Science, politics and the epistemic geographies of climate change in Copenhagen, 2009
This chapter is based on:
Mahony, M., 2013. Boundary spaces: science, politics and the epistemic geographies of climate change in Copenhagen, 2009. Geoforum, 49, pp.29-39.
Introduction
Despite widespread societal agreement on the need for political action to address climate change, so far the achievements of global climate governance have been limited to the rather modest ambition of the Kyoto Protocol. Signed in 1997 following negotiations under the United Nations Framework Convention on Climate Change (UNFCCC), the Protocol committed developed countries (with significant exceptions such as the US) to around a 5% cut in emissions of climate-warming greenhouse gases during the period 2008–2012, as compared to a 1990 baseline (Grubb et al., 1999). The 15th Conference of the Parties to the UNFCCC (COP15) meeting in December 2009 was a crucial moment in political efforts to negotiate a successor treaty to the Kyoto Protocol which would legally commit countries to further emissions reductions post-2012. During the months leading up to the December conference, the city of Copenhagen therefore became a microcosm of the global climate change debate, with a diverse array of actors fuelling a sense of urgency, expectation and hope; Copenhagen became ‘Hopenhagen’65. Part of this anticipation saw the city acting as a key site of science–policy interaction, as a number of scientific actors sought to bring together new and emerging knowledge about the state of the climate, the potential impacts of climate change and possible political and societal responses, with the aim of informing and shaping the political debate.
65
154
The issue of climate change poses unique challenges to the norms and practices of science and democratic politics. Complex mechanisms of physical causation, intractable uncertainties about future changes, the seeming inability of political institutions to deal with global risks; these factors, among others, have seen
conventions of expertise, representation and political authority called into question as societies have collectively or otherwise sought a ‘solution’ to the problem of anthropogenic climate change (Beck 2009; Hulme 2009a; Jasanoff 2010). The physical sciences have exercised a great deal of ‘definitional power’ (Beck 2009, 32) in the climate debate, with organisations such as the IPCC playing a central role in shaping discourse on causation, hazardousness, responsibility and potential solutions. The scientific construction of climate change as a global environmental problem rooted in the universal physical properties of the greenhouse gases has shaped the political space within which actors have responded in technocratic terms of global
environmental managerialism (Demeritt 2001; Miller 2004b; Oels 2005). Yet political contestations over climate change have often focused on scientific arguments, as various actors have sought to shed doubt on the scientifically-delineated need for strident political action (see Oreskes & Conway 2010), while others have called for the insulation of science from the polluting forces of politics (cf. Montford 2010; Mann 2012).
Such arguments reveal tensions inherent to the modernist settlement of science and politics as being wholly separate domains, with the former able to provide the latter with value-free knowledge on which political decisions can be based (Ezrahi 1990; Latour 1993). As I argued in Chapters 2 and 3, work in STS and cognate disciplines has problematised the notion that science operates as an autonomous ‘republic’ (Polanyi 1962), and has instead emphasised the co-production of knowledge and social order. The notion of co-production emphasises how our knowledge and representations of the world are inseparable from our choices about how to live as collectives of human and nonhuman actors (Jasanoff 2004a). Attempts to draw sharp distinctions between the worlds of science and politics therefore tend to mask the complex
interpenetration of epistemic claims and normative commitments (e.g. Demeritt 2001). However, such ‘boundary work’ (Gieryn 1983) can itself be seen as a mode of
155
social ordering, for example to delegate certain forms of authority to science or politics (Ezrahi 1990) in contexts – such as climate change – where complexity and indeterminacy preclude problems being comprehended or ‘solved’ by the activities of any one set of authorised actors (Turnpenny et al. 2009).
This chapter seeks to further explore the boundary dynamics of climate science and politics. The notion of ‘boundary spaces’ is developed to facilitate consideration of the epistemic geographies (the objects, actors, spaces and discourses) of science– politics interactions beyond the conventionally-delineated organisational spaces in which such interactions are subject to formal management. By drawing together literature from STS, geography of science and the geography of organisations, an account is given of the contested spaces of the science–politics relationship in the run-up to the ill-fated international climate change negotiations in Copenhagen. In the following section, the notion of boundary spaces is developed in relation to literatures on the spaces and boundaries of science, with particular reference to examples drawn from the climate change debate.
The geography of science–policy interactions
The lively field of ‘geography of science’ (see Chapter 2) has drawn attention to the significance of locality in scientific knowledge production and to the varied reception supposedly universal knowledge receives in diverse places. For Livingstone (2003, 123), “in the consumption of science, as in its production, a distinctive regionalism manifests itself.” Yet such arguments have a tendency to reify a distinction between spaces of knowledge production and consumption and may overlook the forces of co- production which problematise such distinctions. Along with a “spatially sensitive social constructivism” (Withers 2010a, 67), geographies of science also implicitly adopt a phenomenological spatiality which conceives places as a “distinctive coming together in space” (Agnew 2011, 317) of diverse socio-cultural trajectories (Massey 2005). Place is thus a unique assembly of phenomena and actors where actions unfold through the mobilising of “distant actants that are both present and absent” (Callon & Law 2004, 6); actants that are connected in material networks of
156
argued below, this spatial imaginary may help conceptualise ‘boundary spaces’, where the spaces of knowledge production and consumption intermingle in processes of social ordering.
Social forms at the science–policy interface
For the last 25 years the interaction between science and politics on the issue of climate change has been dominated by the IPCC. Charged with offering scientific knowledge to the nation-state signatories of the UNFCCC, the IPCC has exercised considerable epistemic and definitional power (Hulme & Mahony 2010; Tol 2011; Bjurström & Polk 2011). For many observers the periodic, authoritative consensus statements of the Panel have been instrumental in driving forward the global political process (Edwards & Schneider 1997; Tonn 2007) and public debate (Boykoff 2011). For others, the knowledge mobilised by the IPCC is inflected with localised problem-framings which raise questions about how trust in distant or international scientific practices is to be achieved in diverse political contexts (Biermann 2001, Lahsen 2004, Hulme 2010a; Jasanoff 2011a). The assumption that the IPCC represents disinterested, neutral scientific knowledge (Moss 1995) which can be used to legitimate political decisions has been critiqued by analysts wary of ‘linear model’ understandings of science–policy interactions (e.g. Sarewitz
2004; Carolan 2008). As discussed in Chapter 3, the linear model holds that
authoritative scientific knowledge must always precede effective decision-making, and that the latter is wholly dependent on the former (Beck 2011; Grundmann & Stehr 2012). The linear model thus also reinforces an understanding of science and politics as occupying wholly distinct cultural and physical spaces.
Work in STS has emphasised the diversity of organisations, discourses, and networks which nonetheless straddle the boundaries between science and politics, thus challenging the implicit spatiality of the linear model. In disputing earlier notions of science as a neutral, value-free exercise which can generate wholly impartial yet policy-relevant knowledge, and thus ‘speak truth to power’ (c.f. Wildavsky
1979; Jasanoff & Wynne 1998), studies of the operation of advisory panels (Jasanoff 1990), regulatory science (Jasanoff 1990; Irwin et al. 1997), ethno-epistemic
157
assemblages (Irwin & Michael 2003), and networks at the science–policy interface (Chilvers & Evans 2009) have contributed to understandings of these social processes and forms as instances of co-production. This proposition challenges the notion that sharp distinctions can be drawn between science and politics by drawing attention what Jasanoff calls “the social dimensions of cognitive commitments and
understandings”, without losing sight of “the epistemic and material correlates of social formations” (Jasanoff 2004b, 3).
Boundary spaces
The work of facilitating and managing flows of knowledge, resources, people and material things across the boundary between science and politics has often been bestowed upon what have become known to STS scholars as ‘boundary organisations’ (e.g. Guston 2001; Miller 2001a; Boezeman et al. 2013). The IPCC in many ways fits the description of such organisations, which “exist at the frontier of the two relatively different social worlds of politics and science, but … have distinct lines of
accountability to each” (Guston 2001, 401). Drawing on principal-agent theory, the concept of the boundary organisation highlights the work of authority delegation according to normative principles which may differ across the boundary in question. “The success of the organisation in performing these tasks can then be taken as the stability of the boundary, while in practice the boundary continues to be negotiated at the lowest level and the greatest nuance within the confines of the organisation” (ibid, 401).
The interest in stability as an achievement of ongoing work directs analytic attention towards internal organisational arrangements and practices (Boezeman et al. 2013). But such work arguably also resides within the interactional tradition of co-
productionist inquiry (Jasanoff 2004c). This tradition emphasises that “science and politics operate against a backdrop of an extant natural and cultural order, and highlights the conflicts between competing epistemologies. Under this perspective reliable, credible and authoritative science (and policy) depends on solving problems of social order” (Chilvers & Evans 2009, 358). Boundary organisations are sites where
158
the work of social ordering takes place in ongoing processes of negotiation, translation and accommodation.
This mode of work corresponds to the ‘boundary work’ described by sociologist Thomas Gieryn (1983). In a series of influential studies of the efforts made to define the boundaries of science and to ground the criteria of demarcation between legitimate and illegitimate claims to represent the ‘real’, Gieryn has emphasised the historical, cultural and spatial contingency of settlements of such boundaries (Gieryn 1995; 1999). This points to the never-ending tasks of boundary work in moments of political or epistemic conflict. Using the methodology of what he terms ‘cultural cartography’, Gieryn (1999, xii) suggests that “science is a cultural space: it has no essential or universal qualities. Rather, its characteristics are selectively and
inconsistently attributed as boundaries between ‘scientific’ space and other spaces [e.g. politics] are rhetorically constructed.”
Boundary work often coalesces around ‘boundary objects’ which function as bridges or anchors between different cultural spaces and which are “plastic enough to adapt to local needs and the constraints of the several parties employing them, yet robust enough to maintain a common identity across sites” (Star & Griesemer 1989, 393). A persistent consensus around climate sensitivity estimates (van der Sluijs et al. 1998) and the target of limiting the global mean temperature rise to 2°C to avoid
“dangerous” climate change (Shaw 2010; Randalls 2010) have functioned as objects of boundary negotiation in the climate debate, particularly – as will be shown below – in the boundary spaces enacted in Copenhagen in 2009.
The operations of boundary organisations and boundary objects thus contribute to the construction of science–politics boundaries, while reifying the very possibility of their existence. However, Gieryn’s work encourages us to recognise that boundary work is not confined to formalised institutions charged with the management of science–politics boundaries. Rather, the sites and spaces of boundary work are diverse, often spontaneous, and frequently unexpected. The concept of boundary organisations largely arose in studies of scientific advisory processes in the United States (e.g. Guston 1999; 2000). The concept has now gained a particular popularity
159
among students of comparable processes in the Netherlands (e.g. Pesch et al. 2012; Boezeman et al. 2013). As Miller (2001a) points out, this tying of the concept to nation-state contexts may limit its applicability to transnational or intergovernmental spaces. The “theory has not fully escaped conventional patterns of thought that circumscribe the institutional landscape inhabited by these institutions” (ibid, 484) to what Guston (2000, vx) describes as a “fine, bright line”. The concept thus arguably recapitulates elements of US political culture which seek to identify clear dividing lines between pure science and pure politics, as suggested by Jasanoff (2005a) and as I discuss in Chapter 6. This “overly static view of science and politics” (Miller 2001a, 484) elides differences between institutions in the respective cultural domains – differences which may “stand out more distinctly in international settings…where the scientific and political institutions of myriad countries are brought into immediate contact with one another (ibid, 483). As I argued in the previous chapter, the IPCC is a space where competing understandings of the boundary between scientific and political reasoning have been brought to bear on the production and circulation of scientific assessments. Such diversity in the ‘lines of accountability’ (Guston 2001) enacted between the assessment process and multiple scientific and political
communities is not adequately captured by the concept of the boundary organisation, which assumes largely homogenous cultures of science and politics.
Furthermore, in the context of climate change, the diversity of networks and
assemblages of political and scientific actors engaged with the issue means that the ongoing processes of boundary work are not restricted to the boundaries of
organisations like the IPCC. The profusion of various ‘alternative’ scientific
assessments of climate change (e.g. Biermann 2001, 302), heated public debates in new social media platforms (Koteyko et al. 2012) and the more fundamental
disconnect between a global climate science and locally-embedded forms of meaning and action (Hulme 2010b; Jasanoff 2010) suggest that the boundaries between climate science and politics are contested in a diversity of spaces.
Recent work on the geographies of organisations has sought to destabilise
conceptions of organisations as neatly-bounded, homogenous entities which should be studied in terms of their procedural outputs (Beyes & Steyaert 2011). A turn
160
towards practice has emphasised networks, embodiment, materiality and affect as being constitutive of ‘organisational space’ (e.g. Conradson 2003; Dale 2005). Moving beyond conceptions of space as a passive container of organisational activity,
organisational spaces are associated with and constituted by particular associations of actors and objects coalescing around certain goals, imaginaries (Taylor 2002) and practices (Conradson 2003). Drawing variously on Lefèbvre (1991), actor-network theory (ANT) and non-representational theories, work on the production and generative potential of organisational space (e.g. van Loon 2000; Beyes & Steyaert 2011) draws attention back to the often banal and habitual processes of ordering, as opposed to order-as-product. The concept of the boundary organisation to a large extent shares this concern with the contingency of practice and process. Yet it potentially deflects attention away from the multiplicity of spaces and processes in which the organisation of the science–politics boundary is accomplished (Chilvers & Evans 2009; Irwin & Michael 2003).
We might then emphasise the importance of boundary spaces – the spaces and spacings (Derrida 1981; Beyes & Steyaert, 2011) in and through which the work of organising and negotiating the boundary between science and politics is conducted. This focus has the potential to transcend the latent state-centric functionalism of existing literature on science–policy boundaries (Miller 2001a). It might also respond to the diversity of empirical settings and networks through which such boundaries are contested in the context of a complex issue like climate change (Hulme
2009a; Chilvers & Evans 2009). Following the non-representational critique of the socio-material rigidity of ANT (Thrift 2007), the concept of boundary spaces directs us towards the epistemic and political geographies of boundaries in their emergence and contestation. It also places emphasis on the embodied forces of event and conjuncture over the progressive institutionalisation of stability (Bingham & Thrift 2000). In problematising conventional organisational boundaries, the notion of boundary spaces permits us to consider the interpenetration of different organisational spaces in particular space–time configurations. For example, the dominant position of the IPCC at the science–policy interface has effected a complex geography of connected boundary spaces as norms, procedures, bodies and objects
161
associated with the IPCC have circulated widely through the worlds of scientific assessment and policy advice (e.g. Hulme & Mahony 2010; Perrings et al. 2011).
Boundary spaces should be conceived of as spaces where the co-production of scientific knowledge and social order occurs. Boundaries, as a form of social order, are co-produced with the very knowledge they are mandated to contain and signify (Jasanoff 2004c). In considering boundary spaces, we therefore encounter one example of how geographies of science and ideas about co-production may be brought into fruitful conversation. In a generative and performative sense, boundary spaces are co-produced along with scientific knowledge, political commitments, and normative allocations of authority on the cultural map of late modernity (Gieryn 1999; Chilvers & Evans 2009). This co-production of space resonates with a growing interest among geographers of science in going beyond a simple localism in accounts of scientific practice towards a fuller treatment of the mutual constitution of the epistemic and social spaces of science (cf. Shapin 1998; Powell 2007a; Livingstone 2010).
Science for Copenhagen: two cases
Throughout 2009 the word “Copenhagen” took on a number of new connotative, one might even argue denotative, functions (Barthes 1977). Phrasings such as “the road to Copenhagen” and “countdown to Copenhagen” – common in media coverage of COP15 – elide space and time in anticipation of a particular event66. “Copenhagen” became synonymous with the COP15 meeting; the name of a city became the name of a gathering of political actors at a particular time and place. A new, transient sense of the city’s political salience thus took shape through these constructions of a
particular time-space; one of scientific deliberation, political wrangling, and popular protest. Copenhagen became a key site for the political deliberation of climate change, and the events of December 2009 will likely continue to be seen as a critical discourse moment (Carvalho & Burgess 2005) which led to the transformation of discursive and political positions among political and scientific actors (Bailey 2010).
66
See http://www.roadtocopenhagen.org/ and
162
These transformations included a newly prominent scepticism about the efficacy of top-down, multi-lateral climate policy initiatives (e.g. Prins et al. 2010) and, as will be argued below, changes in how scientific actors perceive and respond to political processes.
The following analysis investigates the epistemic geographies of two boundary spaces which were enacted in the run-up to COP15. Two groupings of scientists and political actors sought to bring together new and emerging scientific knowledge in order to inform the anticipated political debates about climate change mitigation and adaptation. In the months and weeks leading up to the COP15 meeting, two particularly high-profile documents were produced67. The Synthesis
Report (document 72) arising from a scientific conference entitled “Climate Change:
Global Risks, Challenges and Decisions” (also known as the ‘Climate Congress’) held at the Bella Center in Copenhagen from 10th to 12th March 2009 presented key findings and ‘messages’ from an interdisciplinary collection of 58 conference sessions, which were later presented to the Danish Prime Minister.
The Copenhagen Diagnosis (document 114) was a 64-page document produced by 26 prominent climate scientists to communicate the latest policy-relevant findings to decision-makers at COP15. The document, published in November 2009, covers observations of atmospheric greenhouse gas concentrations, extreme events, changes in land use, the cryosphere and oceans, the prospect of “tipping points” in the earth system, and the most up-to-date projections of future changes and emissions trajectories. The central theme of media coverage of the report’s publication was that across these variables, the effects of climate change are occurring faster than estimated in the 2007 IPCC report68.
67 These cases were selected due to their high-profile media coverage (see e.g. Boykoff 2011, 20-28 on