Understanding Agency in Collective Action.pdf

8/16/2019 Understanding Agency in Collective Action.pdf

1/23

Understanding Agency in Collective Action FRANCES CLEAVER

Frances Cleaver is a senior lecturer at the Bradford Centre for International Development, University of Bradford, UK

Abstract Participatory approaches to natural resource managementencompass ideas about the desirability of citizens actively engaging inthe institutions, policies and discourses that shape their access to

resources. Underpinning such approaches are assumptions about thenature of human agency. Purposive individual action is seen asinstrumentally desirable as well as potentially radical andtransformatory. Through participation in collective resourcemanagement it is claimed that people can re-negotiate norms, challengeinequalities, claim their rights and extend their access. This paper drawson insights from theories of structuration, governmentality and genderedempowerment to explore understandings of how individual humanagency shapes and is shaped by social relationships and institutions. Itoutlines six factors that constrain and enable the exercise of agency for

different people; cosmologies, complex individual identities, the unequalinterdependence of livelihoods, structure and voice, embodiment andemotionality. The paper concludes by considering some of theimplications for research and development interventions.

Key words: Participation, Agency, Natural resource management

Introduction

The concept of agency is central to understandings of collective action, but

the assumptions made about the motivations and actions of individuals areoften problematic. In this paper I argue that conceptualizations of agency as deliberate public participation in decision-making and collective actionare unhelpfully narrow. The paper synthesizes previous work of the author and draws on a variety of theoretical positions to outline more complex ways of understanding agency in collective action as deeply relational, andconstituted by routine practice as well as purposive action.

This paper evolved from thinking about the nature of individual andcollective action for the management of communal water resources, muchof it informed by common property resource management theory. Such

theory is underpinned by the assumption that institutions shape individual

Journal of Human Development Vol. 8, No. 2, July 2007

ISSN 1464-9888 print/ISSN 1469-9516 online/07/020223-22 # 2007 United Nations Development Programme


2/23

behaviour in collectively desirable ways. The participation of individuals innegotiation and decision-making around collective arrangements isassumed to positively influence rule-making and resource distribution.Individuals are seen mainly as ‘resource appropriators’ cleverly deploying

sociocultural norms and positions in maximizing their access to and use of resources.However, such theories have been criticized for deficiencies in their

modelling of the links between individual agency and collective action.Heavily reliant on institutions as the analytic lens through which behaviour can be understood, these theories are limited in their modelling of thesocial/historical/political formation and location of individuals (Mollinga,2001; Agrawal, 2005). Derived from methodologically individualistassumptions, institutional theory relies heavily on concepts of theboundedly rational individual who may exhibit behaviour on a spectrum

from ‘‘saint’’ to ‘‘sinner’’, from ‘‘rational egoist’’ to ‘‘conditional co-operator’’ (Ostrom, 2005). While the interaction of ‘endogenous variables’ with individuals is recognized as shaping action, institutionalism (drawingon rational choice theory, game theoretic models and evolutionary theory)explains human behaviour primarily through varying responses toincentives, rules and sanctions.

Concerns about unequal ability to shape decision-making are notprominent in institutional theory, in which people may be categorized asrule makers or rule followers (Ostrom, 2005). Such theories tend to modelagency as decision-making and are largely concerned with the choices thatpeople make rather than their ability to make those choices (Agrawal,2005). Similarly, equity of outcome is not a major concern, the emphasisbeing on the sustainable management of the natural resource, althoughinstitutions are thought to work best when all those affected by their rulesare involved in decision-making.

In development policy and practice, resource managing institutionsare increasingly seen as vehicles of local governance, as manifestationsof community and as the channels through which equity goals can bepursued. Purposive individual acts, particularly when aggregated ascollective action, are assumed to be both instrumentally useful andpotentially empowering. Local institutions such as Water User Associations, supposedly foster the involvement of resource users inmanagement and, more broadly, the potentially transformatory articula-tion of their needs, the negotiation of claims to rights and resources.The concept of water governance, linking levels of analysis from theinternational to the individual, is being widely applied in the water sector (Cosgrove and Rijsberman, 2000; WWAP, 2003) and linked to theachievement of the Millennium Development Goals (DFID, 2005; UnitedNations, 2005; WWC, 2006). There is, however, relatively little analysisof how good water governance manifests at the micro level, how theactions of individuals and communities contribute to it (Franks and

Cleaver, 2007).

F. Cleaver

224


3/23

In recent ‘post-tyranny’ thinking about participation there is aperceived need to link the micro-politics of participation to goodgovernance more broadly and to the workings of the state. This leads toa concern with promoting a radical participatory form of citizenship

(Hickey and Mohan, 2004), in which rights can be claimed from below. While asserting that such citizenship works best within the context of a wider societal project of social justice, this perspective optimistically viewsindividual agency as the range of socio-political practices through whichpeople can increase control over resources and extend their status. (Theextension of the capacity to exercise individual choice is also a centraltenet of the capability approach with which other papers in this volumeare concerned.) Important to such ‘post-tyranny’ thinking are concepts of ‘new spaces’ and ‘new rules’ for the promotion and articulation of suchcitizenship and rights (Cornwall, 2004; Kesby, 2005).

The underlying assumptions made by policy-makers about individualagency are crucial to the effectiveness of policy (Greener, 2002). In thecase of water, the active participation of marginalized people ininstitutions for collective management is assumed to ensure equity goals. With equity concerns in mind, in this paper I draw on a range of theoretical insights to explore better understandings of the numerous ways in which individual human agency is shaped and exercised in respectto collective resource management. In so doing I touch upon someinteresting dilemmas: Why are some individuals better placed to shapepublic decision-making than others and what are the differential costs andbenefits of such participation? Is non-participation an active exercise of

agency or a symptom of crippling structural constraint? By what authority is collective action legitimized, individual claims to resources asserted andmaintained, representations of others made? Critically, how do weunderstand the balance between the individual capacity for autonomousaction and the essentially relational nature of exercising agency in thesocial world? And finally, what are the links between the capacity toexercise agency and the effects of that agency?

Theoretical insights

In this section I draw on a selection of theoretical approaches to enrichthinking about the relationship between individuals, institutions and theoutcomes of collective decision-making over natural resources. Althougheclectic, these theoretical perspectives commonly emphasize the need toplace understandings of agency in wider contexts and frameworks, andthink beyond the assumption of agency only as purposive action. Thefollowing section then draws heavily on the ongoing debate in social theory about the balance between agency and structure, between enablement andconstraint. It supplements this debate with points drawn from contempor-ary theorizing about governmentality (Dean, 1999; Agrawal, 2005) and

about gendered empowerment (Kabeer, 2000; Kesby, 2005).

Understanding Agency in Collective Action

225


4/23

Enablement and constraint

In social theory, agency is seen as the capability, or power to be theoriginator of acts and a distinguishing feature of being human. Reflexivity and the ability to act purposively commonly feature in definitions of

agency. For example, Weber characterized human action as shaped by arationality consisting of self-consciousness, reflection, intention, purpose

and meaning (Rapport and Overing, 2000). Agency is commonly conceptualized as relational; it does not exist in a vacuum but is exercisedin a social world in which structure shapes the opportunities andresources available to individuals, in which appropriate ways of beingand behaving are not simply a matter of individual choice. While mostsocial theorists embrace the duality of individuality and relationality, thereare varying emphases on the balance between enablement and constraint.Debate concerns the extent to which the exercise of agency is generative of

individual transformations and social change, and to how far it is subjectedto the ‘discipline’ imposed by social norms and sanctions.

Giddens (1984) and Long (1992, 2001) emphasize the variationresulting from the exercise of agency. They recognize the structuralconstraints within which agents operate, the non-reflexive nature of mucheveryday practice, and the intended and unintended effects of individualactions. However, they are largely optimistic about the possibilities of individual agency, seemingly assuming that the exercise of reflexive agency and the conscious deployment of various forms of capital by the individualcan overcome the constraints of habitus (Greener, 2002). For example,

Long suggests that the ‘‘individual actor has the capacity to process socialexperience and to devise ways of coping with life, even under the mostextreme forms of coercion’’ (1992, p. 22) and that agency is about theability to choose levels of ‘‘enrolment’’ in the projects of others and toexert influence to enrol others in one’s own project. Thus, even thecompromise of ‘‘partial enrolment’’ is seen as a strategic positioning of actors in response to conflicting priorities and critical events, ofteninvolving ‘‘deferral, accommodation, negotiation, selective appropriationand distantiation or absenteeism’’ (Arce and Long, 2000, p. 3). Giddens(1996), mostly writing of contemporary western society, emphasizes theability of agents to respond to the challenges that society throws at them,to confront and deal with risk.

Bourdieu (1977) emphasizes how strongly habitus (a set of dispositions that incline agents to act and react in different ways) shapesagency and the ways in which hegemonic elites shape habitus. For himindividuals have agency, but the kind of agency they have is partially prescribed by the culture of which they are members. Archer (2000)similarly emphasizes the constraining effects of culture and the difficultiesof acting beyond the roles in which society places us, while Douglas (1987)expounds the many ways in which social institutions (the accepted ways of thinking and doing) constraint us, often invisibly. However, crucially, for

Bourdieu agency is not simply comprised of reflexive action, but strongly

F. Cleaver

226


5/23

constituted by non-reflexive practice; agents are potentially capable of generating change through both purposive acts and routine practices(Greener, 2002).

The co-existence of factors of enablement and constraint is a tension

addressed by theorists exploring the ways in which individual agents areformed and explaining the intermittence and unpredictability of changeeffected through agency (Henriques et al ., 1984; Mc Nay, 2000). Agrawal(2005) favours the term ‘subjects’ over agents for its ‘productive’ambiguities; for him it implies a notion of both creative agents and subjected individuals. Mc Nay rejects a Foucauldian over-emphasis onsubjection of the individual in favour of an analysis, drawing onBourdieu, which shows how creative agents are sometimes able toovercome constraint and generate transformational change. She empha-sizes though that self-conscious creativity is not generalizable but

a complex and evanescent feature of action (Mc Nay, 2000, p. 22). Themultiple and dynamic (and potentially contradictory) positioningof agents is much emphasized in these writings, with Mc Nay sugges-ting ‘‘the need for a more precise and varied account of agency toexplain the differing motivations and ways in which individuals andgroups struggle over, appropriate and transform cultural meaning andresources’’ (2000, p. 4).

Development policy relating to participation in natural resourcemanagement tends towards the optimistic Giddenesque approach toagency, emphasizing the instrumental, empowering and transformatory effects of individual participation in collective action. Institutional theory assumes that behaviour is either rationally strategic or learnt; that peoplechoose either to cooperate or to free ride on the collective good. In thispaper I illustrate some of the multiple ways in which agency is exercisedthrough social relationships, as well as the constraints that limit theopportunities, especially for poor people, to exercise agency.

Allocative and authoritative resources

Collective action for natural resource management is enacted through a variety of bureaucratic and socially embedded institutions. In institutional

processes people draw on various forms of authority as sources of legitimation; the ability to exercise authority (in the name of thecollectivity) is implied in understandings of representative local institu-tions for resource management

The individual exercise of agency is tied into relations of power andauthority. We have seen how, in much social theory, to be an agent (toexercise agency) depends on the ability to act, to choose a course of action(or inaction), the capability to make a difference—in sum, to exercisesome sort of power. For Giddens, individuals may do this by accessing anddeploying various sorts of allocative or authoritative resources. Using the

concept of ‘resources’ as the material and non-material properties of social


227


6/23


7/23

study of the collective governance of forest resources in India. He placesindividual agency in a wider context by illustrating the linkages betweenthe policies of decentralized local government, the emergence of new localized ‘‘regulatory communities’’ (village level forest councils) and the

transformation of perceptions and actions of some of those ‘‘environ-mental subjects’’ who participate in community decision-making. Agency is exercised in a variety of ways in multiple arenas. The

concept of ‘regimes of practice’ central to governmentality implies aplurality of spheres within which interactions take place. Dean suggeststhat there is a plurality of governing agencies and authorities, of aspects of behaviour to be governed, of norms invoked, purposes sought and of effects, outcomes and consequences (Dean, 1999, p. 19). Such complexity may encompass pluralities of scale as in Agrawal’s study and multiplearenas of overlapping networks within which interactions around

governance and self-governance are formed. The concept of plurality finds direct echoes in current thinking in natural resource managementand has relevance both for the opportunities and constraints for individualaction.

Thinking through the nature of complexity of the collective manage-ment of water, Merrey et al. (2007) identify three inter-related aspects of plurality. Polycentric governance involves a various actors and organiza-tions at all levels, institutional pluralism relates to the multiplicity of rulesand procedures applicable to a specific issue (e.g. the congruent ‘modern’and ‘traditional’ laws governing water in many African countries) andmultifunctionality refers to the range of different uses to which water isput (and the associated variety of values attached to these by differentactors). Such analyses of pluralism well indicate the complexity of arenasin which individual agency is exercised.

In previous work on the collective management of water and land Ihave outlined the concept of institutional bricolage; the piecing together through conscious action and non-reflexive practice of institutionalarrangements from existing norms, practices and relationships (Cleaver,2002). I have presented bricolage as a sort of institutional do-it-yourself,rather than a more explicitly conscious and rational form of institutionalengineering or design. Implied in the concept of bricolage is a multiplicity of potential resources to draw on and a variability in the capacity of individuals to act as bricoleurs.

Positions vary as to whether plurality offers opportunity or constraint.Ostrom (2005) and Merrey et al . (2006) seem to imply that plurality in theform of polycentric channels for resource governance provides scope for more negotiability, flexibility, room for manoeuvre. This point isreinforced by those thinking about the potential for ‘new spaces’ toprovide sites for the renegotiation of norms and inequitable socialrelationships (Cornwall, 2004; Kesby, 2005). However, Odgaard (2002)and Maganga (2002), writing of plural systems for the governance of land

and water resources in Tanzania, suggest that these serve to reproduce


229


8/23

inequalities in multiple ways, to consolidate the marginalization of the weak.

Effects of exercising agency

The working of power through plural institutional settings shapes theeffects of agency, the ability of individuals to effect significant difference intheir lives. Power may be exercised and accepted both consciously andunconsciously; the ‘self-disciplining’ of agents; their acceptance of relations of inequality, may ensure the reproduction of power througheveryday acts and relationships; ‘‘Self-disciplining actors often enrolthemselves into the projects of others’’ (Kesby, 2005, p. 2047). Whileseveral writers emphasize the everyday reproduction of relationships of power, there is also a common assumption in the literature that theexercise of agency is not just about choice but about challenging power relations and the ways that things are commonly done. Thus reflexiveagents supposedly can question norms, challenge inequitable distributionof resources, claim and extend their rights. For these writers, agency isseen as empowering, and is potentially transformatory—although, as McNay (2000, p. 5) notes, the individual and innovative ways in whichindividuals respond may hinder as well as catalyse social change.

Not all choices are equal, not all agents are equally placed to makethem and similar actions may have very different outcomes. Agrawal(2005) shows how apparently equitable forest rules have unequal impacton different people, who also respond variously to them because of theopportunities and constraints offered by their subject positions. In thecommunities he studied, the forest guards detected and reported womenand harijans most frequently for violating the forest rules; such offenders were not only more frequently detected, but they paid their fines morepromptly than offending high-caste neighbours. In this case we couldspeculate that the perception by others of the social position of agents,and their dependence on good reputation and relations of patronage,shapes their ‘rule-following’ behaviour.

For an acknowledgement that not all choices are equal we turn togender theorists. For Kabeer the issue is not so much whether an

individual can choose to act in a particular circumstance, but about how much control they have to make strategic decisions that shape their lives.She distinguishes between first-order and second-order choices; theformer referring to decisions fundamental to the shape of a persons life,the second to choices that affect quality of life but do not constitute itsdefining parameters (Kabeer, 2000, p. 28). Importantly, for Kabeer, theexercise of agency is not just about potential ability to exercise choice, butabout the real effects these choices can have. In participatory naturalresource management, the participation of individuals in itself is assumedto lead to benefits to them, although this is demonstrably often not the

case.

F. Cleaver

230


9/23

‘Unpacking’ agency in natural resource management

Let us now draw on some of these theoretical debates and insights toexplore how agency is enabled and constrained in the collective manage-ment of natural resources. The following sections cannot be a compre-

hensive illumination of the workings of agency, but they offer assortedsnapshots of the complexity of the exercise of such agency in social life.Through these illustrations I aim to explore some of the tensions in therelationship between agency, collective action and the achievement of equitable outcomes.

Moral world views

As we have seen, social theorists often emphasize how much of our daily conduct is taken for granted, how many of our interactions so routinized

that we do not often subject them to discursive scrutiny. One of thedebates that arises is the extent to which such actions can be made visibleand be questioned; when and how can the norms be negotiated?Institutionalist theories of common property resource management err towards the strategic; they place great faith in the ability of people tobargain and negotiate, to strategically respond to incentives and toposition themselves in relation to collective norms.

A body of work in natural resource management documents complex and wide-ranging world views within which people, both consciously andunconsciously, place themselves and their actions (Adams et al ., 1997;

Mosse, 1997; Cleaver, 2000). In many African countries a form of moral– ecological rationality links people’s perceptions of their own actions withthe natural and supernatural worlds. For example, in villages where I worked in Zimbabwe, people perceived that rainfall, the associated fertility of the land and the ability to secure livelihoods was critically affected by proper observation of custom, the showing of respect to elders and thelevels of cooperation between people. Negotiations over access toresources, and the rules and sanctions deployed in the management of land and water was strongly shaped by such ‘moral–ecological rationality’.These often unspoken cosmological assumptions influenced individuals’ willingness to abide by collective decisions and to broadly comply withunwritten norms of resource use.

Moral world views often include strongly gendered and socially stratified ideas about proper behaviour and the rightful place of individuals of different social identities. A notable aspect of suchcosmologies (which may include ideas about sacred places, spirit homesand a direct relationship of people with ancestors) is that they link ‘proper’ ways of individuals behaving in the social world with effects inthe natural and supernatural ones. In the Zimbabwean case mentionedabove, moral–ecological rationality was often invoked by elders toexplain drought and deforestation as related to the unacceptable

behaviour of young people (e.g. to condemn young women having


231


10/23

terminations of pregnancy or ‘boys’ for not respecting their fathers). It was also referred to frequently to bemoan the lost social solidarity of thepast, linked to a lost environmental fecundity, a theme strongly reflectedin invocations of similar world views in the Usangu basin in Tanzania

(SMUWC, 2001).But such socially embedded views can be subject to some discursivescrutiny and to manipulation. Mercy Dikito-Wachtmeister studied thereasons for women’s participation or non-participation in village water committees in Zimbabwe. She notes women’s accounts of how belief in witchcraft shaped their unwillingness to speak out, to disagree with othersin matters of water management. She also observed that while fear of witchcraft seemed real to a majority of participants, she suspected some of strategically using this belief to excuse themselves from onerous and time-consuming participatory decision-making. Thus, in reference to a

disagreement about whether a borehole should be used for garden watering or not, she contrasts the views of two women. The firstrespondent claimed: ‘‘[…] I prefer to keep quiet because there are peoplehere who use lots of witchcraft. If you exchange words with these people itmakes their magic work on you. The best thing is to keep quiet, and notgive them a chance’’ (Dikito-Wachtmeister, 2000, p. 219). Another respondent however says of the water committee members: ‘‘I think some of them will want to pretend that the reason they do not do their duties is because of fear of witchcraft, when they are either lazy, or can’tbecause of other responsibilities. Witchcraft has become an excuse for everything. […] They use it to explain everything that has not worked out well or to justify their lack of participation in community things’’ (Dikito- Wachtmeister, 2000, p. 219).

It is often unclear where the boundaries of reflexivity are; when and to what extent people can choose to negotiate the norms. At these rather blurred boundaries there may be room for creative adaptation to occur.For example, in the Usangu plains in southern Tanzania in-migrantpastoralists sought some kind of social acceptance from indigenous Sanguagriculturalist neighbours by participating in Sangu ceremonies. They sought blessing for the well-being of their cattle from the keeper of animportant Sangu shrine, they contributed to ceremonies to ensure good

rain and good grass in the plains. The in-migrant pastoralists, littleintegrated into local society, were nevertheless relatively rich in theresources needed to participate in these ceremonies; making contribu-tions of grain, chickens and even cattle (SMUWC, 2001). So cultural beliefsmay be accepted by many, partially negotiated by some, and provide anauthoritative resource for others.

Complexity of individual identities

People do not exercise property rights over natural resources simply as

‘resource appropriators’, but as people with rich social identities.

F. Cleaver

232


11/23

Individual identities and associated motivations are complex and multi-layered, as are the channels through which resources are accessed.Plurality manifests in agents themselves.

For example, evidence from a number of countries (Odgaard, 2002;

Rao, 2003; Arora-Jonsson, 2005) suggests the complexity of identities andrelationships through which women’s land rights are exercised. Womenmay claim rights to natural resources as legal and equal citizens but alsothrough their subject positions as daughters, wives, mothers as membersof a particular caste or ethnic group. To these women, living their lives within marriages and kin groups, exercising their agency through publicinstitutions may not always be the preferred option. In Arora-Johnson’sstudy of women’s activisim in forest communities in Sweden and India,she found that in both places women saw forest issues as inseparable fromsocial relationships and wider village issues. Change required the exercise

of agency in both domestic and public spheres. An account of irrigation management in Nepal shows how some women drew on their family and gendered identities rather than those asactive citizens, to secure water. Constrained by prevailing ideas aboutproper gender roles, some women did not participate in the irrigationassociation but secured their water partly through the public participationof male members of their own households, partly through the patronageof kin and partly through stealing and cheating. They justified their non-participation by referring to gendered norms; the inappropriateness of them attending meetings held in the evening, the demands of their domestic work (Zwarteveen and Neupane, 1996).

Approaches to participatory natural resource management tend tooverlook the complexity of individual identities to categorize people intotaxonomic groups, and to assume motivations linked to the labels of ‘farmers’, ‘women’, ‘pastoralists’, ‘irrigators’, and so on. Institutionalistunderstandings of collective action are also rule-oriented—they see thecreation of collective rules and the sanctions applied to rule-breaking asessential to effective action. But how much capacity do people withdifferent social identities have to exercise agency in respect of these rules?The agency of individuals is clearly exercised in relation to the perceptionby others of them. Dikito-Wachtmeister (2000) notes the tendency in the villages she studies for more powerful women to be excused from rulefollowing; examples included someone associated with the ruling party, a woman who employed such ‘‘vile language’’ that people were afraid toconfront her, and a well-off woman who administered the church fund for people in misfortune. Social placement of an individual critically affectstheir ability to ‘bend’ collective rules.

Unequal interdependence of connected livelihoods

Participatory approaches to common property resource management tend

to be based on the assumption that groups of individuals can collectively


233


12/23

manage resources on the basis of some sort of equality derived fromcommon livelihood interests. Indeed, interrelatedness is seen as funda-mental to the building of social capital, the moulding of social norms, thegeneration of collective action and, hence, the further building of relations

of trust and cooperation (Pretty, 2003).This assumption of equality of interest of agents is underpinned by the myth that people following similar practices (in relation to naturalresource use and management) may be equally placed to shape thesepractices. This clearly is not the case. For example, studies analysinggender relations show us that even where men and women in a householdor community perform similar tasks and have similar interests in theresource, decision-making power, the ability to mobilize authoritative andallocative resources, and the outcomes of this ability, may be skewed.

The unequal interdependence of people with similar livelihood

interests is complex. Discussing indigenous forest management in Uganda,Nabanoga (2005) notes very similar knowledge about trees, and practicesof tree management between men and women. However, she shows how prevailing cosmologies, myths that women are not significantly involved intree protection and use, and land tenure arrangements that favour menmean that men dominate in decisions about tree planting (even of women’s tree planting). Similarly, participation in tree-related activitiesoften depends on social status, and the outcomes confer status; men’sownership of fig trees was historically an important status symbol, andability to produce bark cloth in quantity conferred social status. By contrastthe production of handicrafts from palm trees, and the collection of particular wild vegetables, conferred social status on women but wereconsidered only the last resort of the poorest and unmarried men(Nabanoga, 2005).

Rao (2003), however, usefully points out that the exercise of agency involves mutuality and interdependence as well as relations of dominationand subordination. Indeed, the two are mutually implicated. Commoninterests are embedded in relations of unequal interdependence, ofteninvolving arrangements of patronage. For many poorer and subordinateindividuals and groups, access to resources is primarily exercised throughinequitable social arrangements. Thus Joshi cites a dalit woman fromIndia, excluded by higher caste neighbours from equitable access to water,explaining why she will not object, even in the context of a gender-focusedparticipatory development intervention: ‘‘My neighbours are important tome, no matter what they do. I need their support for my family’s daily existence.’’ (Joshi et al ., 2003, p. 2)

Positions of relative power do not imply unfettered ability to exerciseagency. Recent research explored the interaction between individuals,community-based workers and local institutions at in participatory development initiatives in Tanzania and South Africa. Although thecommunity-based workers were structurally relatively privileged, they

frequently expressed frustration that community members did not ‘‘play

F. Cleaver

234


13/23

their part’’ properly, absented themselves, declined to cooperate, gavefalse information, and thus constrained the effective exercise of agency by the workers (Understanding Community Action, 2006).

A similar point about skewed interdependence is made in Masaki’s

study of caste-based management of irrigation infrastructure in Nepal(Masaki, 2004). The paper chronicles attempts by members of thelabouring caste to exercise both individual and collective agency inreshaping ‘customary’ labour obligations to be less onerous to them. But while the landowning elite need the lower caste villagers as labourers ontheir irrigated land, the relationship is one of unequal interdependence asthe labourers are unlikely to survive at all without the patronage of thelandowners. Relations of interdependence do offer the possibility of negotiation of rights but such interdependence exists alongside structuralinequality. Immediate livelihood imperatives, combined with unequal

authoritative and allocative resources, may shape the exercise of agency indirections productive of social harmony and conflict avoidance, rather than effective resistance or explicit negotiation of rights.

Relations of mutuality are plural in their manifestations. In muchinstitutionalist theorizing, the focus on resource users and institutionalfora means that the multiple locations of decision-making and theshaping of norms are often overlooked. In particular, intra-householddimensions of the shaping of agency and decision-making are oftenignored. Bolt and Bird (2003) schematize the layers of overlappingfactors (gender, age, birth order, physical ability and relationship tohead of household) that may disadvantage individuals within house-holds; rendering them less likely to shape decisions and to accessresources. Livelihoods research in Tanzania showed how even entre-preneurial women were restricted in their agricultural business activitiesby the location of their marital home and expectations of wifeliness(SMUWC, 2001), while young women unhappily excluded themselvesfrom volunteering for training as Village Activators, citing the influenceof elders as key restrictions on their choice of social and economic roles(Cleaver and Kaare, 1998).

Structure and voiceSocial theorists debate the extent to which structural placement shapesthe ability to exercise agency, even shapes the person it is possible tobecome. Much thinking about the participatory management of naturalresources assumes that if the forum for decision-making is local and therules for access and distribution are fair, then all will (potentially) be ableto participate and to benefit. We have already seen above how this wasnot the case in Agrawal’s forest communities—despite local decision-making arrangements and ‘‘fair’’ rules, patterning of class caste andgender shaped engagement and outcomes for different people (Agrawal,

2005).


235


14/23

The ways in which structure shapes agency are numerous but here I will concentrate on illustrating ways in which structures of gender andclass shape the ability of individuals to articulate in public fora and so toinfluence the management of resources.

The literature on participatory natural resource management isremarkably silent about the effects of structure on norms of articulation,the empowering or disempowering effects of participating through arepresentative at a meeting, and whether strategic mutedness is evidenceof the conscious exercise of agency or merely another manifestation of thehegemonic inequitable social structures. Post-tyranny literature onparticipation and gendered empowerment, however, does tackle theperplexing issue of voice. It recognizes both that participatory spaces areimbued with power relations that may result in the conscious andunconscious self muting of disadvantaged people (Kesby, 2005) and is

optimistic about the possibilities of creative participatory methodologiesfor disrupting ‘business as usual’ and facilitating the articulation of alternative views (Cornwall, 2004).

Hierarchies among women, concepts of proper behaviour and virilocal marriage arrangements constrain the agency of some. One young woman explains constraints on her participation in water management toa researcher: ‘‘I cannot be seen to be taking a lead role at meetingsattended by older women as this could be perceived as being disrespectful.I am a young woman who has just been married here for a few years, so Icannot be speaking often and taking a lead in these things. If you are still young and have young children like me, nobody takes you seriously,nobody respects you’’ (Dikito-Wachtmeister, 2000, p. 221).

However, we have already seen how moral rationalities, concepts of the right way of doing things, may unwittingly shape people’s agency, butmay also be consciously deployed and negotiated. Dikito-Wachtmeister (2000) postulates the transfer of household bargaining models, based onnegotiations of proper gender roles, to the more public domain. The women she studies used the strategy of ‘buttering’ (or flattering) men,both in the household and in public, and invoking gendered norms of desirable behaviour to avoid water management work. Dikito- Wachtmeister documents a meeting in which women consciously anddeliberately ‘buttered’ men, through praise and positive reinforcement of their manliness, in order to persuade one particular man to take on a water management role. They adopted a humble demeanour in keeping with tradition and respect when doing this. The man then asserted hisintention to be strong and strict in the role. Dikito-Wachtmeister notes thestrategic use on other occasions of gendered norms of mutedness; by deliberate and collective non-contributions to a meeting, women wereable to gain the concerned attention of the chairman and an invitation toelucidate their views.

The unequal co-existence of empowerment and subordination in

gendered negotiations and expression of voice is common. Tukai (2005)

F. Cleaver

236


15/23

documents an attempt to empower women in a pastoralist community inTanzania through a participatory water project. The aim of promoting women’s role as active citizens was partially achieved; they increased their representation on the village council, on the water committee and claimed

greater confidence in discussions both at village and household level.However, within the timescale of the project, their increased capacity toexercise agency and voice could not counter the wider relations of patriarchy in which they were embedded. Despite their gains in citizen-ship, they could not secure access to water on better terms with men.

A concern with gendered participation and voice is often translatedinto the counting of numbers of women participating. But this throws littlelight on the gendered dynamics of meetings, or on gendered norms of articulation and who they privilege or oppress. Research in Tanzaniashowed women participating in village meetings on a different basis to

men. Although meetings followed accepted patterns, with men sittingseparately from women, and women speaking less, this could not bemistaken for lack of participation on the part of women. Femaleinformants suggested that they deputed one or two women, known for their eloquence, to speak for them. These were not formal representativesand might be decided upon during the course of the meeting, according toneed. But in this way women felt they could have their voice heard. At arecent village meeting there had been a debate about how to use surplus water funds. Individual men had spoken in favour of funding a beer drink and feast, women had deputed their speakers to argue in favour puttingthe money towards providing another water source. The triumph of water over beer illustrated the potential effectiveness of the women’s way of articulating their interests through representatives (Cleaver and Kaare,1998).

Space and location shapes the ability to exercise agency through voice. In Arora-Jonsson’s study of forest activism, women in India pushedfor formalization of their group as they saw this as necessary for greater recognition and bargaining power, whereas in Sweden it was by maintaining informality that women felt able to sustain a forum for collective articulation (Arora-Johnsson, 2005).

Embodiment

Accessing and using natural resources is often a very physical activity, andthe participation in decision-making involves embodied presentations of self. And yet the enacting of participatory natural resource managementthrough corporeal human conduct is often neglected in analyses; littleattention is given to how embodiment enables and constrains activeparticipation (Jackson, 1998). Bourdieu emphasizes the body in hisdiscussion of the habitus —for him, the habitus moulds the humanphysique and the human body is an important signifier in social

interaction, expressing social status and power as much as communicative


237


16/23

intent. Deportment, ways of dressing and speaking are produced by andreproduce social structure (King, 2005, p. 223). Dikito-Wachtmeister’sexample of the humble demeanour of women imploring a man to take ona role of responsibility and management illustrates this point.

It could be argued that able-bodiedness is key to the effective exerciseof agency in participatory natural resource management. Public participa-tion, accessing resources, and interacting with others is shaped by physicalcapabilities; the healthy and able-bodied are most likely to be able toexercise effective agency in these respects. Research in Tanzania showedpoor families that suffered multiple constraints on their physical ability toproduce livelihoods, to travel and to engage socially, gradually droppedout of associational life and became more and more limited in the ways in which they could use resources. Constraints on able-bodiedness andmobility ranged from diseases and disablement to the limitations imposed

by caring for sick relatives, large numbers of very small children or carryinga difficult pregnancy. People suffering such problems were unable tocontribute to collective action, attend meetings or social events or pay feesto clubs and associations. They were also unable to access distantresources, relied on very local supplies of grass, wood and water, or had topay to buy these in small quantities (Cleaver, 2005). In other words, they were unable to fully exercise their agency in the maintenance of their livelihoods. Embodiment and agency are recursively linked. Campbell and Jovchelovitch (2000) also note the fact that people who lack the power toshape their life courses in strategic ways are significantly less likely to behealthy, a factor that can create a vicious cycle of disempowerment, andpoor health.

Additionally, even for healthy, mobile, able-bodied people, embodi-ment creates some challenges to the exercise of agency. In a recent projectresearching community-driven development in Tanzania and South Africa,community-based workers were asked to keep reflective diaries chron-icling their work. The diaries detailing home visits, community mobiliza-tion and training activities are characterized by daily references to the workers’ own physical state (being hungry, tired, hot, cold or wet) andthe detrimental effects of this on their activities. The diaries illustrate wellthe ways in which both strategic action and everyday practice is shaped by

the experiences of our embodied selves (Understanding Community Action, 2006).

Not all agents have equal command over their physical selves. Naturalresources are often accessed through deploying the labour of others;children and junior wives sent to fetch water, herdsmen employed to takecattle to grazing lands, youth required to collect and carry firewood for their families. Such people are often the ones who abide or break the‘rules-in-use’ relating to the natural resource, and yet may not be involvedin participatory decision-making. How far the effective exercise of agency implies command over one’s own labour, and how far it can be exercised

by commanding the labour of others, is a moot point.

F. Cleaver

238


17/23

Access to natural resources is also enacted through gendered bodies.Taboos around menstruation, pregnancy and childbirth may restrictaccess to natural resources (particularly to water points), collectiveceremonies and public appearances (Upperman, 2000). Additionally,

gendered ideas about strength and physical ability, linked to divisions of labour and livelihood priorities, shape access and participation (Jackson,1998, 2000). For example, in Zimbabwe, men rarely headload water butgenerally deploy some means of transport (a wheelbarrow, cart, donkey or bicycle) while women and girl children carry heavy loads in bucketson their heads. Different physical norms of accessing the resourceentwined with gendered livelihood priorities shape preferences for usingparticular waterpoints; men may be more willing to travel further to waterpoints with plentiful supply (as they could take more water at onetime); women might weigh up the benefits of closeness with theconstraints of queuing and the quality of water—definitions often alsorelated to the amount of work implied (hard water makes washingclothes more laborious).

We have seen above Dikito-Wachtmeister’s analysis of the ways in which women deployed physical expressions of respect and humility of public meetings, and how gendered forms of articulation take differentphysical forms. Dikito-Wachtmeister goes on to speculate about the impactof body size on the effective exercise of agency in participatory decision-making. She notes the concept of a ‘‘big man’’ or ‘‘big woman’’ usedlocally as not only denoting body size but wealth and prestige, andspeculates that large body size may generally bestow authoritativeresources in public decision-making contexts in Zimbabwe.

Emotionality: unconscious motivations of intended actions

Social theorists recognize the importance of unconscious motivations of conscious actions, the unconscious self-disciplining of agents and their internalization of hegemonic norms. They suggest that not all individualacts are the results of conscious strategy, but of habit, routine and the workings of the unconscious mind (Bourdieu, 1977; Giddens, 1984). While Giddens tends towards an emphasis on the nature of agency as

reflexive action, he nonetheless conceives three types of consciousness;the discursive (those actions and beliefs that agents are able to bring intodiscursive scrutiny), the practical (the taken-for-granted everyday prac-tices that are so part of habit routine and precedent that they are rarely subjected to scrutiny) and the unconscious. Henriques et al . (1984) arguefor the important insights that psycoanalysis can offer for understandingindividual action, offering the concepts of irrational thought and theunconscious, the complex intertwining of cognition and action and theinfluence of the agent’s history and experience on present behaviour. Inthinking about participation in natural resource management, there has

been an over-emphasis on reflexive action, on deliberative purposive


239


18/23

strategizing and a relative neglect of the impact of the practical and theunconscious on people’s agency.

Both conscious and unconscious emotions are critical in shapingpeople’s sense of self-efficacy and their social relationships (Myers, 1996),

and therefore the extent to which they publicly engage and assert their rights. Both Kabeer and Agrawal emphasize that, for potential courses of action to become real, they must be both materially possible and beconceived by the agent to be within the bounds of possibility. ‘Imaginedautonomy’ is therefore an important factor in people’s conceptualizationof their own agency. Campbell and Jovchelovitch (2000), writing of thesocial psychology of health, link the individual’s perception of self-efficacy to their placement in wider networks of social support and macro-socialfactors shaping perceived self-efficacy such as ethnicity, class and gender.This layered analysis of the psychology of individuals in communities is auseful antidote to the methodological individualism of bounded ration-ality approaches.

Mumby and Putnam (1992), writing of models of behaviour inorganizations, further critique bounded rationality assumptions aboutindividual motivations and preferences through a productive dialogue withalternative assumptions about ‘‘bounded emotionality’’ (although they reject the supposed dichotomy between ‘rationality’ and ‘emotionality’).They characterize bounded rationality as informed by intentional, reasonedand goal-directed models of human action, ‘‘bounded’’ by the limitations of context and the ability to attain and process information. Institutionally, thismodel translates into the need for role-based hierarchies and rules,

boundaries, the divorcing of physical labour and emotions from decision-making. By contrast (although carefully not counter-posed as an alternative)they suggest a model of bounded emotionality based on the recognition of inter-relatedness, tolerance of ambiguity, community and relational feelingsand embodied self-identities. Some of the desire for relatedness and thephysicality of agency is well represented in the discursive justification of a young (cattle-less) man in Zimbabwe for volunteering to become acommunity grazing policeman. This young man did not directly benefitfrom the rotational grazing of cattle, did not own or control land and wasnot old enough or wealthy enough to ordinarily command respect. He said:

‘‘To be a successful policeman you must catch offenders because thenpeople will say ‘that is a good policeman’ and then they will confide in youand bring problems to you. I am very strict about devoting time to policing,even in the rainy season when there is plenty to do in the fields, because thatis the only way to be a good policeman’’.

Conclusion

This paper has raised, and inevitably left unanswered, a number of questions about agency. Conceptually it leaves unresolved the question of

where the boundaries of agency lie; are all actions constitutive of agency,

F. Cleaver

240


19/23

whether conscious or unconscious, strategic or habitual? In thinking aboutthe implications for the collective management of natural resources, andthe prospects for participatory governance, further questions anddilemmas arise. We need to think further about how far participatory

development initiatives require people to exercise agency within theexisting rules of the game, within the parameters of the programme or project on offer, the priorities of the implementing agency. Additionally, if we are concerned with securing more equitable development for millions,an emphasis on the diverse possibilities of individuals exercising agency isnot enough. Equity concerns and the scale of development goals requireus to recognize the importance of structure in patterning inequality and inshaping the probable impacts on interventions. If we are interested in how institutions that shape access to rights and resources work, and how effective and equitable their outcomes are, we need better understandingsof why and how individuals act, and the balance between empowermentand constraint in such actions. Critically, we need to expand our gazebeyond visible public forums of decision-making to the places whereagency is exercised through practice. In particular, we need analyses thatilluminate various aspects of agency in decision-making, with a rigorousand differentiated scrutiny of the effects of such decision-making ondifferent actors. Indeed, it seems to me that without a focus on the effectsof exercising agency for differently placed people in relation to others, afocus on agency-as-decision-making is ultimately sterile.

References

Adams, W., Watson, E. and Mutiso, S. (1997) ‘Water, rules and gender: water rights in anindigenous irrigation system, Marakwet, Kenya’, Development and Change, 28(4),pp. 707–730.

Agrawal, A. (2005) Environmentality: Technologies of Government and the Making of Subjects, Duke University Press, London.

Arce, A. and Long, N. (2000) ‘Reconfiguring modernity and development from ananthropological perspective’, in A. Arce and N. Long (Eds.), Anthropology, Development and Modernities, Routledge, London.

Archer, M. (2000) Being Human: The Problem of Agency, Cambridge University Press,Cambridge.

Arora-Jonsson, S. (2005) ‘Unsettling the order: gendered subjects and grassroots activismin two forest communities’. Doctoral thesis No. 2005:70, Swedish University of

Agricultural Sciences, Uppsala.Bolt, V. and Bird, K. (2003) ‘The intra-household disadvantages framework: a framework

for the analysis of intra-household difference and inequality’, CPRC Working Paper 32,Chronic Poverty Research Centre, University of Manchester, Manchester.

Bourdieu, P. (1977) Outline of a Theory of Practice, Cambridge University Press,Cambridge.

Campbell, C. and Jovchelovitch, S. (2000) ‘Health,community and development: towards asocial psychology of participation’, Journal of Community and Applied Social Psychology, 10, pp. 255–270.

Cleaver, F. (1996) Community Man of R.W.S in Zimbabwe, unpublished PhD thesis,

University of E.A, p. 212. .


241


20/23

Cleaver, F. (2000) ‘Moral ecological rationality: institutions and the management of common property resources’, Development and Change, 31(2), pp. 361–383.

Cleaver, F. (2002) ‘Reinventing institutions and the social embeddedness of naturalresource management’, European Journal of Development Research, 14(2), pp. 11–30.

Cleaver, F. (2005) ‘The inequality of social capital and the reproduction of chronic poverty’,

World Development , 33(6), pp. 893–906.Cleaver, F. and Kaare, B. (1998) ‘Social embeddedness and project practice: a genderedanalysis of promotion and participation in the Hesawa programme, Tanzania’, mimeo,University of Bradford for Swedish International Development Agency, Bradford.

Cornwall, A. (2004) ‘Spaces for transformation? Reflections in issues of power anddifference in participation in development’, in S. Hickey and G. Mohan (Eds.), Participation: From Tyranny to Transformation, Zed Books, London.

Cosgrove, W.J. and Rijsberman, F.R. (2000) World Water Vision: Making Water Everybody’s Vision, Earthscan, London.

Dean, M. (1999) Governmentality: Power and Rule in Modern Society, Sage, London.DFID (2005) Meeting the Water and Sanitation Millennium Development Goal ,

Department for International Development, London.Dikito-Wachtmeister, M. (2000) ‘Women’s participation in decision-making processes in

rural water projects, Makoni District, Zimbabwe’, Ph.D. thesis, University of Bradford,Bradford.

Douglas, M. (1987) How Institutions Think, Routledge, London.Franks, T. and Cleaver, F. (2007, forthcoming) ‘Water governance and poverty: a

framework for analysis’, Progress in Development Studies [http:www.splash;bradford.ac.uk/projects/] (accessed 15 May 2007), pp. 1–12.

Giddens, A. (1984) The Constitution of Society: Outline of the Theory of Structuration,Polity Press, Cambridge.

Giddens, A. (1996) ‘Affluence, poverty and the idea of a post scarcity society’, Development and Change, 27, pp. 365–377.

Greener, I. (2002) ‘Agency, social theory and social policy’, Critical Social Policy, 22(4),pp. 688–705.

Henriques, J., Hollway, W., Urwin, C., Venn, C. and Walkerdine, V. (1984) Changing the Subject: Psychology, Social Regulation and Subjectivity, Methuen, London.

Hickey, S. and Bracking, S. (2005) ‘Exploring the politics of poverty reduction: how are thepoorest represented?’, World Development , 33(6), pp. 995–1009.

Hickey, S. and Mohan, G. (2004) ‘Relocating participation within a radical politics of development: Critical modernism and citizenship’, in S. Hickey and G. Mohan (Eds.), Participation: From Tyranny to Transformation? Exploring New Approaches to Participation in Development , Zed Books, London.

Jackson, C. (1998) ‘Gender, irrigation and environment: arguing for agency’, Agriculture and Human Values, 15(4), pp. 313–324.

Jackson, C. (2000) ‘Men at work: labour, masculinities and development’, European Journal of Development Research, 12(2), pp. 1–22.

Joshi, D., Lloyd, M. and Fawcett, B. (2003) ‘Voices from the village: an alternative paper for the alternative water forum’, paper presented at the Alternative Water Forum, University of Bradford, 1–2 May [www.brad.ac.uk/acad/bcid/GTP/altwater.html] (accessed 15 May 2007).

Kabeer, N. (2000) ‘Resources, agency, achievements: reflections on the measurement of women’s empowerment’, in Razavi Shahra (Ed.), Gendered Poverty and Well-Being ,Blackwell, Oxford.

Kesby, M. (2005) ‘Retheorising empowerment-through-participation as a performance inspace: beyond tyranny to transformation’, Signs: Journal of Women in Culture and Society, 30(4), pp. 2037–2065.

King, A. (2005) ‘Structure and agency’, in A. Harrington (Ed.), Modern Social Theory: An Introduction, Oxford University Press, Oxford.

Long, N. (1992) ‘From paradigm lost to paradigm regained? The case for an actor

orientated sociology of development’, in N. Long and A. Long (Eds.), Battlefields of

F. Cleaver

242


21/23

Knowledge: The Interlocking of Theory and Practice in Social Research and Development , Routledge, London.

Long, N. (2001) Development Sociology: Actor Perspectives, Routledge, London.Maganga, F. (2002) ‘The interplay between formal and informal systems of managing

resource conflicts: some evidence from South-Western Tanzania’, European Journal of

Development Research, 14(2), pp. 51–70.Masaki, K. (2004) ‘The ‘‘transformative’’ unfolding of tyrannical participation: the corveetradition and ongoing politics in Western Nepal’, in S. Hickey and G. Mohan (Eds.), Participation: From Tyranny to Transformation? Exploring New Approaches to Participation in Development , Zed Books, London.

Mc Nay, L. (2000) Gender and Agency: Reconfiguring the Subject in Feminist and Social Theory, Polity Press, Cambridge.

Merrey, D., Meinzen Dick, R., Mollinga, P. and Karar, E. (2007) ‘Policy and institutionalreform processes for sustainable agricultural water management: the art of the possible’,in D. Molden (Ed.), Water for Food, Water for Life: A Comprehensive Assessment of Water Management in Agriculture, Earthscan Publications Ltd, London.

Mollinga, P. (2001) ‘Water and politics: levels, rational choice and South Indian Canal

Irrigation’, Futures, 33, pp. 733–752.Mosse, D. (1997) ‘The symbolic making of a common property resource: history, ecology and locality in a tank-irrigated landscape in south India’, Development and Change,28(3), pp. 467–504.

Mumby, K. and Putnam, L. (1992) ‘The politics of emotion: a feminist reading of boundedrationality’, Academy of Management Review, 17(3), pp. 465–485.

Myers, D.G. (1996) Social Psychology, McGraw-Hill, New York.Nabanoga, G. (2005) ‘Transgressing boundaries: gendered spaces, species and indigenous

forest management in Uganda’, Tropical Resource Management Papers, No 60,University of Wageningen, Wageningen, The Netherlands.

Odgaard, R. (2002) ‘Scrambling for land in Tanzania: processes of formalisation andlegitimisation of land rights’, European Journal of Development Research, 14(2),pp. 71–88.

Ostrom, E. (2005) Understanding Institutional Diversity, Princeton University Press,Princeton, N.J.

Pretty, J. (2003) ‘Social capital and the collective management of resources, Science, 302,pp. 1912–1915.

Rao, N. (2003) ‘Only women can and will represent women’s interests: the case of landrights’, paper given at a workshop on Gender Myths and Feminist Fables: Repositioning Gender in Development Policy and Practice, Institute of Development Studies,University of Sussex, UK, 2–4 July.

Rapport, N. and Overing, J. (2000) Social and Cultural Anthropology: The Key Concepts,Routledge, London.

SMUWC (2001) SMUWC Final Report: Rural Livelihoods, Sustainable Management of theUsangu Wetland and its Catchment , DFID Project [http://www.brad.ac.uk/acad/bcid/ research/papers/paper2.pdf (accessed 15 May 2007).

Tukai, R. (2005) ‘Gender and access in pastoral communities: re-evaluating community participation and gender empowerment’, paper presented at the ESRC Seminar ‘Access, Poverty and Social Exclusion’, ODI, London [www.odi.org.uk/wpp/News_Events/ESRC](accessed 15 May 2007).

United Nations (2005) Health, Dignity and Development: What Will It Take? , StockholmInternational Water Institute & United Nations Millennium Project, New York.

Understanding Community Action (2006) Research Project Working Papers [[http:// www.brad.ac.uk/acad/bcid/research/poverty/projects/interlink/papers/] (accessed 15May 2007).

Upperman, E. (2000) ‘Gender relations in a traditional irrigation scheme in northernTanzania’, in C. Creighton and C.K. Omari (Eds.), Gender, Family and Work in

Tanzania, Ashgate, Aldershot.


243


22/23

WWAP (2003) Water for People, Water for Life, UNESCO Publishing and Berghahn Books,Oxford.

WWC (2006) Ministerial Declaration of the Fourth World Water Forum, Mexico[http:// www.worldwaterforum4.org.mx] (accessed 15 May 2007).

Zwarteveen, M. and Neupane, N. (1996) ‘Free-riders or victims: women’s non-participationin Nepal’s Chhattis Mauja Irrigation schemes’, Research Report No 7 , InternationalIrrigation Management Institute, Colombo, Sri Lanka.

F. Cleaver

244


23/23

Documents

Understanding Agency in Collective Action.pdf