1 Dancing with storks: The role of power relations in Payments for Ecosystem Services 1 2 3 Hiroe Ishihara, Unai Pascual and Ian Hodge 4 5 Abstract 6 The institutional change induced by payments for ecosystem services (PES) schemes is a ‘messy’ 7 process. The uptake and outcomes of PES schemes cannot be fully explained from a rational 8 choice perspective. The notion of ‘institutional bricolage’ is needed to analyse how actors 9 assemble or reshape their actions by combining new institutions such as a PES scheme within 10 other locally embedded institutions. A case study from Japan is used to illustrate how a PES 11 scheme designed to conserve the habitat of a charismatic and endangered flagship species, the 12 Oriental White Stork, has been reshaped by social actors to fit the locally dominant 13 ‘institutional logic’. We also show how the resulting institutional change is not only able to 14 subvert policy makers’ original assumptions, for instance about how to target and distribute the 15 payments, but can also contribute to the reproduction of unequal power relations. 16 17 Keywords 18 Institutional bricolage, power relations, agency, payments for ecosystem services, 19 environmental governance, common pool resources 20 21 2 1. Introduction 22 The use of economic incentives, such as so-called payments for ecosystem services (PES), 23 continues to gain increasing attention in environmental policy circles as an efficient and 24 potentially equitable tool for environmental governance, including the conservation of 25 biodiversity (Pascual et al., 2014). Here PES is defined broadly as a transfer of resources 26 between actors, which aims to create incentives, subject to clear conditions, to align individual 27 and/or collective resource use decisions with the social interest in the management of natural 28 resources (modified from Muradian et al., 2010). The burgeoning research on PES schemes 29 indicates that i) implementation of PES schemes is not a straightforward process as rational 30 choice models might suggest (e.g Osborne, 2011; Rodríguez-de-Francisco and Budds, 2015; 31 Rodríguez de Francisco et al., 2013; Vatn, 2010), ii) there is a high degree of complexity 32 associated with the trade-offs between cost-effectiveness and other policy goals, such as social 33 equity (e.g. Pascual et al., 2010; Pascual and Phelps, 2014), and iii) the introduction of PES 34 impacts on power relations amongst stakeholders (e.g. Milne and Adams, 2012; 35 Rodríguez-de-Francisco and Budds, 2015). 36 We argue that the introduction of PES schemes is often associated with a layer of complexity 37 because ecosystem service providers are not just suppliers of such services but are also 38 ‘institutional bricoleurs’ who rearrange the standardised PES-logics in order to ‘fit’ their own 39 (local) social context. The term ‘bricoleurs’ implies that actors creatively combine elements 40 from different institutional contexts into a new institutional arrangement (Christiansen, Larke 41 and Lounsbury, 2013). This concept enables us to challenge the view of actors as powerless 42 victims of institutional change. The application of the notion of institutional bricolage is helpful 43 in undertaking power-sensitive analysis of environmental governance and can contribute to the 44 inclusion of power relations into the ecosystem services framework (Berbés-Blázquez et al., 45 2016; Pascual and Phelps, 2014; Van Hecken et al., 2015). We borrow the term ‘institutional 46 bricolage’ from cultural anthropology (Douglas, 1986; Lévi-Strauss, 1966) and organisational 47 studies (Christiansen, Larke and Lounsbury, 2013; Lounsbury and Glynn, 2001; Schneiberg, 48 2002) and adapt it to conceptualise a process in which actors assemble or reshape existing (often 49 local) institutions, such as collective action norms in the management of common pool 50 resources (hereafter CPR), by combining them with a recently introduced PES scheme. We also 51 3 argue that due to institutional bricolage, the introduction of PES can often lead to a ‘messy’ 52 institutional process with unintended consequences.1 53 A case study from a rural community in Japan within Toyoka City is presented to illustrate the 54 application of the idea of ‘institutional bricolage’ and show how it helps to analyse the ways in 55 which a PES scheme may lead to messy institutional change. This PES scheme has been 56 introduced as an incentive to support the conservation of a flagship endangered species, the 57 Oriental White Stork (Ciconia boyciana), which is on the IUCN Red List of Threatened Species 58 and can only be found in Japan, South Korea, China and Russia. More generally, in Japan, PES 59 approaches have been heralded as tackling the problem of under-use, rather than 60 over-exploitation, of CPRs (see Section 3.1. for a description of problems caused by the 61 under-use of CPRs). Since the 1990s, in Toyoka city, the local government has implemented 62 eco-certification schemes for organic farming and eco-tourism to tackle this problem (Hyogo 63 Prefectural Government, 2009). Toyoka city is considered to be one of the success stories in the 64 use of a broadly defined PES scheme, along with Osaki City and Sado City, where similar 65 schemes supporting flagship species have been implemented (Honda, 2008a; Wittmer and 66 Gundimeda, 2010).2 67 The case study from Toyoka city reveals how community residents act as institutional bricoleurs 68 rearranging the PES-logic to ‘fit’ their own local context and to reproduce or change the power 69 relations within their community. It illustrates that the implementation of a PES scheme is not 70 just about incentivising people to align their resource use decisions with broader social interests 71 but also about power struggles among stakeholders, such as between policy makers and local 72 residents as well as among the residents themselves. 73 1 ‘Messy’ institutional process does not imply that deliberate institutional design is necessarily impossible. Each actor will act deliberately to achieve their own goal and use their resources to negotiate the institutional outcome. However, we term this process as ‘messy’ because it is not the deliberation of one actor or one social group, such as the implementer of a PES scheme, which determines the outcome; rather it is determined by the negotiations amongst various stakeholders. 2 Both cities use flagship species similar to that of Toyoka. In Osaki city, the protected species is a goose and in Sado city, it is the crested ibis For details see; http://www.biodic.go.jp/biodiversity/shiraberu/policy/pes/satotisatoyama/satotisatoyama01.html for Osaki city and http://www.biodic.go.jp/biodiversity/shiraberu/policy/pes/satotisatoyama/satotisatoyama03.html for Sado city. http://www.biodic.go.jp/biodiversity/shiraberu/policy/pes/satotisatoyama/satotisatoyama01.html http://www.biodic.go.jp/biodiversity/shiraberu/policy/pes/satotisatoyama/satotisatoyama03.html 4 The next section introduces the central idea of institutional bricolage from a critical 74 institutionalist perspective. Then, Section 3 describes the background and methodology used in 75 the case study and Section 4 analyses how the PES scheme designed for the conservation of the 76 Oriental White Stork in Toyoka city was contested, altered and articulated by the local 77 community. The paper concludes with the main lessons learnt for future PES-based governance 78 of biodiversity through the lens of institutional bricolage. 79 80 2. Setting the scene: Institutional bricolage, agency, and power 81 In recent years the institutionalist school of thought on the study of the commons, often termed 82 critical institutionalism, has emerged to fill the gaps left by mainstream institutionalism led by 83 the work of Elinor Ostrom (e.g. Cleaver, 2012; De Koning and Cleaver, 2012; Hall et al., 2014). 84 The mainstream approach generally understands institutions as ‘the rules of the game’ which 85 define what actors may (permitted), must (obliged) or may not (forbidden) do (Crawford and 86 Ostrom, 1995; North, 1990). It argues that rational actors design appropriate institutions to fulfil 87 certain functions, for instance to solve collective action problems or information asymmetries 88 (Hotimsky et al., 2006). When carefully crafted, the assumption goes, CPR institutions can in 89 principle curb individuals’ selfish incentives to free-ride and enhance collective action to avoid 90 the over-exploitation of the commons (McKean, 1992; Ostrom, 1990; Wade, 1989). 91 Critical institutionalists emphasize that new institutions cannot be separated from the 92 pre-existing social and cultural embeddedness of resource users and thus the messiness that 93 arises from the multiplicity of their interests and the complexity through social interactions 94 (Cleaver, 2002; Fabinyi et al., 2014; McCay, 2002; Mosse, 1997). For this school of thought, 95 resource users do not hold narrowly defined utilitarian and purely instrumental preferences 96 (Mosse, 1997). Instead they are viewed as holding multiple social identities and rationalities 97 (Schnegg and Linke, 2015). This implies that CPR use cannot be understood solely in terms of a 98 narrow desire to optimise a given objective (e.g. income maximization or risk minimization). It 99 can also be strongly influenced by other concerns and interests stemming from various social 100 identities and roles, as well as being associated with norms relating to authority/respect, 101 in-group loyalty and fairness/reciprocity (Chan et al., 2016). It follows that institutional change 102 relating to CPR use can be motivated, for example, by an elite’s desire to reproduce power 103 relations (Hall et al., 2014; Hotimsky et al., 2006). It should also be pointed out that collective 104 action for the conservation of CPR does not necessarily guarantee outcomes that are either 105 socially or ecologically desirable (Ishihara and Pascual, 2009). 106 5 Critical institutionalism draws strongly on sociology and anthropology and institutions are 107 understood as “social arrangements that shape and regulate human behaviour and have some 108 degree of permanency and purpose transcending individual human lives and intentions” 109 (Cleaver, 2012, p. 8). The process of institutional change is understood through the idea of 110 ‘institutional bricolage’ (Cleaver, 2002; Galvan, 2004), not as an issue of ‘crafting’ rules 111 (Ostrom, 1992) or of searching for ‘institutional fit’ (Young, 2002) but by explicitly 112 acknowledging the ‘messy’ process or the ‘unintended’ outcomes that it may produce. The term 113 ‘bricolage’ was originally used by the French anthropologist Levi-Strauss (1966) to describe 114 how the ‘savage’ mind understands different cultures, and later developed by Douglas (1986) in 115 the context of institutions, who argued that to economise on cognitive effort, actors borrow 116 existing familiar institutions as a ‘frame of reference’ to understand new and ‘unfamiliar’ social 117 phenomena. In the study of the commons, the use of the term institutional bricolage refers to “a 118 process through which people, consciously and non-consciously, assemble or reshape 119 institutional arrangements, drawing on whatever materials and resources are available, 120 regardless of their original purpose” (De Koning and Cleaver, 2012, p. 4). 121 This notion of institutional bricolage enables us to better understand human agency and the 122 dynamic power relations involved in the institutional processes that arise in the governance of 123 the commons. At this point, we make two remarks: first, the actors involved in institutional 124 change are not powerless victims of this process, rather they are active ‘improvisers’. Drawing 125 on Sewell (1992), we use the term human agency3 to refer to actors’ capacity to transpose and 126 extend an ‘institutional logic’ to a new institutional context, where ‘institutional logic’ is the 127 shared taken-for-granted social prescriptions that guide individual behaviour in an institutional 128 context (Battilana, 2006). The institutional logic thus embodies belief systems and material 129 practices that represent particular worldviews, valued ends, and the appropriate means to 130 achieve such ends (Christiansen, Larke and Lounsbury, 2013). When actors are faced with new 131 situations, they exercise their agency by extending their existing institutional logic and make do 132 with whatever is at hand to fit the new institutional context (Baker and Nelson, 2005). 133 3 Agency is temporally embedded in the past and is oriented towards the future (Emirbayer and Mische, 1998). On the one hand, because actors are born into a specific social structure, they internalise an institutional logic and follow routines forming habits or habitus (Bourdieu, 1990). On the other hand, actors face unexpected outcomes even when they follow habitus. Then they start to question what they have ‘taken for granted’. In such a circumstance, actors distance themselves from so-called common knowledge (Ishihara and Pascual, 2009), creating a space for them to reflect on their actions in relation to their circumstances. 6 The notion of ‘institutional logic’ in this paper is used in a similar sense to that of ‘habitus’ 134 (Bourdieu, 1990) or ‘common knowledge’ 4 (Ishihara and Pascual, 2009). According to 135 Bourdieu, habitus is similar to a ‘sense of the game’. A player in a game, in our terms an 136 institutional bricoleur, is guided by a ‘sense of the game’; but at the same time, the player is not 137 ruled by it. The player still has room to create new strategies and new moves. This is because 138 actors are not merely embedded in one institutional logic but rather in ‘multiple habitus’ (Seo 139 and Creed, 2002). For instance, an actor, such as a farmer, is not just a ‘farmer’, but is also 140 engaged with different institutions and social groups such as neighbourhood associations and 141 families, which yield different logics to justify various types of farming actions (Agrawal and 142 Gibson, 1999; Cleaver, 2002; Leach et al., 1999). 143 At the same time, some actors have the capacity to impose their institutional logic as the 144 legitimate worldview and by so doing become the dominant group (Bourdieu, 1989). In this 145 case a given institutional logic assumes a dominant role (Battilana, 2006; Reay and Hinings, 146 2005). This leads us to our second remark: power relations are necessarily involved in any 147 institutional change. When actors are faced with a new institutional context, they can generally 148 contest which institutional logic is to be adopted. In other words, the institutional process 149 involves ‘battlefields of legitimacy’5 where both dominant and marginalised actors challenge as 150 to whose institutional logic is to be applied in guiding the process of institutional change. To 151 this end, the use of the term power6 in this paper refers to the power to ‘naturalise’ institutional 152 logics (Haugaard, 2008) so that an institution can present itself as if it is a part of the 153 institutional logic as “the way we do things around here” (Foucault, 1977). 154 Of course, some actors may resist the dominant or naturalised institutional logic. However, the 155 possibility of resistance in the context of institutional change should not be romanticized (Van 156 Hecken et al., 2015). We do not deny the possibility of marginalised actors challenging the 157 naturalised institutional logic when institutions are going through transformation. The dominant 158 actors will try to maintain their power by ensuring predictability but marginalised actors also 159 4 Other authors refer to this aspect of institutions as ‘institutional context’ (Clegg, 2010). We acknowledge that we are not using the term logic in a strict philosophical sense. However, we chose not to use the term ‘context’ in order to avoid confusion with the social and cultural context in which the actors are embedded which includes wider connotations beyond institutional logic. 5 This notion is a modification of ‘battlefields of knowledge’ (Long and Long, 1992) in which different actors fight over meaning and moral affiliation in order to gain dominance in a field or domain. 6 We concur with Haugaard (2008) that we should not look for a definitive definition of power; rather consider it as a ‘family resemblance’ concept in line with Wittgenstein (1968). 7 have an interest in adopting counterhegemonic practices to diminish the legitimacy of the 160 dominant group and thus of the dominant institutional logic (Clegg, 2010). However, obviously 161 not all marginalised actors have the capacity to challenge or resist. Depending on their social 162 position within the community, actors have different levels of access to physical and so-called 163 symbolic resources (Bourdieu, 1990). As a result, they do not all have the same capacity to 164 produce effective claims towards naturalising their own institutional logic (Battilana, 2006). In 165 fact, it is often the case that in contrast to more powerful actors, marginalised actors have to 166 bear the burden of proof (Kohn, 2000). 167 The emerging key message is that institutional bricolage, which focuses on agency embedded 168 inside a set of power relations, enables us to better understand institutional change induced by 169 new environmental governance approaches, such as through PES schemes (Büscher, 2012; 170 Rodríguez-de-Francisco and Budds, 2015; Van Hecken et al., 2015). An institutional logic held 171 by the ‘experts’ who introduce these approaches, i.e. policy makers and international NGOs, is 172 not always shared with other key actors, e.g. local providers of ecosystem services. In such 173 cases, the introduction of a PES approach to solve an environmental ‘externality’ provides the 174 opportunity for local actors, both dominant and marginalised, to seek to exercise their agency in 175 various ways. One way is for dominant actors to use this opportunity, e.g. by designing and 176 implementing PES, to extend the legitimacy of their institutional logic and to reproduce their 177 dominance over marginalised actors. But marginalised actors may also seize this opportunity to 178 legitimise their own claims, such as access to certain ecosystem goods or services, demand for 179 land tenure security, etc. which can challenge the power relations within their community 180 (Hendrickson and Corbera, 2015; Shapiro-Garza, 2013). It is thus important to understand the 181 institutional change induced by the introduction of PES schemes as a ‘messy’ process where 182 bricoleurs continuously fight over the legitimacy of their own institutional logics. In this way, 183 new power struggles arising from the introduction of PES schemes can be interpreted as key 184 social inputs to be considered in their design and implementation (Muradian et al., 2013; 185 Pascual et al., 2014; Van Hecken et al., 2015). 186 It also follows that institutional change is not merely a process guided by rational calculation in 187 order to achieve an optimal solution over an externality problem. Of course, within this 188 mainstream approach, ‘second generation game theorists’ incorporate bounded rationality and 189 other types of rationalities, such as interpersonal altruism, fairness, reciprocity and inequity 190 aversion, into a rational choice model (Narloch et al., 2012; Ostrom and Ahn, 2009). However, 191 the problem with this kind of modelling lies in its strict adherence to methodological 192 individualism (Peacock, 2011). Altruistic individuals may choose to act altruistically because 193 their welfare depends on them taking an altruistic action. Here “the tight link between individual 194 http://en.wikipedia.org/wiki/Altruism http://en.wikipedia.org/wiki/Fair_division http://en.wikipedia.org/wiki/Reciprocity http://en.wikipedia.org/wiki/Inequity_aversion http://en.wikipedia.org/wiki/Inequity_aversion 8 welfare and choice of action” (Sen, 1982, p. 8) is not broken. As a result, the mainstream 195 approach cannot fully explain other types of institutional logic such as individuals acting as part 196 of a ‘community’ with which they share a common ‘identity’ (Martins, 2009). 197 3. Case study and methodology 198 3.1. Case study background 199 Many CPRs in Japan face the problem of under-use, often termed as the ‘Satoyama’ problem 200 (Shimada, 2015). Historically, rural communities in Japan were heavily dependent on CPRs, 201 such as communal forests, shallow seas, paddy fields and irrigation channels and thus developed 202 strict institutions to avoid their over-use. These institutions are well-known for the graduated 203 sanctions (McKean, 1992) linked to various social customs and norms that lead rule breakers to 204 face social ostracism (Aoki, 2001).7 This suppressed the over-use of CPRs. However, following 205 rural depopulation and economic transformation in Japan, especially since the 1960s (Hasumi, 206 1990; Torigoe, 2007), the concern regarding CPRs has shifted from over-use to under-use, 207 bringing different environmental challenges (Shimada, 2015). For example, the under-use of 208 communal forests in Japan (called Iriai) causes a loss of biodiversity (Murota and Mitsumata 209 2004) and ‘co-produced’ ecosystem services (Palomo et al., 2016), such as regulating services 210 associated with water purification and prevention of land-slides (Morimoto, 2014). Additionally, 211 and in conjunction with the under-use problem, important traditional communal organisations 212 and institutions, once prevalent in hamlets and villages, are being eroded (Onda, 2006). Often, 213 these communities feel themselves heavily burdened with the responsibilities of having to 214 continue to manage the CPRs in traditional ways (Mitsumata et al., 2008). This has led many 215 communities to hand over their CPR management to national or local governments to avoid the 216 burden involved (Mitsumata and Inoue, 2010; Murota and Mitsumata, 2004). Other 217 communities have chosen to co-manage their CPRs by involving new actors from outside the 218 local community. For instance, schemes such as the ‘forest volunteer’ system, where volunteers 219 are recruited from urban areas to conduct forest management together with forest owners 220 (Yamamoto, 2003), or the terraced paddy field owner system, where urban residents invest 221 financially and/or provide a workforce for the conservation of traditional terraced paddy fields 222 (Maeda and Takao, 2007; Nakajima, 2007). 223 7 This custom was called ‘Mura-Hachibu’. This custom ostracises the households that violate the CPR institution from various social groups and social event organised by these groups. As Aoki (2001) argues this ostracism had a detrimental effect on the livelihoods of the villagers, the violation of CPR institutions was limited. 9 In this evolving social-ecological context, Toyoka City is considered by many as a success story 224 for CPR management through the introduction of its PES scheme (Wittmer and Gundimeda, 225 2010). The uniqueness of this PES lies in the fact that it has been implemented as part of a 226 policy to conserve the Oriental White Stork, a flagship species. This species, which had become 227 extinct in Japan by the 1980s, was reintroduced in Toyoka City which offered the last available 228 habitat in Japan (Toyoka Municipal Government, 2007a). Oriental White Storks used to live in 229 paddy fields and the surrounding irrigation channels. But the introduction of chemical pesticides 230 and modernization of agricultural infrastructure during the 1950s drastically changed the natural 231 environment, making it less suitable as a stork habitat and adding pressure to other direct drivers 232 such as excessive hunting during the 1920s and 1930s and a decrease in genetic diversity due to 233 the shrinking population. After the reintroduction of the species, various policies were tested to 234 improve the quality of the habitat, mainly targeted at enhancing biodiversity in the paddy fields 235 and surrounding wetlands. 236 Two types of PES have been implemented in Toyoka; an eco-certification scheme and a 237 co-management scheme. The former, called “the Dance of the Storks” (kounotori no mai), 238 incentivized farmers to switch from conventional to organic farming by providing a premium 239 price and some additional subsidies (Honda, 2008b; Kikuchi, 2006; Toyoka Municipal 240 Government, 2007b). Further, this scheme allows consumers to pay a price premium not just for 241 organic products but also for providing Stork habitat. The co-management scheme is still at a 242 nascent stage and to date has only been implemented in Tai Hamlet, which is our case study site 243 (NGO Wetlands Action Circle for Oriental White Storks, 2012). The hamlet receives limited 244 funds from the local government and an NGO in order to conduct activities for stork 245 conservation (for details see section 4.1.). 246 This paper focuses on the co-management scheme introduced in Tai Hamlet as a pilot project by 247 the local Toyoka government and a local NGO. The hamlet is located in the northern part of 248 Toyoka City (see map in Figure 1). In 2013, it consisted of around 50 households with a 249 population of around 200. The hamlet suffers from an ageing and declining population: one 250 third of the population is over 65 years old, well above the national average of 23% (Cabinet 251 Office, 2012). Average annual per capita income is about JPY 3.5 million or ca. USD 43,750,8 252 above the national average of JPY 2.9 million or ca. USD 35,750. Before the 1960s, the 253 residents of this hamlet were heavily dependent on CPRs for their livelihoods, especially the 254 rice paddies and shallow sea fishery. However, residents can no longer sustain their traditional 255 8 Using the conversion rate of USD1= JPY80 (average rate for 2013). 10 livelihoods based on agriculture and fishing, and during the 1960s, the hamlet witnessed 256 significant out-migration of young people to urban areas such as Osaka and Tokyo (Minato 257 Community Centre, 1965). This led to the disbandment of various communal organisations, 258 such as the young men’s and women’s associations. Further, during the 1960s, the remaining 259 residents started to work outside of the hamlet, including in the adjacent Kinosaki area or the 260 city centre of Toyoka. This diversified not only the income sources of the residents but also 261 their interest in CPR institutions and rice cultivation activities.9 In other words, this hamlet has 262 gone through a gradual ‘delocalization’ process, through which it has become more intimately 263 connected to actors outside the local domain (Ojha et al. 2016). In this social context, the hamlet, 264 or at least its leaders, decided to accept the co-management scheme in 2009 by modifying the 265 existing local institution for the management of its CPRs, particularly the terraced paddy fields. 266 267 Figure 1. Location of field study of Tai Hamlet in Toyoka City (Hyogo Prefecture, Japan) 268 269 9 Due to this diverse interest, Tai Hamlet failed to implement farmland consolidation despite three attempts since the 1960s. To implement the consolidation project, all the paddy field owners in the hamlet had to agree to share the cost of the project. However, the hamlet failed to reach agreement due to the different interests among the land owners. 11 3.2. Fieldwork methods 270 Fieldwork was carried out in Tai Hamlet between January and December of 2010 and further 271 short-term field visits were undertaken between 2011 and 2013 to follow up on specific issues. 272 Methods included i) in-depth interviews with different key actors, ii) participant observation of 273 various communal events in the hamlet, and iii) observation of meetings between the hamlet 274 leaders and other stakeholders such as government officials and NGOs and meetings among the 275 residents and the leadership council. The KJ method (Kawakita, 1970, 1967) (discussed below) 276 was used for qualitative analysis of the data collected through these three complementary 277 approaches. Discourse analysis was undertaken of government policy documents related to 278 stork conservation and use was made of a municipal database on agricultural statistics. 279 Two types of interviews were undertaken in Japanese. Firstly, a series of non-structured 280 interviews was conducted with residents of the hamlet, especially residents from older 281 generations and members of the leadership council, 10 regarding the changes in CPR 282 management, their livelihoods, and past as well as current life in the hamlet. The sampling for 283 unstructured interviews used a snowball method. Through these in-depth interviews, we gained 284 basic information on the institutional and organisational structure of the hamlet. 285 Subsequently, strategically targeted interviews were conducted with individuals and groups in 286 relation to each of the various organisations in the community such as the leadership council, 287 neighbourhood groups and fire-fighting group (see a list of communal groups in Appendix 1 288 and an organigram of the hamlet in Appendix 2). At least two group interviews were completed 289 with members of each of the organisations. Initial interviews gathered general information about 290 the organisations as well as information about interactions amongst the residents involved in 291 them. Follow-up interviews were conducted to ask in-depth questions, especially regarding i) 292 the roles that each organisation performs in various communal events and ii) the residents’ 293 motivation for participating in these communal events. Since these organisations are organised 294 by gender and age, these group interviews were utilised to collect life histories from each 295 generation and gender. Most of the interviews were carried out through personal visits, i.e. 296 10 The leadership council is composed of nine members who are elected at the annual meeting by all the households in the hamlet. Once they are elected, they will serve as a ‘leader’ (yakuin) until they decide to retire. The position of ‘hamlet chief’ (kucho) is rotated among these nine members and it is this council which makes the decision regarding the hamlet and negotiates with outside parties, such as NGOs and the Toyoka local government. 12 going either to individuals’ houses or to the location of the activity, since most of the groups 297 held meetings at least once a month. 298 Participant observation was conducted during various communal events associated with 299 introduction of a ‘Storks labour day’, e.g. negotiation between the municipal government and 300 the hamlet, as well as with traditional ‘labour days’, communal festivals, funerals, religious 301 events and annual meetings. This participant observation approach provided an important 302 opportunity to closely observe some of the conflicts inside the community and between the 303 community and other actors. In order to provide additional contextual information, meetings 304 held between the hamlet and the government convened to decide various issues relating to 305 conservation policy were also observed. 306 The KJ method, developed by Jiro Kawakita, was used to systematically summarise qualitative 307 data (Kawakita, 1970, 1967). The method takes three steps. First, cards are generated by 308 transcribing information from the interviews. One statement is transcribed on each card. Next, 309 cards with similar statements are clustered together. Finally, the clusters are schematised to 310 illustrate the relationships between clusters. This method makes it possible to analyse the vast 311 amount of data collected through different fieldwork methods, such as interviews and 312 participant observation, without losing its meaning; however, a weakness lies in that its 313 clustering can be rather arbitrary (Sato, 2008). The method was chosen because it affords the 314 most freedom in coding, which was necessary in order to understand the complexity of the CPR 315 institutions, especially the arrangements for the collective management of paddy fields and 316 other communal resources in Tai Hamlet. The statements quoted in the following sections have 317 been carefully chosen to represent the clusters of statements created through the KJ method. 318 They illustrate the views shared among the residents in the hamlet rather than the authors 319 attempting to rephrase them in their own words. 320 321 4. ‘Fitting’ the new PES scheme to the local institutional logic in Tai Hamlet 322 4.1. Introduction of the co-management scheme through the Storks labour day 323 A co-management scheme, based around a ‘Storks labour day’, was introduced in Tai Hamlet in 324 2009. The ‘labour day’ (Hiyaku), is an institution for managing terraced paddy fields. 325 Cultivating paddy rice requires collective action because it utilises a gravity-fed irrigation 326 system (Onda, 2006). Under the Hiyaku institution, each household has to provide the labour of 327 one person, ideally the head of the household or another adult male, on certain days during the 328 13 year. While there is no formal penalty for non-attendance on labour day, attendance is checked 329 by calling out the name of each household in front of everyone, thus creating peer-pressure that 330 functions as an informal sanctioning mechanism. Further, if there is any absentee, the leaders 331 will go to their house to collect the missing person. 332 The Hiyaku institution is deeply embedded in the social context of the hamlet. Although, the 333 main purpose of the labour day was to maintain various types of infrastructure for rice 334 production, such as irrigation channels and paths to the paddy fields, it also has religious 335 significance. Hiyaku is considered as a day to recreate a spiritual pathway for ancestors so that 336 they may come back to the village during the summer period. Further, the hamlet has a custom 337 called ‘selling-off’ (Haraimon) which is also associated with Hiyaku. Historically, residents 338 planning to migrate out of the hamlet were obliged to sell their properties, including their house, 339 lands and terraced paddy fields because they could no longer participate in Hiyaku and fulfil 340 their collective action duties. It is thus important to note that Hiyaku as a collective action 341 institution goes beyond the management of CPRs within the hamlet; it also establishes the ‘civic’ 342 duties of the community and cements the identity of its members. 343 However, the Hiyaku, as a deeply embedded CPR institution, has faced serious challenges as 344 Tai Hamlet has gradually withdrawn from rice cultivation, which stopped completely in 2006. 345 Rice production was ended for various reasons, including the effects of the Japanese 346 Government’s agricultural policy introduced to prevent excessive supply of rice, a rapidly 347 ageing and declining population, and damage caused by wild animals, such as boar and deer. It 348 was in this context, that a pair of Oriental White Storks arrived in the hamlet in 2007 and, along 349 with other individual birds, began feeding on the abandoned terraced paddy fields. Because this 350 species feeds on the managed paddy fields and surrounding wetlands, in 2008, a local NGO, the 351 Wetlands Action Circle for Oriental White Storks (WACOWS)11 along with the Toyoka 352 municipal government,12 approached the hamlet with the intention of resuming management of 353 terraced paddy fields using volunteers from outside the hamlet. 354 11 WACOW was founded by a former Toyoka government staff member who was head of the department in charge of Oriental White Stork conservation. He is also a close friend of the current mayor who is the prime driver for promoting Oriental White Stork conservation in Toyoka. 12 Both WACOW and the local government are considered as outsiders by the residents of Tai Hamlet. From their perspective, the interests of the NGO and the local government do not coincide with those of the hamlet. For the NGO and the local government, the goal is to conserve Oriental White Storks, whereas for the hamlet, the conservation of Oriental White Storks is a means to an end, to revive their community and to bring back the younger generation. 14 In 2009, the Tai Hamlet leadership council13 officially accepted the invitation to become the 355 pilot site for a co-management scheme for terraced paddy fields for Oriental White Storks and 356 the hamlet agreed to add a new version of Hiyaku, in addition to the continuing traditional 357 labour days, known as Kounotori Biyaku, the Storks labour day. This new ‘labour day’ differs 358 from the traditional days in two ways. Firstly, the new Hiyaku aims to create a habitat for 359 Oriental White Storks, rather than to maintain infrastructure for rice cultivation. Secondly, 360 Hiyaku involves both residents and actors from outside the hamlet, i.e. local NGO and Toyoka 361 local government employees and volunteers from Osaka and Kyoto. 362 The Toyoka local government introduced monetary as well as in-kind payments to compensate 363 for the cost of conservation work for Oriental White Storks, mainly the work on the Storks 364 labour day, and to incentivise community members to participate in the conservation activities. 365 The local government and the NGO also suggested a Storks eco-tourism project in order to raise 366 further income through the conservation of the Storks. The local government and the NGO 367 mobilised funds and brought external experts to help create habitat for Oriental White Storks. 368 When implementing activities inside the hamlet, they always asked for the participation of the 369 leadership council. 370 From the commencement of this co-management scheme, the stakeholders were motivated by 371 different interests. The local government and WACOW were motivated by the conservation of 372 Oriental White Storks and their habitat, whereas the hamlet residents, especially the leaders, 373 were more concerned with sustaining their Hiyaku institution and their community as a whole. 374 As mentioned previously, the Hiyaku institution was suffering from the declining population 375 and the end of rice cultivation. Their desire to continue growing rice and to maintain Hiyaku is 376 apparent in the fact that the residents, not just the leaders, often refused to refer to the terraced 377 paddy field as “abandoned paddy fields kousaku-hokichi”, and instead called them “fallow-land 378 kyukonchi”, despite the fact that they all knew that they would not resume rice cultivation in the 379 near future. This discrepancy in the motivation caused a fierce controversy between the 380 ‘outsiders’, i.e. the local government and WACOW, and the ‘insiders’, i.e. the residents, when 381 the ‘outsiders’ tried to prioritise stork conservation goals over the needs of the community as 382 described in the following section. 383 13 The leadership council conducted all the official negotiations with the Toyoka local government and NGO and took all decisions related to the hamlet. 15 In the following section, based on fieldwork data, we first describe the way external payments 384 were distributed within the community, revealing the influence of the community leaders’ local 385 institutional logic on the payment scheme’s distributional design. We then discuss the conflict 386 arising due to the rejection of the eco-tourism proposal as part of the PES scheme, by local 387 community leaders who adhered to their own/dominant institutional logic. Finally, we argue 388 that, most significantly, community leaders took the opportunity presented by the introduction 389 of the PES to exercise their power to reinforce their dominant institutional logic in the face of 390 resistance from marginalised groups. 391 392 4.2. PES design and its need to ‘fit’ the dominant local institutional logic 393 The hamlet’s involvement in Oriental White Stork conservation, especially through the 394 establishment of the Storks Hiyaku, has brought some financial benefits through: i) a direct 395 financial contribution to the hamlet budget from WACOWS for the collective work that the 396 hamlet undertakes on labour days (e.g. JPY 528,000 ca. USD 6600 in 2012), ii) fees paid by 397 visitors to the hamlet (JPY 500 per visitor amounting to about JPY 130,000 ca. USD 1625 in 398 2012), iii) payments by visitors to use communal facilities such as the community centre when 399 WACOWS or the local municipal government conducts various activities locally.14 Together 400 the annual income received by the hamlet is ca. USD 8,750, equivalent to about 25% of the 401 annual hamlet budget. 402 The funds accruing to the hamlet are distributed through the existing ‘hamlet tax’ (Ku-hi), a 403 monthly payment collected from each household to contribute to the hamlet budget. In 2010, the 404 monthly payment ranged from JPY 2,800 to 5,400 (about USD 35-67.5) per month, depending 405 on the income level and financial conditions of each household. The funds received from the 406 conservation were used to waive the hamlet tax for August as compensation for participating in 407 the Hiyaku in July. That is, each household received JPY 2,800-5,400. This distribution 408 mechanism fits the leaders’ institutional logic that the communal benefits in connection to the 409 communal property must be distributed to the whole community in an egalitarian way. In order 410 to reveal this institutional logic, we look into the conflict that this raised between the hamlet 411 leaders and the local government and WACOW. 412 14 According to the hamlet’s financial report distributed at the annual meeting on 20th December 2012 and some interviews with members of the leadership council. 16 In the Autumn of 2010, the local government and WACOW suggested to the leadership council 413 that the payment, not entirely but partially, should be distributed to individuals. They wanted to 414 incentivise individuals, especially women and young people. Women were the ones mainly 415 involved in implementing various conservation activities other than through the Storks Hiyaku, 416 such as conducting excursions for local schools and tours for visitors. Further, the local 417 government and WACOW were concerned about the sustainability of the co-management 418 scheme itself. The younger generation was deemed to be vital for the survival of this PES 419 scheme over the longer term. However, because the payments were distributed to all the 420 households equally, the women and young people considered them as a payment to the head of 421 household. Some residents felt that they had undertaken the work of managing the paddy fields 422 without receiving any payment or compensation because they were not the head of household. 423 This was especially frustrating for those women who actively participated in activities related to 424 the conservation of the storks. 425 However, when the local government and the NGO suggested distributing the payments to 426 individuals, one of the hamlet leaders refused stating that: 427 “No way! There is no need to pay individuals. They are making money from the 428 communal property. The money should go to the hamlet not to the individual” 429 (Community leader during an interview after the meeting with local government and 430 WACOW on 20th November 2010; own translation).15 431 This statement reveals hamlet leaders’ institutional logic. For them, the terraced paddy fields are 432 communal property and the benefit arising from it must be distributed to the households in an 433 egalitarian way, not to the individuals based on their efforts towards the conservation of the 434 habitat for the storks. Part of the institutional logic, i.e. treating the terraced paddy fields as 435 communal property, is manifested by the fact that two households, which do not own any 436 terraced paddy fields, have nevertheless participated in the Hiyaku institution because ‘they are 437 part of the hamlet’.16 The previously mentioned custom, ‘selling-off’ (Haraimon), provides 438 15 Another leader commented that: “All the payment should go into the hamlet first. If it is to be distributed, it should be to the household not to the individuals, because the households are the ones providing labour for the Hiyaku and other communal duties. To be honest I do not understand why the young people are making a fuss about this. If it is distributed to the household, it is the same. No?” (Interview conducted with community leader on 8th December 2010; own translation). 16 According to the interview conducted with these households which do not own any rice-fields on 11th August 2010. 17 further evidence of this institutional logic. A household that moves out of the hamlet cannot 439 hold rights in the communal property because they are no longer part of the community.17 It 440 should be noted though that the unit of reference for distributional purposes is the household, 441 not the individual. This is also reflected in various institutions within the hamlet, such as the 442 hamlet tax and the Hiyaku institution. It is the household which provides the labour for Hiyaku 443 and pays the tax. 444 The suggestion by the government and WACOW was especially problematic as it went against 445 the institutional logics. The hamlet leaders felt that the local government and the NGO were 446 prioritising stork conservation rather than their own community.18 In other words, the leaders 447 chose to distribute the benefits from the co-management scheme through the hamlet tax system 448 so that they could reinforce the legitimacy of their institutional logic. To this end, the leaders 449 were not just powerless victims but rather actors who were able to utilise their institutional logic 450 to rearrange the PES scheme to ‘fit’ the local social context, and by doing so subvert some of 451 the original intention of the local government and the NGO that introduced the co-management 452 PES scheme. The local government and the NGO had to pull back from their preference for the 453 individual distributional rule, which might have challenged the legitimacy of the local 454 leadership. In fact, policy makers were heavily dependent on the hamlet leadership council for 455 implementing various activities in the hamlet.1920 This reaction by the leaders is not surprising 456 when we consider the fact that they are faced with resistance from young people, who have 457 different interests in the CPR in the context of the ‘delocalization’ of the hamlet. Before going 458 into the issue of resistance to and challenges faced by the local leaders, the next section reveals 459 how far the leaders were willing to go in reinforcing the legitimacy of their institutional logic. 460 461 17 This has been described as the ‘duality of land ownership’ by many Japanese rural sociologists. This refers to a notion strongly held in Japanese rural societies that all land, including the paddy fields, belongs to the hamlet or the village not just to the individual residents or households (Fujimura, 2001; Torigoe, 1993). In these communities, private land ownership co-exists with communal land ownership (Kada, 1997). 18 According to a personal conversation with the hamlet chief on 20th November 2010. 19 It is our understanding that a majority of the local government staff and the NGO leaders were also in a similar position to the hamlet leaders in their own hamlets. Thus, they were more sympathetic towards the leaders than towards the female or young residents. 20 In the end, policy makers came up with a compromise to organize a study-tour to Lake Shinji as a form of in-kind payment. The tour was organised so that the participants would be able to enjoy sightseeing as compensation for the female residents (According to a personal conversation with local government staff on 15th August 2011). 18 4.3. Conflicts over the distributional design of PES under the dominant local 462 institutional logic 463 The economic impacts arising from Oriental White Stork conservation in Toyoka are estimated 464 at USD 12.5 million, or JPY 1 billion per annum (Ohnuma and Yamamoto, 2009). To further 465 expedite the co-management in Tai Hamlet, a proposal was made by the local government and 466 WACOW to develop an eco-tourism project within the hamlet. This proposal to develop stork 467 eco-tourism gained momentum when the local government negotiated with the hamlet over the 468 inclusion of the terraced paddy fields within a wetland area being designated during the 11th 469 Conference of the Parties to the Ramsar Convention in Bucharest (the Lower Maruyama River 470 and surrounding paddy fields). The Ramsar Convention requires protection of wetlands through 471 domestic law (Halls, 1997) which is implemented in Japan through the ‘wildlife reserve’ 472 (choju-hogoku) scheme under the 2002 Wildlife Protection and Hunting Law (Tanaka, 2008). 473 Once the area is designated as a ‘wildlife reserve’, any kind of development project or 474 construction work would require permission from the prefectural government and no hunting 475 would be allowed in the area. The stork eco-tourism project was proposed as compensation for 476 restricting future development options in Tai Hamlet. 477 However, the leaders’ response was ambivalent at best. They argued that the eco-tourism 478 project would only benefit certain individuals, particularly two hostel owners. For the hamlet 479 leaders, this was especially problematic since the hostel owners were themselves members of 480 the leadership council. They feared that if they were seen by the rest of the hamlet to be 481 profiting from the communal property and breaking the local institutional logic, their authority 482 could be undermined and their institutional logic delegitimized. For example, one of the hostel 483 owners explicitly said that he did not want to be seen by other residents as gaining personal 484 benefit: 485 It is not right to make personal profit from the communal property. What do you think 486 others would think? We are already having a difficult time persuading villagers to join 487 the labour days. If I do such a selfish thing, what do you think will happen? (Member 488 of leadership council and hostel owner in his 60s from personal interview on 13th April 489 2011; own translation). 490 The hamlet leadership thus declined the eco-tourism development proposal as part of the 491 broader PES scheme. For the leaders, the financial returns from eco-tourism were seen as 492 illegitimate because they would be profiting personally from the ‘communal properties’. This 493 19 also goes against their institutional logic by which the return from the PES ought to be 494 distributed in an egalitarian way within the community. The two hostel owners, as part of the 495 hamlet leadership, were willing to voluntarily sacrifice an economic opportunity in order to 496 adhere to the local institutional logic. This shows that the leaders are not just economically 497 rational actors aiming at optimizing financial flows for the community from the terraced paddy 498 fields but actors who hold multiple social identities and rationalities within a community context 499 (Schnegg and Linke, 2015). They are capable of reinforcing the legitimacy of the dominant 500 institutional logic. Next, we analyse why such an active re-imposition of the dominant 501 institutional logic was necessary in Tai Hamlet in the context of ‘delocalization’. 502 503 4.4. Adherence to the local institutional logic and reproduction of power relations 504 As discussed previously, the majority of the hamlet’s residents have diversified their income 505 outside the hamlet. This has weakened or more precisely diversified their interests over the 506 maintenance of the traditional CPR and enabled them to challenge the legitimacy of the Hiyaku 507 institution. This is especially the case since the cessation of rice cultivation in 2006. Some 508 residents have openly voiced discontent during the annual community meetings when the 509 schedules for Hiyaku were announced by asking questions such as: 510 Why do we have to continue to maintain things that we do not use any more? We 511 stopped cultivating rice. It is a waste of time and money. How stupid are we to continue 512 carrying out Hiyaku? (Male resident in his 70s during the annual meeting on 3rd 513 January 2010; own translation)21 514 Others adopt a subtler resistance strategy by providing a substitute worker on the labour-day, 515 often an elderly female instead of a male labourer. A few households take even further measures 516 to avoid participating in Hiyaku and other communal obligations by obtaining residence cards 517 21 Other residents express their discontent towards the Hiyaku, one stated that: I understand hamlet’s tradition and the importance of hiyaku and other communal obligations. I want to continue these traditions. But look at the situation. That generation (the generation of the members of leadership council) had nearly 30 people, but ours only has 2. How can we maintain the same level of work? It is not realistic. It is causing the young generation to move away from the hamlet. (according to an interview with a female resident in her 40s, on 15th August 2010: own translation) 20 for their younger generation from different areas of Toyoka. This is because people who are not 518 officially residents are not obliged to participate. 519 The legitimacy of the Hiyaku and its institutional logic have been challenged in other ways too. 520 To illustrate this, we note a fierce controversy that occurred in 2012 over the retirement fee paid 521 to the male fire-fighting group, which functions as an extension of the public Fire Department. 522 Although this controversy is not directly related to the broader PES scheme and the 523 management of the CPR institution, it reveals the more general challenges and resistance that 524 the hamlet leaders face. In the hamlet, those who serve in the fire-fighting group receive a 525 retirement fee from the local Fire Department when they retire at the age of 42. This fee has 526 been paid into the hamlet budget instead of being paid to the individuals involved. The young 527 residents, who serve as members of the fire-fighting group, have argued that individuals have 528 the ‘right’ to receive the retirement fee individually since they are the ones who served as 529 fire-fighters. They also referred to the fact that in other neighbouring hamlets, the payments are 530 made individually. In response, the leaders fiercely accused the young residents of being 531 ‘selfish’, as from their institutional logic, the retirement fee was paid for a communal activity 532 and thus should not be paid to individuals. They argued that it was the hamlet that runs the 533 fire-fighting group, not the particular individuals who serve in it. 534 Although the young residents did not succeed with their claims, this controversy within the 535 hamlet reveals that they are gaining power to negotiate vis-a-vis their leaders in the hamlet. The 536 younger residents do not adhere to the same institutional logic as the leaders and have shown 537 they are capable of challenging it publicly. This is mainly because the younger generations are 538 not ‘educated’ in the communal customs and the institutional logic that runs through them. Due 539 to the rapid fall in population, many communal groups have disappeared or become dormant 540 over the past 50 years, such as the youth association and agricultural association. In the past, as 541 the young male residents grew older, they ‘graduated’ from one group to the next, learning ‘the 542 right way of doing things’. However due to the disappearance of these groups and the 543 engagement with the world outside of the hamlet, the younger generation is now exposed to 544 different types of institutional logic, including those that emphasize individualistic norms.22 It 545 22 But at the same time, the disappearance of communal groups has negatively impacted the younger generations. Their claim was easily silenced by the leaders as ‘ignorant’ due to their lack of ‘education’ in communal customs and the institutional logic. Further, the younger generations are not represented in the leadership council. Previously when the hamlet was cultivating rice, all the young residents joined the agricultural association. If they could prove themselves in this association, they would then be given an opportunity to join the leadership council. However, since this association is now dormant, young people cannot prove themselves, resulting in a stagnation in the membership of the leadership council. In fact, 21 is in this context that the leaders chose to sacrifice their individual gains by forgoing the 546 opportunity presented by the stork eco-tourism project. By showing their adherence to their 547 institutional logic, the leaders are demonstrating that their institutional logic is the ‘right way of 548 doing things’ in Tai Hamlet in the face of the resistance from the younger generation. 549 This controversy shows that the legitimacy of the dominant institutional logic, that the benefits 550 arising from communal activities should be distributed in an egalitarian manner, has been 551 seriously challenged. The introduction of the PES scheme presented an opportunity for the 552 dominant group, the hamlet leaders, to re-assert their institutional logic as the legitimate view of 553 the world vis-a-vis the marginalised group, the younger generation. In order to do so, the 554 dominant group had to negotiate the distribution mechanism and to forego the eco-tourism 555 project. The case of Tai Hamlet illustrates that although the introduction of the PES scheme 556 contributed to the reproduction of power relations within the hamlet, the dominant actors did not 557 have a free hand in doing so. Rather it is a ‘battlefield of legitimacy’ where the dominant and 558 marginalised actors fight over whose logic should be applied to a new institutional context, i.e. 559 the new PES scheme. 560 561 5. Conclusions 562 This paper has applied the idea of institutional bricolage to analyse a situation where PES 563 schemes are introduced, creating a ‘messy’ institutional transformation process in the context of 564 the management of CPRs. By means of a case study of a hamlet community in Japan, in the 565 context of the conservation of the Oriental White Stork, we have shown that the leaders of the 566 hamlet actively incorporated a PES scheme as a new conservation institution, in light of their 567 own local institutional logic. The case study shows that the community leaders used the 568 introduction of the PES scheme as an opportunity to reinforce the eroding legitimacy of the 569 traditional CPR institution and their dominant institutional logic i.e. that the benefits from the 570 management of the CPR should be shared in an egalitarian way among the households of the 571 community instead of being targeted to individuals. At the same time, we revealed that the 572 leaders are faced with severe resistance from marginalised actors, especially from the younger 573 residents. It is in the context of this resistance that the leaders re-asserted their institutional logic 574 as the legitimate worldview. This case study illustrates how the notion of institutional bricolage 575 the current leadership council has not had a new member for more than 10 years and the youngest member is already in his 50s. 22 can help to unravel the agency and power relations involved in the introduction of a collective 576 PES instrument. 577 This connects with Norgaard’s (2010) notion of ‘complexity blinder’. He argued that the PES 578 discourse blinds us to the complexity of ecosystems and ecosystem service co-production. Here 579 we go further by uncovering another layer of social-ecological complexity, i.e. the complexity 580 of power relations. We argue that the introduction of a PES scheme as a new institution in any 581 social context induces a ‘battlefield of legitimacy’ as to who has the right to dictate the new 582 institutional process. Actors produce their own interpretations of the logic of PES and utilise 583 those interpretations to extend their interests and goals, leading to a ‘messy’ institutional 584 process. In this way, dominant actors, in our case study the leaders of Tai Hamlet in Toyoka 585 City, have managed to impose their interpretation of a fair distribution of the payments onto 586 non-dominant actors, such as the younger residents, closing down alternative interpretations of 587 what is just or fair, providing institutional stability, and (re)producing power relations in the 588 community. The study demonstrates that the institutional change induced by a PES is loaded 589 with subtle political struggles between the dominant and marginalised groups and their 590 associated interests. 591 So then, how can the application of this idea of institutional bricolage help to better capture the 592 implications of PES interventions? First, it calls for the analysis of agency and power relations 593 arising from the implementation of PES schemes. This is especially important considering the 594 popularity of such schemes that embody trade-offs between conservation goals and complex 595 social equity considerations (Milne and Adams, 2012; Pascual et al., 2010). The implementation 596 of PES schemes requires an assessment of agency and power relations in local social-cultural 597 contexts, especially with regard to the perspectives, or institutional logics of marginalised actors 598 (Berbés-Blázquez et al., 2016; Shapiro-Garza, 2013). Second, the notion of institutional 599 bricolage contributes to the re-politicisation of the PES discourse which has often been obscured 600 by neoliberal terminology (Milne and Adams, 2012). In this regard, a key challenge lies in 601 effectively communicating to policy makers the social complexity surrounding the need for the 602 re-politicisation of PES schemes and to get beyond purely technical debates about their design 603 and implementation. This calls for research to engage in the co-production of knowledge with 604 stakeholders to improve the understanding and thus the co-design of PES schemes in order to 605 take account of ‘messy’ institutional processes. 606 23 Appendix 1. List of communal groups in Tai Hamlet and number of interviews and 607 participant observations conducted in these groups 608 Communal group Number of interviews Participant observation Leadership council 21 (12 group interviews and 9 individual interviews) Annual meetings Negotiations with the local government Neighbourhood group 9 (2 group interviews for the group in the neighbourhood where the lead author resided, and 7 interviews for each neighbourhood group chief) Seasonal religious events/ cleaning events Agricultural association 5 (2 individual interviews with the representative of the Agricultural association and 3 with individual farmers who cultivated rice until 2006) N.A. Fishery association 5 (2 group interviews with the members, 2 individual interviews with the representative of the Fishery association, 1 individual interview with the holder of fishing rights) Seaweed collection conducted during March and April/ Seaweed festival organised by the community Fire-fighting group 5 (2 group interviews, 3 individual interviews) Monthly training for the fire-fighting group Female fire-fighting group 6 (2 group interviews, 4 individual interviews) Monthly training for the fire-fighting group Night watch N.A. Rotational night-watch activities Elderly club 2 (2 group interviews) Religious group 2 (2 group interviews) Religious events, such as ‘Bon’ during August Funeral group 2 (2 group interviews) Funeral in August 2010 609 610 24 Appendix 2. Organigram of Tai Hamlet 611 612 613 614 615 25 References 616 617 Agrawal, A., Gibson, C.C., 1999. Enchantment and disenchantment: the role of community in 618 natural resource conservation. World Dev. 27, 629–649. 619 Aoki, M., 2001. Community norms and embeddness: A game-theoric approach, in: Hayami, Y., 620 Aoki, M. (Eds.), Communities and Markets in Economic Development. Oxford University 621 Press, Oxford, pp. 97–128. 622 Baker, T., Nelson, R.E., 2005. Creating Something from Nothing: Resource Construction 623 through Entrepreneurial Bricolage. Adm. Sci. Q. 50, 329–366. 624 doi:10.2189/asqu.2005.50.3.329 625 Battilana, J., 2006. Agency and Institutions: The Enabling Role of Individuals’ Social Position. 626 Organization 13, 653–676. 627 Berbés-Blázquez, M., González, J.A., Pascual, U., 2016. Towards an ecosystem services 628 approach that addresses social power relations. Curr. Opin. Environ. Sustain. 19, 134–143. 629 Bourdieu, P., 1990. The logic of practice. Stanford University Press. 630 Bourdieu, P., 1989. Social space and symbolic power. Sociol. theory 7, 14–25. 631 Büscher, B., 2012. Payments for ecosystem services as neoliberal conservation: (Reinterpreting) 632 evidence from the Maloti-Drakensberg, South Africa. Conserv. Soc. 10, 29. 633 Cabinet Office, 2012. White paper for aging society. 634 Chan, K.M.A., Balvanera, P., Benessaiah, K., Chapman, M., Díaz, S., 2016. Why protect 635 nature ? Rethinking values and the environment. Proc. Natl. Acad. Sci. 113. 636 doi:10.1073/pnas.1525002113 637 Christiansen, Larke, H., Lounsbury, M., 2013. Strange Brew: Bridging Logics Via Institutional 638 Bricolage and the Reconstitution of Organizational Identity : Institutional Logics in Action, 639 Part B [WWW Document]. Res. Sociol. Organ. 640 Cleaver, F., 2012. Development through bricolage: rethinking institutions for natural resource 641 management., Development through bricolage: rethinking institutions for natural resource 642 management. Routledge. 643 26 Cleaver, F., 2002. Reinventing institutions: Bricolage and the social embeddedness of natural 644 resource management. Eur. J. Dev. Res. 14, 11–30. doi:10.1080/714000425 645 Clegg, S., 2010. The State, Power, and Agency: Missing in Action in Institutional Theory? J. 646 Manag. Inq. 19, 4–13. 647 Crawford, S.E.S., Ostrom, E., 1995. A grammer of institution. Am. Polit. Sci. Rev. 89, 648 582–600. 649 De Koning, J., Cleaver, F., 2012. Institutional bricolage in community forestry: an agenda for 650 future research, in: Forest-People Interfaces. Springer, pp. 277–290. 651 Douglas, M., 1986. How institutions think. Syracuse University Press. 652 Emirbayer, M., Mische, A., 1998. What Is Agency? Am. J. Sociol. 103, 962–1023. 653 Fabinyi, M., Evans, L., Foale, S.J., 2014. Social-ecological systems, social diversity, and power: 654 Insights from anthropology and political ecology. Ecol. Soc. 19. 655 doi:10.5751/ES-07029-190428 656 Foucault, M., 1977. Discipline and punish: The birth of the prison. Vintage. 657 Fujimura, M., 2001. What it does it mean to have “collective property”? (in Japanese), in: 658 Miyauchi, T. (Ed.), Sociology of Commons. Shinyosha, Tokyo. 659 Galvan, D.C., 2004. The state must be our master of fire: How peasants craft culturally 660 sustainable development in Senegal. Univ of California Press. 661 Hall, K., Cleaver, F., Franks, T., Maganga, F., 2014. Capturing Critical Institutionalism: A 662 Synthesis of Key Themes and Debates. Eur. J. Dev. Res. 26, 71–86. 663 Halls, A.J., 1997. Wetlands, biodiversity and the Ramsar convention: the role of the convention 664 on wetlands in the conservation and wise use of biodiversity, in: Ramsar Convention 665 Bureau, Gland, Switzerland. ISBN. Ramsar Convention Bureau, Gland, Switzerland. 666 ISBN. 667 Hasumi, O., 1990. Anguishes of Japanese rural villages: The national policy and the social 668 changes (in Japanese). Yushindo, Tokyo, Japan. 669 Haugaard, M., 2008. Power and habitus. J. Power 1, 189–206. 670 27 Hendrickson, C.Y., Corbera, E., 2015. Participation dynamics and institutional change in the 671 Scolel Té carbon forestry project, Chiapas, Mexico. Geoforum 59, 63–72. 672 doi:10.1016/j.geoforum.2014.11.022 673 Honda, Y., 2008a. Coexistence with Orietanl White Storks - From a forced coexitence to a 674 communal belongings (in Japanese) . Genjinsha. 675 Honda, Y., 2008b. Re-thinking re-introduction of Oriental White Storks and co-habitation with 676 this birds (in Japanese). Genjin-sha, Tokyo. 677 Hotimsky, S., Cobb, R., Bond, A., 2006. Contracts or scripts? A critical review of the 678 application of institutional theories to the study of environmental change. Ecol. Soc. 11, 679 41. 680 Hyogo Prefectural Government, 2009. Action Plan for the Promotion of Organic Farming in 681 Hyogo Prefecutre (in Japanese). 682 Ishihara, H., Pascual, U., 2009. Social capital in community level environmental governance: A 683 critique. Ecol. Econ. 68, 1549–1562. 684 Kada, Y., 1997. Multiple ownership derived From everyday life practice: Ownership and 685 utilization of communal resources around the Lake Yogo (in Japanese). Japanese Assoc. 686 Environ. Sociol. 3, 72–85. 687 Kawakita, J., 1970. For Creative Ideas 2: KJ method (in Japanese). Chuo-Kouronsha. 688 Kawakita, J., 1967. For Creative Ideas (in Japanese). Chuo-Koronsha. 689 Kikuchi, N., 2006. Revival of the Oriental White Storks (in Japanese). Tokyo University Press, 690 Tokyo, Japan. 691 Kohn, M., 2000. Language, Power, and Persuasion: Toward a Critique of Deliberative 692 Democracy. Constellations 7, 408–429. doi:10.1111/1467-8675.00197 693 Leach, M., Mearns, R., Scoones, I., 1999. Environmental entitlements: dynamics and 694 institutions in community-based natural resource management. World Dev. 27, 225–247. 695 Lévi-Strauss, C., 1966. The savage mind. University of Chicago Press. 696 Long, N., Long, A., 1992. Battlefields of knowledge : the interlocking of theory and practice in 697 social research and development. Routledge. 698 28 Lounsbury, M., Glynn, M.A., 2001. Cultural entrepreneurship: stories, legitimacy, and the 699 acquisition of resources. Strateg. Manag. J. 22, 545–564. doi:10.1002/smj.188 700 Maeda, M., Takao, K., 2007. Connecting rural and urban communities: Terraced paddy field 701 owner system (in Japanese). Maruzen, Tokyo. 702 Martins, N., 2009. Rules, social ontology and collective identity. J. Theory Soc. Behav. 39, 703 323–344. 704 McCay, B.J., 2002. Emergence of Institutions for the Commons: Contexts, Situations, and 705 Events. The Drama of the Commons 361–402. 706 McKean, M.A., 1992. Management of traditional common lands (Iriaichi) in Japan. Mak. 707 commons Work. Theory, Pract. policy 1st editio, 63–98. 708 Milne, S., Adams, B., 2012. Market Masquerades: Uncovering the Politics of Community-level 709 Payments for Environmental Services in Cambodia. Dev. Change 43, 133–158. 710 Minato Community Centre, 1965. History of Minato village (in Japanese). 711 Mitsumata, G., Inoue, M., 2010. Possibility of local commons: New relationship between 712 self-governance and envrionment (in Japanese). Mineruva Shobo. 713 Mitsumata, G., Morimoto, S., Murota, T., 2008. The Frontiers of the Study of Commons: 714 Common World of Mountains/Sea/Rivers (in Japanese). Tokyo University Press, Tokyo. 715 Morimoto, M., 2014. Degradation of ecosystem service caused by the underuse of commons (in 716 Japanese). Sociol. Theory Methods 29, 261–276. 717 Mosse, D., 1997. The symbolic making of a common property resource: history, ecology and 718 locality in a tank irrigated landscape in South India. Dev. Change 28, 467–504. 719 Muradian, R., Arsel, M., Pellegrini, L., Adaman, F., Aguilar, B., Agarwal, B., Corbera, E., 720 Ezzine de Blas, D., Farley, J., Froger, G., Garcia-Frapolli, E., G??mez-Baggethun, E., 721 Gowdy, J., Kosoy, N., Le Coq, J.F., Leroy, P., May, P., M??ral, P., Mibielli, P., Norgaard, 722 R., Ozkaynak, B., Pascual, U., Pengue, W., Perez, M., Pesche, D., Pirard, R., 723 Ramos-Martin, J., Rival, L., Saenz, F., Van Hecken, G., Vatn, A., Vira, B., Urama, K., 724 2013. Payments for ecosystem services and the fatal attraction of win-win solutions. 725 Conserv. Lett. 6, 274–279. 726 29 Muradian, R., Corbera, E., May, P.H., 2010. Reconciling theory and practice: An alternative 727 conceptual framework for understanding payments for environmental services. Ecol. Econ. 728 69, 1202–1208. 729 Murota, T., Mitsumata, G., 2004. Iriai Forest and Commons: Sustainable Shared Forest (in 730 Japanese). Nihon-Hyoronsha, Tokyo. 731 Nakajima, M., 2007. Terraced paddy fields owener system: its development, typology and 732 evaluation (in Japanese). Connect. Rural urban areas 57, 26–35. 733 Narloch, U., Pascual, U., Drucker, A., 2012. Collective Action Dynamics under External 734 Rewards: Experimental Insights from Andean Farming Communities. World Dev. 40, 735 2096–2107. 736 NGO Wetlands Action Circle for Oriental White Storks, 2012. Challenges of Tai hamlet (in 737 Japanese). 738 Norgaard, R.B., 2010. Ecosystem services: From eye-opening metaphor to complexity blinder. 739 Ecol. Econ. 69, 1219–1227. 740 North, D.C., 1990. Institutions, institutional change and economic performance. Cambridge 741 university press. 742 Ohnuma, A., Yamamoto, M., 2009. Economic analysis of returning the white storks to the wild 743 in Toyooka City, Hyogo Prefecture: economic rationale for adopting white stork friendly 744 farming method and the effect of returning the white storks to the wild on the local 745 economy (in Japanese). J. Assoc. Mita 102, 191–211. 746 Ojha, H.R., Ford, R., Keenan, R.J., Race, D., Carias Vega, D., Baral, H., Sapkota, P., 2016. 747 Delocalizing Communities: Changing Forms of Community Engagement in Natural 748 Resources Governance. World Dev. xx. doi:10.1016/j.worlddev.2016.06.017 749 Onda, M., 2006. Reciprocal society: Ethnographical sociology of “Yui”, “Moyai” and “Tetsudai” 750 (in Japanese). Sekai Shisou-sha, Kyoto. 751 Osborne, T.M., 2011. Carbon forestry and agrarian change: access and land control in a 752 Mexican rainforest. J. Peasant Stud. 38, 859–883. doi:10.1080/03066150.2011.611281 753 Ostrom, E., 1992. Crafting Institutions for Self-Governing Irrigation Systems, ICS Press. 754 Institute for Contemporary Studies. doi:10.1002/rrr.3450080314 755 30 Ostrom, E., 1990. Governing the commons: The evolution of institutions for collective action. 756 Cambridge university press. 757 Ostrom, E., Ahn, T.K., 2009. The meaning of social capital and its link to collective action, in: 758 Svendsen, G.T., Svendsen, G.L.H. (Eds.), Handbook of Social Capital: The Troika of 759 Sociology, Political Science, and Economics. Edward Elgar, Cheltenham, Glos, UK ; 760 Northampton, MA, p. xvii, 460 . 761 Palomo, I., Felipe-Lucia, M.R., Bennett, E.M., Martín-López, B., Pascual, U., 2016. Chapter 762 Six – Disentangling the Pathways and Effects of Ecosystem Service Co-Production, in: 763 Advances in Ecological Research. pp. 245–283. 764 Pascual, U., Muradian, R., Rodriguez, L.C., Duraiappah, A., 2010. Exploring the links between 765 equity and efficiency in payments for environmental services: A conceptual approach. 766 Ecol. Econ. 69, 1237–1244. doi:10.1016/j.ecolecon.2009.11.004 767 Pascual, U., Phelps, J., 2014. Social equity matters in payments for ecosystem services. …. 768 Pascual, U., Phelps, J., Garmendia, E., Brown, K., Corbera, E., Martin, A., Gomez-Baggethun, 769 E., Muradian, R., 2014. Social Equity Matters in Payments for Ecosystem Services. 770 Bioscience 64, 1027–1036. 771 Peacock, M., 2011. Sen’s apples: Commitment, agent relativity and social norms. Ration. Mark. 772 Morals 2, 35–47. 773 Reay, T., Hinings, C.R., 2005. The recomposition of an organizational field: Health care in 774 Alberta. Organ. Stud. 26, 351–384. doi:10.1177/0170840605050872 775 Rodríguez-de-Francisco, J.C., Budds, J., 2015. Payments for environmental services and control 776 over conservation of natural resources: the role of public and private sectors in the 777 conservation of the Nima Watershed, Colombia. Ecol. Econ. 117, 295–302. 778 Rodríguez de Francisco, J.C., Budds, J., Boelens, R., 2013. Payment for Environmental Services 779 and Unequal Resource Control in Pimampiro, Ecuador. Soc. Nat. Resour. 26, 1217–1233. 780 Sato, I., 2008. Method for analyising qualitative data (in Japanese). Shiyosha, Tokyo. 781 Schnegg, M., Linke, T., 2015. Living institutions: Sharing and sanctioning water among 782 pastoralists in Namibia. World Dev. 68, 205–214. doi:10.1016/j.worlddev.2014.11.024 783 Schneiberg, M., 2002. Organizational Heterogeneity and the Production of New Forms: Politics, 784 Social Movements and Mutual Companies in American Fire Insurance, 1900-1930, in: 785 31 Lounsbury, M., Ventresca, M.J., Schneiberg, M. (Eds.), Research in the Sociology of 786 Organizations. Emerald Group Publishing Limited, pp. 39–89. 787 doi:doi:10.1016/S0733-558X(02)19002-1 788 Sen, A., 1982. Choice, welfare, and measurement, 1st MIT Pr. ed. MIT Press, Cambridge, 789 Mass. 790 Seo, M.-G., Creed, W.E.D., 2002. Institutional Contradictions, Praxis, and Institutional Change: 791 a Dialectical Perspective. Acad. Manag. Rev. 27, 222–247. 792 Sewell, W.H.J., 1992. A Theory of Structure: Duality, Agency, and Transformation. Am. J. 793 Sociol. 98, 1–29. 794 Shapiro-Garza, E., 2013. Contesting the market-based nature of Mexico’s national payments for 795 ecosystem services programs: Four sites of articulation and hybridization. Geoforum 46, 796 5–15. doi:10.1016/j.geoforum.2012.11.018 797 Shimada, D., 2015. Multi-level natural resources governance based on local community: A case 798 study of semi-natural grassland in Tar??ji, Nara, Japan. Int. J. Commons 9, 486–509. 799 Tanaka, K., 2008. Legal framework concerning wetland conservation and its challenges (in 800 Japanese). Annu. Econ. Reveiw by Nagasaki Univ. 24, 51–74. 801 Torigoe, H., 2007. Researching Rural Community (in Japanese). Nousangyoson Bunka Kyokai, 802 Tokyo. 803 Torigoe, H., 1993. Sociology of households and villages (in Japanese). Sekaishisosha, Tokyo. 804 Toyoka Municipal Government, 2007a. Toyoka environmental economic strategy (in Japanese). 805 Toyoka Municipal Government, 2007b. History of Oriental White Storks (in Japanese). 806 Van Hecken, G., Bastiaensen, J., Windey, C., 2015. Towards a power-sensitive and 807 socially-informed analysis of payments for ecosystem services (PES): Addressing the gaps 808 in the current debate. Ecol. Econ. 120, 117–125. 809 Vatn, A., 2010. An institutional analysis of payments for environmental services. Ecol. Econ. 69, 810 1245–1252. 811 Wade, R., 1989. Village republics. Cambridge University Press. 812 Wittgenstein, L., 1968. Wittgenstein’s Family Resemblance. Philos. Investig. 813 Wittmer, H., Gundimeda, H., 2010. TEEB For local and regional policy makers. 814 32 Yamamoto, S., 2003. Reorganization of irrigation system due to land improvement: A 815 Case-study of Izeya System at Ogi District in Otsu City, Shiga Prefecture (in Japanese). 816 Japanese Assoc. Environ. Sociol. 9, 185–201. 817 Young, O.R., 2002. Institutional interplay: The environmental consequences of cross-scale 818 interactions. The drama of the commons 263–291. 819 820