Social capital in community level environmental governance: A critique

E CO L O G I CA L EC O NO M IC S 6 8 (2 0 0 9) 1 54 9–1 56 2

a v a i l a b l e a t w w w. s c i e n c e d i r e c t . c o m

w w w. e l s e v i e r. c o m / l o c a t e / e c o l e c o n

ANALYSIS

Social capital in community level environmental governance: A critique Hiroe Ishihara⁎, Unai Pascual Department of Land Economy, University of Cambridge, UK

AR TIC LE D ATA

ABSTR ACT

Article history:

Social capital is often claimed to facilitate collective action regarding the management of

Received 4 July 2008

complex environmental goods and services. However, there is little systematic analysis in

Received in revised form

the literature that explains the way social capital aids in fostering collective action. The

3 November 2008

paper integrates ideas from institutional ecological economics, sociology and anthropology

Accepted 3 November 2008

to argue that power relations, involving struggle and resistance, should be acknowledged as

Available online 26 December 2008

they affect collective action. We address the question of why social capital should not be straightforwardly associated positively with common property resource management. To

Keywords:

unravel the complexity of the links between social capital, collective action and common

Social capital

property resource management, the concepts of common knowledge and symbolic power are

Collective action

introduced.

Power relations

© 2008 Elsevier B.V. All rights reserved.

Common knowledge Symbolic power Human agency

1.

Introduction

Since the 1990s, the concept of social capital1 has gathered increased attention in the literature about common pool resource (CPR) management and collective action, especially in relation to sustainable use of natural resources and sustainable development (e.g. Ostrom, 2000; Lehtonen, 2004). Social capital is often defined as ‘features of social organization such as networks, norms and trust that facilitate coordination and co-operation for mutual benefit’ (Putnam, 1993, pp. 35–36). This concept is associated with incentive mechanisms or institutional arrangements to curb individuals' incentive to free-ride regarding the provision of public goods (Ostrom, 2000; Aoki, 2001). It is also generally argued that the existence of

networks among agents and the dense flow of information among them lower the transaction costs of creating collective action (Putnam, 1993; Ostrom, 2000; Pretty and Ward, 2001; Collier, 2002; Paavola and Adger, 2005). However, these institutionalist arguments have also been criticised due to their incapacity to explain why social capital has this intrinsic ability to curb such perverse incentives and lower transaction costs (Harris, 2001; Cleaver, 2003; Mosse, 2006). Critics argue that the institutional angle cannot explain why some communities succeed in creating collective action while others fail to do so despite the existence of networks and a dense flow of information. As governments and markets fail, communities can fail as well (Ostrom, 1990a; McCay and Jentoft, 1998; Jentoft, 2000; Bowles and Gintis, 2002; Acheson,

⁎ Corresponding author. Department of Land Economy, University of Cambridge, 19 Silver Street, Cambridge CB3 9EP, UK. E-mail address: [email protected] (H. Ishihara). 1 This paper does not argue whether social capital should be termed ‘capital’ or not. Solow (1999) and Arrow (1999) argue that social capital is a by-product rather than deliberate investment. Perhaps a better term, following McNiell (2007), would be sociality. However, considering the popularity of the concept and the fact that it is one of the fields where interdisciplinary conversation is taking place, we use the term social capital throughout the paper. 0921-8009/$ – see front matter © 2008 Elsevier B.V. All rights reserved. doi:10.1016/j.ecolecon.2008.11.003

1550

E C O L O G IC A L E C O N O M IC S 6 8 ( 2 0 09 ) 15 4 9 –1 56 2

2006). Social capital can impact negatively on economic or environmental outcomes creating exclusion of some marginalised groups or contributing to rule breaking behaviour by community members as citation (Agrawal, 2001; Portes, 1993; Carroll and Stanfiled, 2003; van Steveren, 2003; Ray and Bijarnia, 2007). The causal relationship between social capital and environmental outcomes through governance of collective action is far from clear and often tautological (Durlauf, 2002; Sobel, 2002; Lehtonen, 2004; Ballet et al., 2007). The current theories of social capital cannot sufficiently open the black box linking collective action and CPR management (Fine, 2001; Harris, 2001). This paper aims to theoretically address the question of how social capital may help to foster collective action making use of the concepts of common knowledge as defined by Chwe (1999) and symbolic power, sensu Bourdieu (1990). By bringing these two concepts together, we argue that the creation of collective action is not just the result of rational calculation about how much to invest in collective action by individuals but also a result of power relations and social structure.2 The existence of power in the creation of knowledge has been fiercely debated between Foucault and Habermas (see Kelly, 1994).3 Foucault (1965) argued that the creation of knowledge/rationality itself is a historical and social construct which legitimises the existing power relations and social structure. Following this line of argument, it can be argued that power structures deprive the voice of the powerless and their ability to express themselves by categorizing their voice as irrational (Guha and Spivak, 1988). Habermas (1984) instead pointed out that it is possible to make a normative distinction between legitimate and illegitimate use of power through communicative rationality. The latter argument has led to the discussion on ‘public sphere’ and discursive democracy which has been recently introduced in environmental evaluation methodologies (e.g. Dryzek, 1990; Goodin, 1996).4 Further, 2 We limit our argument to community level power relations and social structure. In the last section we briefly look into the outside powers that can influence institutional outcomes. Formal institutions such as governments at various levels can have a strong influence on environmental governance and the formulation of social capital and collective action (see: Ostrom, 1990a; Acheson, 2006). However this debate is beyond the scope of this paper. In addition, we use the concept of social structure interchangeably with institutions. Although Hodgson (2000) distinguishes between social structure and institutions by defining institutions as social structure that have ‘recursive downward causation’, i.e. capacity to structure human agency. However, we cannot imagine a social structure that does not structure human agents. Both social structures and institutions shape the human agencies in one way or another. Thus in this paper we use these term interchangeably. 3 Although physical debate did not take place between Habermas and Foucault, Habermas went through many debates with other post-modernist thinkers, e.g. Derrida and Deleuze (Villa, 1992). In terms writing, both authors wrote about their opponents and the authors believe that in this debate between Habermas and Foucault that the issue of power became most apparent. 4 Although discursive democratic approaches provide an alternative to neoclassical monetary valuation approaches such as CVM, the Foucauldian criticism of the ‘public sphere’ claims that not all the members of the community have the same opportunities to voice out their preferences in the ‘public sphere’, i.e. Subalterns cannot speak (Spivak, 1988; Benhabib, 1992; Fraser, 1992; Kohn, 2000).

institutional economists tend to use the Habermasian concept of communicative rationality, whereas anthropologists and sociologist tend to deny rationality itself by relying on the Foucauldian argument. While this debate is outside the scope of this paper we argue that power relations influence the way preferences by individuals are shaped through the creation of so-called ‘common knowledge’ and collective action, an approach not tackled by institutionalist analysis of CPR (Ray and Bijarnia, 2007). In this vein we hope to contribute to the analysis of power within the pluralistic perspective of ecological economics. The paper is structured as follows: In the next section we provide a description of the roadmap linking social capital and collective action using a simple conceptual model. Section 3 introduces the concept of ‘social embeddedness’ which gave birth to the theory of social capital. We discuss the way new institutional economists link CPR management which collective action and social capital. We point out that standard social capital theories suffer from both ‘over-socialised’ and ‘under-socialised’ models of human agency. Then in Section 4 we argue that incentive mechanisms for CPR management crafted by social capital do not necessarily lead to collective action. Instead, we argue that what is first needed is to focus on the ability of social capital to create a ‘common knowledge’. The creation of common knowledge is associated with the capacity to represent individuals' preferences as if they were the preference of the entire community through ‘symbolic power’. It follows that the existence of collective action cannot be considered just the result of pure rational calculation by individual community members. This discussion leads us in Section 5 to explore how common knowledge can be diffused among community members and how it may be challenged by some marginalised groups within the community leading to failure in fostering collective action. Then, in Section 6, we argue that bonding social capital can contribute to the creation of common knowledge whereas bridging social capital may contribute to its sharing. Further, the main point is that different combination of these two types of social capital in different contexts where the interpretation of the common knowledge takes place results in different environmental outcomes. The last section concludes by discussing the main

Fig. 1 – Conceptual framework of the role of social capital in fostering collective action.

1551

E CO L O G I CA L EC O NO M IC S 6 8 (2 0 0 9) 1 54 9–1 56 2

implications of using the argument of common knowledge and symbolic power in the context of CPR management towards collective action and the management of CPR.

2. From social capital to collective action: the critical nodes From an institutional and sociological perspective, this paper argues that social capital can positively contribute to (i) the creation of ‘common knowledge’, identified with the link “L1” in Fig. 1 and (ii) through its sharing among community members (L3), it may successfully lead to collective action in order to solve environmental governance problems (L4). The paper also addresses the often neglected negative pathway between social capital and collective action, shown by the link L1–L5–L6. In addition, such pathway is mediated by ‘symbolic power’ or capacity to represent complex social preferences (L2) in communities affected by collective environmental problems. It is becoming accepted even among ‘super-rational’ game theorists that individual agents do not always act according to their own preferences but that they take into account other agent' preferences and the effects of their actions, thus acknowledging the fact that humans do not behave in the social vacuum and recognizing interdependency (Harsanyi, 1965; Aumann, 1976; Gächter and Fehr, 1999). The understanding of the preferences of others is referred to as ‘common knowledge’ (Chwe, 1999, 2001). In this vein, the creation of common knowledge should not be understood just as a simple process of averaging out the preferences of members of the community, for instance about their individual provision towards collective action according to rational calculation. Rather, it is a process of influencing (or in an extreme case imposing) certain arbitrary forms of preferences, often the one of the dominant group5 in a community. This is due to the fact that human agents in a community are not capable of comprehending all other individuals' preferences and some kind of typification or representation (‘symbolic formula’) is thus necessary (Berger and Luckmann, 1966; Douglas, 1986; Hodgson, 1997). Following Bourdieu (1990), we call this kind of power, symbolic power. It is the power to influence the way social preferences are shaped in the community to then become common knowledge (L2 in the Fig. 1). In this sense, we depart from the traditional view of collective action as the result of individuals' rational calculation about their optimal amount of resource provision of a public good. Instead we stress the political process that leads to collective action and that involves domination and marginalisation of different groups in society (Cleaver, 2003; Mosse, 2006).

5

Here the dominant/marginalised group has a political/social connotation linked with the idea of power. A dominant group is defined as group of agents that seize economic, political and social resources and occupies a superior position in the hierarchical social structure. However, there is always an alternative discourse to the common knowledge and there is a possibility for these alternative discourses to become common knowledge. Legitimacy of common knowledge is challenged through daily struggles (Bourdieu, 1984; Giddens, 1984).

However, it is not enough just to create common knowledge. This has to be diffused and shared among the members of the community (L3). Even when this commonly shared preference regarding the collective provision to a public good effectively becomes common knowledge, some individuals, especially those from marginalised groups may be implicitly forced to act collectively despite their preferences not coinciding with that of common knowledge. In such cases, the possibility to challenge the common knowledge appears (L5), leading to a failure in the creation of collective action (L6). The argument about the possibility of some marginalised groups challenging the common knowledge is important and is often overlooked despite the fact that the history of CPR management is full of resistance and struggles (Ray and Bijarnia, 2007; Bardhan and Ray, 2008). While some members of the community may realise that their preferences are not adequately represented by common knowledge, often some level of internalisation or sharing of common knowledge occurs. However, this does not eliminate the possibility of common knowledge being misinterpreted by individual members of the community (Taylor, 1993; Hodgson, 1997). We refer to the process of sharing common knowledge as ‘weintention’ following Tuomela (1995, 2007). It allows to address the role of human agency towards collective action without falling into the pitfalls of ‘over-socialisation’ and ‘undersocialisation’. In this sense, we point out that individuals in a community are not just passively influenced by common knowledge but they also positively create the world they live in (Giddens, 1984; Taylor, 1993). We also argue that while

Table 1 – Main concepts as roadmap between social capital and collective action for CPR management Concepts Symbolic formula

Definition

Frame of reference that provide agents with understanding of the world Common Symbolic formula that represents knowledge the generalised preference of the others Symbolic Power that legitimise the certain power ‘symbolic formula’ as common knowledge Internalisation A long-term process of consolidating and embedding one's own beliefs, attitudes, and values, when it comes to moral behaviour We-intentions An individual agents' attribution of an intention to a community that the agents believe is reciprocally held by other agents in the same community Human Ability of the human agents to agency change and transform social structure ‘Bridging’ social capital ‘Bonding’ social capital

Seminal literature Douglas (1986)

Chwe (1999, 2001) Bourdieu (1990) Berger and Luckmann (1966)

Tuomela (1995, 2007)

Giddens (1984) Bourdieu (1990) Sewell (1992) Weak ties that bridges difference Granovetter inside the community (1973) Strong ties that serves the interest Granovetter of certain group of the community (1973)

1552

E C O L O G IC A L E C O N O M IC S 6 8 ( 2 0 09 ) 15 4 9 –1 56 2

‘bonding’ social capital contributes to the creation of common knowledge, ‘bridging’ social capital contributes to its sharing (Granovetter, 1973; Putnam, 2000; Woolcok and Narayan, 2000; Adger, 2003; Newell et al., 2004; Ray and Bijarnia, 2007). However, both bonding and bridging social capital can also lead to failure as well as success in fostering collective action, in turn leading to changes in social structure. Table 1 summarises the main concepts used in the paper which form the building blocks of the conceptual model that connects social capital and collective action in the context of the governance of complex environmental public goods such as CPRs.

3. Revisiting the link between social capital and institutions 3.1.

Social capital and social embeddedness

Since the 1960s, gradually social scientists have recognised that the social structure, such as social networks and interpersonal relations, which have been later epitomized as social capital, greatly influences various institutional outcomes such as economic development and environmental management. Granovetter's (1973, 1985) seminal work had a strong impact in this regard. Granovetter put forward the argument of ‘social embeddedness’ in order to take into account the influence of social structure in economics. Influenced by Granovetter and other economic sociologists, some institutional economists have conceptualised economic activity as an institutional process (Williamson, 1981; North, 1990). Ecological economics has also recognized the ‘embeddedness’ of the economic system not only within the broader ecological system but also within social systems (Spash and Villena, 1998; Paavola and Adger, 2005; Røpke, 2005). The analysis of the economic system as embedded in social structures and the relationships between the latter and environmental outcomes has been greatly enriched by incorporating the notion of social capital, especially in the CPR management literature (Ostrom, 1990b; Lehtonen, 2004; Ballet et al., 2007). But what was the implication of Granovetter's proposition in relation to the coined concept of social embeddedness? In his work, Granovetter takes into account two sister disciplines in social science, namely sociology and economics. He juxtaposes the mainstream utilitarian economics with its tenets of the almost robotic human agent with stable preferences seeking utility maximisation and the sociological view that tends to regard human agents as basically constructed by a social structure through the internalisation of symbolic systems, social norms, and customs. Granovetter criticised the utilitarian economic model as embracing the ‘under-socialised’ model of human agency which ignores the fact that human rationality is restricted by the social environment that the agent is embedded in. He also criticised the sociological view as being ‘over-socialised’ because it disregards the role of rationality or creativity in human agency. However, Granovetter recognised that both disciplines share a key common trait in that they see human agents as atomized

actors. He eloquently pointed out that “fruitful analysis of human action requires us to avoid automation implicit in the theoretical extremes of under- and over-socialised conception. Actors do not behave or decide as atoms outside the social context, nor do they adhere as slave to a script written for them by the particular intersection of social categories that they happen to occur. Their attempts at purposive action are instead embedded in concrete, ongoing system of social relations” (Granovetter, 1985, pp. 487). It is through the concept of social embeddedness that Granovetter attempts to avoid both the ‘over-socialised’ approach of generalized morality and the ‘under-socialised’ one of impersonal, institutional arrangements, thus trying to find some space for a more realistic human agency model. Later Giddens (1984) and Archer (2003) further developed Granovetter's idea of human agency.6 Human agency arises from the agent's control of resources, implying the capacity to reinterpret or mobilise an array of resources creating a new position inside the social structure (Giddens, 1984; Bourdieu, 1990; Sewell, 1992). Hence, human agency is structuring as well as structured. It is structured because agents are born into certain social structure, however it is structuring because agents have the ability to transform the social structure. But of course, agents also vary in the extent of their control of social relations, i.e. agency (Goffman, 1967; Lin, 2001). Despite the emergence of new theories of social capital, especially after Putnam's (1993) work which is also said to follow Granovetter's, most of these theories keep on falling into the pit-falls of either the ‘under-socialised’ or ‘over-socialised’ models.

3.2.

Social capital and institutions

Since Hardin's (1968) views on “the tragedy of commons" the possibility for collectively managing CPR has been analysed extensively. More recently, game theory has also been applied taking into account a social capital dimension. The motivation mainly comes from the assumption that focusing the attention only on the dominant free-rider behaviour leaves no place for cooperative rule unless it is imposed and enforced by an outside power. However, empirical evidence demonstrates that coordination norms do exist inside communities that may reinforce the expectation of collective behaviour leading a critical mass of individuals to adopt solutions based on cooperative strategies (Ostrom, 1990b; Runge, 1992). Moreover, behavioural economics shows that it is possible to increase cooperation under multiple equilibria by introducing approval incentives (Gächter and Fehr, 1999).7 The institutionalist view has championed the argument of interdependency. That is, any dominant strategy would fail to capture the essence of the motivation of individual decisions if the importance of the expectation of others' behaviour is not analysed. In this sense, decision making regarding the use of CPRs involves interdependent choices. It is not only the benefit and the cost of using it which becomes a function of 6

Bourdieu term this as ‘habitus’. The terms agency sensu Giddens and ‘habitus’ sensu Bourdieu are close (Sewell, 1992). 7 Gächter and Fehr (1999) note that approval incentive by itself is not enough to sustain cooperative strategy, however, with other incentives agents are able to act collectively.

E CO L O G I CA L EC O NO M IC S 6 8 (2 0 0 9) 1 54 9–1 56 2

the aggregate action of individuals. Rather the decision to use (or overuse) the resource is also affected by the expected decisions of others due to the subtractrability of CPRs (Paavola and Adger, 2005). The concept of social capital has also been introduced to explain how communities are able to impose certain coordination norms without relying on external coercive regulation. But how did the notion of social capital come to play a role in the CPR debate? Here we briefly review the history of the origin and emergence of the social capital concept, previous to becoming popular after Putnam's (1993) contribution and the economic literature that has followed since, also in ecological economics (e.g. Lehtonen, 2004; Rodríguez and Pascual, 2004).

3.3. The origin and institutionalist emergence of social capital It is possible to trace back the notion of social capital to Aristotle, but the current use of the term is due to Pierre Bourdieu's (1986) seminal work (Portes, 1998; Carroll and Stanfiled, 2003).8 Bourdieu defined it as: “aggregate of the actual or potential resources which are linked to the possession of the durable network of more or less institutionalized relationship of mutual acquaintance and recognition – or in other words, to in a group – which provides each of its members with backing of the collectively-owned capital, a ‘credential’ which entitles them to credit, in various senses of the world” (Bourdieu, 1986, pp. 248). Clearly, Bourdieu's position is that social capital transforms the contingent relationships into durable obligations subjectively felt, e.g., feelings of gratitude, respect, friendship, etc. It follows that for Bourdieu, social capital functions as a mechanism to impose coordination norms, understood as durable obligations, because it produces mutual knowledge and recognition. In Chwe's (1999) term this is known as ‘common knowledge’. However, Bourdieu's contribution has been mostly ignored by those who follow the rational choice model and game theoretical analysis (Harris, 2001; Morotomi, 2003).9 This is especially problematic when one wants to include variables proxying social capital in empirical, e.g., econometric analysis, as Bourdieu did not bring the concept to bear such analysis (Durlauf, 2002; Sobel, 2002). We concur with the argument that this has resulted in the ‘mystification’ of the functions of social capital and its de-politicisation (Fine, 1999; Harris, 2001). Influenced by Granovetter, Coleman (1988) became the first rational choice theorist to use the concept of social capital and assumes rationality behind the actions of individual agents but superimposes a social and institutional structure on them.

8

Portes (1998) traces back the origin of the concept to Marx and Durkheim. 9 This may be due to the following three facts; (i) Bourdieu's main focus is on cultural capital which he argues extensively in relation to his key concepts, ‘habitus’ and ‘fields’, rather than on social capital (Bourdieu, 1984), (ii) Bourdieu's concept of capital is rather chaotic, especially for economists (Fine, 2001), and (iii) Bourdieu does not make clear the argument regarding the relation between social capital, cultural capital and symbolic capital, and at one point these three concepts become undistinguishable (Swartz, 1997).

1553

In order to explain the effect of this social/institutional structure, Coleman states that “social capital is defined by function. It is not single entity, but variety of different entities having two characteristics in common: they all consist of some aspect of social structure, and they facilitate certain actions of the actor within the structure” (ibid, pp. S.98). However, it was Putnam's (1993) influential work which popularised the idea of social capital. For Putnam, social capital refers to those “features of social organization, such as trust, norms, and networks that can improve the efficiency of society by facilitating coordinated action” (Putnam, 1993, pp. 167). The concept of social capital, which was introduced to explain institutional performance in his first book, later makes an unexpected turn as he also refers to it as the object to be fostered by public policy (Putnam, 2000; Morotomi, 2003). It is this new conceptualisation of social capital that has made his argument most popular (Fine, 2001). Since then the idea of social capital has become detached from a social structure and it is conceptualised as an object that can be crafted and manipulated by agents to obtain favourable economic and political outcomes. This trend has been further enhanced by the CPR management literature most notably by Ostrom (2000) for whom the concept of social capital is mainly interpreted as consisting of coordination norms and used to explain the potential for collective action inside communities. In this vein, social capital is understood as “shared knowledge, understanding, norms, rules and expectation about the pattern of interaction that a group of individuals bring to recurrent activity” (Ostrom, 2000, pp. 176) and as “an attribute of individual and of their relationship that enhance their ability to solve collective– action problem” (Ostrom and Ahn, 2002, pp. xiv). Ostrom's view goes further to argue that “to create social capital in a self-conscious manner, individuals must spend time and energy working with one another to craft institutions — that is the set of rules that will be used to allocate the benefit derived from an organized activity and to assign responsibility for paying the cost” (Ostrom, 2000, pp. 178 emphasis added). Ostrom thus argues that human agents have the capacity to craft and invest in social capital to provide the (socially and ecologically) efficient amount of collective action according to their rational calculation. Ostrom's view has been reinforced in game theory following the idea that social capital is part of institutional arrangements that agents have free hand in crafting according to their rational calculations. For example, Aoki's (2001) applies a ‘linking game’ to exemplify how in the Edo period (1603–1868) in Japan farmers coordinated their actions to maintain the functioning of the irrigation system in a context in which individually farmers have an incentive to free ride on such communal effort. Since in that period it was technically difficult to punish cheaters in order to foster collective action coercively, the social exchange game of ostracism was linked to the irrigation game. Aoki argues that the linkage between both games enables individuals to act collectively, especially when the cost of ostracism is large enough. Another important strand of the institutionalist approach puts more emphasis on social capital in relation to its ability to lower transaction costs (Katz, 2000; Pretty and Ward, 2001; Paavola and Adger, 2005). For example, Pretty and Ward (2001)

1554

E C O L O G IC A L E C O N O M IC S 6 8 ( 2 0 09 ) 15 4 9 –1 56 2

define social capital as “the structure of relations between actors and among actors' that encourages productive activities. Theses aspects of social structure are called social capital because it acts as a resource for individuals to use to realize their personal interest. Local institutions are effective because ‘they permit to carry on our daily life with minimum repetition and costly negotiations’ (ibid, pp. 211, emphasis added). Within this tradition, it is commonly assumed that social capital is useful as it lowers the transaction cost of acting cooperatively for individuals' mutual benefit. In summary, social capital theories, especially stemming from game theoretical approaches, assume that social capital is a resource that can be invested in order to circumvent the incentives of free riding and to lower the transaction cost associated with collective action. Implicitly the main assumption is that individuals rationally calculate the cost and benefit of collective action and self-consciously invest in social capital. However, these theories cannot explain why social capital has the ability to circumvent the incentive to free-ride (Cleaver, 2000, 2003) or why indeed it can lower transaction costs (Ballet et al., 2007). Rather their arguments become circular (Portes, 1998; Durlauf, 2002; Sobel, 2002), i.e., “a successful group succeeds because it has social capital, but the evidence that the group has social capital is its success” (Sobel, 2002, pp. 146). A theory that explains why agents feel more obliged to comply with ‘coordination norms’ when there are interpersonal networks or dense information flows is missing.

3.4.

Under- and over-socialised human agency

In institutional analysis of social capital (e.g. Knack and Keefer, 1997; Narayan and Pritchett, 1999; Guiso et al., 2007), the idea of social capital is often reduced to institutional arrangements and treated as an exogenous variable that can explain the existence of collective action (Cleaver, 2000, 2003). However, the capacity of social capital to create collective action and positive institutional outcomes resides in the fact that it is part of a hierarchical social structure and its power to impose certain cooperative strategies as a legitimate way to act.10 The rational choice theory which has decoupled social capital from social structure, cannot explain how the common understanding of the game is formed and shared in the community. We argue that it is not only the rational calculation of individual cost and benefit that allows agents to act collectively but the understanding of the preferences of other community members. In fact, this approach ignores the influence of power relations in creating collective action and CPR management (Adler and Kwon, 2002; Pérez-Cierra and Lovett, 2006; Ray and Bijarnia, 2007) and thus holds to the under-socialised model of human agents with stable preferences (van Steveren, 2003). In other words, this under-socialised rational approach to social

10

For example, Ray and Bijarnia (2007) has shown that democratic system, i.e. higher equality in power among members of the community, are less effective in terms of creating collective action, although higher inequality may lead to bleach in the rules and norms created for collective action.

capital ignores the role of the social structure and its coercive character (Mosse, 1997; Cleaver, 2003; Mosse, 2006). At the same time, paradoxically, when game theory is applied to problems such as CPR management, it also suffers from an over-socialised model of human agency. This is because it is often assumed that once self-enforcing incentive mechanisms are built, agents would automatically follow the commonly agreed rules/norms and reach an efficient solution (Hodgson, 1997). However, empirical evidence shows that CRP management is also associated with resistance and struggle despite the existence of networks and a dense flow of information (Cleaver, 2000). By reducing the notion of social capital to mere institutional arrangement, much is lost since the standard theories cannot consider negative aspects of social capital, for instance on economic and environmental outcomes, such as creating social exclusion, reproducing unequal social structures (Portes, 1993; Carroll and Stanfiled, 2003; Dolfsma and Dannreuther, 2003). Social capital should be considered to forge a link between social structure11 and agents rather than just being a part of an institutional arrangement per se (Paavola and Adger, 2005). Further, if social capital is to be a factor explaining collective action, its interpretation as the density of the networks and information flows is clearly insufficient. We align with Cleaver's (2000) view that the existence of networks and information flows do not necessarily lower the transaction cost nor does it guarantee that agents would follow an institutional arrangement. Thus there is still a need to identify the nature of networks and information flows that can lower transaction costs and how this comes about.

4. Creation of common knowledge: the first node between social capital and collective action 4.1. Common knowledge as the generalised preference of the community: node L1 Notwithstanding that the current social capital theory does not sufficiently explain the creation and potential direction of collective action, we argue that the key concepts of ‘common knowledge’ and ‘symbolic power’ help to regain the explanatory power of social capital. We have argued that community members may tend to act collectively only when they can formulate a common knowledge that acting collectively creates benefits that outweigh the costs of such effort. Here we define common knowledge as the set of understanding embedded in social structure, which enables to put resources to a particular use following Newell et al. (2004), thus enabling agents to take into account the preferences of the other agents it the community (Chwe, 1999, 2001). In the case of CPR management, taking other agents' preferences into consideration is especially important as the decisions of heterogeneous interests by community members are interdependent (Vatn, 2005). While it is often assumed that in such cases social capital plays an important role in reducing transaction costs and thus facilitating collective action, we

11

See footnote 2.

E CO L O G I CA L EC O NO M IC S 6 8 (2 0 0 9) 1 54 9–1 56 2

4.2.

1555

Symbolic power for common knowledge: node L2

argue that if social capital is to be instrumental in this sense, it would be mainly through its ability to create common knowledge (Sobel, 2002; Mosse, 2006). Social structure and rationality are not mutually exclusive when agents are strategically rational, that is, when they make decisions assuming that others are also rational (Chwe, 1999). However, considering the preferences of others in a rational way has its own limitations. Even in small communities where CPR is at stake, the preferences of individual members are often highly heterogeneous and varied according to their gender, cast, ethnicity, etc. Agents with different social positions have different preference structures (Leach et al., 1999; Agrawal and Gibson, 2001). Given our cognitive limitations, it is not possible to deal with all such different preferences in a mechanistically rational way (Hodgson, 1997). To avoid this overload, as human agents we economize our cognitive capacity by relying on what Douglas (1986) calls ‘symbolic formulae’.12 For an individual member of the community this represents the generalised preferences of the other members and as such the preference of the whole community needs to be created as common knowledge. In other words, we need to socially construct a ‘symbolic formula’ as common knowledge that carries with it the notion that ‘our’ benefit (i.e., the benefit for the community as whole) exceeds ‘our’ cost of engaging in collective action rather than rationally calculating the preference of others. This goes beyond the argument of bounded rationality and embraces the notion that common knowledge needs to be socially constructed as similarly pointed out by Berger and Luckmann (1966) and Bourdieu and Wacquant (1992). Social construction implies that agents interacting together form, over time, typifications or ‘symbolic formula’ of each other's preferences, and that these typifications eventually become “habitualised" into reciprocal roles played by the actors in relation to each other (Berger and Luckmann, 1966). It is important to note that this process of social construction is neither a process of averaging out the existing preferences in the community nor a purely rational calculation, as often modelled in game theoretical approaches. Rather it is a process of political struggle over who has the legitimacy to represent their preference as common knowledge. It is not the actual cost and benefit of collective action but the perceived ones that matter (Maruyama, 2003). Moreover, through this social construction process, the boundary of communities are also imagined and created (Anderson, 1983). Collective action is formed upon such common knowledge, and the frictions or relationships that may arise between dominant and marginalised groups, each with their symbolic formulae of each others' preferences. These symbolic formulae form the necessary ingredients in order to understand to what extent it is possible to create collective action for CPR management or how sustainable such social capital may be. Let us turn to the way symbolic power creates common knowledge (c.f. link L2 in Fig. 1).

Whose calculation of benefits and costs is represented as common knowledge out of the manifold preferences existing in a heterogeneous community? The answer should focus on those who have the power to impose the legitimate vision of the social world and of its division (Bourdieu, 1990; Swartz, 1997).13 This is where the notion of symbolic power comes in, interpreted by Bourdieu (1990) as “world-making power”. As mentioned in the previous section, not all preferences are represented in common knowledge. Symbolic power is necessary to create such common knowledge. It allows the arbitrary preferences of the dominant group in the community to be represented as common knowledge due to the advantageous position the dominant group holds within the social structure. By disguising as ‘common’, the preference structure and ordering of the dominant group overrides alternative preference orderings as illegitimate. Fundamentally, it is the capacity to ‘impose’ such common knowledge that ultimately enables social capital to reduce the transaction cost that arises in collective action. This is so especially when there is a heterogeneous set of preferences within the community, and these are associated with groups that are hierarchically related in terms of the power they have to impose what is perceived as ‘common’. In this sense, it is the symbolic power that plays a key role in reducing transaction costs. Further, this implies that the existence of the network and information flows cannot guarantee the reduction of such transaction costs. Rather, networks and information flows may even become instrumental and enable the dominant group to spread their vision of the costs and benefits associated with collective action. In other words, this may leave some marginalised groups within the community not being able to represent their real preferences towards collective action. But, at the same time this situation creates possibilities for challenging the articulation of the community's preferences by the dominant group. In fact, the legitimacy of the common knowledge is often challenged (Bourdieu, 1984). At the same time it is worth noting that there is no guarantee that the preferences that gain such common character would achieve the level of collective action needed for the desired (ecological or economic) outcome. In this sense, the preference which becomes common knowledge is arbitrary. By legitimizing the arbitrary symbolic formula as common knowledge, other preferences become ‘unthinkable’ or ‘irrational’ (Bourdieu, 1971). For example, if we come back to Aoki's (2001) linking game, there are numerous games in society and which game to be linked to which one is arbitrary. Deciding that the irrigation game is to be linked to the social exchange game might be the result of rational calculation; but at the same time, it might be also the result of the fact that this specific type of linking game gives advantage to the dominant group. In this sense, the

12 This concept is very similar to Veblenian's notion of ‘habit of thought’ (Hodgson, 1997, 2000).

13 In similar way, Habermas (1986) argues ‘the fundamental phenomenon of power is not the instrumentalization of another's will, but the formulation of a common will in a communication directed to reach agreement’ (ibid, pp.76, emphasis added).

1556

E C O L O G IC A L E C O N O M IC S 6 8 ( 2 0 09 ) 15 4 9 –1 56 2

symbolic power jettisons other possible liking games as being illegitimate and irrational. For this reason, in the context of the problem of collective action for CPR management, the creation of common knowledge is not just the result of rational calculation, but mostly the result of a hierarchical power relation inside the social structure, which in turn legitimises certain groups in the community and monopolises symbolic power (Mosse, 1997; Cleaver, 2000, 2003; Mosse, 2006).

5.

Sharing and challenging common knowledge

5.1. Collective action needs sharing of common knowledge: nodes L3 and L4 For common knowledge to function as an incentive mechanism or institutional arrangement for collective action, it is not enough to create it but it also has to be shared or diffused effectively among the members of the community. In other words, common knowledge has to be internalised as a longterm process of consolidating and embedding one's own beliefs, attitudes, and values (Berger and Luckmann, 1966). However, we do not term this process as internalisation. Rather we call it the creation of ‘we-intention’ following Tuomela (1995, 2007) recognising that incentive mechanisms do not directly lead human action. Instead human action is determined by a complex interplay of common knowledge, symbolic power and human agency. We-intention is interpreted as an individual agent's attribution of an intention to the whole community when agents believe such attribution is reciprocally held among the majority of other agents in the same community (Tuomela, 1995, 2007; Davis, 2002). Translating this argument to the issue of collective action for CPR management, ‘we-intention’ would then imply that most individual members in the community would agree with the common knowledge that ‘our’ benefit exceeds ‘our’ cost of collective action. However, in reality all individual agents deliberate on the content of common knowledge, thus allowing room for interpretation/misinterpretation and for challenge to occur within the community. But for collective action to be successful a majority of members must share the common knowledge and consider it to be the legitimate symbolic formula. By terming the sharing process as creation of ‘we-intention’, we have attempted to avoid an over-socialised model of human agency.

5.2. Challenging common knowledge and the failure of collective action: nodes L5 and L6 Despite that community members tend to share common knowledge as ‘we-intention’, it does not usually go without challenge from inside. Let's assume, for illustrative purposes, that the dominant group has the interest of fostering collective action. Even in this case, where the dominant group monopolises symbolic power to impose its preferences as common knowledge to marginalised groups in the community, the dominant group would not be automatically free to foster collective action. For example, when there is high

inequality inside a community, there tends to be a high occurrence of rule/norm violation (Agrawal, 2001; Varughese and Ostrom, 2001; Pérez-Cierra and Lovett, 2006; Ray and Bijarnia, 2007). This is due to the sharing process of common knowledge being imperfect thus allowing for the interpretation/misinterpretation of common knowledge to occur and the alternative discourses to survive. Firstly, not all members can always interpret the common knowledge as the dominant group would wish, but rather they can interpret and misinterpret its content according to the context they face. Secondly, the alternative discourses may lead to the challenge of common knowledge, especially when there is a strong ‘bonding’ social capital within marginalised groups. With regard to the first point, Hodgson (2000) quotes Wittgenstein to argue that, “there are all sorts of interpretive problem involved in moving from the existence of rule to behaviour that follows the rule. The feeling that one is being guided by rules does not guarantee that the rules are being followed” (ibid, pp. 61). Human agents are not mere rule/norm follower but rather interpreters or improvisers of the common knowledge (Taylor, 1993). Community members interpret and misinterpret the common knowledge according to their position inside a certain social structure. It is thus possible that community members displace the original content of common knowledge and create a new interpretation of it since they have their own socialisation process (Sewell, 1992; Lin, 2001). This implies that whether common knowledge is shared by individuals as ‘we-intention’ depends on the context of the interaction between them. Hence, even if common knowledge exists, this does not guarantee that community members would act accordingly. The second point, namely that some members may have alternative discourses which may in turn challenge the common knowledge is of great importance. It is possible that marginalised groups do not share the common knowledge and hence do not accept its legitimacy. According to their alternative discourse, marginalised groups may challenge the arbitrary nature of common knowledge, especially when they have to bear a disproportionate cost associated with it (Taylor, 1993).14 However, these cases are not frequent as it is difficult to reflect the voice of the marginalised groups within the public discourse that favours the dominant group's interests (Spivak, 1988). It is important to note that human agency varies according to the position that human agents occupy in the social structure. Often, dominant groups have more freedom in terms of improvising new interpretation or even challenging the common knowledge due to their privileged social positionality (Sewell, 1992; Lin, 2001). However, this does not mean that the arbitrariness of common knowledge cannot be disclosed. Disclosure does occur and thus the legitimacy of common knowledge is implicitly negotiated in everyday social life practice (Willis, 1977; Bourdieu, 1990; Cleaver, 2003). In this section we have argued for social capital to create collective action, common knowledge needs to be widely and

14

It should be noted that this challenge does not arise from a neoclassical rationality idea. It is the dominant group who defines what is ‘rational’ and what is not (Foucault, 1965; Bourdieu, 1990).

E CO L O G I CA L EC O NO M IC S 6 8 (2 0 0 9) 1 54 9–1 56 2

1557

mutually shared as ‘we-intention’ among the community members. We have termed this process of diffusion as the creation of ‘we-intention’, not as internalisation. Although internalisation of common knowledge does occur, it is only partial leaving room for agents to misinterpret the content of common knowledge. Also not all community members internalise the common knowledge effectively. Moreover, common knowledge is susceptible to challenges from within the community with the possibility of failing to foster collective action.

6. Common knowledge and ‘bonding’ vs. ‘bridging’ social capital 6.1.

Nodes L1–L3–L4 and the two types of social capital

Whether common knowledge is shared or challenged in the community depends on the type of social capital, i.e. ‘bonding’ social capital and ‘bridging (linking)’ social capital. The former refers to strong horizontal ties inside the community and may become the basis for narrower sectarian interest, whereas the latter refers to intra-communitarian (weaker) ties that gives legitimacy to common purpose, such as collective action (Granovetter, 1973; Putnam, 2000; Woolcok and Narayan, 2000; Adger, 2003). In standard social capital theory, it has been argued that bridging social capital has the capacity to leverage resource, ideas and information from formal institutions beyond the community and can have a positive effect on economic and environmental outcomes. But we argue that bridging and bonding social capital contribute to different processes in the creation of collective action. While bonding social capital

Fig. 2 – The creation of nodes L1–L3–L4. Notes to Figure 2: Link 1 shows the power to challenge held by the MG due to the existence of bridging SK between MG and DG. It results in challenge to the common knowledge. Link 2 shows the power to impose common knowledge held by the DG due to the existence of bridging SK between DG and MG. It results in effective sharing of common knowledge. Net result: Lesser amount of negotiation due to fact that power held by DG is greater than that of the MG. It results in the success of collective action.

Fig. 3 – Social capital as tyranny. Notes to Figure 3: Outside powers refer for example to state discourse / development discourse by donors. Link 1 shows the existence of bridging SK between DG and outside power. It results in empowering DG. Link 2 shows the decrease in power to challenge DG. It results in more violation of common knowledge. Link 3 shows the increase in power to impose the common knowledge to MG. Net result: Decrease in negotiation, reproduction of unequal social structure and less effective collective action.

contributes to the creation of common knowledge, bridging social capital can be said to contribute to the sharing of common knowledge (Sobel, 2002; Newell et al., 2004). Moreover, both types of social capital can have negative and positive impacts on institutional outcomes. Newell et al. (2004) argue, that within certain groups, members need to develop strong bonds with each other so they can share the sense of purpose and some common knowledge to achieve collective action. Bonding social capital is necessary for assimilating the cognitive framework and sharing the understanding (Becker, 2001). Without the existence of bonding social capital, even the dominant groups cannot formulate the common knowledge and claim for its legitimacy. Further, bridging social capital may contribute more to the sharing of common knowledge; group members need to use their social capital to bridge for the public good of the community (Newell et al., 2004). Bridging social capital facilitates the dominant group to impose common knowledge since the heterogeneous community becomes better interconnected. The transaction cost of sharing common knowledge may be reduced, thus helping the dominant group's discourse to reach out to marginalised ones and thus facilitating collective action (assuming this favours the dominant group's position in the community). This would be specially so when both common knowledge and any alternative discourse have a similar understanding of the potential of collective action. This would be the case when the dominant group's power to impose common knowledge outweighs the power of marginalised groups to negotiate that common knowledge. This is depicted in Fig. 2. Hence, bridging social capital can strengthen the dominant group's symbolic power, which in turn may enable to disguise the arbitrariness of common knowledge. By playing ‘politics of recognition’ (Taylor, 1994)

1558

E C O L O G IC A L E C O N O M IC S 6 8 ( 2 0 09 ) 15 4 9 –1 56 2

marginalised groups are forced to give recognition of common knowledge in exchange of some of the benefits of collective action.15 When the dominant group holds a strong bridging social capital that reaches out to policy makers, (e.g., governments or donors), the resulting policy claims for a participatory approach and use of social capital may exacerbate the social exclusion and hierarchical social structure inside the community (Agrawal, 2001; van Steveren, 2003). This may be a pervasive situation specially when social capital, both bonding and bridging, is too costly to maintain by the poor who are on the verge of survival (Cleaver, 2000, 2003). This situation is depicted in Fig. 3. However, at the same time, bridging social capital is also instrumental to collect the necessary information by the dominant group to form common knowledge (Sobel, 2002; Newell et al., 2004). This also implies that there is a possibility for marginalised groups to influence the creation of common knowledge and for their social inclusion leading to positive effects on various institutional outcomes (Narayan and Woolcok, 2000). But at the same time, it is worth mentioning that it takes time for communities to share common knowledge (Dolfsma and Dannreuther, 2003).

6.2.

Nodes L1–L5–L6 and two types of social capital

Social capital is a double-edged sword (Portes, 1993, 1998; Carroll and Stanfiled, 2003; Dolfsma and Dannreuther, 2003; van Steveren, 2003; Ballet et al., 2007). Both bonding and bridging social capital can contribute in their own way not only to the success but also failure of collective action. For example, bonding social capital can function as a mechanism of exclusion (Portes, 1993, 1998; Agrawal, 2001) or ‘distinction’ if we follow Bourdieu (1984). A strong bonding social capital inside a dominant group can lead to exclusion of the marginalised group by not allowing their voice to be represented in common knowledge (Guha and Spivak, 1988) leading to inefficient management of CPR (Agrawal, 2001; Varughese and Ostrom, 2001; Pérez-Cierra and Lovett, 2006; Ray and Bijarnia, 2007). At the same time, bonding social capital may support the survival of alternative discourse and at times may contribute to fostering the challenges to common knowledge. This is so as bonding social capital may more easily enable the marginalised groups to form their own alternative discourse which, in turn help them to realise that their perceived costs and benefits from collective action is different from that implied by common knowledge. In other words, we argue that bonding social capital enables the marginalised group to form “subaltern counter-public” which enables the powerless to form their voice by forming their own public sphere (Benhabib, 1992; Fraser, 1992). Thus, bonding social capital may disclose the arbitrary nature of common knowledge within the marginalised group and by doing so it may increase the transaction cost of collective action as challenging common knowledge can more easily surface. If bonding social capital within marginalised groups is strong, it can allow formation of 15 Typical example is the patron-client relationship (Eisenstadt, 1984).

Fig. 4 – The creation of nodes L1–L5–L6. Notes to Figure 4: Link 1 shows the power to challenge held by the MG due to the existence of bridging SK between MG and DG. It results in challenge to the common knowledge. Link 2 shows the power to impose held by the DG due to the existence of bridging SK between DG and MG. It results in sharing of the common knowledge. Net result: Greater amount of negotiation due to the fact that power held by MG is greater than power held by DG. It may result in failure in collective action if this is the DG's wish. new common knowledge that includes the preference of those who were excluded from it (Fig. 4).16 In addition, bridging social capital has an ambiguous character, it can be said to facilitate tyranny and empowerment. It can function as catalysis for collective action but it can also be an obstacle. It can promote alternative discourses to common knowledge by helping to link the community to an outside discourse which may be more powerful than common knowledge. It is not hard to imagine that marginalised groups such as women within male-dominated communities being ‘empowered’ by outside feminist discourses which enable them to create a coalition of discourses (Hajar, 1995; Appadurai, 2001) and thus allow them to expose the arbitrariness of the male-dominated common knowledge. The point is that discourses can form various coalitions and oppositions according to the context (Hajar, 1995; Ockwell and Rydin, 2006) producing various possibilities for collective actions (Fig. 5). But we need to be aware that the capacity of bridging social capital resides in its vertical social relations (Woolcok, 1998). Moreover, disclosing the arbitrariness of common knowledge at the community level may also lead to subjugation to another common knowledge at the global scale as for instance many feminist groups in the Third Word would argue (Abu-Lughod, 2002, 2006).

16

However there is a difference in the level of human agency according to the positionality of the agents inside the social structure. Often the dominant groups have more freedom to create new common knowledge than marginalised group (Sewell, 1992; Lin, 2001).

E CO L O G I CA L EC O NO M IC S 6 8 (2 0 0 9) 1 54 9–1 56 2

1559

that describes the interface between society and ecology (McCay and Jentoft, 1998).

7.

Fig. 5 – Social capital as empowering. Notes for Figure 5: Outside power at global level, e.g., global discourse on feminism or human rights. Link 1 shows existence of bridging SK between MG and outside power. It results in empowering MG. Link 2 shows increase in power to challenge by MG. Link 3 shows decrease in power to impose by DG. Net result: Increase of negotiation. It results in transformation of unequal social structure and potential failure of collective action.

6.3.

Contextuality of social capital

Whether common knowledge is shared or challenged depends highly on contextual factors (Adler and Kwon, 2002); for example it depends on (i) the availability of bonding and bridging social capital, (ii) the social distance between the dominant and marginalised groups and (iii) the character of collective action (Ostrom, 1990a). First, what kind of social capital is available to which group within the community determines the possibility of collective action. If the dominant group monopolises social capital, the resistance by marginalised groups becomes more difficult. However this is often not the case as monopolising all social capital is unlikely. The different combination of bonding vs. bridging social capital inside the dominant and marginalised groups ultimately determines whether common knowledge is shared or challenged. Second, the social distance between the dominant and marginalised groups, i.e., the socially constructed boundaries in the community matters (Dolfsma and Dannreuther, 2003). If both are close enough, the marginalised group might not be considered as an intruder when managing the CPR. In this case, it is easier for them to influence common knowledge, presenting themselves as ‘co-ventures’ or socially integrated ‘we’ (Etzioni, 1988; McCay and Jentoft, 1998), bringing a positive institutional outcome. Third, what kind of task is required by the dominant group effects the creation and sharing of common knowledge (Adler and Kwon, 2002). If the dominant group is required to facilitate collective action among different heterogeneous groups, sharing common knowledge becomes a greater challenge. Thus, whether the common knowledge is shared or challenged is the outcome of a complex interplay between bridging and bonding social capital, social structure and the resources at stake. What is required is a ‘thick’ description of the context (Geertz, 1973)

Is social capital an ‘anti-politics machine’?

In this paper we have argued that current theories of social capital a la Putnam, are insufficient to explain why social capital can foster collective action and have challenged the widespread notion that it is almost invariably instrumental in lowering the transaction cost that arise in the process. In addition, we have argued that this does not just deprive social capital theories of their explanatory power in creating collective action but it also results in their falling into the pitfalls of either ‘over-socialised’ or ‘under-socialised’ models of human agency. In this sense we concur with Harris (2001) in that social capital theory becomes an anti-politics machine17 given that the theories are de-politicised even if it is the very political nature of social capital that can be instrumental in creating collective action. We have thus attempted to bring in the concepts of common knowledge and symbolic power to capture the political nature of social capital which also creates some needed room for human agency. We have identified the main processes by which social capital may contribute to create collective action for managing complex environmental goods such as common property resources: first, the creation of common knowledge (L1 c.f. Fig. 1) and second, the diffusion of common knowledge (L3). Here we interpret common knowledge as the understanding of the preferences of others. It is the common knowledge which enables agents to act together and to bear the costs of collective action. In the face of limited cognitive capacity, such common knowledge is socially constructed by generalising preferences in turn represented by symbolic formulae (Hodgson, 1997). This has led us to address the question of whose preferences should be represented in that common knowledge. We have argued that the process of creation of common knowledge is not a simple process of averaging out the preferences of all community members. Rather it is a process where some dominant group has more power to impose its preference as common knowledge than other groups. This power is termed as symbolic power following Bourdieu (1990). For collective action to be successful it is not sufficient to create common knowledge but it has to be shared or internalised among the community members. In this paper we have termed this process as the creation of ‘we-intention’ to leave room for human agency to play a part where common knowledge is shared and the majority of the community members agree to bear the cost of collective action. But since common knowledge is susceptible to misinterpretation and challenges (L5), collective action may fail (L6). By differentiating ‘weintention’ (Tuomela, 1995, 2007) from internalisation, the

17 This concept is coined by Ferguson (1994) who has analysed why the poverty alleviation policies in Lesotho were renewed despite its ‘failure’ in reaching its objectives. He argued that “uncompromisingly, reducing poverty to a technical problem, and by promising technical solutions to the suffering of the powerless and oppressed people, the hegemonic problem of ‘development’ is the principle means through which the problem of poverty is depoliticized in the world today” (ibid, pp.256).

1560

E C O L O G IC A L E C O N O M IC S 6 8 ( 2 0 09 ) 15 4 9 –1 56 2

paper has recaptured human agency and avoided falling into the pitfall of ‘over-socialised’ models of human agency. We have argued that social capital is very political and contested and that collective action is a complex interplay of common knowledge, symbolic power and human agency (Cleaver, 2003; Mosse, 2006). It is important to understand the power relations inside the community when designing a policy of CPR management, especially as the dominant groups are not necessarily the primary CPR users and the holders of the knowledge about CPR management (Agrawal, 2001). In other words, relying on existing common knowledge may not provide the socially and ecologically efficient amount of collective action or may force the marginalised groups to bear a disproportionate cost of the collective action (Boyce, 2006), which in these cases social capital would serve only the objectives of the powerful in detriment of the disadvantaged groups of the community. As a result of following mainstream social capital theories, we become blind to key political aspects of social capital for promoting collective action. More importantly policies based on such anti-politics machine are not only likely to fail in promoting sustainable use of CPR but they also endorse and encourage the reproduction of current social structure, which is not capable of yielding neither optimal ecological or social outcomes. A note of caution is deserved though. In this paper we do not wish to contribute to the Foucauldian criticisms of social capital in which social capital and its policy relevance is associated with tyranny (Cook and Kothari, 2001) or with the trends of deconstructing development policy (Ferguson, 1994). Instead, noticing the power relation involved in the formation and use of social capital, we have aimed to bring to the fore a discussion over social structure, social capital and human agency. We concur with Gale (1998) that ecological economics is still lacking a suitable concept of power. Especially a concept of power that incorporates not only the material aspects but also the symbolic aspects is necessary (Bardhan and Ray, 2008). Currently it is mainly sociologists and anthropologists, especially Foucauldians, who discuss power relations, especially the symbolic aspect of power, rather than economists (Ray, 2008). Due to a ‘de-constructionist’ tendency, anthropologists/ sociologists tend to be reluctant to generalise their arguments which unfortunately does not allow them to permeate into economics (McNiell, 2007). This has lead to the literature linking social capital, collective action and CPR to be overly dominated by political scientists and economists (Fine, 2001).18 We hope that our approach based on integrating the ideas of common knowledge and symbolic power is a step towards revising this dominance and to promote a more pluralistic understanding of CPR management. The future challenges still lie in advancing in the construction of an interdisciplinary notion of power and incorporating aspects of power into games while at the same time acknowledging institutional

18 However, this predominance should not be considered as only the result of ‘colonization of other social sciences by economics’ (Fine, 2001) but also the result of the anthropologists/ sociologists' inability to communicate their argument to other disciplines.

arrangements. While some sociologists have already modelled power as a game (e.g. Goffman, 1967; Elias, 1978), the communication between economists who deal theories of social capital while engaging in games about collective action and CPR management and sociologists/anthropologists who primarily deal with power relations linked with CPR has not been fruitful (Bardhan and Ray, 2008). We believe that ecological economics is in an advantageous position to lead this interdisciplinary challenge.

Acknowledgement We would like to thank Arild Vatn and Elinor Ostrom for their constructive comments to previous drafts of this paper.

REFERENCES

Abu-Lughod, L., 2002. Do Muslim women really need saving? Anthropological reflections on cultural relativism and its others. American Anthropologist 104, 783–790. Abu-Lughod, L., 2006. Writing against culture. In: Lewin, E. (Ed.), Feminist anthropology: Reader. Blackwell Publishing, Malden, MA. Acheson, J., 2006. Institutional failure in resource management. Annual Review of Anthropology 35, 117–134. Adger, N., 2003. Social capital, collective action, and adaptive action to climate change. Economic Geography 79, 387–404. Adler, P.S., Kwon, S.-W., 2002. Social capital: prospect for new concept. Academy of Management Review 27, 17–40. Agrawal, B., 2001. Participatory exclusion, community forestry and gender: an analysis for South Asia and the conceptual framework. World Development 29, 1623–1648. Agrawal, A., Gibson, C.C., 2001. The role of the community in natural resource reservation. In: Agrawal, A., Gibson, C.C. (Eds.), Communities and the environment: ethnicity, gender, and the state in community-based conservation. Rutgers University Press, New Brunswick. Anderson, B., 1983. Imagined communities: Reflections on the origin and spread of nationalism. Verso, London. Aoki, M., 2001. Community norms and embeddedness: a game-theoric approach. In: Hayami, Y., Aoki, M. (Eds.), Communities and Markets in Economic Development. Oxford University Press, Oxford, pp. 97–128. Appadurai, A., 2001. Deep democracy: urban governmentality and the horizon of politics. Environment and Urbanization 13, 23–43. Archer, M.S., 2003. Structure, agency, and the internal conversation. Cambridge University Press, Cambridge. Arrow, K.J., 1999. Observation on social capital. In: Dasgupta, P., Serageldin, I. (Eds.), Social capital: A multifaceted perspective. World Bank, Washington D.C, pp. 3–6. Aumann, R.J., 1976. Agreeing to disagree. The Annals of Statistics 6, 1236–1239. Ballet, J., Sirven, N., Riquiers-Desjardin, M., 2007. Social capital and natural resource management: a critical perspective. The Journal of Environment and Development 16, 355–374. Bardhan, P., Ray, I., 2008. Economists, anthropologists, and the contested commons. In: Bardhan, P., Ray, I. (Eds.), Contested commons: Conversation between economists and anthropologists. Blackwell Publishing, Oxford. Becker, M., 2001. Managing dispersed knowledge: organizational problems, managerial strategies, and their effectiveness. Journal of Management Studies 38, 1037–1051.

E CO L O G I CA L EC O NO M IC S 6 8 (2 0 0 9) 1 54 9–1 56 2

Benhabib, S., 1992. Models of public sphere: Hannah Arendt, the liberal tradition and Jurgen Habermas. In: Calhoun, C. (Ed.), Habermas and public sphere. MIT Press, Cambridge M.A. Berger, P., Luckmann, T., 1966. The social construction of reality: A treatise in the sociology of knowledge. Doubleday, Garden City N.Y. Bourdieu, P., 1971. The thinkable and the unthinkable. Times Literary Supplement. Bourdieu, P., 1984. Distinction: a social critique of the judgment of taste. Harvard University Press, Cambridge M.A. Bourdieu, P., 1986. Forms of capital. In: Richardson, J.G. (Ed.), Handbook of theory and research for the sociology of education. Greenwood Press, New York. Bourdieu, P., 1990. The logic of practice. Stanford University Press, Stanford Calif. Bourdieu, P., Wacquant, L.C.J.D., 1992. An invitation to reflexive sociology. University of Chicago Press, Chicago. Bowles, S., Gintis, H., 2002. Social capital and community governance. The Economic Journal 112, 419–436. Boyce, J.K., 2006. Inequality and environmental protection. In: Baland J.M., et al. (Ed.), Inequality, Cooperation and Environmental Sustainability. Princeton University Press, Princeton NJ. Carroll, M.C., Stanfiled, J.R., 2003. Social capital, Karl Polanyi, and American social and institutional economics. Journal of Economic Issues 37, 397–404. Chwe, M.S.Y., 1999. Structure and strategy of collective action. The American Journal of Sociology 105, 128–156. Chwe, M.S.Y., 2001. Rational ritual: culture, coordination, and common knowledge. Princeton University Press, Princeton N.J. Cleaver, F., 2000. Moral ecological rationality, institutions and the management of common property resources. Development and Change 31, 361–383. Cleaver, F., 2003. Reinventing institutions: brokerage and the social embeddedness of natural resource management. In: Benjaminsen, T.A., Lund, C. (Eds.), Securing Land Rights in Africa. Frank Cass, in association with EADI, European Association of Development Research and Training Institutes, London. Coleman, J., 1988. Social capital in the creation of human capital. The American Journal of Sociology 94, S95–S120. Collier, P., 2002. Social capital and poverty: microeconomic perspective. In: Grootaert, C., Bastelaer, T.V. (Eds.), The role of social capital in development: An empirical assessment. Cambridge University Press, Cambridge. Cook, B., Kothari, U., 2001. Participation: the new tyranny? Zed Books, London. Davis, J., 2002. Collective intentionality and individual behaviour. In: Fullbrook, E. (Ed.), Intersubjectivity in economics: Agents and structures. Routledge, London, pp. 11–27. Dolfsma, W., Dannreuther, C., 2003. Subject and boundaries: contesting social capital-based policies. Journal of Economic Issues 37, 405–413. Douglas, M., 1986. How institutions think. Syracuse University Press, Syracuse, N.Y. Dryzek, J.S., 1990. Discursive democracy: politics, policy, and political science. Cambridge University Press, Cambridge. Durlauf, S.N., 2002. On the empirics of social capital. The Economic Journal 112, 459–479. Eisenstadt, S.N., 1984. Patrons, clients, and friends: Interpersonal relations and the structure of trust in society. Cambridge University Press, Cambridge. Elias, N., 1978. Game model. In: Elias, N. (Ed.), What is sociology? Columbia University Press, New York. Etzioni, A., 1988. The moral dimension: Toward a new economics. Free Press, New York. Ferguson, J., 1994. Anti-politics machine: “development,” depoliticization, and bureaucratic power in Lesotho. University of Minnesota Press, Minnesota. Fine, B., 1999. The developmental state is dead: long live social capital? Development and Change 30, 1–19.

1561

Fine, B., 2001. Social capital versus social theory: political economy and social science at the turn of the millennium. Routledge, London. Foucault, M., 1965. Madness and civilization: a history of insanity in the age of reason. Pantheon Books, New York. Fraser, N., 1992. Rethinking the public sphere: contribution to actually existing democracy. In: Calhoun, C. (Ed.), Habermas and public sphere. MIT Press, Cambridge M.A. Gächter, S., Fehr, E., 1999. Collective action as social exchange. Journal of Economic Behavior & Organization 39, 341–369. Gale, F., 1998. Theorizing power in ecological economics. Ecological Economics 27, 131–138. Geertz, C., 1973. Thick description: toward an interpretive theory of culture. In: Geertz, C. (Ed.), The Interpretation of Cultures. Basic Books, New York. Giddens, A., 1984. The constitution of society: outline of the theory of structuration. University of California Press, Berkeley. Goffman, E., 1967. Interaction ritual; essays in face-to-face behavior. Aldine Pub. Co., Chicago. Goodin, R., 1996. Enfranchising the Earth and its alternatives. Political Studies 44, 835–849. Granovetter, M., 1973. The strength of weak ties. American Journal of Sociology 78, 1360–1380. Granovetter, M., 1985. Economic action and social structure: the problem of embeddedness. American Journal of Sociology 91, 481–510. Guha, R., Spivak, G.C., 1988. Selected Subaltern studies. Oxford University Press, New York. Guiso, L., Sapenza, P., Zingales, L., 2007. Social capital and good culture. EUO Working papers, ECO2007/57. Habermas, J., 1984. The theory of communicative action. Beacon Press, Boston. Habermas, J., 1986. Hanna Arendt's communication concept of power. In: Lukes, S. (Ed.), Power. New York University Press, New York. Hajar, M., 1995. Politics of environmental discourse. Clarendon Press, Oxfird. Hardin, G., 1968. Tragedy of the commons. Science 162, 1243–1248. Harris, J., 2001. Depoliticizing development: the World Bank and social capital. Anthem, London. Harsanyi, J.C., 1965. Bargaining and conflict situations in the light of a new approach to game theory. The American Economic Review 55, 447–457. Hodgson, G., 1997. The ubiquity of habits and rules. Cambridge Journal of Economics 21, 663–684. Hodgson, G., 2000. Structures and institutions: Reflection on Institutionalism, Structurism theory and Critical Realism. Workshop on Realism and Economics. Jentoft, S., 2000. The community: a missing link of fisheries management. Marine Policy 24, 53–59. Katz, E.G., 2000. Social capital and natural capital: a comparative analysis of land tenure and natural resource management in Guatemala. Land Economics 76, 114–132. Kelly, M., 1994. Critique and power: recasting the Foucault/Habermas debate. MIT Press, Cambridge M.A. Knack, S., Keefer, P., 1997. Does social capital have economic impact? A cross-country investigation. Quarterly Journal of Economics 112, 1252–1288. Kohn, M., 2000. Language, power and persuasion: toward a critique of deliberative democracy. Constellation 7, 408–429. Leach, M., Mearns, R., Scoones, I., 1999. Environmental entitlements: dynamics and institutions in community-based natural resource management. World Development 27, 225–247. Lehtonen, M., 2004. The environmental–social interface of sustainable development: capacities, social capital, institutions. Ecological Economics 49, 199–214. Lin, N., 2001. Social capital: A theory of social structure and action. Cambridge University Press, Cambridge. Maruyama, Y., 2003. Multiplicity of nature and the universal discourse: case study of Japanese Monkeys (in Japanese). Kagaku Gijutsu Shakai-ron Kenkyu 2, 68–78.

1562

E C O L O G IC A L E C O N O M IC S 6 8 ( 2 0 09 ) 15 4 9 –1 56 2

McCay, B., Jentoft, S., 1998. Market or community failure? Critical perspective on common property research. Human Organization 57, 21–29. McNiell, D., 2007. Social capital or sociality? Methodological contrast between economics and other social sciences. In: Ioannides, S., Nielsen, K. (Eds.), Economics and the social sciences: Boundaries, interaction and integration. Edward Elgar, North Hampton M.A., pp. 163–184. Morotomi, T., 2003. Environment (in Japanese). Iwanami Shoten, Tokyo. Mosse, D., 1997. The symbolic making of a common property resource: history, ecology and locality in the tank-irrigated landscape in South India. Development and Change 28, 467–504. Mosse, D., 2006. Collective action, common property, and social capital in South India: an anthropological commentary. Economic Development and Cultural Change 54, 698–724. Narayan, D., Pritchett, L., 1999. Cents and sociability: household income and social capital in rural Tanzania. Economic Development and Cultural Change 17, 871–897. Narayan, D., Woolcok, M., 2000. Social capital: implication for development theory, research and policy. The World Bank Research Observer 15, 225–249. Newell, S., Carole, T., Huang, J., 2004. Social capital and knowledge integration in EPR project team: the importance of bridging and bonding. British Journal of Management 15, S43–S54. North, D.C., 1990. Institutions, institutional change and economic performance. Cambridge University Press, Cambridge. Ockwell, D., Rydin, Y., 2006. Conflicting discourses of knowledge: understanding the policy adoption of pro-burning knowledge claims in Cape York peninsula, Australia. Environmental Politics 15, 379–398. Ostrom, E., 1990a. Chapter 5 Analyzing institutional failures and fragilities. In: Ostrom, E. (Ed.), Governing the commons: The evolutions of institutions for collective action. Cambridge University Press, Cambridge. Ostrom, E., 1990b. Governing the commons: the evolution of institutions for collective action. Cambridge University Press, Cambridge. Ostrom, E., 2000. Understanding social capital: learning from the analysis and experience of participation. In: Dasgupta, P. (Ed.), Social capital: a multifaceted perspective. World Bank, Washington D.C. Ostrom, E., Ahn, T.K., 2002. Introduction. In: Ostrom, E., Ahn, T.K. (Eds.), Foundation of social capital. Edward Elgar Pub., Northhampton M. A. Paavola, J., Adger, N., 2005. Institutional ecological economics. Ecological Economics 53, 353–368. Pérez-Cierra, V., Lovett, J.C., 2006. Power distribution, the external environment and common property forest management: a local user group model. Ecological Economics 59, 341–352. Portes, A., 1993. Embeddedness and immigration: notes on the social determinants of economic action. American Journal of Sociology 98, 1320–1350. Portes, A., 1998. Social capital: its origins and application in modern sociology. Annual Review of Sociology 24, 1–24. Pretty, J., Ward, H., 2001. Social capital and the environment. World Development 29, 209–227. Putnam, R., 1993. Making democracy work: civic traditions in modern Italy. Princeton University Press, Princeton N.J. Putnam, R., 2000. Bowling alone: the collapse and revival of American community. Simon & Schuster, New York. Ray, I., 2008. Cooperative conversation: outcomes and process in economics and anthropology. In: Bardhan, P., Ray, I. (Eds.),

Contested commons: Conversation between economists and anthropologists. Blackwell Publisher, Oxford. Ray, S., Bijarnia, M., 2007. Power relations and institutional outcomes: a case of pastureland development in semi-arid Rajasthan. Ecological Economics 62, 360–372. Rodríguez, L.C., Pascual, U., 2004. Land clearance and social capital in mountain agro-ecosystems: the case of Opuntia scrubland in Ayacucho, Peru. Ecological Economics 49, 243–252. Røpke, I., 2005. Trends in the development of ecological economics from the late 1980s to early 2000s. Ecological Economics 55, 262–290. Runge, F.C., 1992. Common property and collective action in economic development. In: Bromley, D. (Ed.), Making common work: Theory, practice and policy. ICS Press, San Francisco Calif., pp. 17–40. Sewell, W.H., 1992. A theory of structure: duality, agency and transformation. American Journal of Sociology 98, 1–29. Sobel, J., 2002. Can we trust social capital? Journal of Economic Literature 15, 139–154. Solow, R.M., 1999. Notes on social capital and economic performance. In: Dasgupta, P., Serageldin, I. (Eds.), Social capital: A multifaceted perspective. World Bank, Washington D.C., pp. 6–12. Spash, C.L., Villena, M.G., 1998. Exploring the approaches of institutional economics to environment. University of Cambridge Press, Cambridge. Spivak, G.C., 1988. Can the subaltern speak? Marxism and the interpretation of culture. In: Nelson, C., Grossberg, L. (Eds.), Marxism and the interpretation of culture. University of Illinois Press, Urbana. Swartz, D., 1997. Culture & power: the sociology of Pierre Bourdieu. University of Chicago Press, Chicago. Taylor, C., 1993. To follow a rule. In: Calhoun, C., LiPuma, E., Postone, M. (Eds.), Bourdieu: critical perspectives. University of Chicago Press, Chicago. Taylor, C., 1994. The politics of recognition. In: Taylor, C. (Ed.), Multi-culturalism: Examining the politics of recognition. Princeton University Press, Princeton N. J. Tuomela, R., 1995. The importance of us: A philosophical study of basic social notions. Stanford University Press, Stanford Calif. Tuomela, R., 2007. The philosophy of sociality: The shared point of view. Oxford University Press, Oxford. van Steveren, I., 2003. Beyond social capital and poverty research. Journal of Economic Issues 37, 415–423. Varughese, G., Ostrom, E., 2001. The contested role of heterogeneity in collective action: some evidence from community forestry in Nepal. Word Development 29, 747–765. Vatn, A., 2005. Institutions and the environment. Edward Elgar Pub., Cheltenham. Villa, D.R., 1992. Postmodernism and the public sphere. American Political Science Review 86, 712–721. Williamson, O., 1981. The economics of organization: the transaction cost approach. American Journal of Sociology 87, 548–577. Willis, P., 1977. Learning to labour: How working class kids get working class jobs. Saxon House, Farnborough Eng. Woolcok, M., 1998. Social capital and economic development: toward theoretical synthesis and policy framework. Theory and Society 27, 151–208. Woolcok, M., Narayan, D., 2000. Social capital: implications for development theory, research, and policy. The World Bank Research Observer 15, 225–249.