Locating international REDD+ power relations

Geoforum 66 (2015) 64–74

Contents lists available at ScienceDirect

Geoforum

journal homepage: www.elsevier .com/locate /geoforum

Locating international REDD+ power relations: Debating forestsand trees in international climate negotiations

http://dx.doi.org/10.1016/j.geoforum.2015.09.0080016-7185/� 2015 Elsevier Ltd. All rights reserved.

E-mail address: [email protected]

Elana Wilson RoweNorwegian Institute of International Affairs, Norway

a r t i c l e i n f o

Article history:Received 4 January 2015Received in revised form 7 September 2015Accepted 12 September 2015

Keywords:Climate negotiationsREDD+RainforestsGlobal governancePower relations

a b s t r a c t

The Reduction of Deforestation and Forest Degradation initiative (REDD+) was initially hailed widely as asmart and cost-effective way to mitigate climate change and has moved quickly compared to otherstrands of the United Nations Framework Convention on Climate Change (UNFCCC) negotiations. Muchof the initiative’s original appeal – and a good deal of subsequent controversy around it – relates to fram-ing the world’s tropical forests as carbon sinks and compensating developing countries that manage toreverse or avoid deforestation. REDD+ negotiations can thus be seen a site where the standard divisionsbetween Annex 1 and non-Annex 1 (‘developed’ and ‘developing’) were being challenged and interro-gated by the negotiating parties and the broader network of actors around the climate regime. This articlesuggests that such complex and changing global governance policy fields need to be analysed as ‘places’in their own right, populated by actors engaged in field-specific power relations that may not reflectinternational hierarchies or power relations manifested in other international settings. Based in a uniqueset of interviews supplemented by primary data analysis, this article unpacks the power relations ofREDD+ negotiations by examining how those involved seek to assume competence, designate and recog-nize leadership, and shape outcomes. In tracing the dynamics of claiming competence, the ‘competition’between two disciplinary milieus around forests as an international policy object and also delegates’shifting between reliance on expert knowledge and political ‘know-how’ in the negotiations themselvesare identified. To understand the politics of recognition – that is to have a claim to competence or posi-tion acknowledged by others – the perceived qualities and resources of recognized leadership are exam-ined and the absence of global superpowers amongst REDD+ leadership is problematized and discussed.Finally, in terms of wielding influence over outcomes, the fate of two quite similar ideas – one that hasbecome incorporated into REDD+ methodology and another that is failing to be – further illustrate howthe field is marked by internal power practices and that not all actors are equally well-positioned toachieve desired outcomes.

� 2015 Elsevier Ltd. All rights reserved.

1. Introduction

The Reduction of Deforestation and Forest Degradation initiative(REDD+) was initially hailed widely as a smart, fast and cost-effective way to mitigate climate change and has moved quicklycompared to other strands of the United Nations FrameworkConvention on Climate Change (UNFCCC) negotiations (den Bestenet al., 2014, 46). Much of the initiative’s original appeal – and a gooddeal of subsequent controversy around it – relates to framing theworld’s tropical forests as carbon sinks and compensatingdeveloping countries thatmanage to reverse or avoid deforestation.

Developing countries were among REDD+’s earliest proponentsand the idea was on the whole warmly received by developed

countries, as the initiative demonstrated a novel willingness fromdeveloping countries to take on concrete mitigation responsibilityin the global climate regime. Such a move was seen as a potentialresolution to the enduring thorny issue of who should assumeresponsibility for ameliorating global climate change. The 1990sKyoto Protocol enshrined a division of labour within the globalclimate regime between Annex 1 (‘developed’ countries withbinding emissions reductions targets) and non-Annex 1(‘developing’ countries without binding targets). While developedcountries certainly bore the historical responsibility for globalcarbon emissions, this division (aka ‘the Kyoto firewall’) has comeunder increasing scrutiny as non-Annex 1 countries, like Chinaand India, have joined the ranks of major carbon emitting states.

REDD+ negotiations can thus be seen a site where the standarddivisions between Annex 1 and non-Annex 1were being challenged

E. Wilson Rowe /Geoforum 66 (2015) 64–74 65

and interrogated by the negotiating parties and the broadernetwork of actors around the climate regime (NGOs, indigenouspeoples organizations, business actors). This article provides insightinto the power relations of this changing policy field, examininghow policy actors have mobilizedmany different kinds of resourcesto advance their (evolving) positions in REDD+ negotiations. Suchan approach is grounded in the conviction that global governancepolicy fields need to be understood as ‘places’ in their own right,entailing local knowledges and norms and populated by actorsengaged in field-specific power relations that may not reflectassumed international hierarchies (e.g. great powers and smallstates). Given the solid vein of rigorous and critical research thathas been done on the power relations entailed in implementing(and resisting) REDD+ at the national and local scales, it is timelyto complement these with a study of REDD+ international negotia-tions that brings into focus the dynamics of the negotiations them-selves. This ‘how question’ about REDD+ negotiations also serves asan important complement to studies that have unpacked anddenaturalized the outputs (the ‘what’) of REDD+ global debates(e.g. discourses around and reframings of forests, policydocuments/decisions). Without attention to the ‘how’ question ofinternational politics, we are at risk of misreading the global-levelpolitics of REDD+ as a solely realist, great power game. The articlejoins a growing literature from geographers focusing on the con-crete practices of those who ‘construct, embody and narrate. . .geopolitics’ in global governance settings (Foxhall, 2012: 236,see also Jones and Clark, 2015; McConnell et al., 2012; Hakli andKallio, 2014).

I first review existing research around REDD+ and argue for theimportance of paying attention to the ‘local’ power relationsenacted in sites of global governance. I utilize a ‘field analytical’approach drawn from the work of Pierre Bourdieu to facilitate anopen analytical stance on power relations – and on the types ofpower resources that matter (material wealth, expertise, moralauthority, forests/territory and so on) – in sites of globalgovernance. After discussing methods, the broad lines of REDD+development as a policy field is reviewed, drawing upon a wealthof existing research and supplemented with insights from originalinterviews. I then present a case study drawn from a unique set ofinterviews with REDD+ negotiations actors and supported by areview of primary resources (UNFCCC submissions, Earth News Bul-letins). Taking an analytical cue from studies of diplomacy, I unpackthe power relations of REDD+ negotiations by examining howthose involved seek to assume competence, designate and recog-nize leadership and shape outcomes. To address the dynamics ofclaiming competence in REDD+, the ‘competition’ between twodisciplinary milieus around forests as an international policy objectand also delegates’ strategically shifting ‘cross-boundary’ utiliza-tion of expert knowledge and political ‘know-how’ in the negotia-tions themselves are explored. To understand the politics ofrecognition – that is to have a claim to competence or positionacknowledged by others – the qualities of acknowledged leader-ship are examined. Finally, in terms of wielding influence over out-comes, we trace the success and failure of two novel ideasintroduced into REDD+ negotiations (safeguards and non-carbonbenefits) and compare and contrast the constellation of actorsand the resources brought to bear in each instance.

2. ‘Placing’ REDD+ at the global level: conceptual and empiricalbackground

Existing research on REDD+ explores the practical andmethodological complexities of assigning a carbon value toavoided deforestation (Angelsen et al., 2009; Oekreke andDooley, 2010), the power politics and local environmental and

human rights challenges involved in implementing REDD+ in prac-tice in various forested landscapes and ‘forest countries’ (Cavanaghand Benjaminsen, 2014; Evans et al., 2014; Leggett and Lovell,2011; Luttrell et al., 2014; Shapiro-Garza, 2013; McAfee andShapiro, 2010), and the broader ideational/discursive and socialrepercussions of assigning a market logic and carbon value tostanding forests (standardization, simplification, green govern-mentality) (Backstrand and Lövbrand, 2006; den Besten et al.,2014; Forsyth and Sikor, 2013; Gupta et al., 2012; Lovell, 2014;Stephan, 2013; Stephan et al., 2014). These studies draw uponand tie into the work of geography scholars on the contours andconsequences of neoliberal commodification of nature at localand national scales (Robertson, 2006; Matulis, 2013; McCarthyand Prudham, 2004; McAfee and Shapiro, 2010; Shapiro-Garza,2013).

A good deal of important research has also been done on thestructure of and challenges facing international climate negotia-tions. Some of this is from a more prescriptive perspective (i.e.Keohane and Raustiala, 2010; Freestone and Streck, 2005), whileregime scholars have illustrated how new layers of complexityare making the negotiation process increasingly complex(Biermann et al., 2010; Oberthür and Stokke, 2011; van Asseltand Zelli, 2013). A number of recent publications taking aFoucauldian ‘governmentality’ approach to the problem of climatechange examine the effects of international climate negotiations,business strategies and national climate policies on the productionof new objects of governance – like carbon markets and rainforestsas carbon reservoirs – and the new kinds of rationalities, stake-holders and routines associated with this (see contributions inStripple and Bulkeley, 2014 for a key selection of thesecontributions).

However, we still know relatively little about the political land-scape in which a new ‘global’ framing of tropical forests has beennegotiated. As discussed above, REDD+ challenges the existingdivision between Annex 1 and non-Annex 1 within the climateregime and, for that reason, continuity with the few other detailedstudies of the social dynamics of climate negotiations (Dimitrov,2010; Lahn, 2013; Wilson Rowe, 2013) should not be assumed.The current state of knowledge reflects a broader weakness inglobal governance scholarship, namely that what happens in the‘engine room’ of global politics is frequently overlooked(Neumann et al., 2015; Holmes, 2013; Jones and Clark, 2015;Adler-Nissen and Pouliot, 2014). In other words, the meetingspaces of international diplomacy are frequently glossed over asstructured in keeping with extant global hierarchies and with pre-determined interests playing out amongst the representatives ofstates.

In a climate context, to take one example, this is reflected in thehigh levels of media and scholarly attention paid to the climatepreferences of the USA and China. To understand the broad direc-tion of the climate regime – the possibilities and constraints ofreaching a new climate agreement, for example – a focus on majorpowers makes sense. However, in climate change negotiations,there is much to suggest that exercising power and determiningoutcomes in international climate governance is not solely a greatpower game. For example, the Alliance of Small Island States(AOSIS), representing 39 states particularly exposed to rising sea-levels and extreme weather events, has been recognized as a ‘ma-jor player’ in the UNFCCC negotiations. This despite its memberstates comprising less than 1% of world territory and population(Betzold et al., 2012). Norway, a small state by most definitions,has played a high-profile role in promoting anti-deforestationmeasures by putting large amounts of money and expertise intothe processes surrounding REDD+(Lahn and Wilson Rowe, 2014).

Writing about the early days of land use/carbon sink negotia-tions at the UNFCCC negotiations, Lövbrand (2009) shows that

66 E. Wilson Rowe /Geoforum 66 (2015) 64–74

traditionally hegemonic actors did not dominate negotiations. Sheargues this was due to the lack of consensual knowledge aboutthe notion of terrestrial ‘carbon sinks’ and ongoing argument overthe ‘moral rightness’ of the sink concept as an offset for othercarbon reduction measures. This article’s case study of REDD+negotiations in recent years, during which such an epistemic chaoswas not a defining feature, still supports Lövbrand’s point aboutthe absence of consistent superpower (USA and China/G-77) dom-inance (although I find that superpower weight is indeed thrownabout at critical junctures). This suggests that the global gover-nance sites can be – and should be – studied as ‘places’ in theirown right, including locally produced power relations.

In thinking about the place-ness of global government, unpack-ing the notion of a ‘policy field’ is useful. Pierre Bourdieu’s work onvarious forms of capital and structures of power and dominancedescribed a ‘field’ as delineating a particular realm of interactionwith its own rules about behaviour within it. Agents in a fieldoccupy unequal positions and control over economic, social, andsymbolic resources that are significant within that field (allowing‘player to play the game more or less successfully’) (Pouliot,2010: 34; see also Hakli and Kallio, 2014). Sending (2015), in keep-ing with international relations scholars who have extrapolatedthe field concept to global governance settings, argues that fieldsin international relations as: ‘organized around concepts of gover-nance on which actors hold different conceptions. What unitesthem is a ‘‘thin” interest in what is at stake in the field, specifiedas the observed investment of time and resources in what the fieldis about.’ In this relational and practice-oriented tradition withininternational relations, power is not a single capacity (personalor national wealth, control of means of violence and so on) butrather something that is enacted in performance. Thus, the keypuzzle about power is not about seeking a universal definition ofwhat it is, but rather seeking to see what power does in a givencontext (Guzzini, 2005, p. 495). To put it another way, thisapproach allows us to be analytically open in terms of what kindsof competencies may result in REDD+ diplomatic performancesthat influence outcomes. In some instances, it may be knowledgeand expert authority, in others it could be material wealth/donorstatus or the possession of forest resources.

Coming to terms with what power does in the REDD+ policyfield is essential to understanding how the REDD+ policy fieldworks and why it produces the outcomes and effects that it does,thereby addressing the ‘how’ lacuna in existing research identifiedabove. In the case study that follows, I employ a typology of powerpractices synthesized by Adler-Nissen and Pouliot (2014) tounpack the dynamics of local power relations in the REDD+ policyfield. These scholars argue that it ‘requires constant work to turnstructural assets into power in practice’ and focus on threedimensions of this constant work in modern global governance(Adler-Nissen and Pouliot, 2014, 6). Firstly, there is the work ofclaiming competence,which involves positioning yourself as a ‘com-petent player’, in part by shaping the rules of engagement andthereby privileging particular types of knowledge and action.Secondly, engaging in social negotiation involves seeking positiverecognition for one’s positions and concerns (ideally successfully)vis-à-vis other actors in the negotiations. Finally, an actor is ableto wield influence through the deployment of such recognizedcompetence. However, the result of the competence claim andnegotiation need not be immediate and the three steps likely occursimultaneously rather than directly linked in a causal fashion.

This typology’s emphasis on competence lends itself well toexamining the field of land use negotiations, which are markedby a highly technical language. However, the conceptual interde-terminance of ‘competence’ keeps the analysis open to competenceclaims beyond the better studied dynamics of science-based expertperformance, such as competence on negotiating process and

precedent or insight into the political viability of certain proposals.For example, in the negotiation strand concerning how developedcountries account for carbon sinks and sources from land usechange (LULUCF), many of the negotiators are highly educatedexperts (forestry, climatology) engaged in expert discourse withother delegates. They are also responsible for following andpromoting particular national mandates. LULUCF negotiators arewell aware of the intersection of expert norms and national politicsin their policy field and negotiating these crosscurrents success-fully is of great importance to actors’ credibility and influence innegotiations (Wilson Rowe, 2013).

Using Adler-Nissen and Pouliot’s typology to structure the casestudy analysis thus promotes an attention to a diversity of powerresources that can be activated locally in REDD+ diplomacy andnegotiations. This allows us to expand upon the many strongexisting studies that focus primarily on the dynamics of expert/knowledge politics and boundary work in international climatesettings devoted more explicitly to questions of science (e.g. theIPCC) (Adler, 2005; Demeritt, 2001; Beck, 2012; Fourcade, 2009;Haas, 1992; Lövbrand, 2009; Mahony, 2013; Miller, 2004). Wereturn to this typology of power as an analytical window ontothe case study material after a brief foray into methods and areview of the broader institutional backdrop for REDD+.

2.1. Methods

The case study that follows draws upon a unique set of 23 inter-views and also utilizes primary written sources. Given the empha-sis placed in this article on the enactment and performance ofpower, interviews were essential for gaining insight into how the‘game’ is played, as analysis of formal written documents, suchas UNFCCC submissions, is unlikely to reveal informal relationsand practices. Semi-structured interviews were conducted with23 actors involved in UNFCCC negotiations. Interviews were iden-tified by a snowball method, with current interviewees suggestingadditional interviewees at my request. This is standard practice inassembling ‘elite’ interview sets and was absolutely essential in thecase of this research, as the UNFCCC does not specific which mem-bers of national delegations have participated in particular negoti-ating streams. Consequently, the first interviewees were primarilyfrom UNFCCC observer NGOs, who were invaluable sources ofinformation about further contacts with both current and past del-egates. Many requests for interviews were not answered (40%response rate to email requests), therefore those who did agreeto interviews may be a self-selecting group (persons more inter-ested in research, from countries that give a certain freedom to del-egates and so on). Nonetheless, by continuing to add interviews tothe set when needed, representation from developed and develop-ing country official delegates, as well as NGOs, was secured. Allinterviewees were asked questions from an interview guide cover-ing identification of leaders and laggards in REDD+, key ‘turningpoint’ moments in the negotiations, and the nature of expertiseand competence in the broader REDD+ policy field. Interviewswere conducted by the author alone (primarily in English), 19 wereconducted in person with persons stationed in or passing throughon business in Washington (DC), Brussels and Oslo, while fourwere conducted with voice Skype (Skype was an option for allpotential interviewees contacted, and thus the interview samplewas not limited by the interview locations used). The author tookextensive notes throughout the interview and produced a write-upimmediately thereafter. Quotations are only used when the authorwas confident in having a near-verbatim record of particularstatements.

A weakness with any snowball method sample of interviews isanother such sample of interviewees – even, say, starting with thesame initial interviewee – would likely roll further in different

E. Wilson Rowe /Geoforum 66 (2015) 64–74 67

ways at different points in time. Decisions about how to use theinterview transcripts were thus carefully weighed. Interviewwrite-ups were coded for content from context-generated codesand similarly coded statements were aggregated. Groupingstatements with the same general codes (e.g. leaders, laggards,moments of conflict) allowed for an assessment of the robustnessof intersubjective understandings, identification of disagreementand also the identification of statements that, while interesting,seemed to represent an outlier or highly individualized opinion.A reference to an interview in the text is indicated with an ‘I’ forinterview and a number assigned to the individual interview (seeAppendix A for an anonymized overview of persons interviewed).In order to highlight where intersubjective understanding betweenthe interviewees exists, several interview numbers may be cited. Inthe case of a quotation, the quote comes from the interview citedfirst, with subsequent citations indicating a general agreementwith the statement highlighted. The interviews were also used tocarry out a mini-quantitative exercise, presented in the case studybelow, that presents the frequency with which certain countrieswere highlighted as leaders.

Primary sources were also analysed. Written submissions (dis-cussion and position papers) made to the UNFCCC negotiationswith REDD+ subject matter between the years of 2005 and 2014were identified, covering the official introduction of REDD+ tothe negotiation agenda through to the time of writing of this arti-cle. 134 submissions were tallied and organized in Excel accordingto who authored them (both singly and as multiple authors), withthe aim of understanding if the capacity to produce written text inadvance of the negotiations mattered for which countries wereperceived of as influential or leading. A comprehensive overviewof mentions of REDD+ in Earth Negotiations Bulletin (ENB) between2005–2014 was also compiled. This independent summary ofnegotiations is cited when used and served to verify and checkassertions made during the interviews, particularly if a specificincident or issue was mentioned by only a few interviewees. TheUNFCCC setting description also draws upon previous participantobservation/interview research of another strand of UNFCCCnegotiations (Wilson Rowe, 2013).

3. REDD+ evolution and institutional backdrop

Deforestation has a clear global public goods element in that for-ests play an important role in the regulation of the Earth’s climate(absorbing Co2 and breaking it down in photosynthesis) and areshelter for 80% of the world’s terrestrial species (Humphreys,2010, p. 135–136). Nonetheless, states failed to even commencenegotiations for a forest convention when a flurry of environmentalconventions were being negotiated in the 1990s due to a stalemateover developing countries’ sovereignty concerns and developedcountries’ unwillingness to reward or finance developing countries’efforts (Dimitrov, 2005). Since, there have been a plethora of public/private initiatives, intergovernmental forums, commissions, panelsand, eventually, the negotiation of the Non-legally Binding Instru-ment on All Types of Forests (NLBI) (Humphreys, 2009).

The climate community’s gradually intensifying focus on forestshappened somewhat divorced from the ongoing international workon forestsmentioned above (andwewill see below that this becamerelevant in terms of turf wars over relevant expertise and attendingframings of forest landscapes). In the early days of land use negoti-ations, attention was primarily focused on if and how developed(Annex 1) countries should account for carbon sinks and sourcesfrom the land use sector (Backstrand and Lövbrand, 2006) andwhether developing country reforestation initiativeswould be eligi-ble for financial support under the Clean Development Mechanism(CDM). Avoided deforestation had been excluded from the Kyoto

Protocol CDM over concerns about the integrity of the climateregime and about the feasibility of accurate and reliable monitoringand accounting of forest carbon (Lövbrand, 2004, 2009; Stephan,2013). However, forests as carbon sinks and a source of emissions(deforestation) was included in the IPCC’s very first assessment(Stephan, 2013), with more specific land use accounting practicesprovided byGoodGovernance Guidance for LandUse, LandUse Changeand Forestry (IPCC, 2003) and sourcebooks from the intergovern-mental UN organization, Global Observation of Forest and LandCover Dynamics (GOFC-GOLD) (Lovell, 2014).

The ‘Coalition for Rainforest Nations’ (CfRN) reintroduced thenotion of avoided deforestation (or ‘RED’ as it was called then) asa climate mitigation measure in UNFCCC negotiations in 2005.Shortly thereafter, the idea received a important push forwardfrom the Stern Review, which highlighted avoided deforestationas a quick and relatively cheap way to reduce GHG emissions(Buizer, 2014). Countries in the Congo Basin took the CfRN’sproposal further, convincing negotiating parties that it would bepossible to also account adequately for credits in reducing forestdegradation, which resulted in a new ‘D’ in the lengthened ‘REDD’acronym. Subsequently, a group of countries with ‘low but stableforest cover’ (India, China) promoted the notion of including theenhancement of forest stocks, even in the absence of a deforesta-tion threat (giving us REDD+) (Buizer, 2014; den Besten et al.,2014, p. 43; Oekreke and Dooley, 2010; Hiraldo and Tanner,2011). Although the question of whether REDD+ carbon credits willactually function as part of an offset market remains hotly debated(I1; I2; I3; I6; I16; I7; I8; I11; I15), the entire structure of the REDD+ framework, with its emphasis on monitoring and valuing units ofcarbon dioxide equivalent stored in forests, taps into the broader‘win–win’ storylines of market solutions that shape the climateregime more broadly (Backstrand and Lövbrand, 2006; Hiraldoand Tanner, 2011; Lohmann, 2005; Nielsen, 2014).

Although this article’s focus is on the UNFCCC negotiationsthemselves, it is worth noting that REDD+-inspired projects havebeen actively implemented in bilateral and multilateral aid rela-tions. These outside forums were cited by interviewees as havinga direct impact on the negotiations in some instances, particularlyin providing alternative dialogue forums and generating real-worldexamples of what was workable in REDD+(and what was not) (I5;I7; I9; I11; I12; I13; I15; I17; I15). Key outside initiatives includethe Forest Carbon Partnership Facility (FCPF), which was estab-lished by the World Bank in 2010 and has pursued an active ‘readi-ness program’ for potential REDD+ countries, working to ensurethat these countries build and/or document the capacity neededto be eligible for REDD+ projects. A number of negotiators fromthe UNFCCC, for example from the EU Commission, USA, Mexicoand Norway, also represent their countries in the capacity ofdonors to the FCPF steering board (den Besten et al., 2014). Like-wise, the Norwegian Climate and Forest Initiative was the firsteffort to truly put financing on the table for REDD+ measures andled to the implementation of results-based financing agreementswith Indonesia (2010) and Brazil (2009), among other countries.Many of the countries also met between 2010–2014 to engage indialogue about REDD+ in a non-binding format as part of theREDD+ partnership (Reinecke et al., 2014, p. 37).

NGOs and other civil society actors have been official observersin the UNFCCC negotiations and actively work to influence dele-gates via lobbying. The Environmental Defense Fund was amongstthe first actors to propose that Brazil be recompensed for avoideddeforestation, which is an important conceptual departure fromthe active reforestation projects included under the CDM(Stephan, 2013). The role of NGOs in developing the REDD+ policyfield at the international level remains understudied. However, inthe case study of REDD+ negotiations to which the article nowturns, the capacity of NGOs to influence the REDD+ policy space

68 E. Wilson Rowe /Geoforum 66 (2015) 64–74

becomes evident in efforts to introduce safeguards for biodiversityand local users of forests into the REDD+ policy space.

4. Local power relations in international REDD+ negotiations

4.1. Setting the stage

The UNFCCC negotiations are divided into multiple tracks andmultiple kinds of negotiating events. The annual ‘Conference ofthe Parties’ is by far the most high profile event in a negotiationyear as it is the highest decision-making body under the UNFCCC.Heads of state or top-level ministers negotiate issues that haveproved intractable at other levels of the negotiations and newpolitical courses are mapped out (or fail to be). However, muchof the work negotiating REDD+ – and the climate regime more gen-erally) occurs in the meeting rooms and buzzing hallways of a largeconference hotel near the UNFCCC Secretariat in Bonn, Germany.The official members of state delegations stand out by way of thepink cord on their nametag necklaces, which gives them accessto all negotiations, in contrast to the yellow cord of NGOs andothers who can observe negotiations only until their invitationsare rescinded.

Since 2005, REDD+ has been actively negotiated under the Sub-sidiary Body on Scientific and Technical Advice (SBSTA), which ismandated to provide guidance on scientific and technological mat-ters. From 2008, REDD+ has also been negotiated in a workstreamunder the Ad Hoc Working Group on Long-Term cooperativeAction (AWG-LCA), which is tasked with negotiating outcomesrelating to a post-2012 binding climate agreement to replace theKyoto Protocol. The idea is that scientific and technical questionsrelating to how REDD+ should be implemented (i.e. referencelevels, norms and practices of monitoring and reporting) are han-dled under the SBSTA while ‘political’ issues like funding, carbonmarkets and questions that tie into the emerging structure of apost-2012 agreement are handled under the AWG-LCA. In practice,it is frequently the same negotiators or sets of negotiators movingfrom room to room, albeit with different facilitators and reportingchannels. The dynamics of this traversing of technical and politicalstreams and roles is critical to understanding what kind of compe-tence matters in the negotiations, as discussed below.

Agenda items adopted by the COP (high-level meetings) areallocated a certain amount of negotiating hours by the relevantChair (of SBSTA or AWG-LCA) in the two-week negotiating sessionsheld at regular intervals throughout the year. Expert meetings andworkshops are often called for to discuss outside the time-constrained and more formal framework of the negotiations them-selves questions or areas where there is little agreement and sen-sitive issues (I20; I22). However, interviewees frequentlymentioned memorable experiences of ignoring the allocated timelimits, including staying up all night in order to reach agreementon a key text. One European NGO observer had this memory:

One time the NGO observers were sent out. When we cameback the next day, the negotiators had been discussing all nightand they looked terrible. But they were clapping as theythought they had finalized the text. Brazil then approachedthe co-chair and quietly pointed out that the drafts had gottenmixed up and that they had been editing the last few hours in adocument that was not the most recent version of the text. Sothey had to use 2 more hours sorting the messy document backout (I17).

REDD+ negotiations have been described by many intervieweesas unique because (1) within REDD+, the forestry sector has an ownnegotiations stream (unlike transport/energy), (2) there are no for-mal alliances which characterize the negotiations in other UNFCCC

negotiations, and (3) there is quite a high level of participation fromboth developed and developing countries (I8; Allan and Dauvergne,2013). Several interviewees also commented on the well-mannered, informal and generally collegiate interactions betweendelegates to REDD negotiations (I17; I18; I23), even with progressbeing slow, disappointingly slow or ‘brutally slow’ (I6; I7; I20).

Mitigating climate cuts to the heart of business as usual forinternational society, states, local communities and the individual– making it no surprise that power is at play in these negotiations.However, as argued above, we still know too little about whatcounts for power and how power is exercised in the REDD+ policyspace. Below, we draw upon Adler-Nissen and Pouliot’s (2014)typology of competence, recognition/leadership and influence overoutcomes to put three aspects of the performance of power intohigher relief.

4.2. Claiming competence: what kind of knowledge mattersfor REDD+?

Scholarly work in geography and international relations hasshown how experts can play a role in policymaking beyond thatof providers of factual input into international policy processes(Haas, 1992; Adler, 2005), even working to delineate the veryboundaries of new global governance fields and thereby settingthe basic parameters for policy debate (see Beck, 2012; Demeritt,2001; Fourcade, 2009; Neumann and Sending, 2010). A shared fea-ture of this body of research is a focus on the early days of policyfield formation and/or upstream explicitly scientific institutions(like the IPCC), which may reinforce the assumption that expertisematters most at early junctures and when communicated as scien-tific consensus. When we look closely at REDD+ negotiations andwiden our gaze to catch different kinds of competence-claiming(including but not limited to expert politics), it is indeed clear thatclaiming competence is not a one-time act completed in the earlydays of a policy field, but rather a constant and ongoing effort.

Firstly, REDD+ negotiations are marked by an ongoing tensionover what should be considered a technical or policy issue. It isimportant to note that, for REDD+ delegates, defining an issue aseither political or technical likely entails an act of power that priv-ileges certain kinds of competence and knowledge. As one NGOobserver put it, ‘there were technical issues like deciding aboutwhether tree plantations should count and what kind of baselineshould be used. And there were political issues. . .Wait, they wereprobably all political at the end of the day’ (I5; also I6; I23). Forexample, early in the RED(D+) negotiations, the USA repeatedlyissued clear requests that the Coalition for Rainforest Nationsproposal be followed up in SBSTA with a focus on technical consid-erations. Brazil and Papua New Guinea (among others) lobbied tohave the negotiations placed in the Subsidiary Body on Implemen-tation (SBI) for policy/political consideration before technicalconsiderations (ENB, 2005a,b, 2006). The US position prevailed,meaning that several years would be spent on methodology beforea more policy track was opened as part of the Bali Action Plan in2008. Classifying an issue as technical rather than political canbe an important step in muting potential dissent around principlesand engaging parties in negotiations that move a policy fieldforward, but towards a still unknown endpoint.

Several negotiators also underlined the ways in which technicalknowledge could ‘get in the way’ in a more generalized sense atvarious points in the negotiations. For example, certain delegateswere described by other delegates as ‘technical experts with ahobby horse they would pursue endlessly at the expense of thebigger picture’ (I5), ‘divorced from the broader context on financeand bigger issues’ (I23) and pushing back against importantsimplifications wanted at the political level (I14). Another formernegotiator from a developing country suggested that a certain

E. Wilson Rowe /Geoforum 66 (2015) 64–74 69

absence of expertise in REDD+ resulted in a clearer system and bet-ter outcomes than in the expert-heavy negotiations strand ondeveloped countries land based sources and sinks (LULUCF):‘REDD+ is better because there was a learning process and this isreflected in REDD. . .plus many developing countries sent only 1representative – and that would have to be a political representa-tive – who agreed to rules and transparency that a technical guymay not have!’ (I7; also I11). In other words, REDD+ policy fieldparticipants do not exclusively embrace technical expertise andinput as a reliable source for politics (or as a haven from politics).In fact, seeing REDD+ issues in a more purely political light andseeking political agreement was somewhat easier than navigatingthe recurring and somewhat intractable methodological problems.

Secondly, throughout the interviews, it became clearer thatthere had been a kind of ongoing turf war between disciplines interms of whose technical knowledge was going to count at all. Thisis important as studies of expert politics are often of policy fields inwhich one expert community or scientific discipline is dominantlypositioned and in agreement amongst themselves (Haas, 1992).The forestry community, which had been dominant in other inter-national forestry negotiations and agreements, was shocked to seehow REDD+ developed. One informant from the intergovernmentaldevelopment sector put it this way: ‘REDD+ promotes a satelliteview of forests that is top-down and distant. Foresters have workedfor years to decentralize forest management to the communitylevel and now REDD+ is encouraging a recentralization’ (I9; alsoI1; I2). A developed country negotiator characterized the disso-nance between possibly relevant knowledge communities aroundREDD+ this way:

you definitely see a fracturing. Internationally, foresters or for-est conservation people were those negotiating on forestsresulting in the UN Forum on Forests (UNFF) and so on. . .theysaw REDD as very naive and were more aware of the barriers.On the community side, you had economists and consultantsdoing opportunity cost analyses and getting excited about theamount of carbon stored. They didn’t account for costs of landtenure issues and cadastre systems. They didn’t account forpolitical or preparatory costs. It isn’t just forests going to palmoil in a simple conversion, there are a lot of dynamics at play.In the past 5 years, the climate community has learned whatthe forest community knew already (I11).

As a representative of an intergovernmental lending body put it,‘foresters have been kicked out of REDD, it has been taken overby market economists and remote sensing specialists. The UNFFis now an empty shell, everyone looks to the UNFCCC for forest-related funding’ (I9; see also I10).

That these two knowledge communities – surely with a wholerange of actors and perspectives stretching between them – bothhave a claim in the REDD+ policy field becomes particularly man-ifest in questions of units. From outside the negotiations, foresterscontested the UNFCCC focus on tonnes of carbon dioxide equiva-lent rather than a more straightforward measure of hectare changein forest cover (I1; I2; I3). An interviewee at a major intergovern-mental environmental funding body argued:

REDD+ is fixated on units of carbon dioxide equivalent. Thefocus was on getting REDD+ to the market ASAP and it waslinked to the CDM and shaped for too long by a sidelined, intro-verted group of negotiators who obsessed over methods. Thesepeople were used to being able to make calculations – howmuch carbon does a diesel engine emit? My kid could calculatethat. And they just wanted to switch it to forests, but theyunderestimated how complex it is. Why not use hectare mea-surements? If you are not selling it on the market, why keepcarbon units? (I1).

The question of units – and the market-oriented focus entailedin the practices of converting forests into carbon equivalent units– has also been complex for the community of NGOs active in landuse negotiations at the UNFCCC. While several major Washington-based NGOs remain important parts of the REDD+ network, inter-viewees from the European NGO sector have purposely distancedthemselves from the policy field as REDD+ moves from acapacity-building phase to a potential market/offset phase, giventheir organizations’ principle stand against offsetting emissionsreductions commitments (I14; I9; I17). As one NGO REDD+ negoti-ations observer summed up the situation, ‘CAN International [themain climate NGO network] doesn’t touch REDD+ because thereis no common ground amongst its constituent organizations’ (I17).

From inside the UNFCCC negotiations, a focus on carbon unitsand market mechanisms was, by contrast, a taken for granted pointof departure based on the structure of the Kyoto Protocol and itsassociated mechanisms. A negotiator from a developed countryput it this way:

a lot of people wanted emissions to be the unit because theUNFCCC was all about emissions. In order to create a crediblemechanism and get real finance we would need to be precise.When it stays in the realm of development assistance, whichis where REDD+ is now mostly, a lack of precision is ok. But ifthere is hope of REDD achieving any scale, it will have to be pre-cise and credible enough for private finance (I11; I6).

Furthermore, members of the climate ‘community’ argued that afocus on carbon units was a helpful way of avoiding a definitionalbattle over what a forest is: ‘It is also easier to agree on what atonne of carbon is (even if it is harder to do the math) than what‘forest’ is – countries have different definitions of what qualifies’(I6; also I8; I18).

Claiming competence as a means of exercising power seemsindeed to be a task that requires constant work. This is evidentin the oscillation of REDD+ between technical and more explicitlypolitical framings. Consequently, success in negotiations involvesinfluencing how particular problems are defined and then switch-ing convincingly between more expert-style problem solvingnorms and forwarding a political mandate when appropriate. Del-egates do not seem to work to achieve a lasting ‘expert authority’or agreement on all aspects of REDD+. Rather, they need to exercisea kind of ‘epistemic savvy’, drawing upon and deploying expert andtechnical knowledge and political interests when appropriate (andonly then) in order to reach a lasting political consensus. Further-more, the question of whose expert knowledge should matterwhen expertise is called for has been less settled in this case thanin other cases covered in existing literature. REDD+ initially privi-leged a specific kind of professional expertise, often provided byconsultants and accountants, who arguably have a privileged posi-tion in governing the new carbon economy (Bock, 2014; Lovell andMacKenzie, 2011). However, as will be discussed in the section onwielding influence below, the actors that were initially marginal-ized as REDD+ was launched as a new international approach toforests and climate mitigation certainly did regain some staturein the debate on safeguards after the first heady days of carbonmarket dreams subsided.

4.3. Social negotiation: recognizing leaders

Having reviewed two examples of the complexities of claimingcompetence, we turn now to the second analytical window ontothe local enactment of power in REDD+ negotiations. How doREDD+ actors achieve recognition from others within the policyfield? Which actors were perceived as mattering the most, andwhat qualities and resources seemed to matter in gaining this sta-

70 E. Wilson Rowe /Geoforum 66 (2015) 64–74

ture? To gain purchase on this line of inquiry, I elected to ask allinterviewees an open-ended question about their understandingsof leadership in the climate negotiations setting (both who is a lea-der and what counts as leadership). Interestingly, but perhapsunsurprisingly given states’ privileged position in the UNFCCC, allinterviewees elected to focus on national delegates, although thequestion had been posed in an open-ended fashion. Primary sourcematerials – the UNFCCC submissions – supplement the analysisand help us bring into focus differences between formal and infor-mal contributions. This is because the production of written sub-missions can be seen as an intentional investment of financialand human capital in the REDD+ process that is extrinsic to thenegotiations themselves (prepared in advance in state capitalsfunded by national resources).

As can be seen from the results table below (Table 1), there is noone defining leadership quality. Countries can be seen as influen-tial by being a ‘forest country’, otherwise useful in negotiations,financially engaged in REDD+ more broadly and so on. However,it is clear that some countries were certainly seen by intervieweesas more ‘leading’ than others, suggesting that the internationalREDD+ policy field does indeed have an internal hierarchy.

In Table 1 above, 9 developed countries and 10 developingcountries were identified by interviewees as leaders within REDD+. Norway and Brazil stand out as the most frequently acknowl-edged leaders. On the one hand, this is unsurprising. Norway isundoubtedly the largest contributor to REDD+-related efforts andwith 1.9 billion USD in contributions vastly outstrips the next lar-gest contributor, the USA with 176 million USD in REDD+ contribu-tions (Voluntary REDD+ Database, 2013). Brazil’s 5 million squarekilometres of forest cover outstrips all countries of the developingworld (and all the developed countries with the exclusion of Rus-sia’s 8 million square kilometres of taiga) (The World Bank, 2014).

On the other hand, as previous sections have discussed, stand-ing forests have not always had this value and developed countrieshave not always had the opportunity to get ‘climate credit’ forwork on deforestation in the developing world. Rather, these coun-tries’ leading standing in REDD+ should be seen as the outcome ofsuccessful efforts to shape practices of competence and set stan-dards for recognition, thereby generating novel sources and prac-tices of power.

The qualities identified by interviewees in their discussion ofleadership also highlight many factors can be considered intrinsic

Table 1Leaders and associated qualities mentioned.a

Country # ofmentions

Qualities mentioned

Norway 15 ‘major donor/money’ (x4); ‘shot caller/central to the field’speaking time’; ‘arranging/chairing workshops and assess

Brazil 9 Strong negotiators (x5); ‘lots of technical expertise and rposition in negotiations (‘things won’t happen if Brazil doecapacity’ ‘largest forest country’

Indonesia 7 ‘quiet but effective’ (x2); ‘active’Australia 5 ‘agenda setter on remote sensing’; ‘active before changeEU 5 ‘prepared and working effectively to promote agreement

because they are speaking for all EU countries’; ‘chairs vadown by the complexity of establishing positions within

USA 5 Key player, ‘vocal and gets a lot of speaking time’ ‘activeMexico 4 ‘wants to be a climate leader’Philippines 4 Strong on safeguards (x2); chairing SBSTABolivia 3 ‘voluble’; ‘persistent and willing to block negotiations whCanada 3 Chair of SBSTA; ‘experienced, long-term negotiator’UK 3 None mentioned

a Countries with only 1–2 mentions have been removed from the table for increasedNetherlands, Philippines, Tanzania, Thailand, Tuvalu, and Uganda. 2 mentions were givZealand, Papua New Guinea, Switzerland, and Vietnam. Only one specific quality descripwas speaking on behalf of the CfRN. 4 interviewees did not respond to the question on

to the negotiations themselves, as well as extrinsic factors relatingto the country’s policies or resources more broadly. Intrinsic to thenegotiations themselves, individual negotiators’ levels of activityin terms of making useful or ‘relevant’ interventions was fre-quently noted and often linked to negotiators’ years of experience.As one negotiator from a developed country put it: ‘influence in thenegotiations is possessed by the persons themselves rather thanthe qualities of the country they are representing. Networks, howmuch you know and how well you understand matter a lot forthe efficacy of the interventions you make’ (I22). A capable andlarge negotiating team with the capacity to be flexible and gener-ate positions and responses quickly was also seen as a strength,with the ability to ‘draft text’ in the negotiations as a particularlyprized capacity mostly held by developed countries (I16; I18;I23). Likewise, the commissioning of reports perceived as relevantby others, funding of related workshops and side events, and theassumption of chairing responsibilities all merited multiplementions (I20; I21).

Again, some of these qualities – like the ability to attend theUNFCCC with a large and prepared delegation with the capacityto take on chairmanship roles or to commission reports or fundworkshops – tie back to the material capacities of a given state.However, if we take a formal submission to the UNFCCC as a reflec-tion of a state’s asset base and interest in setting the agenda, wesee that these formal contributions do not always tally with theleadership status a state achieves (Chart 1). For example, Japanand Costa Rica submitted the most paper submissions from devel-oped and developing countries respectively and were not amongthe main leaders mentioned (Table 2).

By contrast, the Philippines appears in the leadership table(Table 1) with a solid margin, but has only submitted onesingly-authored formal submission to the UNFCCC. The country’slong-term REDD+ delegates were mentioned frequently ininterviews as having individuals who had strong knowledge ofthe process (both amongst delegates and the broader NGO commu-nity) and trusted as good facilitators (I5; I7; I8; I11; I14; I18). Thissuggests that intrinsic characteristics mentioned by the intervie-wees’ (delegates’ experience, level of knowledge, understandingof the negotiations) do indeed matter for positioning countries asleaders within the UNFCCC setting. In the case of the Philippines,one might say that they established themselves as trustedexperts on the negotiations themselves with a strong understand-

(x6); ‘generally active/well-prepared in negotiations’ (x5); ‘vocal and gets a lot ofment reports that help find consensus’ (x3)espect from other developing countries’; ‘takes domestic action’ (x2); strongsn’t want it to work’) (x2); ‘one of the few developing countries with text drafting

of government’ (x2)’; ‘strong on safeguards but doesn’t really stand out otherwise’; ‘has weightrious working groups and events’; ‘supportive of Norway but limited or slowedthe EU’’

ere other countries would have read the writing on the wall and given up’ (x3)

readability. Single mentions were given to Costa Rica, Chile, China, Finland, Ghana,en to Colombia, the Democratic Republic of Congo, Germany, Guyana, Japan, Newtion was attached to any to these countries, namely that PNG had influence when itleaders and laggards.

Chart 1. Comparing leadership mentions from the interviews to UNFCCC writtensubmissions (2005–2014).

E. Wilson Rowe /Geoforum 66 (2015) 64–74 71

ing of others’ interests within and outside the negotiating room,the climate regime, and REDD+ precedents. Thus one could arguethat, while material and structural capacities (Norway’s donor dol-lars and Brazil’s rainforests) do indeed matter for positioning one-self as a leader in this policy field, there are also roles that can beplayed entirely ‘locally’ to generate an authority intrinsic to thenegotiations themselves.

4.4. Wielding influence

The question of influence is perhaps one of the broadest ques-tions that can be posed about global governance or politics moregenerally. And we know from existing scholarship on REDD+reviewed above that critical influence is exerted by the most natu-ralized, unquestioned and backgrounded features of the policyfield, such as the simplification of a complex forest landscape intocarbon units. Here, however, to keep focus on the local power prac-tices of the REDD+ negotiations themselves, I have elected to focusthe analysis on two active, concrete efforts to introduce new ideasinto the REDD+ policy space – one successful (‘safeguards’) and onefailing (‘non-carbon benefits’).

We turn first to the introduction of ‘safeguards’ (the provision ofassurances and monitoring of human and indigenous rights,including attention to multiple use values and users of forests, inREDD+ projects). As discussed above, REDD+ initially involved aconception of forests that privileged a single forest function as

Table 2‘‘Leaders” single and multi-authored submissions 2005–2014, plus one developed andone developing country with a high number of submissions that were not mentionedas leaders (in italics).

Country mentioned as climate leader inTable 1 (control countries in italics)

Singlyauthoredsubmissions

Multi-authoredsubmissions

Australia 5 1Bolivia 6 3Brazil 6 0Canada 1 0Costa Rica 5 7European Commission

(and EU member states)8

Indonesia 7 3Japan 8 0Mexico 3 1Norway 4 0Philippines 1 3United States 8 0

the point of focus – the storage (or emission) of carbon – and alsoprivileged the expertise of economists and remote sensing special-ists. The subsequent period – 2008–2011 – was characterized bydebate over the notion of human and indigenous rights and a con-cern that forests were being too narrowly conceived in negotia-tions, purely as carbon stock rather than biodiversity for example(den Besten et al., 2014, p. 44). Although these human rights andbroader environmental concerns exceeded what many negotiatorsthought was the proper climate mandate of the UNFCCC and manywere concerned that a broadened REDD+ agenda would, in thewords of a former developed country negotiator, ‘sink the boat’(I11; also I22; den Besten et al., 2014), the country delegatesallowed the REDD+ envelope to expand to include a broader rangeof concerns and assurances. It is widely acknowledged that NGOpressure was important in bringing about this change, althoughkey developed countries like Norway were early out in providingsupportive submissions and workshops relating to safeguards(I22). As one developed country negotiator from a major powercountry recollected:

the safeguard issue was largely driven forward by the NGOcommunity. . .Also some of the negotiators were actually fromthe NGO sector and pushed these interests. So much of whathappens in negotiations depends on the personalities of thoseat the table. They may not represent the country’s total inter-ests. At the same time developed countries did want to beresponsive to community concerns (I11; also I22).

The introduction of safeguards can be understood as a resur-gence of sustainable forestry and development communities withlong track records on forestry issues – who could wield an exper-tise that likely became essential and more broadly acknowledgedas REDD+ projects hit the ground (and ran into difficulties) inspecific forest contexts.

Not all new ideas introduced into REDD+ gain the traction thatsafeguards did, as the troubled journey of a recent effort to furtherexpand the REDD+ envelope illustrates. A coalition of NGOs anddeveloping countries has engaged in an effort to include into thenegotiations a discussion of possible methods to reward/incen-tivize countries for the effective management of non-carbon bene-fits of forests – in other words, for good forest governance and theimplementation of the safeguard regime described above. Thiscoalition describes such benefits as encompassing a ‘wide rangeof positive [social, environmental and governance] outcomesresulting from REDD+ activities beyond those associated withavoided carbon dioxide emissions’ (Conservation Internationalet al., 2014). Developed country negotiators had a fairly negativeview on the initiative driven forward by the countries of WestAfrica, Sudan and Tanzania. One developed country negotiatorcharacterized the issue of non-carbon benefits as ‘not very criti-cal. . .we all know forests are more than carbon but it isn’t clearwe need to push more on this in the UNFCCC’ (I8). Another put itthis way:

these countries support it because they are not getting a pieceof the pie from REDD+ and they want a separate methodologyfor non-carbon benefits. The [developing] forest countries don’twant it and the developed countries don’t want it and theissue’s proponents can’t manage to present it in a coherentand logical manner. . .it falls outside the traditional mitigation/adaptation scope of the UNFCCC. They tried to do so late in Junebut presented it too late at the SBSTA for anything to be con-cluded. They are also supported by civil society actors to someextent. The idea can’t be dismissed because there is a reason-ably large coalition of countries backing it, but it is unlikely theywill dare to hold up the broader negotiations with this issue(I20).

72 E. Wilson Rowe /Geoforum 66 (2015) 64–74

Thus, while the dismissing or combating the idea would cause fric-tion, it seems it will be allowed to languish so long as the dominant‘forest’ developing countries and developed countries do notengage with resources and a willingness to flesh out the notion.

Comparing the trajectory of these two introduced ideas isinstructive and shows the intricate interplay between the threedimensions of enacting power that have been studied here. Whilethe policy field participants were willing to include safeguards asrequirements for REDD+ projects, they were unwilling to adopt anew idea that would further merit separately or reward financiallythe good governance of forests. One may argue that non-carbonbenefits might have their day – the idea has been introducedrecently and is clearly underdeveloped. On the other hand, the lackof development may also indicate that actors rich in relevant com-petence, be it expert knowledge or political capital, or key leadersin the REDD+ local hierarchy have not been amenable to joiningthe coalition. This is a point of contrast to the case of safeguards,which was driven forward by a coalition of NGOs with a wealthof established experience in forestry politics and some of theAnnex 1 REDD+ donor countries acknowledged by REDD+ field par-ticipants as leaders.

5. Conclusion

The three analytical windows above from Adler-Nissen andPouliot’s (2014) typology of diplomatic power – claiming compe-tence, social negotiation and wielding influence – have given ussome insight into the local practices of power at play in recastingforest landscapes as carbon sinks at the international level. Draw-ing upon Bourdieu’s definition of a field and the work of global gov-ernance scholars who have extrapolated this concept to theinternational level, I have argued that existing research on REDD+, while rich and varied, leaves unanswered some key ‘how’ ques-tions about the ‘place’-specific power practices enacted withinREDD+ as an international policy field. I have focused on the webof relations and actors who are engaged in and around UNFCCCnegotiations to come to grips with this empirical lacuna.

In the discussion of claiming competence, it was shown thatexpertise and technical knowledge are not simply the upstreaminputs in a ‘science to policy’ linear process. REDD+ has yet to befully dominated by one community of knowledge over anotherand defining what should count as relevant knowledge has compli-cated and also made multiple the ways in which a successful com-petence claim can be forwarded. Secondly, the process of definingwhich kinds of issues are technical and which are political is initself an act of power. Delegates are left needing to display a highlevel of ‘cross-boundary savvy’ in order to be successful in claimingcompetence as both experts and explicitly political actors. Havingtechnical and political knowledge is not enough, timing is alsoessential. It is knowing when and how to deploy certain kinds ofcompetence that brings recognition.

Interview # Anonymized description

1 Intergovernmental environmental o2 Intergovernmental environmental o3 NGO representative, Washington-ba4 International forestry civil servant,5 NGO, long-time negotiations observ6 NGO representative, Washington-ba7 Former developing country REDD+

It is interesting to note that the US was the motor for definingREDD+ primarily as a technical issue in the early days of REDD+,despite the opposite preference of developing countries. However,after this initial agenda-setting moment, the US is rarely men-tioned again by REDD+ field participants as a key actor to be con-sidered or a leader. This further reinforces the point that we cannotsimply assume that global hierarchies (e.g. superpowers/middlepowers/small states) are reflected at all levels of global governance.In terms of wielding influence, however, it is crystal clear in theopposing fates of the safeguard idea and the non-carbon benefitidea that not all actors are equally well-positioned to achievedesired outcomes.

In other words, the absence of superpowers does not equal theabsence of power politics or perceived hierarchy. The dynamics ofrecognition show us that structural resources, like Norway as amajor donor or Brazil as a forest power, do indeed matter for beingseen by other policy field participants as leader. More surprising, isthat a country like the Philippines also gains a high level of recog-nition based on resources intrinsic to the negotiations themselves(networks, knowledge of process and precedence).

Miller and Edwards (2001) injunction to consider climate poli-tics as a location where a changing global order is being negotiatedseems to remain valid. And, as this article has shown, power rela-tions and resources of power can indeed be ‘local’ to an interna-tional policy field. Extending a geographer’s traditionalcommitment to the specificity of place and bringing to bear anopen analytical stance about what can count for power in theselocal-international contexts is essential to closing knowledge gapsabout the politics of the ‘engine rooms’ of global governance.

Acknowledgements

The author is grateful to three anonymous reviewers and Geo-forum editor Thomas Perreault for useful comments on the draftarticle. I am very grateful to all of the interviewees, who took timeout from busy schedules to speak with me. Colleagues at the Nor-wegian Institute of International Affairs gave valuable input at apublication seminar and Ms. Wrenn Yennie Lindgrenn providedkey research assistance. Bård Lahn at the University of Oslo com-mented on a draft of the article and shared generously of hisREDD+ and UNFCCC expertise over several conversations. Theresearch and writing of this article was funded by the Global Re-ordering: Evolution through European Networks (GR:EEN) project(#266809), which was funded under the Cooperation category ofthe EU seventh framework programme call Europe facing a risingmulti-polar world (FP7-SSH-2010-4.1-1).

Appendix A:. Anonymized overview of interviewees

Date of interview

rganization civil servant 19 May 2014rganization civil servant 19 May 2014sed 19 May 2014Washington-based 23 May 2014er, over Skype 27 May 2014sed 28 May 2014negotiator 28 May 2014

Appendix A (continued)

Interview # Anonymized description Date of interview

8 Active developed country negotiator 29 May 20149 Intergovernmental lending agency representative 29 May 2014

10 Intergovernmental lending agency representative 29 May 201411 Former developed country negotiator, over Skype 31 May 201412 Intergovernmental lending agency representative 3 June 201413 Intergovernmental lending agency representative 3 June 201414 NGO, long-time negotiations observer, active, over Skype 25 June 201415 European NGO, over Skype 1 October 201416 European NGO, negotiations observer 2 October 201417 International NGO, negotiations observer 8 October 201418 Developed actor negotiator 13 October 201419 Developed country negotiator 20 October 201420 Developed country negotiator 21 October 201421 Developed country negotiator 28 October 201422 Developed country negotiator 28 October 201423 Developing country negotiator 11 November 2014

E. Wilson Rowe /Geoforum 66 (2015) 64–74 73

References

Adler, E., 2005. Communitarian International Relations: The Epistemic Foundationsof International Relations. Routledge, London and New York.

Adler-Nissen, R., Pouliot, V., 2014. Power in practice: negotiating the internationalintervention in Libya. Eur. J. Int. Relat. http://dx.doi.org/10.1177/1354066113512702.

Allan, J., Dauvergne, P., 2013. The global South in environmental negotiations: thepolitics of coalitions in REDD+. Third World Quart. 34 (8), 1307–1322.

Angelsen, A., Brockhaus, M., Kanninen, M., Sills, E., Sunderlin, W.D., Wertz-Kanounnikoff, S. (Eds.), 2009. Realising REDD+ National Strategy and PolicyOptions. Indonesia Center for International Forestry Research (CIFOR), Bogor,<http://www.cifor.org/publications/pdf_files/Books/BAngelsen0902.pdf>.

Backstrand, K., Lövbrand, E., 2006. Planting trees to mitigate climate change:contested discourses of ecological modernization, green governmentality andcivic environmentalism. Glob. Environ. Polit. 6 (1), 50–75.

Beck, S., 2012. Between tribalism and trust: the IPCC under the ‘public microscope’.Nat. Cult. 7 (2), 151–173.

Betzold, C., Castro, P., Weiler, F., 2012. AOSIS in the UNFCCC negotiations: fromunity to fragmentation? Clim. Policy 12 (5), 591–613.

Biermann, F., Pattberg, P., Zelli, F., 2010. Global Climate Governance Beyond 2012:Architecture, Agency and Adaptation. Cambridge University Press, Cambridge.

Bock, S., 2014. Politicized expertise – an analysis of the political dimensions ofconsultants’ policy recommendations to developing countries with a case studyof McKinsey’s advice on REDD+ policies. Innovat.: Eur. J. Soc. Sci. Res.

Buizer, M., 2014. Climate change and deforestation: the evolution of an intersectingpolicy domain. Environ. Sci. Policy 35, 1–11.

Cavanagh, C.J., Benjaminsen, T., 2014. Virtual nature, violent accumulation: the‘spectacular failure’ of carbon offsetting at a Ugandan National Park. Geoforum56, 55–65.

Conservation International, Environmental Defense Fund, National WildlifeFederation, Rainforest Alliance, The Nature Conservancy, and Union ofConcerned Scientists, 2014. Clarifying the Role of Non-Carbon Benefits inREDD+. Available online at: unfccc.int/resource/docs/2014/smsn/ngo/405.pdf(accessed 26.11.14).

den Besten, J.W., Arts, B., Verkooijen, P., 2014. The evolution of REDD+: an analysisof discursive institutional dynamics. Environ. Sci. Policy 35, 40–48.

Demeritt, D., 2001. The construction of global warming and the politics of science.Ann. Assoc. Am. Geogr. 91, 307–337.

Dimitrov, R.S., 2005. Hostage to norms: states, institutions and global forest politicsglobal. Environ. Polit. 5 (4), 1–24.

Dimitrov, R.S., 2010. Inside Copenhagen: the state of climate governance. Glob.Environ. Polit. 10 (1), 18–24.

ENB, 2005a. UNFCCC COP 11 and COP/MOP 1. Earth Negotiat. Bull. 12 (284).ENB, 2005b. UNFCCC COP 11 and COP/MOP 1. Earth Negotiat. Bull. 12 (287)

(International Institute for Sustainable Development).ENB, 2006. COP/MOP2. Earth Negotiat. Bull. 12 (309) (International Institute for

Sustainable Development).Evans, K., Murphy, L., de Jong, W., 2014. Global versus local narratives of REDD: a

case study from Peru’s Amazon. Environ. Sci. Policy 35, 98–108.Fourcade, M., 2009. Economists and Societies: Discipline and Profession in the

United States, Britain and France, 1890s–1990s. Princeton University Press,Princeton, NJ.

Forsyth, T., Sikor, T., 2013. Forests, development and the globalization of justice.Geogr. J. 179 (2), 114–121.

Foxhall, A., 2012. Constructing, practicing and narrating Russian (geo)politicalidentity. Geopolitics 17 (1), 235–241.

Freestone, D., Streck, C., 2005. The EU and Climate Change. In: Macroy, R. (Ed.),Reflections on 30 Years of EU Environmental Law, A High Level of Protection?The Avosetta Series (7). Europa Law Publishing, Groningen, pp. 87–106.

Guzzini, S., 2005. The concept of power: a constructivist analysis. Millennium 33(3), 495–522.

Gupta, A., Lövbrand, E., Turnhout, E., Vijge, M., 2012. In pursuit of carbonaccountability: the politics of REDD+ measuring, reporting and verificationsystems. Curr. Opin. Environ. Sustain. 4, 726–731.

Haas, P.M., 1992. Introduction: Epistemic Communities and International PolicyCoordination. International Organization 46(1) Knowledge, Power, andInternational Policy Coordination, 1–35.

Hakli, J., Kallio, K.P., 2014. The global as a field: children’s rights advocacy astransnational practice. Environ. Plan. D: Soc. Space 32, 293–309.

Hiraldo, R., Tanner, T., 2011. Forest voices: competing narratives over REDD+. IDSBull. 42 (3), 42–51.

Humphreys, D., 2010. International forest politics. In: Kütting (Ed.), GlobalEnvironmental Politics: Concepts, Theories and Case Studies. Routledge,London, pp. 135–150.

Humphreys, D., 2009. Discourse as ideology: neoliberalism and the limits ofinternational forest policy. For. Policy Econ. 11, 319–325.

Holmes, M., 2013. The force of face-to-face diplomacy: mirror neurons and theproblem of intentions. Int. Org. 67 (4), 829–861.

IPCC, 2003. Good Governance Guidance for Land Use, Land Use Change and Forestry,Intergovernmental Panel on Climate Change.

Jones, A., Clark, J., 2015. Mundane diplomacies for the practice of Europeangeopolitics. Geoforum 62, 1–12.

Keohane, R., Raustiala, K., 2010. Toward a post-Kyoto climate change architecture: apolitical analysis. In: Aldy, J.E., Stavins, R.N. (Eds.), Post-Kyoto InternationalClimate Policy: Implementing Architectures for Agreement. CambridgeUniversity Press, Cambridge, UK.

Lahn, B., 2013. Klimaspillet: En fortelling fra innsiden av FNs klimatoppmøter [TheClimate Game: a story from inside the UNs top climate meetings]. FlammeForlag, Oslo.

Lahn, B., Wilson Rowe, E., 2014. How to be a ‘front-runner’: Norway andinternational climate politics. In: de Carvalho, B., Neumann, I.B. (Eds.), SmallStates and Status Seeking: Norway’s Quest for International Standing.Routledge, Oxford, pp. 126–145.

Leggett, M., Lovell, H., 2011. Community perceptions of REDD+: a case study fromPapua New Guinea. Clim. Policy 12 (1), 115–134.

Lohmann, L., 2005. Marketing and making carbon dumps: commodification,calculation and counterfactuals in climate change mitigation. Sci. Cult. 14 (3),203–235.

Lövbrand, E., 2004. Bridging political expectations and scientific limitations inclimate risk management – on the uncertain effects of international carbon sinkpolicies. Clim. Change 67, 449–460.

Lövbrand, E., 2009. Revisiting the politics of expertise in light of the Kyotonegotiations on land use change and forestry. For. Policy Econ. 11, 404–412.

Lovell, H., MacKenzie, D., 2011. Accounting for carbon: the role of accountingprofessional organizations in governing climate change. Antipode 43 (3), 704–730.

Lovell, H., 2014. Measuring forest carbon. In: Stripple, J., Bulkeley, H. (Eds.),Governing the Climate: New Approaches to Rationality, Power and Politics.Cambridge University Press, Cambridge, UK, pp. 175–196.

74 E. Wilson Rowe /Geoforum 66 (2015) 64–74

Luttrell, C., Resosudarmo, I.A.P., Muharrom, E., Brockhause, M., Seymour, F., 2014.The political context of REDD+ in Indonesia: constituencies for change. Environ.Sci. Policy 35, 67–75.

Mahony, M., 2013. Boundary spaces: science, politics and the epistemic geographiesof climate change in Copenhagen 2009. Geoforum 49, 29–39.

Matulis, Brett Sylvester, 2013. The narrowing gap between vision and execution:neoliberalization of PES in Costa Rica. Geoforum 44, 253–260.

McAfee, K., Shapiro, E.N., 2010. Payments for ecosystem services in Mexico: nature,neoliberalism, social movements and the state. Ann. Am. Assoc. Geogr. 100 (3),1–21.

McCarthy, J., Prudham, S., 2004. Neoliberal nature and the nature of neoliberalism.Geoforum 35 (3), 275–283.

McConnell, F., Morea, T., Dittmer, J., 2012. Mimicking state diplomacy: thelegitimizing strategies of unofficial diplomacies. Geoforum 43, 804–814.

Miller, C.A., 2004. Climate science and the making of global political order. In:Jasanoff, S. (Ed.), States of Knowledge: The Co-Production of Science and SocialOrder. Routledge, London, pp. 46–66.

Miller, C.A., Edwards, P. (Eds.), 2001. Changing the Atmosphere: Expert Knowledgeand Environmental Governance. MIT Press, Cambridge.

Neumann, I.B., Sending, O.J., Pouliot, V. (Eds.), 2015. Diplomacy: The Making ofWorld Politics. Cambridge University Press, Cambridge, UK and New York, NY.

Neumann, I.B., Sending, O.J., 2010. Governing the Global Polity. The University ofMichigan Press, Ann Arbor, MI.

Nielsen, T.D., 2014. The role of discourses in governing forests to combat climatechange. Int. Environ. Agreements 14, 265–280.

Oberthür, S., Stokke, O.S. (Eds.), 2011. Managing Institutional Complexity: RegimeInterplay and Global Environmental Change. MIT Press, Cambridge, MA.

Oekreke, C., Dooley, K., 2010. Principles of justice in proposals and policyapproaches to avoided deforestation: towards a post-Copenhagen climateagreement. Glob. Environ. Change 20 (1), 82–95.

Pouliot, V., 2010. International Security in Practice: The Politics of NATO-RussiaDiplomacy. Cambridge University Press, Cambridge.

Reinecke, S., Pistorius, T., Pregernig, M., 2014. UNFCCC and the REDD+ partnershipfrom a networked governance perspective. Environ. Sci. Policy 35, 30–39.

Robertson, M., 2006. The nature that capital can see: science, state and market inthe commodification of ecosystem services. Environ. Plan. D: Soc. Space 24,367–387.

Sending, O.J., 2015. The Politics of Expertise: Competing for Authority in GlobalGovernance. University of Michigan Press, Ann Arbor, MI.

Shapiro-Garza, E., 2013. Contesting the market-based nature of Mexico’s nationalpayments for ecosystem services programs. Geoforum 46, 5–15.

Stephan, B., 2013. How to trade ‘not cutting down trees’. In: Methmann, C.P., Rothe,D., Stephan (Eds.), Deconstruction the Greenhouse: Interpretive Approaches toGlobal Climate Governance. Routledge, London, pp. 57–71.

Stephan, B., Rothe, D., Methmann, C., 2014. The third side of the coin: hegemonyand governmentality in global climate politics. In: Stripple, J., Bulkeley, H. (Eds.),Governing Climate: Rethinking Rationality, Power and Politics. CambridgeUniversity Press, Cambridge, UK, pp. 59–76.

Stripple, J., Bulkeley, H., 2014. Governing the Climate New Approaches toRationality, Power and Politics. Cambridge University Press, Cambridge, UK.

The World Bank, 2014. ‘‘Forest Area” The World Bank Data Online<http://data.worldbank.org/indicator/AG.LND.FRST.K2/countries/1W?display=default>.

Van Asselt, H., Zelli, F., 2013. Connect the dots: managing the fragmentation ofglobal climate governance. Environ. Econ. Policy Stud. 16, 137–155.

Voluntary REDD+ Database, 2013. ‘‘Overview” Voluntary REDD+ Database Website.<http://reddplusdatabase.org/#introduction>.

Wilson Rowe, E., 2013. A sylvan superpower: Russian forests in internationalclimate negotiations. Geoforum 48, 216–224.