, James. 2002. User committees: A potentially damaging second wave of decentralisation. Draft paper, accessed via www.panchayats.org

Moore, Mick and James Putzel. 1999. Thinking strategically about politics and poverty. IDS Working Paper 101. Brighton: Institute of Development Studies.

Mosse, David. 2002. Adivasi migrant labour support: A collaborative programme, consultant’s report. Western India Rainfed Farming Project. India: DFID.

———. 2003a. The rule of water: Statecraft, ecology and collective action in south India. Delhi: Oxford University Press.

———. 2004a. Cultivating development: An ethnography of aid policy and practice London: Pluto.

———. 2004b. On the margins in the city: Adivasi seasonal labour migrants in western India. Paper for Livelihoods at the Margins Conference, SOAS, 8th & 9th July.

Nelson, Nici and Sue Wright. 1994. Power and participation. London: Intermediate Technology Publications.

Pepera, Sandra. 2002. Can PRSPs influence the national politics of poverty reduction? Presentation for ODI/DESTIN conference Putting Politics Back into Development: Are We Getting There? London, May 22..

Sahlins, Marshall. 1999. Two or three things that I know about culture. Journal of theRoyal Anthropological Institute of Great Britain and Ireland 5 (3): 399–421. Saxena, N.C. 2001. Issues in panchayats. Draft paper accesses via www.panchayats.org.

Scott, James. 1985. Weapons of the weak: Everyday forms of peasant resistance. New Haven: Yale University Press.

Tendler, Judith. 1997. Good government in the tropics. Baltimore and London: John Hopkins University Press.

Wade, Robert. 1982. Th e System of administrative and political corruption: Canal irrigation in south India. Journal of Development Studies 18 (3): 287–328.

Weber, Max 1964 (1947). The theory of social and economic organization. Ed. Talcott Parsons. New York: Free Press of Glencoe; London: Collier-Macmillan.

[ top]