Article Nature and Space

ENE: Nature and Space Scalar fixes of environmental 0(0) 1–25 ! The Author(s) 2019 Article reuse guidelines: management in Java, sagepub.com/journals-permissions DOI: 10.1177/2514848619844769 Indonesia journals.sagepub.com/home/ene

Martin C Lukas and Michael Flitner University of Bremen, Germany

Abstract This paper analyses the emergence and fixing of scales in struggles over environmental issues. Using the example of watershed and coastal management in Java, we show how political framings of environmental matters and struggles over resources are linked to scalar regimes. We conceptualise these regimes as scalar fixes in which scales of intervention and scales of knowledge production are bound by environmental narratives and social–ecological processes to produce lock-in effects for prolonged periods of time. In our empirical case, particular scales were central in providing ‘problem closure’ and legitimising interventions while precluding other problematisations. Sedimentation of the Segara Anakan lagoon, first desired to support conversion into a rice bowl, was later framed as threat caused by upland peasants. The lock-in of interpretive framings and scales of observation and intervention, which was linked to politics of forest control, impeded debate on the various causes of sedimentation. With our newly defined concept of scalar fixes we contribute to understanding environmental narratives and related knowledge, providing a complement to the micro-perspectives on the stabilisation of knowledge claims currently discussed in cultural and political ecology. In doing so, we offer an approach to scalar analysis of environmental conflicts linking environmental narratives with the material social–ecological processes enrolled.

Keywords Scale, watershed management, knowledge claims, historical political ecology, Citanduy River

Introduction The role of narratives and myths or, more broadly, interpretive framings of environmental matters has been intensely debated in the cultural and political ecology literature over the past two decades – empirically and conceptually (e.g. Bryant, 1998; Forsyth, 2008; Forsyth and Walker, 2008; Leach and Mearns, 1996; Zimmerer, 2007). Interpretations and valuations are part and parcel of the societal relationships with nature, and our knowledge of both nature and society is seen as situated, incomplete and uncertain to

Corresponding author: Martin C Lukas, University of Bremen, Sustainability Research Center (artec), Leibniz Centre for Tropical Marine Research, Enrique-Schmidt-Str. 7, Bremen 28359, Germany. Email: [email protected] 2 ENE: Nature and Space 0(0) varying degrees. This insight has led to a rising interest in the conditions and circumstances under which knowledge claims are being made, how they are deployed, how they gain authority and how they are contested. In dealing with these questions, cultural and political ecologists have found theoretical inspiration in the fields of environmental discourse and policy studies (Hajer, 1995; Roe, 1994) that have directed their interest to issues of discursive ‘problem closure’ as well as methods to analyse the underlying narratives (Forsyth, 2003: Ch. 2). In recent years, the debates around scientific knowledge and expertise were enriched by contributions from Science and Technology Studies that challenge the boundaries between scientific, technological and political issues and focus on the practical making and corroboration of claims (Goldman et al., 2011; Lave, 2012; Whatmore, 2002, 2009). A number of studies on environmental issues have shown how expert knowledge contributes to powerful narratives that bind together specific political and ideological constructions (Bixler, 2013; Davis, 2005), and how such expertise is co-produced with social identities, discourses, and political, economic and social institutions (Jasanoff, 2004; Sneddon et al., 2017: 677). In this paper, we add to the debates around the role of narratives and knowledge claims from a different angle, namely the process and politics of scale-making. We show how a focus on the scales implied and addressed in problematising environmental issues can improve our understanding of the ways in which knowledge claims are established and brought to bear. In particular, such a focus can elucidate why certain narratives are long- lasting and persistent, even if the underlying knowledge claims are doubtful or the related practices lead to unsustainable outcomes. We engage with the scale debate in this context for two reasons: First, we noticed early on in our research on watershed governance in Java that the definition of environmental problems, just as their explanation, was often couched in terms of spatial relations. The focus of political, scientific and practical attention seemed strongly directed and limited to distinct geographical areas and time scales, neglecting or avoiding other spatial and temporal scales and perspectives (Lukas, 2017). We also found discrepancies and mismatches between social–ecological processes and development- oriented interventions and the production of environmental knowledge regarding the questions at stake. Some scalar narratives were broadly accepted with little to no research or other scientific backing; they provided ‘problem closure’ and legitimised interventions for decades while impeding other problematisations. This triggered our interest in theoretical approaches that explicitly address spatial and scalar relations in the context of corroborating knowledge claims. Second, questions of watershed management have a quasi-natural connection to scale- making: Water-based processes unfold over large areas connecting places far apart from each other, and the drivers of these processes may operate far beyond the boundaries of watersheds. Related management approaches are not just constrained by lack of reliable knowledge; they are typically confronted with fragmented interests and polities, rendering the popular approach of ‘integrated management’ and according policy prescriptions ambitious or even naı¨ve (Blomquist and Schlager, 2005). In line with such arguments, several authors have called for specific attention to scale in watershed management, often referring to bio-physical and social processes and scalar politics (Molle, 2007; Norman et al., 2012; Swyngedouw, 2013; Venot et al., 2011). Watersheds can be seen as arenas of the politics of scale, where scales are constructed to favour, legitimise or exclude particular analyses and solutions (Molle, 2007). Hence, research in watershed contexts has become a fruitful ground for the development and refinement of scale concepts. Engaging with the ongoing scale debate, we examine Lukas and Flitner 3 political struggles over resources and environmental management in the catchment of the Segara Anakan lagoon in Java, Indonesia. Taking a political ecology perspective, we critically explore the unfolding of spatially shifting management interventions in the lagoon-watershed region along with related ways of knowledge production that we analyse in their scalar logic. In the following section, we briefly summarise key lines of recent work on scale and scalar politics to define our points of departure. Building on Rangan and Kull (2009), among others, we develop a new heuristics of scalar politics that focusses on two dimensions of scale-making. The scales of observation and the scales of intervention, as we term them, are shaped and articulated by interpretive framings and at the same time anchored in material social–ecological processes and relations. We see the material-cum-semiotic articulation of these two scales as crucial in explaining the formation and persistence of semi-stable arrangements and legitimation patterns for interventions. Drawing on a second strand in the scale debate, we conceptualise these arrangements as scalar fixes, thus giving a new theoretical underpinning to this latter notion. In the third part of the paper, we elaborate on our understanding of scalar fixes in exploring the empirical case of the Segara Anakan lagoon and its catchment, a long- standing priority area for watershed and coastal management in Java. We focus our account on critically developing two consecutive scalar fixes that prevailed in the second half of the 20th century. The first scalar fix was dominated by river and agricultural development experts, consultants and policy-makers framing the lagoon and the adjacent lower river basin as unproductive areas to be converted into a ‘rice bowl’, with lagoon sedimentation seen as a desirable process. They turned the lower river basin into the scale of intervention of the first large U.S.-Indonesian development project, an important endeavour related to nation building, agricultural development and other political goals. The ideas for this endeavour emerged at different (observational) scales, partly far beyond the region. The second scalar fix enrolled some new actors, including forest agencies, and centred on the framing of the lagoon as a valuable ecosystem and of lagoon sedimentation and upland degradation as major threats. It moved the lagoon and upland farmers’ plots into the political spotlight and turned them into hotspots of intervention. Entanglements between watershed conservation and state politics of forest control played important roles in shaping this second scalar fix; narrow scales of observation limiting knowledge production in line with the interpretive framings were key in its stabilisation. We analyse the formation of these scalar fixes, the causes and mechanisms leading to their persistence, and the processes replacing and challenging them. This also demonstrates how such scalar fixes can confine debates and lead to a systematic neglect of major material processes through scalar lock-in effects, resulting in simplistic assumptions about these processes and undermining the effectiveness of long-standing interventions. In the final section we summarise our conceptual contribution, draw conclusions from our empirical case and direct attention to the temporal dimensions of scalar fixes.

Scales, framings and scalar fixes It was more than a decade ago that Marston et al. (2005: 422) called to ‘expurgate scale from the geographic vocabulary’ in favour of a ‘flat ontology’. Yet, the concepts of scale and scalar politics continue to be debated, and political ecology has remained a fruitful field for such discussions. There is neither a need nor the place here for a broader review of the scale debate with its ramifications and diverging positions. With regard to political ecology, Neumann (2009) has given an overview of the main lines of thought and key theoretical 4 ENE: Nature and Space 0(0) distinctions, not without regretting ‘a continuing problem with the conflation of terminology and concepts, so that even in the political ecology of scale literature, scale, level, site, network and assemblage are not clearly distinguished’ (404). This observation is reason enough to briefly sketch our main tenets. With large parts of the literature we share the view that scale should neither be understood as ‘external fact awaiting discovery’ (Delaney and Leitner, 1997: 94) nor as level of socio-political organisation. Instead it is conceived of as an outcome of contingent socio-political processes and thus as open for contestation. We agree with Moore (2008) that scale research should focus on the processes ‘through which specific scalar configurations solidify in consciousness and practice, and the effects these developments have upon social, political and cultural relations’ (213). Social actors can strategically influence scalar configurations in order to pursue particular political or economic outcomes or legitimise certain activities (Bolin et al., 2008; Brown and Purcell, 2005; Kurtz, 2003; Penning-Rowsell and Johnson, 2015; Smith, 1992). Along these lines, we can define the politics of scale for our purposes with Zulu (2009) as referring to processes in which ‘different actors construct, modify and contest the spatial extent, content, and resolution of information and decisions’ (688), if we understand the latter term to comprise ensuing social action. This definition, however, still leaves open at least two important questions: first, in which way those processes of scale-making occur, in particular how related knowledge claims are established to legitimise certain activities, and second, how it is possible that such processes can have stable outcomes over extended periods of time, despite their constructedness and modifiability. To deal with the first question in a political ecology context, we found the work of Rangan and Kull (2009) a useful starting point. Building on Lefebvrian concepts, they see scale as ‘produced by three moments of social action’ which they call the ‘operational, observational and interpretative moments in the production of scale’ (Rangan and Kull, 2009: 30/37). The operational moment or operational scale is then understood as equivalent to or produced through patterns of interaction of both social activity and biophysical processes; spatial practice that can be ‘reflected in distinctive socialized ecologies or landscapes’ (Rangan and Kull, 2009: 37). More simply, Kull and Rangan (2015) later referred to operational scale as ‘relat[ing] to empirical phenomena in nature and society’ (487). The observational scale, in contrast, is produced in the process of defining the limits of what is observed, of measuring, controlling and planning.

Governments, policy-makers, scientists and social researchers produce observational scale by focusing on particular social groups or biophysical entities, delimiting the spatio-temporal extent of their activities or movements, determining the resolution of the data for analysing their behaviour and specifying the disposition [...] of the entities or phenomena chosen for surveillance, control, or study. (Rangan and Kull, 2009: 38) While seeing this distinction as useful, we want to highlight that both types of scale imply and relate to material phenomena and processes in manifold ways. Thus, they do have geographical references and may relate to one or more geographical scales. Contrary to Rangan and Kull (2009), we see in this characteristic an important difference to the ‘interpretive moment’, which the authors referred to as a separate ‘interpretive scale’ that is ‘produced as a normative hierarchy or ordered range of values that serve as the context and means by which ‘truth-making’ occurs in scientific and policy discourse’ (40). We argue that such discourses and their underlying contexts can, but do not necessarily have characteristics related to size, level and relation or hierarchical ordering that suggest specific links to the scalar lexicon (cf. Sayre and Vittorio, 2009). Scientific endeavours, Lukas and Flitner 5 political interventions or everyday practices may be imbued with symbols, models and storylines. They all entail processes of meaning-making and, in a Foucauldian perspective, they can be understood as constitutive part of discourses. This meaning-making, which we refer to as interpretive framing, may or may not have references to geographical scale. Specific spatial–temporal units and political levels can be an integral part of such framings (cf. Kurtz, 2003; Towers, 2000 [Flitner 2007]). Yet, interpretive framings do not necessarily have scalar characteristics. Hence, in contrast to Rangan and Kull (2009), we do not treat processes of framing as a separate ‘interpretive scale’.1 Based on this revision of Rangan and Kull (2009) and our empirical example below, we propose a heuristics of scalar environmental politics which positions the scales of observation and intervention between the interpretive frames and the material processes, being largely conditioned by these two moments (Figure 1). Accordingly, interpretive framings can shape the scales of observation used to explore environmental matters and constitute or justify particular modes and scales of intervention aimed at altering the processes in question. In our empirical case, such interventions comprise, for example, upland conservation measures in the context of watershed management. Obviously, they are at the same time part of material social–ecological processes. And while interpretive framings may be considerably influenced by specific actors, they will in most cases also be linked to scientific facts and different kinds of knowledge practices regarding the complex social–ecological issues at stake. This link between interpretive framings and the scales of observation is one important starting point to deal with the second question raised above: how interpretive framings and processes of scale-making can have quite stable outcomes that persist over longer periods of time. We propose a renewed approach to the notion of ‘scalar fixes’ to describe such semi-

Interpreve framings

Scales of Scales of F1,2… intervenon observaon

Social-ecological processes (interacng across various scales)

Figure 1. Heuristic model of scalar fixes. Note that both, scales of intervention and scales of observation, are substantially connected to material social-ecological processes as well as interpretive framings. The intersection zone of scalar fixes (F1,2 ...) is not to denote spatial congruence or overlap, but a potential lock-in of scales, interpretive framings and social-ecological processes that may diverge in their spatial references. Interpretive framings may or may not have scalar characteristics. 6 ENE: Nature and Space 0(0) stable configurations of particular interpretive framings of environmental issues along with related scales of observation and intervention. The term ‘scalar fix’ was introduced to geographical debates to explain temporal fixity and change in the scalar organisation of capitalist production and capital circulation (e.g. Brenner, 1998). Building on this literature, Cohen and Bakker (2014) elaborated the notion of an ‘eco-scalar fix’, defining it as ‘the process of rescaling and reorganizing governance [...] and thereby displacing conflicts and crises, often through the construction of (purportedly ‘‘natural’’) ecological (watershed) scales, which simultaneously depoliticize and repoliticize governance’ (132). Yet, while we share this interest in the strategic construction of ‘social– ecological scales’, our empirical case will show how politically driven interpretive framings of environmental issues can just as well lead to a complete disregard of scales that seem quite ‘natural’ as units of observation and intervention in social–ecological processes (e.g. the entire watershed). In other words, a scalar fix in environmental matters does not necessarily imply any naturalising of social and political objects such as units of governance (cf. Cohen and Bakker, 2014). In line with these arguments we develop the notion of scalar fix to refer to semi-stable configurations of particular interpretive framings of social–ecological processes and related scales of observation and intervention. We hereby use the term ‘fix’ primarily in the meaning of stability or lock-in effect. But the second meaning of the term which refers to problem solving (cf. McCann, 2003: 162) is also present. It is the processes of establishment, articulation and solidification of different scales that are in the centre of our interest in this paper, and the role of interpretive framings in this regard. In a similar vein, Sievanen et al. (2013) described such framings of observation and intervention as ‘scalar narratives’:

Scalar narratives associate places, spaces, and processes at particular social, institutional, and geographical scales to explain events, attract attention and funding, and in the process, shift social relations by enrolling and excluding different forms of knowledge, spaces, and groups of people. (208) ‘Scalar narratives’ or interpretive framings (which are not necessarily scalar, as discussed above) thus can be the starting point for the formation of particular scalar configurations and, as we will show, their confining of the scales of observation can play a crucial role in the solidification of such configurations as scalar fixes. Accordingly, the scales of observation are themselves contestable, and they play a critical role in challenging established interpretive framings, expertise, and related modes and scales of political intervention. In the following section we corroborate our claims with an empirical example. We will explore the scalar (re-)configurations of watershed and coastal management in Java in a historical perspective and critically analyse two consecutive scalar fixes that have dominated related discourses, research and interventions over the past decades.

Scalar politics of environmental management in Java Using the example of the Segara Anakan lagoon and its catchment area, we illustrate how political interpretative framings determine and confine the scales of observation, resulting in repeated confirmation of simplistic narratives about social–ecological processes, which in turn justify particular scales and types of intervention. The lock-in of interpretative framings with certain scales of observation and intervention constrained the effectiveness of management initiatives for decades and even made them run counter to their stated purpose. To provide a background, we first sketch what has been framed as major environmental issue in this region since the 1980s. Since then, conservation of the Segara Anakan lagoon Lukas and Flitner 7

Figure 2. Map of the Segara Anakan lagoon and its catchment area with the major scales of environmental management-related intervention. has dominated environmental management-related discourses and action. This shallow coastal lagoon forms the estuary of the Citanduy River and a few smaller rivers (Figure 2). The lagoon comprises one of the largest remaining expanses of mangrove forest in Java. With its variety of habitats, its biological diversity and its function as a major nursery ground for ocean shrimp it is regarded as ecologically valuable, and it forms the economic backbone for many people in the adjoining villages of Kampung Laut (see, e.g. Olive, 1997; White et al., 1989; Yuwono et al., 2007). Sediment input, mainly from the Citanduy, but also from the Cikonde and Cibeureum Rivers, has drastically reduced the size of the lagoon’s water surface area from more than 8000 hectares in the mid-19th century to about 2000 hectares at present (Lukas, 2014a, 2017). Sedimentation is said to have contributed to overfishing and mangrove degradation, which in turn undermine the economic viability of fishing, once the dominant livelihood strategy in Kampung Laut (Dudley, 2000; Olive, 1997; Sudjastani, 1982). Since the 1980s, the critical conjuncture of dwindling marine resources, increasing exploitation thereof and ecological degradation has attracted much political interest nationally and internationally, turning the lagoon into a national hotspot of research and political interventions. The framing of lagoon sedimentation as a problem that needs to be addressed and that is mainly caused by unsustainable farming practices in the uplands has provided the political rationale for particularly extensive upland conservation efforts in the lagoon’s catchment. Proposed interventions have included the construction of river diversions to reduce sediment input, lagoon dredging and upland conservation. Whereas the construction of river diversions did partly not materialise for political reasons, dredging had only temporary effects. The upland conservation measures appear not to have been appropriately tailored to the alleged goal of reducing sedimentation, and their effects are uncertain and rather limited (for a more comprehensive discussion, see Lukas, 2015b). 8 ENE: Nature and Space 0(0)

Combining a review of management interventions and related political processes and insights from our own research in the region over the past years,2 we critically explore in the following sections the links between interpretive framings of environmental matters and specific scales of observation and intervention. The first constellation or ‘scalar fix’, which dominated in the late 1960s–1970s, is based on the framing of lagoon sedimentation as a desirable process, creating land for agricultural reclamation. The second ‘scalar fix’, which developed in the late 1970s and is still relevant at present, is based on the framing of lagoon sedimentation as anthropogenic threat, mainly driven by unsustainable upland agriculture.

Scalar fix 1: River basin development Looking at the contemporary main-stream framing of environmental matters in the lagoon- watershed region and the substantial political and scholarly efforts devoted to the lagoon over the past three decades, one can easily presume that the region was selected as one of Indonesia’s priority areas for watershed management for the sake of lagoon conservation. In fact, however, the only value that river and agricultural development experts, consultants and policy-makers attributed to the Segara Anakan lagoon prior to the 1980s was its potential for agricultural reclamation. In line with this framing, they regarded sedimentation as desirable process that should be accelerated. Corresponding development plans, put forward in 1948, 1971 and 1975, proposed river engineering measures aiming at accelerating lagoon sedimentation and draining of the lagoon for agricultural reclamation of its ‘unproductive’ swamp and brackish water areas (PRC-ECI, 1987). Although these plans to reclaim the entire lagoon did not materialise, the lower Citanduy river basin became the scale of implementation of agricultural development projects of national and international political importance from the late 1960s onwards. At that time, the river basin was invested with political meanings shaped far beyond the watershed through various political dynamics and interests. The post-Second World War struggle between world political systems; the international emergence of river basin development as modernisation strategy; the national government’s proclaimed development trilogy; its commitment to raise agricultural production and achieve national rice self-sufficiency; and its imperative to consolidate state power and counter political instability and insurgencies in the area: The combination of these factors, which we discuss in more detail below, turned the Citanduy river basin into the scale of intervention for the first large U.S.-Indonesian development project. This ‘Citanduy Project’, funded by the U.S. Agency for International Development (USAID) and the Asian Development Bank (ADB), marked the beginning of a series of large development interventions in the area and of ‘big D’ Development3 in Indonesia. In the 1960s, parts of Southeast Asia were political and military battlefields in the fight for a communist versus capitalist world order, a fight in which Indonesia played an important role. After the country’s communist movements had been purged in huge massacres in 1965–66 (Cribb, 1990, 2001), Suharto’s anti-communist New Order regime paved the way for a broader U.S.-Indonesian economic partnership, providing an opportunity for the U.S. to promote the western liberal model of modernisation and expand its influence in the region. The U.S.’ engagement in Indonesia was first constrained by its reluctance to support an authoritarian, nepotistic and pervasively corrupt military regime that seemed to provide limited perspectives for democratic development (Simpson, 2008). However, Indonesia’s economic decay, its regime’s high expectations of U.S. assistance and the course of the Vietnam War created the political momentum for a turnaround: ‘The worse things got in Vietnam, the more President Lukas and Flitner 9

Johnson wanted to invest in Indonesia’s success, just as administration officials, the press, and Congress wanted to see Indonesia as a validation of the American stance in Indochina’ (Simpson, 2008: 241). Indonesia was to become a symbol of development-oriented engagement of the U.S. in Southeast Asia. In collaboration with the IMF, World Bank, ADB and private investors, the U.S. aimed at paving the way for Indonesia’s modernisation, unavoidably led by a military regime (Simpson, 2008), which would probably not lead the country to democracy but at least create political stability by reducing poverty and social unrest. River basin development was a favoured modernisation strategy at that time. Based on the political rhetoric surrounding the alleged success of the Tennessee River Valley engineering project in the 1930s, U.S. overseas assistance and diplomacy contributed to a massive worldwide growth of river development initiatives between the 1940s and 1960s (Ekbladh, 2002; Molle, 2009). Large-scale river engineering and reclamation projects had served as development and nation building strategies earlier (e.g. Biggs, 2010 about the Mekong basin) and were now considerably pushed in line with other engineering solutions to ‘‘underdevelopment’’. That this strategy was to materialise in Indonesia was related to U.S. President Johnson’s personal commitment to river development.4 These intertwined national and international processes produced the idea of initiating a USAID-supported river basin development project. The selection of the Citanduy basin as the scale of implementation for this endeavour had clear political motivations: The attention of the national government to this part of Java was ‘prompted by political instability in the area, into the early 1970’s, as a result of Darul Islam activity’ (USAID, 1985: 5). Parts of the Citanduy region belonged to the strongholds of the Dar’ul Islam, a radical Islamic movement that fought for Indonesia to become an Islamic State between the late 1940s and 1960s (Dengel, 1986; Jackson, 1980; Van Dijk, 1981). In the hills northeast of the project’s designated target area, the Dar’ul Islam insurgency and the state’s counter insurgency had displaced the population from a number of villages in the 1950s, which prompted the state to design a land swap and incorporate the former village land into the state forest territory. The deal left many villagers landless, triggered conflicts and led many villagers to join communist movements; the latter were crushed in the wake of Suharto’s seizure of national political power in 1965/66 (Lukas, 2014b). The followers of Dar’ul Islam were still ‘deeply wary of the Suharto regime’s commitment to what they viewed as a secular path to modernisation’ in the 1970s, and correspondingly, ‘[m]any army leaders still feared [...] a revival of movements like Darul Islam’ (Simpson, 2008: 254–255). River basin development as a state-dominated large-scale intervention was in line with the government’s strong interest to bring the Dar’ul Islam and the communist movements in the Citanduy region under control and prevent their resurgence and to consolidate state control in this politically fragile area (USAID, 1985). Moreover, given the disruptions in the region with large numbers of refugees in search of land,5 a project focussing on agricultural development was a perfect strategy for economic recovery and crisis management. It was also in line with the New Order regime’s proclaimed development trilogy (Trilogi Pembangunan), which aimed at national stability, economic growth and equitable development,6 and its commitment to achieve national rice self-sufficiency through the Green Revolution. This commitment strengthened the Ministries of Agriculture and Public Works, both important agents of the projects in the Citanduy basin. Hence, various national and international political dynamics and interests led to the choosing of the lower Citanduy basin as scale of intervention for the first large 10 ENE: Nature and Space 0(0) internationally funded development project in Indonesia (Table 1). In the political context of the U.S. and the Indonesian governments’ interests and in line with the prevailing international development paradigm, the problematisation of the landscape was that of an unproductive swamp land, with the lower river basin along with the Segara Anakan lagoon and its mangroves a potential ground for agricultural development and modernisation. Hence, planners and policy-makers regarded lagoon sedimentation as desirable. The knowledge and ideas relevant for this development endeavour largely derived beyond the region with little reference to the social–ecological set-up: the major scales of observation relevant to the New Order government and project planners were the hills north of the lagoon as hotspot of political insurgencies by Dar’ul Islam and communist movements, and the nation state, which was economically on the brink and to be consolidated politically, developed economically and made self-sufficient in rice production. Far across the Pacific Ocean, furthermore, the Tennessee River basin stood as a shining example of Development. The interpretive framings along with these scales of observation and the lower river basin as scale of intervention together formed a stable, politically coherent arrangement – a scalar fix – that dominated the region throughout the 1970s. In material terms, this fix included and enrolled certain ecological components and processes as well as technical items, such as irrigated rice fields, high yielding rice varieties, concrete irrigation channels and sediments (Table 1). Once established as result of the political interests and developments described above, the persistence of actors and organisations (including consultants and river basin authorities), of a development paradigm focussing on the intensification of irrigated agriculture, and of related funding contributed to the stability of this scalar fix. As a material outcome, it transformed the lower river basin from a seasonally inundated, sparsely populated forest swamp land into a ‘rice bowl’, and formerly meandering rivers into straightened and embanked channels (for a documentation, see Lukas, 2017). While policy-makers, river and agricultural experts, consultants and project implementers established the lower river basin as a politically important scale, institutionally perpetuated by a newly founded river basin authority (BBWS) under the Ministry of Public Works, the uplands and the lagoon ecosystem and their material linkages with the lower river basin through flows and retention of water and sediments barely received any attention. They remained beyond the experts’ scales of observation. This fundamentally changed from the late 1970s onwards.

Scalar fix 2: Lagoon and upland conservation The framing of social–ecological processes in the area and, with them and through them, the scales of observation and intervention underwent a radical shift between the late 1970s and early 1980s (Table 1). This shift resulted from a conjuncture of developments at different levels, most notably (1) the river basin authorities’ and consultants’ recognition of sedimentation as a threat to the newly reclaimed agricultural areas of the lower river basin, and (2) emerging global environmental discourses. Both fitted together and provided the rationale for new interventions, focussing on upland and lagoon conservation. Hence, political attention gradually shifted from the lower river basin to the lagoon and to the uplands, which both became major scales of intervention. Politics of state control over forest territories shaped this new scalar fix and contributed to its persistence over several decades through politically confining the scales of observation. In the following we analyse the emergence and solidification of this new scalar fix. Lukas and Flitner 11

Table 1. Scalar fixes of coastal and watershed management in the region of the Segara Anakan lagoon and its catchment area.

Scalar fix River basin development Lagoon and upland conservation

High time Late 1960s and 1970s 1980s–2000s Political context ‘Big D’ development Sustainable development Vietnam War Emerging environmentalism and greening Non-military, development oriented of aid engagement of USA in SE Asia Global degradation discourses Interest of President Suharto’s New State forestry as part of cleptocratic rule Order Regime to consolidate state Entanglement of state forest and power, counter political insurgencies watershed management and foster economic development Green Revolution Main actors Government of Indonesia Government of Indonesia U.S. Government, USAID U.S. Government, USAID Asian Development Bank (ADB) Asian Development Bank (ADB) Ministry of Public Works with the river Segara Anakan Management Authority basin authority (BBWS, Balai Besar (BPKSA, Badan Pengelola Kawasan Wilayah Sungai Citanduy) Segara Anakan) Ministry of Agriculture State forest corporation; Watershed International consultants (PRC management authority (BPDAS, Balai Engineering Consultants, Inc.) Pengelolaan Daerah Aliran Sungai) Construction companies under the Ministry of Forestry (MoF) Rice farmers in the lower Citanduy and International consultants (PRC lagoon area Engineering Consultants, Inc.) Researchers concerned with social and ecological changes in the Segara Anakan lagoon Upland farmers Interpretive framing Lagoon and swamp lands of the lower Lagoon is regarded as highly valuable river basin regarded as unproductive ecosystem, and lagoon siltation can and areas that are to be converted into a should be stopped; ‘unsustainable ‘rice bowl’; lagoon siltation is a upland agriculture’ is the main driver of desirable process siltation Scales of intervention Lower Citanduy river basin (waterworks, Upland farmers’ private plots (tree irrigation channels, settlements) planting and terracing) Segara Anakan lagoon (dredging, river diversions, investor-based aquaculture development) Scales of observation Area of political insurgencies (Dar’ul Lower river basin (flooding, dry season Islam, communist groups) water shortages, sedimentation) Nation state (political instability, Segara Anakan lagoon (‘hotspot’ of economic decay; aim of increased research) agricultural production and rice self- Upland farmers’ plots (‘unsustainable sufficiency) farming practices’) Lagoon and lower river basin as potential rice bowl Enrolled materialities Irrigation systems, irrigation water, rice Trees, field terraces, dredging gear, fish fields, high-yielding varieties, sediments diversity and reproduction, mangroves 12 ENE: Nature and Space 0(0)

Shifting framings, shifting scales of observation and intervention. In line with emerging global environmental discourses and related changes in development paradigms and the greening of aid, the Indonesian government strengthened environmental conservation policies. These broader political changes shaped the way in which landscapes and environmental processes were framed. They also provided opportunities for new interventions which were to ‘solve’ hydrological issues that partly arose from the previous interventions. River basin authorities and consultants, who had already been engaged in the interventions of the first scalar fix, framed these hydrological issues – seasonal flooding, dry season water shortages and sedimentation of irrigation channels in the lower river basin – as being a result of upland degradation (see PRC-ECI, 1975, 1987). This framing of upland degradation as a threat to the newly established ‘rice bowl’ in the lower river basin served as the first political imperative for upland interventions. Hence, intervention shifted into the uplands, while the lower river basin – previously the major scale of intervention – served as scale of observation, and a key element in the narrative to justify the shift in attention. Over the course of the project, river basin authorities and consultants also identified lagoon sedimentation as a threat to the new agricultural areas in the lower river basin, because it impaired drainage (PRC-ECI, 1987). At the same time and in line with emerging global environmental discourses, researchers ‘discovered’ the lagoon as a unique ecosystem and an important nursery ground for ocean shrimp – a valuable fisheries resource (see Bird et al., 1982). As a consequence, policy-makers, consultants and project implementers regarded sedimentation no longer as desirable, but as a threat – a menace to the irrigation schemes in the lower river basin, for the lagoon’s biodiversity, the offshore fishers and also for the residents of the lagoon villages of Kampung Laut, who had previously received hardly any political attention and now, all of a sudden, were portrayed as victims of environmental ‘disruption’.7 This clear-cut, dominant framing of the lagoon as a valuable ecosystem threatened by sedimentation pushed the lagoon into the spotlight of political attention and turned it into the major scale of observation and intervention, with long-standing research, internationally and nationally funded management interventions, and the establishment of the Segara Anakan Management Authority (BPKSA).8 Numerous research projects have since documented ecological and social conditions and changes in the lagoon – most of them designed to produce knowledge supporting the dominant framing of the lagoon ecosystem and residents threatened by sedimentation. In line with this, development interventions, most notably the ADB-funded ‘Segara Anakan Conservation and Management Project’, aimed at environmental conservation and improving livelihoods. Just as the lower river basin had earlier served as symbol for national agricultural development and modernisation, the lagoon was now to become a symbol of environmental conservation. However, the achievements of three decades of research and interventions are rather limited. Engineering measures to reduce sediment input (river diversions) and lagoon dredging only partly materialised, led to social conflicts, and had negative or only temporary effects. The shrimp ponds officially designed as alternative livelihood option, but established and managed by external investors, hardly benefitted local residents, whose portrayal as victims had served as a rationale for the intervention (Reichel, 2007). Residents soon plundered the ponds, which were then largely abandoned. Project and research reports related to the lagoon fill entire book shelves, but the basic issues persist. Besides sedimentation, poverty and mangrove degradation, these issues include a fundamental problem that has not been the focus of political framings, much research and debate to date: struggles over resources (see Heyde, 2016; Khuriyati, 2010). Lukas and Flitner 13

Framing lagoon sedimentation as a threat, which required upland conservation, including agricultural extension to poor upland farmers, was in line with environmental discourses and newly emerging development paradigms, which put emphasis on supporting the marginalised, rural poor. This politicisation of upland agriculture moved farmers throughout the hill and mountain areas of the watershed, who had so far received little attention under the New Order regime, into the centre of agricultural development. The strong political attention to upland conservation was also associated with a powerful role of the new Ministry of Forestry (MoF), which split from the Ministry of Agriculture in 1983. Watershed management, largely focused on upland conservation, was institutionalised under the MoF and has been entangled with the centralised management and control of Java’s valuable teak and other forests through the state forest corporation, the successor of the colonial forest empire that has been in charge of the state forests in the whole of Java since 1978 – an issue to which we return below. Following feasibility studies and pilot projects, the USAID-funded ‘Citanduy II’ project was implemented from 1981 to 1988. It fostered and supplemented the MoF’s upland conservation programmes carried out also in other parts of Java until recently. The focal scale of intervention was upland farmers’ private plots, and the approach was to encourage and assist them in terracing their land and increasing tree cover. Interestingly, despite the fact that the lower river basin and the lagoon were referred to as major rationale for Citanduy II, a supposedly ‘integrated’ watershed project,

integration [...] had not occurred at all [...] [and] mission interest and attention has been so dominated by the upland element that the lowlands and the watershed concept are forgotten; ‘watershed’ has come to be synonymous with upland catchment areas rather than the original concept of all that lies between geographic divides. (USAID, 1985: 7) Debates, observation and intervention were fixed to the scale of upland farmers’ private plots, producing a lock-in, a scalar fix that has dominated watershed management for decades. The effectiveness of the upland conservation efforts has repeatedly been questioned. Their failure to markedly reduce downstream sediment yields (Diemont et al., 1991; Purwanto, 1999; Schweithelm, 1988) has been attributed to procedural shortcomings in project implementation (ADB, 2006), site-insensitive one-type-fits-all upland agricultural packages (Purwanto, 1999; USAID, 1985), failure to raise farmers’ awareness for erosion (ADB, 1996) and national policies encouraging cassava cultivation (Huszar and Cochrane, 1990; USAID, 1984). Yet, our research has identified a more fundamental issue that has undermined the effectiveness of watershed interventions: the continuous, decade-long confinement of the scales of observation and intervention to upland farmers’ private plots and the consequential neglect of numerous other causes of high river sediment loads (Lukas, 2015b). The next section shows how watershed and land use experts’, policy-makers’ and project implementers’ dominant interpretive framing of watershed issues has limited the scales of observation, thereby producing a lock-in that has confined debates, knowledge production and the modes of intervention and undermined the environmental outcomes of the latter. The subsequent section explores how politics of state control over forest resources have contributed to the consolidation and stability of this scalar fix.

Interpretive framings confining and being reinforced by the scales of observation. Looking beyond the effectiveness of the conservation measures on farmers’ private plots and widening our view to the entire catchment area directs attention to the critical role of the scales of observation in the formation and solidification of the new scalar fix. Tellingly, most research on erosion and related mitigation strategies in Java since the 1980s has focused on the scale of upland 14 ENE: Nature and Space 0(0) farmers’ plots (e.g. Barbier, 1990; van Dijk et al., 2004; Donner, 1987; Palte, 1989; Repetto, 1986). That is, the scales of observation remained confined within the scope of the dominant interpretive framing leading to a ‘problem closure’ as described by Hajer (1995). Once river and watershed authorities and consultants had discursively established upland farmers’ private plots as the major cause of seasonal flooding and sedimentation and turned the same into their preferred scale of intervention,9 observation was confined to the same scale, i.e. upland farmers’ private plots. Such observation would inevitably confirm the existence of erosion on these plots and hence the suitability of the chosen scale of intervention.10 This results in a highly specific lock-in effect in which interpretive framings and the scales of observation and intervention form a stable, seemingly logic and coherent scalar fix – ‘fix’ in the meaning of both stability or lock-in and problem solving. It was not before the late 1990s and early 2000s that a few scholars started to direct attention to other sediment sources, particularly roads, trails and settlements, in two small sub-catchments in East and West Java (Nibbering and Graaff, 1998; Purwanto, 1999; Rijsdijk, 2005; Rijsdijk et al., 2007a, 2007b). Yet, these findings have to date barely been acknowledged in political debates over watershed management in Java (Lukas, 2017). In the Citanduy region, hydrological issues in the lower river basin and lagoon sedimentation served as rationale for particularly intensive interventions in parts of the watershed. Yet, strikingly, the scales of observation did not cover the whole watershed. That is, there was barely any systematic, open-ended, watershed-wide research on the various drivers of lagoon sedimentation, let alone on the links between upland agriculture and water flows. After the first years of implementing upland conservation measures meant to reduce sedimentation and flooding, an assessment of the ‘Citanduy II’ project had to acknowledge that ‘[l]ittle has been learned about the sources of erosion. [...] [T]here is much opinion but little fact’ (USAID, 1985: 9). This barely changed in the following decades, as our interviews showed. In the absence of concrete evidence, experts, project implementers and policy-makers ‘established’ the drivers of lagoon sedimentation through interpretive framings. Untested narratives became, to repeat Klein’s (2002) apt formulation, ‘institutionalised as facts’ (191), guiding research and interventions. While no watershed-wide research on sediment sources and their causes had been undertaken, many sources claimed ‘poor’, ‘destructive’ or ‘unsustainable’ farming practices in the uplands as major causes of lagoon sedimentation (see Lukas, 2015b). Our interviews and discussions with representatives of governmental organisations and scholars at universities showed that these framings not just represented solidified ‘expertise’ of a few leading forest and watershed management organisations, but were broadly shared by scholars and policy-makers and continued to narrow their scales of observation. These framings seem just another formulation of the Himalayan Degradation Theory (Forsyth, 1996; Ives and Messerli, 1989), which obscured the view of the complexity and uncertainty characterising watershed processes in the Himalaya through simplistic assumptions focusing on population growth and upland agriculture as causes for all ills downstream. Besides erosion on upland farmers’ private plots, our research has identified a range of historical and contemporary drivers of lagoon siltation largely neglected to date. These drivers include coffee cultivation, timber extraction, plantation development and in- migration in the late 19th and early 20th century; erosion on contested state forest and plantation lands; state forest management practices; slope cuts to enlarge agricultural fields; agriculture in riparian zones; erosion from roads, trails and settlements; volcanic eruptions; and the river channel and floodplain modifications that were pushed in the Lukas and Flitner 15 frame of the river basin development projects of the first scalar fix since the early 1970s (for more information on these drivers see Lukas, 2017). How is it possible that the one-sided blaming of upland farmers’ agricultural practices continues to this day, while the other drivers of lagoon siltation have hardly been mentioned in debates, let alone examined or addressed? As noted above, experts’, project implementers’ and policy-makers’ dominant interpretive framing confined the scales of observation to upland farmers’ private plots. The knowledge produced in turn served to reinforce and confirm this interpretive framing and to justify the focus of interventions on the scale of farmers’ private plots. Why was this seemingly logic and politically coherent scalar fix not seriously questioned, despite the interventions’ limited effectiveness in reducing sediment yields? Why were observational scales not adjusted to encompass other dimensions of the social–ecological processes that seem quite ‘natural’ as units of observation (i.e. the entire watershed)? First, the links between land use and erosion, the complexity of sediment transport dynamics in river basins, the difficulty to estimate effects of watershed modifications on sediment loads and the time lags of such effects (Bruijnzeel, 2004; Hamilton, 1992; Hamilton and Pearce, 1991; Walling and Collins, 2008; Walling and Fang, 2003) make watershed management a messy policy field, marked by limited knowledge and uncertainty. This tends to impede immediate critical inquiry into established discourses and modes of intervention. Sayre’s (2017) analysis of the history of rangeland science similarly directs attention to the difficulty and limited effort to align what we call the scales of observation to ecological processes that operate and interact across various scales with high levels of variability and uncertainty. Second, the framing of lagoon sedimentation as a threat caused by upland farmers perfectly fit to environmental and development discourses, putting emphasis on soil degradation and poverty eradication and offering related funding. Debates on other causes of lagoon sedimentation, such as the river and floodplain modifications in the frame of the earlier river basin development projects or the state forest corporation’s management practices, did not fit in the political context, could not serve as an imperative for political action, were not politically opportune and would only have distracted from the newly consolidated clear-cut upland-lagoon discourse that was to serve as a basis for upland rehabilitation projects. Third, as we explore now, politics of state control over forests played a considerable role in producing the lock-in of interpretive framings and the scales of observation and intervention.

Scalar fix as hegemonic discourse: Role of conflicts over forest lands. According to our research, the lock-in effect described above with the scalar fixation of watershed debates, knowledge production and intervention on upland farmers’ private plots is related to long-standing conflicts over access to and control of forest land. Almost one quarter of Java’s land area is classified as state forest, mainly involving production and protection forests managed by a parastatal enterprise. The access to and control of these forests are disputed between the state forest corporation and villagers in many localities. As a politics in these disputes and based on the misleading claim that tree cover per se minimises erosion and balances stream flows,11 foresters have hegemonically framed state forests managed by professional foresters as an epitome of watershed conservation and villagers’ cultivation practices as a threat to sustainable watershed management. This framing has played a major role in the formation and consolidation of the lock-in described above. 16 ENE: Nature and Space 0(0)

Java’s state forests started to be demarcated and increasingly tightly managed by the colonial forest administration in the 19th century; villagers’ forest uses were thereby increasingly criminalised (for a comprehensive historical account, see Peluso, 1992). The state foresters’ rationale of sustaining timber production was soon complemented by concerns over upland degradation. The growing dependency of lowland agriculture on steady water flows for irrigation and the observation of deforestation and erosion on former coffee and tea plantations in the uplands sparked calls for forest preservation and prompted the expansion of state forest territories in the early 20th century (e.g. Coster, 1936; De Haan, 1936; Kerbert, 1916; see Lukas, 2015b). These calls were based on the assumption of forests generally functioning like a sponge, balancing stream flows over time. Although this conception is simplistic (Bruijnzeel, 2004; Calder and Aylward, 2006), it has served as important justification of tight state control over Java’s forests (Galudra and Sirait, 2009), well in line with the state’s and its forest corporation’s interest in the forests’ productive resources. The designation of almost a quarter of Java’s land area as state forest territory, and its exclusive, coercive management by the state forest corporation during Suharto’s New Order Regime led to a situation aptly characterised by Peluso (1992) as ‘Rich forests, poor people’. Peasants, who had very limited rights to forest resources, put enormous pressure on forest land. Despite policing, they were often able to access and use forest resources – without an opportunity to control and manage these forests, which resulted in widespread forest degradation (Nibbering, 1988; Peluso, 1992; Smiet, 1990). Faced with this ‘social economic pressure on the forest [that] increases each year’ (Bratamihardja, 1992: 96), the state forest corporation launched first ‘social forestry’ programmes in the 1980s (Peluso, 1992). In addition, the corporation was ‘trying to neutralize pressure keep[ing] in mind environment[al] conservation’, for example, by supporting forest margin villagers to ‘improve their environment and the productivity of their own land’ (Bratamihardja, 1992: 96). In other words, the state, with its watershed authorities under the MoF and its forest corporation, turned farmers’ private plots into a privileged scale of intervention as part of its policy to defend its claim to forest lands. The state forest corporation follows this approach until today, as our interviews with the head and other representatives of the corporation revealed. It runs programmes encouraging forest margin villagers to plant more trees on their private lands and supporting them in adopting alternative livelihood strategies, hence creating alternatives to using resources from the state forest territory and thereby defending the formal forest land tenure status quo. With their politics of forest control, the state and its forest corporation substantially contributed to the establishment and consolidation of the second scalar fix of environmental management. The institutional and discursive entanglement of state forest and watershed management has also effectively prevented the state forest corporation’s own management practices, which contribute to river sediment loads through regular clear cuts and to dry season water shortages through dense monocultures of pine (Lukas, 2017), from being debated in the frame of watershed management discourses; they simply remained beyond the common scales of observation. Furthermore, ironically, the struggles over forest and plantation lands have themselves contributed to river sediment loads: contested state forests and plantations comprise some of the most erosion-prone lands in the watershed of the Segara Anakan lagoon (Lukas, 2014b, 2015b, 2017). Since the late 1990s, these struggles over forest and plantation lands have entailed contestations of the dominant scalar fix and challenged the hegemonic role of state forestry. In particular, villagers and land reform movements now frame questions of just access to land and forest resources at a larger time scale, taking their displacements and dispossessions in the 1950s and 1960s into account (Lukas, 2014b, 2019). In doing so, some Lukas and Flitner 17 of them lay claim to the former role of the forest administration by portraying peasants occupying state forest land as ‘sustainable managers of fragile upland environments’ (Peluso et al., 2008: 398). In fact, villagers have in many parts of Java effectively claimed access to and control of the forests’ understories and thereby transformed them (Lukas and Peluso, 2019; Peluso, 2011). Furthermore, watershed management has started to change from mainly state-led, centralistic, command-and-control approaches to a larger diversity of non-state actors, initiatives and approaches (Lukas, 2013). Thus, with the broader shifts in political power in Indonesia and the state forest administration with its forest enterprise in a crisis of legitimacy, the scalar fix that has dominated watershed management for three decades seems to loosen its grip.

Concluding remarks The case of the Segara Anakan lagoon and its catchment provides a prime example illustrating the establishment of scalar fixes of environmental management in the context of political constellations and interests. It also highlights the persistence of these scalar fixes over longer periods of time, and their tendency, using the words of Sievanen et al. (2013), to ‘discursively sidelin[e] and even mask [...] contentious issues’ (214). Both the watershed and the lagoon have been profoundly transformed by natural processes and a multitude of socio- political dynamics, and both have been invested with meanings changing over time, implying and resulting from particular scales of observation and intervention. One important insight from this empirical case (particularly from the second scalar fix) is the crucial role and political character of observational scales. It comes as no surprise that the key agents of interventions tend to choose scales of observation that confirm rather than challenge their objectives and framings. It is more disturbing to see that only few scientists were able or willing to challenge the confinements that came along with these dominant interpretive framings. The outlines of a ‘knowledge controversy’ (Whatmore, 2009: 592–593), in which scientific knowledge production is contextualised in its social and political networks, did not develop until the early 2000s. For several decades, scientists analysing watershed degradation and sediment sources in Java have limited themselves to particular scales of observation (farmers’ private plots) in line with hegemonic framings, thus affirming the established scales and patterns of intervention through intentionally or unintentionally including or excluding certain social–ecological phenomena and processes. As a consequence, upland interventions aimed at reducing lagoon sedimentation, seasonal water shortages and flooding were based on unrealistic expectations and ignored various other causal factors which did not fit into the strategic interpretive framings and preferred modes of intervention of foresters, government officials and project implementers. In this context, it is important to emphasise that the perspective of scalar fixes urges us to see the scales of observation not as boundaries of our research, but as subjects of critical inquiry. Investigating the intricate relationships between interpretive framings and the scales of observation and intervention can improve our understanding of political struggles over environmental management. This includes a new, political look at the ambiguous roles of expert organisations and researchers. Critically questioning simplifying narratives and (scalar) patterns of knowledge production and intervention provides insight into the mechanisms and actor constellations causing their stabilisation. In contrast to Cohen and Bakker’s (2014) approach to ‘eco-scalar fixes’, we do not narrow the focus on the state or the political-administrative system as a quasi-autonomous actor in a well-defined jurisdictional space. The processes and empirical case we are concerned with are 18 ENE: Nature and Space 0(0) less clear-cut and more polymorphous. In addition to and often aligned with the objectives of powerful state organisations (Ministries of Forestry, Public Works, and Agriculture; watershed authorities; state forest corporation), they involve a range of actors, including international development organisations and consultants, and scientific communities. The empirical case presented directs attention also to the temporal aspect of scales in the shaping and consolidation of interpretive framings. For example, the framing of seasonal flooding of the newly established irrigation schemes in the lower river basin as being a result of unsustainable upland agriculture is based on a short-term observation that does not cover the period prior to the conversion of the seasonally inundated swamp forests into agricultural land. Similarly, the framing of lagoon sedimentation as an outcome of unsustainable upland agriculture neglects the roles of earlier social–ecological processes, in particular transformations during colonial times (deforestation, coffee cultivation, plantation development, and in-migration) and interventions in the 1960/70 s (river channelling and embankments). The state forest corporation’s claim to forest lands also neglects historical time scales and thereby disregards displacements of villagers linked to forest land expansions in the 1950/60 s (Lukas, 2014b) and earlier modes of forest access and control prior to the establishment of colonial forestry in the 19th century (Peluso, 1992). That is, there also exists a politics of temporal confinements of observation, knowledge production and intervention. Implicitly, the notion of scalar fixes that dominate knowledge production and political action for certain periods of time directs attention to historical developments and such temporal confinements. Only few authors have explicitly dealt with temporal issues in the scale debate so far, describing a politics of temporal scale (McCann, 2003; Zia, 2013) or pointing to future obligations (Bolin et al., 2008) and the recurrent character of rescaling processes over time (Brown and Purcell, 2005). More recently, and related to our argument, Sayre (2017) directed attention to a failure of knowledge production to adequately cover relevant time scales of ecological processes. The analysis of social–ecological change often requires such a time-sensitive perspective due to the temporal variability of the processes it is concerned with. Our conceptualisation of scalar fix as a temporary lock-in of interpretive framings of social–ecological processes and related scales of observation and intervention is well applicable to other contexts, beyond our empirical case presented here. For example, Sayre’s (2017) analysis of the history and politics of rangeland science could be thought through this conceptual lens, possibly yielding additional insight into the stabilisation and persistence of knowledge claims, the links between political framings and observational scales, and related (prescribed) management modes. Similarly, reconsidering the Himalayan Degradation Theory (Ives and Messerli, 1989) in the light of scalar fixes may provide new insight into the mechanisms of knowledge (re-)production and stabilisation and ensuing politics. Besides stabilisation, our empirical case also demonstrates that scalar fixes are not made to last; they can be challenged or replaced by new scalar fixes. The shift from the first to the second scalar fix resulted from a conjuncture of emerging global environmental discourses, the greening of international aid, the strengthening of national conservation policies, new research interest on coastal ecosystems, and river managers’ and consultants’ identification of sedimentation as a threat. The second scalar fix is being challenged as a result of a sudden political shift (the fall of President Suharto) and subsequent renegotiations of resource access and control, pushed by villagers. Thus, the starting points, agents and processes of breaking up established scalar fixes are highly diverse; and only conjunctions thereof may suffice to induce transformations – a finding of interest to the emerging field of regime and transition theorists (see, e.g. Geels, 2002; Lawhon and Murphy, 2012; Markard et al., 2012). Lukas and Flitner 19

The persistence of certain explanations of land, forest and watershed degradation in Java has been the focus of this paper and the starting point of our theoretical interest. The notion of scalar fixes provides a conceptual approach to better understand such persistent narratives and to challenge them. The political changes and struggles over land and forest resources since the fall of the New Order regime demonstrate how even such engrained interpretive framings and related practices can be challenged successfully that seemed unchangeable for decades.

Highlights

. The paper proposes a conceptual approach to better understand and challenge persistent environmental narratives and modes of political action. . It conceptualises scalar fixes as semi-stable configurations of interpretive framings of social-ecological processes and related scales of observation/intervention. . The lock-in of interpretive framings with particular scales of observation and intervention provides problem closure and impedes open debate. . The paper analyses two scalar fixes that have dominated watershed and coastal management discourses, research, and interventions in Java. . It illustrates how scalar fixes constrain the effectiveness of environmental management, and how they can be challenged or replaced.

Declaration of conflicting interests The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The research was part of the Indonesian-German research programme SPICE II þ III (Science for the Protection of Indonesian Coastal Marine Ecosystems) and sponsored by the German Federal Ministry of Education and Research (Grant No. 08F0391A and 03F0644B). It was also supported by the Bremen International Graduate School for Marine Sciences (GLOMAR), funded by the German Research Foundation (DFG) within the frame of the Excellence Initiative by the German federal and state governments to promote science and research at German universities.

ORCID iD Martin C Lukas http://orcid.org/0000-0001-8718-6300

Notes 1. Obviously this position implies also a distancing from the Lefebvrian triad on which Rangan and Kull build their model, more specifically the distinction between espace ve´cu and espace perc¸u that we find problematic in the interpretive practice-centred approach that we follow. 2. The research of our group combined remote sensing and historical cartography with social- scientific research methods, focussing on land use and land cover changes and environmental governance in the Segara Anakan lagoon and its catchment area (Lukas, 2014a, 2014b, 2015a, 2015b, 2017, 2019; Heyde, 2016). 20 ENE: Nature and Space 0(0)

3. Hart (2001) distinguished ‘big D’ Development, i.e. the ‘post-second world war project of intervention in the ‘‘third world’’’, from ‘little d’ development as a ‘geographically uneven, profoundly contradictionary set of historical processes’ (650). 4. During his visit to the Commission overseeing the U.S.-funded Mekong River development project, U.S. President L.B. Johnson is quoted declaring: ‘I am a river man. All my life I have been interested in rivers and their development’ (Press Release ECAFE/88, 17 May 1961, Memo, Ortiz-Tinoco, nd, Vietnam Country File, National Security File, Box 202, LBJL; quoted in Ekbladh, 2002). A few years later Johnson added: ‘We should take some of our ambitious plans which haven’t been working in other countries, [...] and put them into action in Indonesia’ (Simpson, 2008: 241). 5. Interviews in the frame of a series of case studies in the catchment of the Segara Anakan lagoon (Lukas, 2015b). 6. The last of these aims remained an illusion in the light of continuous nepotism, corruption and political repression. 7. It is an interesting irony that, while policy-makers, consultants and project implementers have consistently viewed lagoon sedimentation as a threat to the ecosystems and the livelihoods of lagoon residents since the early 1980s, some of the residents in the lagoon in fact see sedimentation as a chance to expand agriculture on emerging land and hence actively accelerate this process – a fact that has only recently started to be recognised by the state authorities in charge of river and lagoon management (Heyde, 2016). 8. The name of the Segara Anakan Management Authority (Badan Pengelola Kawasan Segara Anakan, BPKSA) was changed a number of times in the 2010s, and it was finally integrated into the district fisheries agency (Heyde, 2016: Ch. 3). 9. This focus on farmers’ agricultural plots was in line with simplistic watershed degradation hypotheses in other parts of the world and well aligned with the emerging international development paradigm putting emphasis on the marginalised rural poor. 10. Some of the upland farmers’ private agricultural plots are in fact an important sediment source, but only one among many others (Lukas, 2015b). 11. This widespread assumption has been challenged, as the links between land use, erosion and stream flows vary widely between localities. The knowledge of these relationships is controversial and the subject of revision and refinement (Bruijnzeel, 2004; Calder, 2005; Calder and Aylward, 2006; Hamilton, 1992). In Java, some of the monocultural teak and pine forests managed by the state forest corporation in fact have detrimental effects on erosion rates, land slide frequency and dry season water flows (Lukas, 2017).

ORCID iD Martin C Lukas http://orcid.org/0000-0001-8718-6300

References ADB (1996) Report and recommendation of the President to the Board of Directors on Proposed Loans and Technical Assistance Grant to the Republic of Indonesia for the Segara Anakan Conservation and Development Project. Mandaluyong: Asian Development Bank. ADB (2006) Indonesia: Segara Anakan conservation and development project. Completion Report. Mandaluyong: Asian Development Bank. Barbier EB (1990) The farm-level economics of soil conservation: The uplands of java. Land Economics 66: 199–211. Biggs D (2010) Quagmire. Nation-building and nature in the Mekong Delta. Seattle and London: University of Washington Press. Bird ECF, Soegiarto A., Soegiarto KA and Rosengren N (1982) Proceedings of the Workshop on Coastal Resources Management in the Cilacap region, Gadjah Mada University, Yogyakarta, Lukas and Flitner 21

Indonesia, 2024 August 1980. Jakarta: The Indonesian Institute of Sciences and the United Nations University. Bixler RP (2013) The political ecology of local environmental narratives: Power, knowledge, and mountain caribou conservation. Journal of Political Ecology 20: 273–285. Blomquist W and Schlager E (2005) Political pitfalls of integrated watershed management. Society & Natural Resources: An International Journal 18: 101–117. Bolin B, Collins T and Darby K (2008) Fate of the verde: Water, environmental conflict, and the politics of scale in Arizona’s central highlands. Geoforum 39: 1494–1511. Bratamihardja M (1992) Social forestry in Java, experiences and prospects. In: Simon H, Fattah ADS, Sumardi, et al. (eds) One century of sustainable forest management with special reference to teak in Java. Yogyakarta: Perum Perhutani & Faculty of Forestry, Gadjah Mada University, pp. 95–101. Brenner N (1998) Between fixity and motion: Accumulation, territorial organization, and the historical geography of spatial scales. Environment and Planning D: Society and Space 16: 459–481. Brown JC and Purcell M (2005) There’s nothing inherent about scale: Political ecology, the local trap, and the politics of development in the Brazilian Amazon. Geoforum 36: 607–624. Bruijnzeel LA (2004) Hydrological functions of tropical forests: Not seeing the soil for the trees? Agriculture, Ecosystems & Environment 104: 185–228. Bryant RL (1998) Power, knowledge and political ecology in the third world: A review. Progress in Physical Geography 22: 79–94. Calder IR (2005) Blue Revolution. Integrated land and water resource management. London: Earthscan. Calder IR and Aylward B (2006) Forest and floods: Moving to an evidence-based approach to watershed and integrated flood management. Water International 31: 87–99. Cohen A and Bakker K (2014) The eco-scalar fix: Rescaling environmental governance and the politics of ecological boundaries in Alberta, Canada. Environment and Planning D: Society and Space 32: 128–146. Coster C (1936) Typen van stervend land in Nederlandsch-Indie¨. 1. Bevolkings-theetuinen in de Preanger. Tectona 29: 781. Cribb R (1990) The Indonesian killings of 1965–1966: Studies from Java and Bali. Victoria: Centre of Southeast Asian Studies, Monash University. Cribb R (2001) How many deaths? Problems in the statistics of massacre in Indonesia (1965–1966) and East Timor (1975–1980). In: Wessel I and Wimho¨fer G (eds) Violence in Indonesia. Hamburg: Abera Verlag, pp. 82–98. Davis DK (2005) Indigenous knowledge and the desertification debate: Problematising expert knowledge in North Africa. Geoforum 36: 509–524. de Haan JH (1936) Uit de geschiedenis der indische boschhydrologie. Tectona 29: 79–82. Delaney D and Leitner H (1997) The political construction of scale. Political Geography 16: 93–97. Dengel HH (1986) Darul-Islam. Kartosuwirjos Kampf um einen islamischen Staat Indonesien. Stuttgart: Franz Steiner Verlag Wiesbaden. Diemont WH, Smiet AC and Nurdin (1991) Re-thinking erosion on Java. Netherlands Journal of Agricultural Science 39: 213–224. Donner W (1987) Land use and environment in Indonesia. London: C. Hurst & Company. Dudley RG (2000) Segara Anakan Fisheries Management Plan. Segara Anakan Conservation and Development Project, Components B & C. Consultant’s Report. Ekbladh D (2002) Mr. TVA: Grass-roots development, David Lilienthal, and the rise and fall of the Tennessee valley authority as a symbol for U.S. Overseas Development, 1933–1973. Diplomatic History 26: 335–374. Flitner M (2007) La¨rm an der Grenze: Flugla¨rm und Umweltgerechtigkeit am Beispiel des Flughafens Basel-Mulhouse. Stuttgart: Franz Steiner Verlag. Forsyth T (1996) Science, myth and knowledge: Testing Himalayan environmental degradation in Thailand. Geoforum 27: 375–392. Forsyth T (2003) Critical Political Ecology: The politics of environmental science. London: Routledge. Forsyth T (2008) Political ecology and the epistemology of social justice. Geoforum 39: 756–764. 22 ENE: Nature and Space 0(0)

Forsyth T and Walker A (2008) Forest guardians, forest destroyers. The politics of environmental knowledge in Northern Thailand. Seattle and London: University of Washington Press. Galudra G and Sirait M (2009) A discourse on Dutch colonial forest policy and science in Indonesia at the beginning of the 20th century. International Forestry Review 11: 524–533. Geels FW (2002) Technological transitions as evolutionary reconfiguration processes: A multi-level perspective and a case-study. Research Policy 31: 1257–1274. Goldman MJ, Nadasdy P and Turner MD (2011) Knowing nature: Conversations at the intersection of Political Ecology and Science Studies. Chicago, IL: University of Chicago Press. Hajer MA (1995) The Politics of environmental discourse: Ecological modernization and the policy process. Oxford: Clarendon Press. Hamilton LS (1992) The protective role of mountain forests. GeoJournal 27: 13–22. Hamilton LS and Pearce AJ (1991) Biophysical aspects in watershed management. In: Easter KW, Dixon JA and Hufschmidt MM (eds) Watershed Resources Management. Studies from Asia and the Pacific. Singapore: ISEAS (Institute of Southeast Asian Studies, Singapore) and East-West Center Honolulu, pp. 33–52. Hart G (2001) Development critiques in the 1990s: culs de sac and promising paths. Progress in Human Geography 25: 649–658. Heyde J (2016) Environmental governance and resource tenure in times of change: Experience from Indonesia. Doctoral Thesis, University of Bremen, Germany. Huszar PC and Cochrane HC (1990) Subsidisation of upland conservation in West Java: The Citanduy II Project. Bulletin of Indonesian Economic Studies 26: 121–132. Ives JD and Messerli B (1989) The Himalayan dilemma: Reconciling development and conservation. New York: Routledge. Jackson KD (1980) Traditional authority, Islam, and rebellion: A study of Indonesian political behavior. Berkeley: University of California Press. Jasanoff S (2004) States of knowledge: The co-production of science and the social order. London: Routledge. Kerbert HJ (1916) De Praktijk van de Boschreserveering. Tectona 8: 823–837. Khuriyati SF (2010) Babad tanah harapan: Babak-Babak penguasaaan tanah timbul Nusakambangan. Yogyakarta: INSIST Press. Klein J (2002) Deforestation in the Madagascar highlands – Established ‘truth’ and scientific uncertainty. GeoJournal 56: 191–199. Kull CA and Rangan H (2015) The political ecology of weeds: A scalar approach to landscape transformations. In: Bryant RL (ed.) The International Handbook of Political Ecology. Northampton: Edward Elgar, pp. 487–500. Kurtz HE (2003) Scale frames and counter-scale frames: Constructing the problem of environmental injustice. Political Geography 22: 887–916. Lave R (2012) Bridging political ecology and STS: A field analysis of the Rosgen Wars. Annals of the Association of American Geographers 102: 366–382. Lawhon M and Murphy JT (2012) Socio-technical regimes and sustainability transitions: Insights from political ecology. Progress in Human Geography 36: 354–378. Leach M and Mearns R (1996) The lie of the land: Challenging received wisdom on the African environment. Oxford: Currey. Lukas MC (2013) Political transformation and watershed governance in Java: Actors and interests. In: Muradian R and Rival LM (eds) Governing the Provision of Ecosystem Services. Dordrecht and New York: Springer, pp. 111–132. Lukas MC (2014a) Cartographic Reconstruction of Historical Environmental Change. Cartographic Perspectives 78: 5–24. Lukas MC (2014b) Eroding battlefields: Land degradation in Java reconsidered. Geoforum 56: 87–100. Lukas MC (2015a) Neglected treasures. Linking historical cartography with environmental changes in Java, Indonesia. Cartographica 50: 141–162. Lukas and Flitner 23

Lukas MC (2015b) Reconstructing contested landscapes. Dynamics, drivers and political framings of land use and land cover change, watershed transformations and coastal sedimentation in Java, Indonesia, Ph.D. thesis, University of Bremen, Germany. Lukas MC (2017) Widening the scope: linking coastal sedimentation with watershed dynamics in Java, Indonesia. Regional Environmental Change 17: 901–914. Lukas MC (2019) Tenure contestation: An important cause of land cover change (under review). Lukas MC and Peluso NL (2019) Transforming the classic political forest: Contentious territories in Java. Antipode, forthcoming. McCann EJ (2003) Framing space and time in the city: Urban policy and the politics of spatial and temporal scale. Journal of Urban Affairs 25: 159–178. Markard J, Raven R and Truffer B (2012) Sustainability transitions: An emerging field of research and its prospects. Research Policy 41: 955–967. Marston SA, Jones JP and Woodward K (2005) Human geography without scale. Transactions of the Institute of British Geographers 30: 416–432. Molle F (2007) Scales and power in river basin management: The Chao Phraya River in Thailand. Geographical Journal 173: 358–373. Molle F (2009) River-basin planning and management: The social life of a concept. Geoforum 40: 484–494. Moore A (2008) Rethinking scale as a geographical category: From analysis to practice. Progress in Human Geography 32: 203–225. Neumann RP (2009) Political ecology: Theorizing scale. Progress in Human Geography 33: 398–406. Nibbering JW (1988) Forest degradation and reforestation in a highland area in Java. In: Dargavel J, Dixon K and Semple N (eds) Changing tropical forests – Historical perspectives on today’s challenges in Asia, Australasia and Oceania. Canberra: Centre for Resource and Environmental Studies, pp. 155–177. Nibbering JW and Graaff DJ (1998) Simulating the past: Reconstructing historical land use and modeling hydrological trends in a watershed area in Java. Environment and History 4: 251–278. Norman ES, Bakker K and Cook C (2012) Introduction to the themed section: Water governance and the politics of scale. Water Alternatives 5: 52–61. Olive CA (1997) Land use change and sustainable development in Segara Anakan, Java, Indonesia: Interactions among society, environment and development. Doctoral Thesis, University of Waterloo, Canada. Palte JGL (1989) Upland farming on Java, Indonesia. A socio-economic study of upland agriculture and subsistence under population pressure. Doctoral Thesis, University of Utrecht, The Netherlands. Peluso NL (1992) Rich forests, poor people. Resource control and resistance in Java. Berkeley and Los Angeles: University of California Press. Peluso NL (2011) Emergent forest and private land regimes in Java. The Journal of Peasant Studies 38: 811–836. Peluso NL, Suraya A and Rachmann NF (2008) Claiming the grounds for reform: Agrarian and environmental movements in Indonesia. Journal of Agrarian Change 8: 377–407. Penning-Rowsell EC and Johnson C (2015) The ebb and flow of power: British flood risk management and the politics of scale. Geoforum 62: 131–142. PRC-ECI (1975) The Citanduy River Basin Development Project. Master Plan. Denver, Colorado and Banjar, Indonesia: Engineering Consultants, Inc, Ministry of Public Works and Electric Power, Directorate General of Water Resources Development, Directorate of Rivers and Swamps. PRC-ECI (1987) Segara Anakan engineering measures study. Main Report. Denver, Colorado, and Banjar, Indonesia: PRC-Engineering Consultants Inc, Ministry of Public Works, Directorate General of Water Resources Development, Republic of Indonesia. Purwanto E (1999) Erosion, sediment delivery and soil conservation in an upland agricultural catchment in West Java, Indonesia. A hydrological approach in a socio-economic context. Doctoral Thesis, Vrije Universiteit, The Netherlands. 24 ENE: Nature and Space 0(0)

Rangan H and Kull CA (2009) What makes ecology ‘political’?: Rethinking ‘scale’ in political ecology. Progress in Human Geography 33: 28–45. Reichel C (2007) Aspekte und Konflikte maritimer Ressourcennutzung und lokale Strategien zum Erhalt des natu¨rlichen Schutzes vor Naturgefahren. Ein Forschungsbericht aus der Segara Anakan Lagune, Central Java und dem Taka Bonerate Archipel, Su¨d-Sulawesi, Indonesien. Unpublished report of the field research for a Magister thesis, FU Berlin, Germany. Repetto R (1986) Soil loss and population pressure on Java. Ambio 15: 14–18. Rijsdijk A (2005) Evaluating sediment sources and delivery in a tropical volcanic watershed. In: Proceedings of symposium S1 held during the seventh IAHS scientific assembly at Foz do Iguac¸u, Brazil, April 2005, pp.16–23. Wallingford: IAHS. Rijsdijk A, Bruijnzeel LAS and Prins TM (2007a) Sediment yield from gullies, riparian mass wasting and bank erosion in the upper Konto catchment, East Java, Indonesia. Geomorphology 87: 38–52. Rijsdijk A, Bruijnzeel LAS and Sutoto CK (2007b) Runoff and sediment yield from rural roads, trails and settlements in the upper Konto catchment, East Java, Indonesia. Geomorphology 87: 28–37. Roe E (1994) Narrative policy analysis: Theory and practice. Durham, NC: Duke University Press. Sayre NF (2017) The politics of scale. A history of rangeland science. Chicago and London: The University of Chicago Press. Sayre NF and Vittorio AD (2009) Scale. In: Kitchin R and Thrift N (eds) The International Encyclopedia of Human Geography. Oxford: Elsevier, pp. 19–28. Schweithelm J (1988) Hydrology, soil erosion and sedimentation in the Citanduy basin: A watershed management perspective. Lashburn: Resources Management International, Inc. Sievanen L, Gruby RL and Campbell LM (2013) Fixing marine governance in Fiji? The new scalar narrative of ecosystem-based management. Global Environmental Change 23: 206–216. Simpson BR (2008) Economists with guns: Authoritarian development and U.S.-Indonesian relations, 1960–1968. Stanford, CA: Stanford University Press. Smiet AC (1990) Forest ecology on Java: Conversion and usage in a historical perspective. Journal of Tropical Forest Science 2: 286–302. Smith N (1992) Geography, difference and the politics of scale. In: Doherty J, Graham E and Malek MH (eds) Postmodernism and the Social Sciences, 1st ed. London: Macmillan, pp. 57–79. Sneddon CS, Magilligan FJ and Fox CA (2017) Science of the dammed: Expertise and knowledge claims in contested dam removals. Water Alternatives 10: 677–696. Sudjastani T (1982) The marine fishery resources of the Cilacap region. In: Bird ECF, Soegiarto A., Soegiarto KA and Rosengren N (eds) Proceedings of the Workshop on Coastal Resources Management in the Cilacap region, Gadjah Mada University, Yogyakarta, Indonesia, 20–24 August 1980. Jakarta: The Indonesian Institute of Sciences and the United Nations University, pp. 44–50. Swyngedouw E (2013) Into the sea: Desalination as hydro-social fix in Spain. Annals of the Association of American Geographers 103: 261–270. Towers G (2000) Applying the political geography of scale: Grassroots strategies and environmental justice. The Professional Geographer 52: 23–36. USAID (1984) Citanduy II – Organizational difficulties hinder implementation of an ambitious integrated rural development project. Audit Report No. 2-497-84-04. Manila: Regional Inspector General for Audit. USAID (1985) Citanduy II Assessment. Special Evaluation. Washington, DC: USAID. van Dijk C (1981) Rebellion Under the Banner of Islam: The Darul Islam in Indonesia. The Hague: Nijhoff. van Dijk A, Bruijnzeel LA and Purwanto E (2004) Soil conservation in upland Java, Indonesia: Past failures, recent findings and future prospects. 13th International Soil Conservation Organisation Conference. Conserving soil and water for society: sharing solutions, Brisbane. Venot J-P, Bharati L, Giordano M, et al. (2011) Beyond water, beyond boundaries: Spaces of water management in the Krishna river basin, South India. The Geographical Journal 177: 160–170. Walling DE and Collins AL (2008) The catchment sediment budget as a management tool. Environmental Science & Policy 11: 136–143. Lukas and Flitner 25

Walling DE and Fang D (2003) Recent trends in the suspended sediment loads of the world’s rivers. Global and Planetary Change 39: 111–126. Whatmore SJ (2002) Hybrid Geographies: Natures Cultures Spaces. London: Sage. Whatmore SJ (2009) Mapping knowledge controversies: Science, democracy and the redistribution of expertise. Progress in Human Geography 33: 587–598. White AT, Martosubroto P and Sadorra MSM (1989) The coastal environmental profile of Segara Ankan-Cilacap, South Java, Indonesia, Manila: International Center for Living Aquatic Resources Management (ICLARM) on behalf of the Association of Southeast Asian Nations/ United States Coastal Resources Manageement Project, Technical Publications Series 4. Yuwono E, Jennerjahn TC, Nordhaus I, Riyanto EA, Sastranegara MH and Pribadi R (2007) Ecological status of Segara Anakan, Indonesia: A mangrove-fringed lagoon affected by human activities. Asian Journal of Water, Environment and Pollution 4: 61–70. Zia A (2013) Post-Kyoto Climate Governance: Confronting the politics of scale, ideology and knowledge. Oxon: Routledge. Zimmerer KS (2007) Cultural ecology (and political ecology) in the ‘environmental borderlands’: Exploring the expanded connectivities within geography. Progress in Human Geography 31: 227–244. Zulu LC (2009) Politics of scale and community-based forest management in southern Malawi. Geoforum 40: 686–699.