Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Power, science and nature in the Great Bear Rainforest : an actor-network analysis of an integrated natural… Page, Justin Lawrence Roy 2010

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Notice for Google Chrome users:
If you are having trouble viewing or searching the PDF with Google Chrome, please download it here instead.

Item Metadata


24-ubc_2010_fall_page_justin.pdf [ 24.38MB ]
JSON: 24-1.0071350.json
JSON-LD: 24-1.0071350-ld.json
RDF/XML (Pretty): 24-1.0071350-rdf.xml
RDF/JSON: 24-1.0071350-rdf.json
Turtle: 24-1.0071350-turtle.txt
N-Triples: 24-1.0071350-rdf-ntriples.txt
Original Record: 24-1.0071350-source.json
Full Text

Full Text

       POWER, SCIENCE, AND NATURE IN THE GREAT BEAR RAINFOREST: AN ACTOR-NETWORK ANALYSIS OF AN INTEGRATED NATURAL RESOURCE MANAGEMENT PROJECT   by   JUSTIN LAWRENCE ROY PAGE   MES, York University, 2000 B.A., University of British Columbia, 1995   A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF  DOCTOR OF PHILOSOPHY  in  THE FACULTY OF GRADUATE STUDIES  (Sociology)       THE UNIVERSITY OF BRITISH COLUMBIA  (Vancouver)  September 2010  © Justin Lawrence Roy Page, 2010  ii Abstract  This dissertation explores the potential contribution of actor-network theory to the investigation of power and hierarchy, science and politics, and the relationship between nature and society in integrated natural resource management (INRM) projects. INRM consists of natural resource management approaches that seek to devolve power and authority from governments and experts to stakeholders, take account of people as part of ecosystems, and directly link conservation and development.  While INRM projects represent an important evolution in resource management, they come with particular sets of problems. Specifically, (1) the devolution of decision-making authority to communities provokes issues of power and hierarchy as groups vie to ensure that their interests are adequately taken into account, (2) critiques of expert-led processes shift responsibility for knowledge production to stakeholder groups, thus raising questions about the relationship between science and politics, and (3) attempts to link ecology and economy require a difficult re-conceptualization of the link between nature and society. Actor-network theory (ANT) avoids presuppositions about power, science, nature, and society in order to study how they are produced as effects of networks, thus offering unique conceptual tools to study INRM as a complex, contingent, and innovative network-building process.  A qualitative case study of the “Great Bear Rainforest” agreement on British Columbia’s west coast is undertaken to explore these issues in INRM.  Analysis of interviews with 34 individuals from environmental organizations, forestry companies, First Nations, consultancies and local and provincial governments, as well as analysis of textual material, reveals how environmentalists (1) generated power by building a network of activists, bears, forest products customers and forestry companies, (2) simultaneously deployed science and politics in their network-building activities and (3) moved away from attempts to purify networks into “nature” and “society,” working instead to directly link ecosystem integrity and human well-being in a new, common “collective” of humans and nonhumans.  The research provides significant detail and analysis of a particular case of INRM that will be of use to INRM practitioners, advocates and activists.  Additionally, the research demonstrates the applicability of ANT to the investigation of power, science, and nature in INRM projects.  iii Table of Contents   Abstract ............................................................................................................................... ii	
   Table of Contents............................................................................................................... iii	
   List of Figures ..................................................................................................................... v	
   Acronyms........................................................................................................................... vi	
   Acknowledgements........................................................................................................... vii	
   Dedication ........................................................................................................................ viii	
   1  Introduction..................................................................................................................... 1	
   1.1  Research Objectives................................................................................................. 6	
   1.2  Overview of the Study ........................................................................................... 15	
   2  Integrated Natural Resource Management and Actor-Network Theory....................... 20	
   2.1  Introduction............................................................................................................ 20	
   2.2  From “Fortress Conservation” to “Integrated Natural Resource Management” ... 21	
   2.3  Integrated Natural Resource Management in BC .................................................. 27	
   2.4  Key Controversies in Integrated Resource Management: Defining Research Objectives ..................................................................................................................... 29	
   2.5  Researching Integrated Natural Resource Management........................................ 35	
   2.6  INRM as Network-Building: The Promise of Actor-Network Theory.................. 42	
   2.6.1  ANT, INRM and Power.................................................................................. 48	
   2.6.2  ANT, INRM and the Relationship Between Science and Politics.................. 57	
   2.6.3  ANT, INRM and the Relationship Between Humans and Nonhumans.......... 63	
   2.7  Summary ................................................................................................................ 71	
   2.8  Outline of the Analysis .......................................................................................... 73	
   3  A Method and not a Theory .......................................................................................... 77	
   3.1  Methodological Strategies ..................................................................................... 77	
   3.2  Analytic Procedures ............................................................................................... 87	
   3.3  Validity .................................................................................................................. 91	
   4  Problematizing the Coast: Shifting the Terrain of BC’s Wilderness Conservation Movement ......................................................................................................................... 94	
   4.1  Finding British Columbia’s “Forgotten Coast” ..................................................... 94	
   4.2  Constructing the Object of Environmental Politics ............................................. 100	
   4.2  Translation 1: The Coastal Temperate Rainforest ............................................... 103	
   4.3  Translation 2: Unprotected Watersheds............................................................... 111	
   4.4  Translation 3: Stories and Images........................................................................ 121	
   4.5  Assembling a Panorama of the Coastal Forests................................................... 130	
   5  Generating Power: Spokespeople, Interessment and the Market Campaign .............. 142	
   5.1  The Canadian Rainforest Network....................................................................... 146	
   5.2  Conservation Biology and Spokespersons for the Rainforest.............................. 153	
   5.3  Mobilizing Spokespersons for the Rainforest...................................................... 165	
   5.4  The Impossible Locus of Local Blockades.......................................................... 172	
   5.5  The Market Campaign ......................................................................................... 181	
   5.6  The Spectre of Consumer Boycott: Transforming Companies Into  Activists .... 189	
   5.7  Conclusion ........................................................................................................... 202	
    iv 6  Enrolling Engos, Forestry Companies and First Nations in a Common Matter of Concern ........................................................................................................................... 209	
   6.1  Negotiating With the Enemy ............................................................................... 212	
   6.2  Developing a Common Matter of Concern.......................................................... 223	
   6.3  Enrolling Others in the Matter of Concern .......................................................... 230	
   6.4  Conclusion ........................................................................................................... 245	
   7  Mobilizing Allies and Reconciling Interests............................................................... 247	
   7.1  Reconciling Conservation, Development and Justice.......................................... 252	
   7.2  CIII: Rendering Economy Commensurable With Ecology ................................. 267	
   7. 3  Turning the Matter of Concern Into a Matter of Fact ......................................... 280	
   7.4  Conclusion ........................................................................................................... 293	
   8  Conclusion .................................................................................................................. 298	
   Bibliography ................................................................................................................... 312	
   Academic Works Cited ............................................................................................... 312	
   Research Materials...................................................................................................... 328	
   Appendix: Research Materials........................................................................................ 334	
   A.1  Letter of Introduction .......................................................................................... 334	
   A.2  Example Follow Up Letter.................................................................................. 336	
   A.3  Consent Form...................................................................................................... 337	
   A.4  Interview Guide................................................................................................... 340	
   A.5  Example of Tailored Interview Guide – Turning Points (TP) ............................ 349	
   A.6  UBC Research Ethics Board Certificate of Approval......................................... 352	
     v List of Figures   Figure 1: Coast Land Use Zones ........................................................................................ 3	
   Figure 2: Original Global Distribution of Coastal Temperate Rain Forests .................. 107	
   Figure 3: Coastal Watersheds, South Coast ................................................................... 116	
   Figure 4: Coastal Watersheds, Mid and North Coasts ................................................... 117	
   Figure 5: McAllister's "Great Bear" ............................................................................... 124	
   Figure 6: Canada's Rainforest – Worth Saving .............................................................. 132	
   Figure 7: Blockade at Roderick Island, May 1997......................................................... 174	
   Figure 8: Chain of Custody ............................................................................................ 184	
   Figure 9: New York Times Ad, Dec 1998, CRC ........................................................... 194	
   Figure 10: The Markets Campaign: Sample Letter to Business..................................... 198	
   Figure 11: The Markets Campaign: Sample Letter From Business ............................... 199	
   Figure 12: EBM "Levels" ............................................................................................... 264	
   Figure 13: Risk at Different EBM Levels ...................................................................... 266	
   Figure 14: Conservation Financing Equation................................................................. 276	
     vi  Acronyms   Acronym Full Name  AAC Annual Allowable Cut ANT Actor-Network Theory BC British Columbia CAD Conservation Area Design CBNRM Community-Based Natural Resource Management CCLRMP Central Coast Land Resource Management Plan CFCI Coast Forest Conservation Initiative CIII Conservation Investments and Incentives Initiative CIT Coast Information Team CLUA Coast Land Use Agreement CRC Clayoquot (later Coast) Rainforest Coalition CRN Canadian Rainforest Network EBM Ecosystem-Based Management EBMH Ecosystem-Based Management Handbook EBM WG Ecosystem-Based Management Working Group EI Ecological Integrity ENGO Environmental Non-Governmental Organization FAN Forest Action Network G2G Government-to-Government GBR Great Bear Rainforest HWB Human Well-Being ICDP Integrated Conservation and Development Projects INRM Integrated Natural Resource Management Interfor International Forest Products IWA Industrial Wood and Allied Workers of Canada JSP Joint Solutions Project Kit-Git-Pit Kitasoo-Gitga’at Protocol Implementation Team LOI Letter of Intent LRF Land and Resource Forum LRMP Land and Resource Management Plan LUP Land Use Plan MOF  Ministry of Forests OPP Obligatory Point of Passage PIMC Plan Implementation Committee RAN Rainforest Action Network RONV Range of Natural Variation RSP Rainforest Solutions Project WCWC Western Canada Wilderness Committee WFP Western Forest Products   vii Acknowledgements   I first began to develop a proposal for this research project in the spring of 2006.  From that time forward, I received valuable feedback and suggestions from my dissertation supervisor, Dr. Ralph Matthews, and committee members Dr. Thomas Kemple and Dr. Teresa Satterfield.  I benefited from presenting early drafts and ideas at conferences, including the Annual Meeting of the Canadian Sociological Association in 2008 and the Annual Meeting of the American Sociological Association in 2009.  In the Fall of 2009, I had the opportunity to discuss Bruno Latour’s ideas in Dr. Stephen Petrina’s seminar on the influential anthropologist and sociologist.  Discussions with seminar participants helped me sharpen my understanding of Latour’s work.  Maryam Nabavi provided amazing encouragement, sympathetic audience, and constructive feedback, for which I am most grateful. The best guidance that I received in developing this dissertation was from the actors themselves.  I would like to acknowledge and thank all thirty-four individuals who agreed to give their time, materials, knowledge, thoughts and passion to help me understand how the Great Bear Rainforest agreement was put together, as well as others I talked to more informally. Others who contributed in some way to this work include Rebecca Page and Dr. Renisa Mawani.  Finally, I would like to acknowledge the support of the Social Sciences and Humanities Research Council for their financial support and thank all copyright holders for permission to reproduce their materials in my dissertation.   viii Dedication     For my parents, Robyn and Larry Page, whose love and support provided this dissertation’s condition of possibility and for Maryam Nabavi, my ongoing inspiration   1 1  Introduction   It is February 7th, 2006.  You sit down in front of the television or spread out a newspaper, maybe you’re listening to the radio in your car. You might be in Vancouver, or Montreal, or New York.  Maybe you’re in England, or Austria, perhaps Japan.  You learn that an agreement has been reached to protect the world’s “largest remaining temperate coastal rainforest.”  Maybe you haven’t heard of this place.  You’re told that it is called the “Great Bear Rainforest” by environmentalists and that it stretches along British Columbia, Canada’s central and north coasts.  The size of new protected areas is translated into local terms: depending on where you are, they are almost the size of New Jersey, twice the size of Yellowstone, or three times the size of Prince Edward Island. The entire area is twice the size of Belgium, they say.  The agreement involves an “unprecedented collaboration between First Nations, industry, environmentalists, local governments, and many other stakeholders” (Government of British Columbia, 2006).  It protects, reporters note, the habitat of species like grizzly bears, wolves, salmon and the “elusive spirit bear.”  It is “British Columbia’s gift to the planet,” says one environmentalist (Smith, 2006).  “Spirit bear,” you think, “that sounds pretty interesting.”  A wilderness of close to five million acres, almost the size of New Jersey, in what is commonly called the Great Bear Rain Forest or the Amazon of the North will be kept off limits to loggers in an agreement that the disparate parties describe as a crossroads in their relations. – New York Times (Krauss, 2006a)  2  Ending a decade-long environmental battle once dubbed the "War of the Woods," British Columbia is set to announce Tuesday the creation of a park twice the size of Yellowstone along a vast coastal swath where grizzly bears and wolves now prowl under thousand- year-old cedar trees. – Washington Post (Struck, 2006)  An improbable assemblage of officials from the provincial government, coastal Native Canadian nations, logging companies and environmental groups will announce an agreement to preserve the home of the Spirit bear, which is also the largest remaining temperate coastal rain forest. – Herald Tribune (Krauss, 2006b)  Reading on, you learn that environmentalists, First Nations, logging companies, and the Government of British Columbia (BC) have agreed to protect 1.8 million hectares while putting in place strict, lighter touch forestry practices for 4.6 million hectares, bringing the total area of the agreement to 6.4 million hectares.  This is a little difficult to visualize, so they provide a map (see Figure 1).  Reporters note that the 100 protected valleys represent about one-third of the area, while around two-thirds remains open to logging.  However, logging and other activities on this land will be guided by something called ecosystem-based management (EBM).  3  Figure 1: Coast Land Use Zones Source: Integrated Land Management Bureau (2010)   4 EBM, they say, is a way of managing land use that minimizes impact on the environment, thereby ensuring that the use of resources remains sustainable.  Someone is quoted as saying that EBM guides forestry planning on the basis of what you should leave, not what you should take.  That is, it shifts the focus from logging parameters such as fibre volume to ecosystem characteristics such as bear dens, tree snags and buffers around streams.  The new parks, in addition to 600,000 hectares already in parkland, will create a network of protected areas encompassing 1.8 million hectares -- an area three times the size of Prince Edward Island -- in 100 pristine river valleys. – The Vancouver Sun (Hamilton, 2006)  Logging will be allowed in many areas in the Great Bear Rainforest, but it will take place under "ecosystem-based management."  The new logging approach is supposed to protect the environment while permitting up to 50 per cent of the timber to be removed from some areas. – Globe and Mail (Hume, 2006)  [Ecosystem-based management] integrates ecological, economic and social purposes and is designed to work as a management and planning regime that first looks at what is needed to be left in place to allow for a healthy ecosystem and then looks at what can be taken out. – Rainforest Solutions Project (Rainforest Solutions Project, ND)  Descriptors such as historic, unprecedented, and landmark are being used throughout the news stories in reference to the agreement so as to suggest that this  5 agreement represents an entirely new way of resolving resource conflict.  The “war in the woods” is over, say the newspapers, TVs, and radios.  The agreement appears to reconcile the interests of environmentalists in protection, forestry companies in continued operations, and First Nations in decision-making and economic development.  “This agreement brings an end to the long-standing resource-use conflicts over this land,” says Dallas Smith, a First Nations leader (Government of British Columbia, 2006). “This is a revolution in the way that we approach forestry in British Columbia,” says Merran Smith, an environmental representative.  “It is a revolution where communities are leaders in their own destinies. Where logging practices have a lighter touch on the land and conservation comes first” (Smith, 2006). Along these lines, you hear something about the development of a new economy, a conservation economy. Apparently, environmentalists have raised $60-million, supplemented by the Provincial and (probably) Federal governments to a total of $120- million, to support the development of a regional economy that creates employment while protecting wilderness.  First Nations seem to be keen on this idea. Art Sterritt, the leader of a group called Coastal First Nations talks about a diversified economy, about First Nations’ access to forestry, but also the development of new opportunities like ecotourism, fishing lodges, and shellfish. He says that, “For First Nations [the agreement] is a new beginning.  It means that we’re going to be able to develop an economy that’s sustainable, and that’s really what’s important about this” (as cited in Forsythe, 2006).     6 1.1 Research Objectives    Reading or hearing all of this while sitting in front of your newspaper or TV, you may have a number of questions. Where did this “historic” and “unprecedented” agreement come from?  What are “ecosystem-based management” and the “conservation economy?”  Why are non-governmental groups making decisions about public resources? Since when did environmentalists, forestry companies, and First Nations “collaborate”? Who really made these decisions?  Was the decision based on science?  Is the area a “pristine wilderness” or a working landscape?  Is it protected or is it going to be developed? These are legitimate questions, ones that many people who came across the Great Bear Rainforest (GBR) announcement may have had. Their sheer variety reflects the complexity and diversity of the case.  The Great Bear Rainforest agreement draws together a wide range of actors, elements and practices, including environmentalists, forestry companies, First Nations, the Provincial Government, bears, trees, coastal temperate rainforests, ecosystems, forestry practices, and economies.  These elements feed into a complex and nuanced decision that avoids top-down, expert-led decisions to either protect land or to develop it.  The usual categories of understanding do not seem to apply.  The land is neither transformed into a giant park, nor into a working forest.  Some valleys are protected, but many more are open to development; moreover, people are considered to be important parts of the coastal ecosystems, rather than in necessary conflict with the coastal forests. The conservation economy seeks to avoid a trade-off between ecology and economy: conservation will provide jobs and economic  7 development will conserve the forests.  While the Provincial Government convened the press announcement, it is not the primary decision-maker: rather, multiple stakeholders have played a part in producing the agreement.  It may seem difficult to understand this case, given its complexity and novelty. However, despite the fact that the agreement is indeed “unprecedented” for British Columbia, many of its central features are informed by developments within the conservation field over the past 30-40 years.  Many conservation scholars and practitioners argued that conventional conservation projects – those that try to protect areas of land by excluding human use – are too top-down, neglect the rights of local people who depend on natural resources for their livelihoods, and result in species loss through ecological fragmentation (Hulme, 1999).  In response, advocates called for community-based forms of decision-making, resource management that takes into account entire landscapes and the human populations that reside in them, and the linkage of conservation and development in positive rather than zero-sum terms (Hulme, 1999). These “win-win-win” proposals seeking to protect ecosystems, foster economic development, and promote justice and inclusion of local communities, were given a big push with the 1980s focus on sustainable development.  Collectively, these proposals can be referred to as “integrated natural resource management” (INRM), which may be defined as an approach to resource management that seeks to devolve decision-making to stakeholders, broaden management to human-nonhuman landscapes, and reconcile conservation and development.   8 There was the ‘80s, all the campaigns on Vancouver Island to save each last valley that was left, to try to save them and so much effort put into it.  And then people sort of woke up and realized, “Okay, well here we have this region to the North, where there are 100s of valleys,” you know, 100 large ones and hundreds of small ones, so, “My gosh, what are we gonna do to protect it?” [E – NT: 078]  It took a while for people to hash it out from, “Okay, we want no logging” – which everybody might want in their hearts – but there was enough people to say, “well, we’re not going to achieve that, and is that really what we want?”  [E – NT: 102]  It took me a long time to realize that our conceptions of wilderness, you know the basic conception of wilderness, that it’s a place untouched by humans, is really wrong. (Berman as cited in The National, June 16, 1999)  The reality is we’ve been out and visited those First Nation communities and they’re fucking in trouble, like, it’s not good out there.  These people, something has to shift. [E - KI1: 30]  We came up with this, I think, at the time it felt like what we’re coming up with was a big vision.  You know, we’re going to protect the majority of all the intact valleys and we’re going to transform the logging practices, and we’re going to create a new kind of economy.  [E - NT: 102]   9  INRM represents an important evolution in resource management.  However, the approach comes with particular sets of problems.    First, devolution of decision-making authority to communities is not a simple or straightforward process.  Communities are not homogenous or bounded, but are internally fractured and connected in multiple ways to actors and entities outside their boundaries (Agrawal & Gibson, 1999).  Devolution of decision-making to stakeholders therefore takes place in a context of power and hierarchy as groups vie to ensure that their interests are adequately taken into account.  Secondly, challenges to expert-led decision-making involved in INRM not only bring more groups – both lay and expert – into contact with one another, but also dispute an assumed separation between science and politics.  Epistemic questions – questions about research priorities, data collection methods, and the value and purpose of knowledge – are political questions that must be addressed directly rather than merely as social impacts of science.  Lastly, attempts to link ecology and economy involve not only practical but also conceptual issues.  If under conventional conservation, ecology is defined as nonhuman and economy as destructive to nature, how can these “spheres” be redefined to directly link ecology and economy in positive rather than zero-sum terms? These difficulties have been explored by several research paradigms, including common property theory (e.g. Ostrom, 1990), social learning (e.g. Keen, 2006), political ecology (e.g. Brown, 1998), social constructionism (e.g. Steins, 1999), and resilience theories (e.g. Walker et al., 2002).  Existing research tells us much about the role of local institutions in resource management, the importance of social capital, the role of common frameworks of understanding, the political and economic causes of unequal outcomes and the importance of resilience in coupled social-ecological systems.  However, existing  10 research says little about the role of INRM as a complex, contingent and innovative process.  That is, existing research paradigms tend to explain INRM outcomes with reference to particular principles, forces and structures – whether they are local institutions, global structures, cognitive frameworks, representational practices, or resilience – which are assumed a priori, rather than exploring INRM as a process that produces new principles, forces and structures.  Researchers propose lists of conditions that need to be met if INRM projects are to be successful, draw on assumed structures of power to explain outcomes, and assume given groups with given interests.  This approach is certainly highly valuable; however, it does not lend itself to an examination of process and innovation.  The latter approach would inquire into how conditions, structures, interests and the like are produced in INRM projects. This is not to say that the latter approach is superior to or should replace the former approach, but that it can provide a useful complement, especially in the case of INRM which is still so new and innovative with respect to the very categories that are relied upon to provide explanations. INRM projects provoke a reconsideration of accepted ways of understanding nature, society and their relationship.  When new groups and coalitions form and interests shift and change, how can predefined groups and interests provide an explanation?  When new economies are invented, how can economic structures provide an explanation? When the actors propose new principles, how can predefined principles be used to evaluate outcomes?  When the actors produce new mixtures of humans and nonhumans, how can we explain the outcomes as consequences of social construction?  When the actors design new political institutions, how can there be offered only explanations relying on assumed political structures?  When the actors are busy redefining nature and  11 society in order to render them compatible, how can an over-arching theory of coupled social-ecological systems explain everything there is to be known? As Latour (2005b) argues, when institutions, groups, interests and the like are established and stable, it makes sense to treat them as given and draw on the valuable concepts and theories that have been developed to explain them.  However, “in situations where innovations proliferate, where group boundaries are uncertain, when the range of entities to be taken into account fluctuates [conventional forms of explanation are] no longer able to trace actors’ new associations” (Latour, p. 11).  The question that conventional perspectives are not well equipped to answer is how INRM projects emerge and evolve, and how they generate new groups, interests, and institutions?  How are a wide variety of public and private decision-makers, stakeholders, agencies, and nonhuman species drawn together in a common project, and how are their interests aligned?  How do they generate power to create change?  How do they create new knowledges?  How do they design new economies?  How do they work to reconcile ecology and economy?  How can we take into account the wide variety of elements that go into an INRM project while withholding presuppositions about factors that influence outcomes?  Think of Roscoe Inlet [an inlet on the central coast] as the hub of a wheel, with spokes that reach all the way to the premier’s office, the corporate headquarters of logging multinationals, the foreign offices of some of B.C. wood exporters’ largest customers, the influential command centres of the most affluent philanthropic foundations in the U.S., environmental groups and First Nation chiefs and their councils. It has brought long-  12 standing combatants to the table to discuss, on more or less equal terms, the future of what is now widely recognized as one of the last great terrestrial conservation opportunities on the globe. (Findlay, 2007)  This study does not attempt to promote INRM as a superior approach to resource management, nor provide guidelines, principles or conditions that will lead to successful INRM.  Rather, the purpose of this study is to engage in an in-depth exploration of the Great Bear Rainforest case in order to learn how this particular example of INRM emerged, developed, and evolved, and to elucidate how the processes involved in the production of new groups, new interests, new knowledges, new representations, new institutions, and new relationships with nonhumans arose. How was this particular place designated as a conservation opportunity?  How did various groups come together to make decisions about it?  How were these processes contested?  How did actors go about trying to reconcile the interests of multiple human and nonhuman groups?  My central focus is on the key processes of environmentalists, examining how they sought to redefine the central and north coasts as a scientific and political reality, how they went about convincing others to adopt their emerging vision of reality, and how their vision shifted and changed as a result.  Specifically, my study has three primary objectives:  • To examine how environmentalists gained the power to influence resource decision-making for coastal BC. • To examine how environmentalists articulated science with politics in their attempts to influence the management of the GBR.  13 • To examine how environmentalists specified the conceptual link between nature and society as they sought to reconcile ecology and economy.  In this study, I draw on conceptual tools developed by Latour and his colleagues, which are collectively referred to as actor-network theory (ANT) (Callon, 1980, 1986; Callon & Latour, 1981; Callon & Law, 1982; Law, 1986; Latour, 1987, 1988).  I adopt this approach due to its ability to study innovative processes without making any presuppositions about what it may find.  As I explain more fully in Chapter 2, ANT researchers avoid given categories of explanation – such as micro and macro, local and global, subject and object, agency and structure, reality and representation, and nature and society – to instead learn from the actors themselves how they fabricate their worlds. The primary focus of ANT scholars is on finding, exploring and examining how associations are established between entities and how the emerging network acts as a collective entity to produce new worlds.  This approach promises to provide unique insights into power, science and nature. First, a network approach explores power as an effect of networks, rather than an attribute of groups held on the basis of position in social structures (Law, 1992).  That is, power is not seen to be determined in the last instance by economic structures (such as ownership of the means of production), ideological structures (such as collective forms of belief), cultural structures (such as distinction and taste), normative structures (such as functional norms and roles), or epistemic structures (such as the linkage between discourse, knowledge and power).  Moreover, power is not considered to be something that is held by particular groups and wielded on the powerless.  Rather, power is  14 conceptualized as the ability to achieve an outcome that can be realized only through the support of a network. In this understanding, power can be generated and taken away via constructing and intervening in networks.  This view of power is similar to common understandings: some interests may be realized while others may not be realized, some groups may be included and others may be excluded, some included groups may be coerced into participating in projects that meet the needs of others.  However, there are key differences: interests are re-defined through network formation, no one is in ultimate control of network Second, ANT scholars (who originated in the field of science and technology studies) explore the production of knowledge about the natural world as a political process that involves strategic decisions, enlistment of support, and rhetoric; symmetrically, they examine the political process of determining the structure of the social as performed through many nonhuman objects and devices (Latour, 1998, 1991). For ANT researchers, science and politics do not pertain to two separate domains – nature and society – but to a single world constructed by networks (Latour, 2005b). Following from this, and third of all, ANT provides tools to examine how humans and nonhumans are associated with one another without drawing on concepts of nature and society as explanatory resources.  ANT scholars examine nature and society – if they exist at all – as outcomes of networks rather than their source.  That is, they study how the actors go about conceptually purifying the networks that they construct, ignoring their heterogeneity and placing them into discrete categories of society and nature.  The exciting implication of this approach for INRM – which is exceptionally innovative when it comes to re-specifying the link between nature and society – is that ANT can explore  15 how such projects reshuffle humans and nonhumans in ways that do not correspond to the concepts of nature and society.  1.2 Overview of the Study    By means of an examination of the processes through which issues of power, science and politics, and ecology and economy are dealt with in the GBR case, this study contributes to our understanding of key issues in integrated natural resource management. As noted above, the purpose of this study is not to derive sets of principles or guidelines that can be applied in the design of new INRM projects.  Rather, the goal is to explore the emergence of an INRM project and its innovative processes while focusing on issues of power, science and nature.  In particular, I will demonstrate that issues of power can be understood as processes of network formation and intervention that take place outside of official stakeholder processes.  I will show that the production of knowledge for resource management is a simultaneously scientific and political process in which all actors that have a stake in what knowledge is produced can work together to articulate a form of “civic science.”  Finally, I will show that the question of linking ecology and economy involves fundamentally reworking both categories in order to render them commensurable with one another. I explore these issues by examining how the GBR agreement emerged and evolved as a process of network construction.  The research reported in the following pages involves an in-depth, qualitative exploration of a case of INRM, based on 34  16 interviews with key participants in the GBR process, purposefully selected from environmental nongovernmental organizations (ENGOs), forestry companies, First Nations organizations, and mediators.  It is also based on analysis of textual and audiovisual material associated with the case – including agreements (Land and Resource Management Plans, government-First Nations, ENGOs-industry, etc.), terms of reference, work plans, legal orders, presentations, reports (technical, organizational, workshops), scientific articles, ecological data, newsletters, press releases, news stories, public and customer information materials, campaign materials, histories and timelines, minutes and agendas of meetings, maps, videos, films, and radio interviews.  The analysis details the work of environmentalists in enrolling a variety of human and nonhuman actors in their vision for the region, translating the interests of others into their own.  It shows how environmentalists generated power by severing other groups from their networks and drawing them into their own.  The study details the material and discursive means by which environmentalists went about constructing the network, showing how they intertwined science with politics.  Additionally, the research shows how environmentalists eventually abandoned attempts to place the networks they convened into pre-existing concepts of “nature” and “society,” choosing instead to find ways to simultaneously promote ecological integrity and human well-being.  Finally, the study shows that, while environmentalists worked hard to convene a network in order to realize their interests in protecting BC’s coastal forests, they lost ownership and control of the project’s direction as they enrolled other groups.  Thus, environmentalists were also defined and translated by the network that they sought to create.  17 The study explores the value of a network perspective in analyses of INRM.  As such, it does not examine issues of concern to other approaches, such as trust, social capital, social learning, resilience, and so on.  Instead, this study attempts to complement other approaches by examining some of the ways in which a particular INRM project emerged and evolved, rather than drawing upon already established concepts, categories, and principles to explain and evaluate the case.  As such, findings are restricted to this specific purpose.  Nevertheless, the study contributes empirically to studies of INRM, providing substantial detail on the case of the GBR.  This study will be a useful resource for practitioners and advocates of INRM as they seek to learn about the processes involved in developing such a project.  Additionally, the research provides a contribution to INRM research and theory, demonstrating the value of a network approach to the topic.  In particular, the study provides unique insight into issues of power and hierarchy, science and politics, and the relationship between society and nature.  Finally, the study contributes to ANT scholarship by providing a full-scale ANT study of INRM. In the next chapter (Chapter 2), I review the emergence of integrated natural resource management as a critique and response to conventional approaches to conservation.  In this review, I point out the main problems associated with INRM (revolving around power, science, and nature) thereby specifying my research objectives. Touching on existing research paradigms employed in INRM research, I introduce the promise of actor-network theory as a complementary approach.  I review the main features of ANT, address some of its critiques and show how it can be applied to the study of INRM.  I end the chapter with a preview of the analysis contained in Chapters 4- 7, which will be presented after a discussion of research methods in Chapter 3.   Chapter  18 4 examines how environmentalists redefined the central and north coasts, materially and discursively, into a format that they hoped would generate interest in ushering in a campaign to save them.  In this analysis and throughout, the material and discursive and not considered to be separate.  I look at how environmentalists’ production of meaning was produced through material means; moreover, I look at how new knowledges, meanings and truths produced new material outcomes for the coastal forests and the people connected to them.  Chapter 5 looks at the efforts of environmentalists to get other groups – the BC wilderness preservation movement, forestry companies and First Nations – involved in their project.  Chapter 6 analyses the negotiations and pushback involved as these other groups conditionally accepted their roles in the emerging network.  Chapter 7 examines the procedures invented by the actors to reconcile their interests and to construct a common world.  In the Conclusion, I summarize the main findings before returning to the issues raised in the introductory chapter and evaluating the ability of ANT to address them. By the end of the analysis, readers will have learned much about how the GBR project was originally conceived and how it shifted and changed as it evolved.  They will have travelled with the actors as they pulled together concepts and data from BC’s coasts and from around the world and (re) defined the GBR into a new type of forest; with bears as they became the representatives of the coastal forest, protesting outside of forestry companies’ offices and travelling around North America and Europe; with environmentalists as they traced and intervened in the commodity chain linking forests, forestry companies, and retail customers; with forestry companies as they joined environmentalists in a joint project; with First Nations as they translated the emerging  19 network into their own terms; and with lay stakeholders and experts as they worked to redefine ecology and economy.  By the end of the account, readers will have witnessed the emergence of a network and will have had the opportunity to examine in detail the work that it performed to generate power, intertwine science and politics, and redefine the relationship between society and nature.  Readers will have encountered a full-scale ANT study with which to evaluate the value of this approach to the study of INRM’s particular problems.   20 2  Integrated Natural Resource Management and Actor- Network Theory   2.1  Introduction    In this chapter, I examine research on natural resource management and actor- network theory.  I find that natural resource management has shifted away from top- down, expert-led attempts to protect areas of land by excluding human use and occupation, and towards integrated approaches to conservation that seek to include stakeholders and reconcile human livelihoods with biological conservation.  I also look at the issues that are associated with the newer approach to conservation, including power, the relationship between science and politics, and the specification of the link “between” nature and society.  These issues define the research objectives of my case study of the GBR.  Noting that conventional theoretical approaches to the study of natural resource management generally do not look at integrated projects as complex, contingent, and innovative processes, I propose to use actor-network theory as a complementary research approach.  I argue that ANT has the ability to shed light on the process of integrated natural resource management and to capture the ways in which such projects construct new realities.  While ANT scholarship can be drawn on to inform many different forms of natural resource management, the innovative nature of INRM – particularly its tendency to question an enforced separation between nature and society – make it a very useful development to study from an ANT perspective.  Working through criticisms of  21 the approach, I discuss how ANT can help to provide unique insight into issues of power, science, and nature in INRM.  I end the chapter with a preview of the analysis to come.  2.2  From “Fortress Conservation” to “Integrated Natural Resource Management”   Traditional conservation projects attempted to protect biodiversity by excluding local populations from protected areas, prompting the characterization of this approach as “fortress conservation” or “fences and fines” approach1 (Brown, 2002, p. 6).  The approach has been extensively critiqued for its negative impacts on both people and ecosystems.  One primary aspect of fortress conservation that has been critiqued is its reliance on unquestioned and apolitical “neo-Malthusian” beliefs (Adger et al., 2001; Bryant, 2001; Robbins, 2004).  These beliefs – which underpin the contemporary environmental movement’s focus on limits (Nordhaus & Shellenberger, 2007) – suggest that human demands on natural resources, if left unchecked, will inevitably lead to the collapse of those resources (Malthus, 2005 [1798]; Hardin, 1968).  This approach has been criticised as being apolitical (but secretly political)2 since its logic rests on  1 The mirror belief held by development activists is that conservation produces a direct threat to local livelihoods. 2 No attempt to control human behaviour with respect to the environment can be considered apolitical (Latour, 2004, p. 28).  Indeed, Malthus’ own thesis was written as a polemic against charity for the poor: his idea was that charity might be misplaced since it would result in overpopulation of dependent populations, and eventually national bankruptcy.  Hardin’s thesis is written as a polemic against open-access resources, but ignores the fact that the allocation of property rights to previously “common” land – as in the case of the enclosures of the commons in England – alienated land held in common in order to place it in private hands. It also ignores the fact that collectives have in many  22 supposedly impersonal laws that direct human behaviour (Robbins, 2004).  Critics have argued that fortress conservation alienates land from local resource users, distributes the costs and benefits of conservation unequally, and generally impoverishes, disenfranchises, and disempowers already disadvantaged groups (Robbins, 2004).  Often, according to critics, conservation projects are implemented in a top-down, exclusionary, expert-driven manner (Holling, 1996), thus further alienating local groups and leading to inevitable conflict (Adger, 2001; Brown, 2002).  Moreover, researchers in the social sciences and humanities have pointed out that the object of conservation – “pristine wilderness” – is a myth created by western conservationists while places designated for protection often have long histories of human occupation and use (Cronon, 1996b). Critiques have not been limited to issues of social equity and social ideas, however. Critics have also pointed out the inadequacy of fortress conservation from the perspective of ecological integrity.  They’ve argued that this strategy for wilderness preservation does little to address land use outside of protected areas, thereby resulting in ecological fragmentation (MacArthur & Wilson, 1967; Diamond, 1975; Harris, 1984; Saunders, 1991).  Protected areas become biological islands unconnected with one another and too small to sustain important species.  Consequently, rather that preserving biodiversity, protected areas can perversely contribute to its decline.  instances developed effective formal and informal rules for managing common property resources (Agrawal, 2001; Berkes, 1989; Dietz, 2003; Ostrom, 1990).  More recently, the “politics of limits” at the core of contemporary environmentalism have been under attack from a variety of positions, for both their inability to deal with new environmental problems and their inadequate focus on social justice (Braun and Castree, 1999; Latour, 2004; Nordhaus and Shellenberger, 2007; White and Wilbert, 2009).   23 In response to these perceived failings, new approaches to conservation have begun to emerge since the 1980s and early 1990s.  These approaches have been bolstered through the articulation of the concepts of “sustainable development” and “public participation” by the UN World Commission on Environment and Development (Brundtland, 1987) and Agenda 21 of the 1992 UN Conference on Environment and Development (United Nations. Dept. of Public Information, 1993).3  As Brown (2002) notes, “traditional, top-down exclusionary approaches to protected areas […] are not sustainable, and are not conducive to notions of sustainable development” (p. 6).  In response, Brown continues, “new approaches to the design and management of protected areas, primarily in terms of attempts to integrate protected areas into the economic and social context locally, regionally and nationally, have been designed and implemented” (p. 6).  These new approaches – which fall under the category of what I will refer to here as “integrated natural resource management” (following Bellamy et al, 1999) – can be roughly grouped into the terms that have come to be associated with the concept of sustainability, which is commonly represented as resting on three “pillars”: the social, the economic, and the ecological (Blackburn, 2007).  In the area of integrated conservation and development, these pillars are translated as “populist,” “neoliberal,” and “ecological,” respectively (see, for example, Blaikie & Jeanrenaud, 1997, and Hulme & Murphree, 1999).  3 I do not wish to imply that these developments constitute a “context” within which new approaches to conservation developed: rather, I recognize that they are articulated with them in traceable ways.  The concept of sustainable development can be critiqued and its articulation in particular places mapped (see, for example, Bryant, 1991; Escobar, 1996; Selman & Wragg, 1999; Ellis & Waterton, 2005).  However, for the purposes of this review, I am treating these concepts as “black boxes” (see Latour, 2005: 11).  24 Populist approaches to wilderness protection focus on the participation of impacted social groups in resource management decision-making.  Projects go under a variety of names, such as collaborative resource management, collaborative conservation, collaborative planning, participatory resource management, participatory environmental governance and grass-roots ecosystem management.  Perhaps the most general term is community-based natural resource management (CBNRM) (Conley & Moote, 2003, p. 372). Conservation projects of this type have grown out of dissatisfaction with entrenched conflict and the adversarial nature of resource management (Coughlin et al., 1999, pp. 2-3).  Particularly in the US, resource management decisions tended to become bogged down in lengthy court challenges (Wondolleck et al., 1994).  In response, advocates for CBNRM have argued that citizen participation in decision-making processes would ensure that a full range of values were considered upstream, focusing efforts on “win-win-win” outcomes that would thereby prevent later conflict (Weber, 2000, p. 238).  This approach, part of a wider trend in participatory environmental governance (Bulkeley & Mol, 2003), seeks to devolve decision-making authority from centralized governmental departments to a wide range of stakeholders. Ecological approaches to conservation have developed in the fields of conservation biology and landscape ecology.  Conservation biologists have argued that protected areas were too small and isolated to adequately preserve biodiversity (MacArthur & Wilson, 1967; Diamond, 1975; Harris, 1984; Saunders, 1991) while landscape ecologists have problematized the idea of stable, pristine, nonhuman environments by drawing attention to processes of disequilibrium, flux and chaos (Worster, 1990; Zimmerer, 2000).   These developments converge on the concept of  25 ecosystem management (EM) (now more commonly referred to as ecosystem-based management [EBM]).  EBM involves a shift from resource management – which focuses on maximizing the outputs of a single value (e.g. wood fibre) – to the management of the processes and functions that maintain healthy ecosystems (Cortner & Moote, 1999, p. 37).  While the principles underlying this shift date at least as far back as Aldo Leopold’s (1949) “land ethic,” the concept did not gain saliency in resource management until the mid-to-late 1980s (Grumbine, 1994, p. 28; Czech, 1997, p. 671).  As noted by Pavlikakis and Tsihintzis (2000, p. 258), “as sustainability has become an explicitly stated goal by many governmental, public or private resource management agencies, E[B]M has arisen as the new approach focusing on sustainable development.” Healthy ecosystems are considered to be supportive of a range of social, economic and ecological values: thus, ecosystem management can support the goals of sustainable development. Importantly, in this conception, the role of humans in relation to nature shifts.  As noted by Grumbine (1994), “along with defining the ecosystem management approach as a new policy framework there appears to be a parallel process of redefining the fundamental role of humans in nature” (p. 28). This new role is enabled by a reconsideration of the “balance of nature” view inherent in the fortress conservation approach.  In particular, flux, disturbance, and disequilibrium have become more prevalent ways of describing ecological processes in ecology (Holling, 1973; Worster, 1990), leading some to describe the emergence of a paradigm shift in ecology (Botkin, 1990; Berkes, 2004).  Moreover, just as natural events, such as storms, fires and insect outbreaks cause disturbance in forests, so do humans (Worster, 1990; Walker et al., 2002).  This view, in addition to the recognition that biodiversity conservation requires  26 the management of large areas that include human populations, results in the commonly held perspective among EBM advocates that humans are part of ecosystems (CIT, 2004, p. 9). Economic, or neo-liberal approaches to conservation focus on mechanisms that link ecology and economy in an attempt to “address biodiversity conservation objectives through the use of socio-economic investment tools” (Hughes & Flintan, 2001, p. 4). Blaikie and Jeanrenaud (1997) have referred to these approaches as “neo-liberal” since they are linked with wider developments in which governments devolve traditionally centralized decision-making to market forces. However, relaxation in government control has opened up new avenues for environmental governance among non-state actors (Cashore 2002, p. 506).  Moreover, one key set of initiatives, referred to as Integrated Conservation and Development Projects (ICDPs), ties themselves to social equity and justice at the same time as it promotes ecological conservation.  As such, they involve “an almost complete convergence with sustainable development thinking” (Hughes & Flintan, 2001, p. 4).  ICDPs originated with the World Wildlife Fund’s “Wildlands and Human Needs Program,” launched in 1985.  The program explicitly sought to address the shortcomings of traditional approaches to wilderness conservation by ensuring that conservation projects addressed the development needs of local populations.  ICDPs have been implemented primarily in developing countries and no projects of that name are found in North America (although CBNRM projects share similar beliefs and objectives). They range from attempting to provide local populations with the infrastructure to meet their basic needs, to providing alternatives to resource-dependent livelihoods, to ensuring that local populations have a stake in resource management decision-making.  27 2.3  Integrated Natural Resource Management in BC   While the GBR agreement is represented as being “unprecedented,” some of the trends of INRM have been well established in BC for some time.  Environmentalists and First Nations have long criticized conventional resource management decision-making in BC.  Forestry resource decision-making has historically been centralized in a “compact” between the Provincial Government and forestry companies since the 1940s when the policy of “sustained yield” was instituted (Marchak, 1983; Wilson, 1998). Environmentalists and First Nations began to challenge forestry policies in the late 1960s and early 1970s, arguing that cutting rates were far from sustainable, that resource decision-making over public lands was undemocratic, and, in the case of First Nations, that jurisdictional issues were not settled (Williams et al., 1998).  By the 1980s, these disputes erupted into what the news media dubbed the “war in the woods” (Hayter, 2003).  Entrenched conflict connected to a series of logging blockades and public protests prompted the Provincial Government to experiment with more inclusive and participatory forms of resource decision-making such as advisory committees, task forces and community consultation (Tanis et al., 2004, p. 62).  Informed by the concept of “sustainable development” (McManus, 2002, p. 853), these early attempts were reformulated by the Provincial Government into the Commission on Resources and Environment (CORE, see Owen, 1998).  Subsequently, due to dissatisfaction with CORE’s poor progress, the Land and Resource Management Planning (LRMP) process superseded CORE (Jackson & Curry, 2004, p. 33).  In addition to goals of reducing conflict and promoting sustainability with respect to the management of BC’s natural  28 resources, the LRMP process was specifically designed to foster stakeholder participation in resource management decision-making (Government of British Columbia, 1996). Included in critiques of BC’s resource management “compact” has been its narrow focus on timber to the exclusion of wider ecological and human interests (Cashore, 2001).  Some began to call for wider, more holistic approaches to forestry management (Hammond, 1991, 1997; Hoberg, 2001, p. 63) following the lead of the US, where ecosystem management has been widely adopted, including by the US Forest Service and the Bureau of Land Management (Szaro et al., 1998; Pavlikakis & Tsihintzis, 2000).  However, adoption of EBM in Canada has been sparse.  According to a recent review (Nantel et al., 2003), very few ecosystem management applications have been reported in peer reviewed literature and the authors were able to identify only 30 projects “that met the criteria of ecosystem management of forest lands” (p. 4).  The application of the concept in BC is restricted to the recommendations of the Clayoquot Sound Scientific Panel (Cashore et al., 2001, p. 250). However, there are also differences between the situation in BC and wider developments in resource conservation – particularly with respect to the question of local livelihoods. First, threats to biodiversity do not arise from local populations, but from forestry companies and forestry workers located elsewhere in the province, as well as from customers of BC forest products located elsewhere in the world.  Nevertheless, economic development opportunities for local people – First Nations – are a key part of the GBR solution.  Second, while local First Nations are economically disadvantaged in comparison with other Canadians and often live in conditions similar to those in the Global South, Canada is a developed country – a first for ICD projects which have taken  29 place exclusively in developing countries.  Third, while First Nations are economically disadvantaged, recent legal developments have recognized considerable rights and title, and the Provincial Government has (lately) begun to recognize First Nations as key decision-makers regarding resources in their traditional territories.  2.4  Key Controversies in Integrated Resource Management: Defining Research Objectives    While INRM represents an important evolution in resource management, it comes with problems of its own.  It is worthwhile to examine these problems in order to see what an in-depth analysis of the GBR case can illuminate about INRM as a whole.  One of the key critiques of integrated resource management has been focused on its simplistic and under-theorized view of community and participation.  In particular, some of the early projects were criticized for assuming that communities are homogenous, and for thereby neglecting to take into account internal conflict and unequal power relations (Brosius et al, 1998; Agrawal, 1999).  Moreover, by defining communities as small, bounded entities, projects neglected to take into account the influence of important nonlocal actors (Brown, 2002, p. 10).  The consequences of this oversight, according to critics, is the tendency for projects to exacerbate existing power differences (Brosius et al., 1998) and, consequently, for the interests of some parties to become co-opted during the process (e.g., Modavi, 1996).  Indeed, according to critics, stakeholder processes have often been perversely top-down in practice, with predefined agendas and outcomes that result in a passive role for groups with little power (Brown, 2002, p. 11).  One of the key  30 questions for inherently unequal stakeholder processes is representation: who is empowered to speak on whose behalf (Coughlin, 1999, p. 9).  Often, “stakeholders” are predefined with their interests assumed as clear and fixed (Daniels, 1996), leading to a neglect of the importance of group formation and network building (Mahanty, 2002). While BC has experimented with various forms of stakeholder involvement in natural resource management decision-making since the 1980s, BC environmentalists have long been critical of such attempts.  In particular, environmentalists have argued that processes such as the LRMP scatter the environmental movement into numerous, unconnected processes; sap the energy and resources of environmental organizations; lead to public apathy since they see that environmentalists are part of decision-making processes; co-opt environmentalists’ demands due to the need for compromise and support for consensus decisions; and distract environmentalists from direct activism, all the while logging continues in the areas under discussion (Salazar & Alper, 1996; Wilson, 1998). Therefore, one of the key issues that need be explored in integrated resource management is how the actors negotiate power.  Conservation projects can assume neither that communities are homogenous, nor that power differences can be resolved by the structure of the project itself.  For this reason, it is useful to learn from the actors themselves how they resolve issues of power inequality. How do groups engage in stakeholder processes if they feel that they have an unequal ability to influence decisions? How do groups gain power to influence decision-making?  Who are the relevant actors? Are all relevant actors “local” or are some stakeholders located in distant yet connected places?  Are official processes the most appropriate places to conduct negotiations if  31 stakeholders do not have the same degree of influence?  These questions need to be explored in order to understand how “stakeholders” participate in CBNRM.  The first objective of this study is to examine how environmentalists gained the power to influence decision-making beyond that afforded to them by the LRMP process as it was originally designed.  A second key controversy with integrated management is how to integrate multiple knowledges in resource management projects (Selman & Wragg, 1999, p. 650; Folke et al., 2005).  Integrated management rejects top-down decision-making by “experts” in recognition of scientific uncertainty and in favour of the democratization of science (Cortner & Moote, 1999, p. 84; Ludwig, 2001; Berkes, 2004).  Yet, according to Agyeman and Angus (2003), “despite the rhetoric, many local sustainability projects or attempts to encourage the development of sustainable communities are still top-down, expert led processes that fail to genuinely engage local citizens” (p. 346).  For example, critics have charged that new approaches to resource management such as ecosystem- based management are so complex that only experts have the ability to make authoritative claims (Cooperrider, 1996).  Perhaps the reason for this is that the practice of “civic science” (Lee, 1993) has been poorly defined (Wilson et al., 1999; Bäkstrand, 2003, p. 24).  In one sense of the term, civic science – the attempt to “to connect natural science and politics in the pursuit of sustainability” (Plummer, 2006, p. 10) – involves bringing well-defined knowledge and technical information (or “the facts”) to environmentalists, policy-makers and a range of stakeholders to inform resource management decision-making processes.  In this sense, science plays an advocacy role by imparting knowledge and information; but this knowledge and information may still be  32 delivered in a top-down, expert led manner.  In another sense of the term, civic science involves the participation of multiple groups in asking and answering key research questions (Cortner, 2000, p. 27; c.f. Latour, 2004, and Callon et al., 2009).  In this sense, science becomes an “increasingly interactive process between lay and expert people, reconnecting science and its cultural context” (Warburton, 1998, p. 3, cited in Agyeman & Angus, 2003, p. 355).  In particular, according to Cortner (2000) “the civic science model democratizes expert cultures.  It emphasizes learning among participants, and is highly collaborative” (p. 27). The articulation of science with politics has become increasingly prominent in BC natural resource management.  For one, environmentalists have learned a great deal about the science and economics of forest management through participation in stakeholder processes (Wilson, 1998, p. 211).  But more importantly, according to Kranjc (2002), the 1990s saw BC become the site of civic science as conservation biologists teamed up with environmentalists to advocate the protection of core habitats and alternative land use planning processes in Clayoquot Sound and the GBR.  Kranjc’s analysis suggests that the scientific reality of coastal ecosystems was brought to light for political purposes, and thus that science and politics were externally linked together.  As such, it conforms to the first sense of civic science given above since it treats science and politics as internally distinct from one another.  By contrast, I will explore the second sense of civic science given above, and examine the extent to which the GBR case intertwines science and politics – from problem definition to collaborative knowledge projects. According to the conventional view of both scientific objectivity and expert-driven models of resource policy-making, science and politics need to be kept separate from one another – even if  33 the former is to direct the course of the latter.  However, criticism of the top-down model entails a reworked understanding of the relationship between these two “spheres.”  Is it necessary to keep science and politics separate from one another?  Is it possible?  Is it desirable?  If science and politics are intertwined, how can their intertwining be done properly? The second objective of the study is to examine how environmentalists articulated science with politics in their attempts to influence the management of the GBR. A third issue in integrated resource management has to do with unclear assumptions about the conceptual linkage “between” conservation and development (Salafsky & Wollenberg, 2000; Brown, 2002).  While integrated resource management explicitly moves away from models that separate people and environment via fortress conservation, early alternatives linked conservation and development indirectly through the provision of substitute economic opportunities (Wells & Brandon, 1992; Salafsky & Wollenberg, 2000; Brown, 2002)4. By contrast, more direct linkages were made in the 1990s (Western & Wright, 1994) with projects such as nontimber forest product harvesting or tourism enterprises that provide opportunities for local people to directly benefit from biodiversity and for livelihoods to drive conservation (Salafsky & Wollenberg, 2000, p. 1425). The shift from indirect to direct linkages between conservation and development involves not just a practical but also a profound conceptual shift.  Under the fortress conservation approach, ecology and economy are  4 For example, the UNESCO Man and the Biosphere Program’s Biosphere Reserves included the idea of “buffer zones,” or economic development zones surrounding protected areas.  These zones are designated for economic opportunities – such as coffee plantations and agricultural production – that provide alternatives to direct exploitation of resources in protected areas.  34 defined as mutually exclusive: economic development, by its very nature, entails ecological destruction; correspondingly, ecological protection, by its very nature, entails the destruction of economic development opportunities and local livelihoods.  Indirect linkages between conservation and development represent an alternative approach to conservation, but keep these basic set of assumptions in place: biodiversity must be protected from economic activities; by providing substitute activities outside of reserves, pressure will be removed.  The shift to direct linkages, by contrast, fundamentally alters basic beliefs about ecology and economy, suggesting that ecology can be promoted via economic development, and that economic development can proceed by means of conservation.  This approach builds on the “win-win-win” approach envisioned by the Brundtland Commission, but veers sharply away from a trade-off model that is implicit in the Commission’s report.  Rather than finding a “balance” between two opposing forces, the direct linkage approach to conservation and development consider these two things to be in a positive sum relationship.  Thus, the key question here is: to what extent do projects such as these avoid dualistic forms of thinking about society and nature? In BC, conflicts between advocates of wilderness preservation and proponents of economic development through forestry are very entrenched (Marchak, 1983; Wilson, 1998; Stansbury, 2000).  These conflicts have often been analysed as resting on a dualistic logic that pits pristine nature against society, ecology against economy.  For example, Willems-Braun (1997) argues that forestry companies’ and environmentalists’ representational practices frame the conflict over BC’s forests as an opposition between jobs and the environment.  Similarly, Stefanick (2001) argues that actors in the conflicts over BC’s forests frame humans as separate from and superior to nature and,  35 simultaneously, as dependent on a non-exploitative relationship with nature.  Doyle et al. (1999) argue that the primary frame deployed by the forestry industry in the early 1990s was one of “trees versus jobs,” contending that this frame is connected to a more general “nature versus people” frame.  Finally, Rossiter (2004) argues that environmentalists constructed an image of “pristine nature” that “leaves no room for human economy, technology or politics” (pp. 151-52).  The key question concerns the extent to which actors in the GBR reproduce dualistic framings of the relationship between society and nature and the extent to which they avoid such framings to propose new types of relationships ‘between’ ecology and economy.  The third objective of the study is to examine how environmentalists specified the conceptual link between nature and society while seeking to reconcile ecology and economy.  2.5  Researching Integrated Natural Resource Management    A number of research perspectives have been deployed to study integrated natural resource management, including common property, social learning, political ecology, social constructionism and social-ecological systems.  These perspectives have afforded numerous insights into resource management issues.  For example, common property researchers have shown that private property and state management are not exhaustive forms of resource management and have documented many cases in which local institutions have emerged to effectively and sustainably manage resources (McCay & Acheson, 1987; Berkes, 1989; Ostrom, 1990; Matthews, 1993; Agrawal, 1999). This  36 insight is of special significance in the current trend to decentralize resource governance and develop local stakeholder forms of governance (Agrawal, 2002, p. 41; Blomquist, 2010).  Building on acknowledgement of the role of social capital in local resource management (Pretty, 2003), social learning theory suggests that strong networks and trust enable groups to explore and challenge their beliefs as they work toward a common understanding of the problem, agreement, and collective action (Daniels, 1996; Maarleveld & Dangbegnon, 1999; Shusler et al., 2003; Bouwen, 2004; Pahl-Wostl, 2004, 2007).  As a corrective to the tendency of common property, social capital, and social learning approaches to focus primarily on the local level, political ecologists have examined the influence of external structures and forces on resource management projects (Cockburn & Ridgeway, 1979; Blaikie & Brookfield, 1987; Peet & Watts, 1996; Bryant & Bailey, 1997), focusing in particular on how the unequal distribution of costs and benefits of conservation projects reproduce existing economic and political inequalities (Paudel, 2006), such as the displacement and dispossession of local people (Geisler, 2003; Rangarajan & Shahabuddin, 2006).  While political ecology originally tended to focus on the material determinants of unequal outcomes, some have attempted to avoid structuralist and determinist accounts by focusing on the role of language in resource management (Bryant, 2001, p. 162; see also Braun & Wainwright, 2001) and how particular visions of nature and the environment are imposed on landscapes and people (Escobar, 1998; Peet & Watts, 2004).  This latter focus is the main interest of social constructionist analysis of nature and resource management, a paradigm well- represented in environmental sociology (Greider & Garkovich, 1994; Hannigan, 1995; Eder, 1996; Macnaghten & Urry, 1998;), geography (Evernden, 1992; Castree & Braun,  37 2001; Castree, 2005), anthropology (Escobar, 1995; Neuman, 1998; Tsing, 2001), history (Cronon, 1983; Worster, 1994; Oelschlaeger, 1996) and other cognate fields (Cronon, 1996a).  Finally, resilience theory has redirected attention away from “command and control” approaches to resource management (Holling, 1996) and toward recognition of the linkages between social and ecological systems and the need for resilience in the face of inevitable change and system disturbance (Berkes & Folke, 1998; Gunderson & Holling, 2001; Holling, 2001; Gunderson & Prichard, 2002; Berkes et al., 2003; Folke, 2006). While these research paradigms are rich, diverse and very productive, they tend to leave certain elements of INRM unexplored.  In particular, they tend to focus on explaining outcomes rather than exploring processes (Parkins & Mitchell, 2005, p. 531), and to put forward factors, principles or theoretical forces that are posited to influence outcomes in any given case, thereby neglecting to base explanations on close analysis of the complex and contingent features of particular cases (Steins, 2001, p. 18).  For example, the common property approach lists up to 35 principles that must be in place for INRM to be a success (Agrawal, 2001, p. 1651), thereby neglecting to study the contingent nature of particular cases.  Social learning perspectives examine particular cognitive mechanisms whereby groups achieve a common understanding without analyzing processes of conflict and discord – which often take place before consensus is achieved – including processes of coercion involved in getting others to see the problem in a common way.  Political ecology relies on a priori assumptions about the political factors that are involved in producing environmental changes (Vayda, 1999), thus neglecting to focus on how factors important to changes are constructed in particular  38 cases.  Social constructionism assumes that INRM can be analysed purely as a social process of language use and representational practice, without taking into account the role of nonhumans (Benton, 1994; Dunlap & Catton, 1994; Martell, 1994; Murphy, 1994; Soulé & Lease, 1995; Dickens, 1996). The common shortcomings of these approaches, although small, are important since they deflect attention away from integrated natural resource management as an innovative and complex process that is highly contingent on local circumstances (Vayda, 1999; Steins, 2001; Heikkila, 2005).  INRM brings together a wide and complex variety of elements that need to be connected with one another in ways that support sustainable outcomes.  Multiple agencies, public and private decision-makers and stakeholders of all sorts – some of whom are nonlocal (Brown, 2002; Mahanty, 2002) – must be stitched together in ways that are sensitive to issues of power and inequality (Selman & Wragg, 1999b, pp. 649-650; Harrington et al., 2008, p. 201).  Moreover, these groups are connected not only with each other but also with multiple nonhumans, such as bears, trees, and aquatic ecosystems (Woods, 1997, p. 322).  Indeed, INRM crosses multiple jurisdictional, spatial, temporal, and conceptual boundaries, linking such disparate things as economies, livelihoods, rights, sciences, politics, ecosystems, governance, equity, and values – across entire regions and with a view to achieving future states of sustainability. The result is a form of “hybrid governance” in which states, markets, and civil society actors all participate (Lemos & Agrawal, 2006; see also Cashore, 2002).  For these reasons, it is fair to say that “resource management is a heterogeneous network” (Holne, 1999, p. 5).  Seeing INRM in this way entails a significant emphasis on relationships, negotiation, and network building (Mahanty, 2002, p. 1383; Ali-Kahn & Mulvihill, 2008,  39 p. 1991).  Questions arise with respect to how networks are assembled, how elements become connected with one another, and the transformations they undergo.  Moreover, actors and issues will be unique to each case, drawing attention to the contingent and uncertain nature of INRM (Mehta et al., 1999; Steins, 2001; Vayda, 1999).  This entails that attempts to predict and explain cases with a priori theoretical presuppositions, or with lists of characteristics and success factors derived from other studies, should be balanced with attention to the details of individual cases and how they unfold.  Indeed, the uncertain, complex, and contingent nature of individual INRM plans suggests that the latter strategy might be given precedence over the former, at least to counteract the opposite tendency of existing research.  In other words, “the current emphasis in the conservation arena on what conservation programs do, should be balanced with attention to how conservation programs are undertaken” (Mahanty, 2002, p. 1383).  A focus on how conservation programs evolve and develop can provide an understanding of how and why the characteristics and factors associated with particular projects are formed and why actors come together to produce them.  According to Heikkila (2005, p. 584) “scholars who study large-scale collaborative efforts have focused more on defining the characteristics of collaboration and the factors that make them successful than on what brings actors together in the first place […] In other words, the factors that support the formation of these institutions are not well understood.”  One way of assessing how resource management projects develop and unfold is to follow a conservation proposal from its very inception and investigate the ways in which actors attempt to actualize it.  This would involve focusing on how an “intervention proponent  40 [initiates] interest in a core set of ideas and practices and build[s] a network of actors to sustain the idea, technology or practice” (Mahanty, 2002, p. 1375).  Therefore, an approach is required that would study INRM as an innovation – that is, something not yet well known and that must therefore be explained in its own terms. INRM projects not only assemble a complex, contingent and heterogeneous array of entities, domains, and issues, but also produce new worlds: new social groups, new shared understandings, new knowledges, new economies, new relations with nonhumans. Accordingly, the approach would have to avoid making assumptions about society, nature, or their relationship.  As mentioned above, the range of actors is often wider than usually taken into account in INRM studies (Mahanty, 2002, p. 1370; Morris, 2004, p. 178; Brown, 2002).  Consequently, “a framework of analysis is required that allows a diversity of actors, and the interactions and relations between them, to be investigated” (Morris, 2004, p. 178).  Moreover, as networks of actors come together in support of conservation projects, they often form new groups and coalitions.  As a result, actors’ perceptions, attitudes, and even interests shift and change.  As a corrective to the common “assumption that the participants' interests are largely fixed and that the best one can hope for is a grudging compromise that perhaps satisfies no one,” (Daniels, 1996, p. 74) studies of INRM can focus on how interests are formed and changed through interactions.  INRM also focuses on combining and reconciling the “interests” of humans and nonhumans.  Accordingly, research can focus on how interests – whether human or nonhuman – are produced and how they shift and change.  In the next section, I will present the main concepts and features of an approach little known to INRM – termed “actor-network theory” – which promises to account for  41 the complexity and contingency of innovative conservation projects in ways not well accounted for by other approaches.  As I will explain, ANT is able to focus on important actors without ignoring nonlocal actors; to give an account of power that is sensitive to the contingencies of the case; to focus on joint knowledge production and constructions of nature without reducing them to cognitions or discourse; and to account for nonhumans in ways other than those dictated by a global, overarching theory.  In particular, ANT can adequately address INRM’s fundamental challenge to the fortress conservation view that nature is separate from society – without presuming the factors that enable it to do so a priori. In this discussion, I do not wish to suggest that ANT and IRNM are coterminous, or that the latter somehow derives from the former.  ANT concepts can be used to illuminate the processes involved in many forms of natural resource management, past and present.  However, some features of INRM – particularly its innovations with regard to power, science, politics, nature and society – help to render explicit what might otherwise remain implicit (or “black-boxed”) in other approaches.  As preceding sections have indicated, INRM renders particular assumptions problematic: that power is held on the basis of particular social structures (since the purpose of INRM is to redistribute power), that nature and society are separate ontological domains (since INRM argues against separating nature and society), and that science and politics pertain to separate realities (since in INRM they are brought together in the form of “citizen science”).  As I will show in the next section, ANT provides conceptual tools that explore alternatives to these assumptions.   42 2.6  INRM as Network-Building: The Promise of Actor-Network Theory    Actor-network theory is the name given to a set of post-structuralist ideas and concepts developed in the field of science and technology studies by Bruno Latour, Michel Callon, and John Law to explain how knowledges, technologies, and actors are produced (Callon, 1980, 1986; Callon & Latour, 1981; Callon & Law, 1982; Law, 1986; Latour, 1987, 1988)5.  Reacting against what these authors considered to be over- socialized accounts of science and technology (Bloor, 1999, p. 83) – namely, accounts that failed to consider the specific power of science (Murdoch, 1997, p. 734) – Latour, Callon, and Law developed a novel form of analysis that allows researchers to avoid a methodological choice between internalist and externalist explanations of knowledge and technology; that is, a choice between explanations resting either on the internal cognitive and technical content of science and technology or explanations deferring to the external social, economic, and political “context” (Latour, 1983).  In particular, early studies of Louis Pasteur’s knowledge of the anthrax bacillus (Latour, 1988), France’s failed experiment with the electric vehicle (Callon, 1986), biologists’ attempt to domesticate scallops (Callon, 1968), and the methods of navigation in long-distance maritime travel (Law, 1986), all sought to explain science and technology as outcomes of interactions  5 As its main proponents are often at pains to explain, ANT is not a coherent framework or theory that can be “applied” to the study of social and natural life, but a loose collection of ideas that shifts and changes depending on how they are taken up (Latour, 1999a; Law, 1997).  Perhaps for this reason, and also due to the fact that they bypass usual conceptual distinctions, the concepts have been confusing for critics. leading one critic to speak of  Latour’s work as “obscurantism raised to the level of a general methodological principle” (Bloor, 1999: 97)  43 among heterogeneous networks of actors (Law, 1992, p. 2).  For example, “instead of clearly dividing science from the rest of society,” Latour’s (1988) study of Pasteur “makes no a priori distinction among the various allies” – including microbes, farmers, laboratories, social movements and veterinarians – that were assembled to produce knowledge about anthrax. The key idea supporting this move is the ontological proposition that things – whether “facts,” technologies, or actors – emerge as outcomes of processes that assemble heterogeneous elements into stable networks.  In the words of Law (2008),  Actor-network theory is a disparate family of material-semiotic tools, sensibilities and methods of analysis that treat everything in the social and natural worlds as a continuously generated effect of the webs of relations within which they are located. It assumes that nothing has reality or form outside the enactment of those relations. […] [T]he actor-network approach thus describes the enactment of materially and discursively heterogeneous relations that produce and reshuffle all kinds of actors including objects, subjects, human beings, machines, animals, “nature,” ideas, organisations, inequalities, scale and sizes, and geographical arrangements. (p. 142)  ANT thus replaces concepts of social actors and nonhuman things with the concept of the “actor-network”: entities are produced through the actions of the elements that comprise them; those elements, in turn, can be ascribed with “actor” status only by virtue of their connections with other elements in the network. Although this novel approach associated with ANT was originally developed in the field of science and technology studies, numerous other fields have realized its potential, including (inter alia) anthropology, economics, feminism, geography, organizational sociology, social psychology, media studies, performance studies, political theory, sociology of globalization, sociology of tourism, and the philosophy of science  44 (Saldanha, 2003, p. 421).   When it comes to recognizing the potential of ANT to study interactions among people and the environment, geographers have been at the forefront (Bingham 1996; Hinchliffe, 1996; Thrift, 1996; Murdoch, 1997, 1998; Whatmore, 1997, 2002; Demeritt, 1998; Castree, 2001, 2003; Braun, 2002).  Environmental sociologists, on the other hand, have been slow to take up the perspective, with the main bulk of advocacy coming from a geographer (Murdoch, 2001; but see also Lockie & Kitto, 2000, and Lockie, 2004, for sociological advocacy of ANT in agricultural research).6  Examples of ANT-inspired research relevant to INRM are few; those that do exist come from geography (Davies, 2002), environmental science/studies (Holn, 1999; Harrington et al., 2008), collaborations between geography and environmental science/studies (Comber et al., 2003; Beveridge & Guy, 2009), planning (Hillier, 1998; Kitchen, 2000; Selman, 2000; Selman & Wragg, 1999a, 1999b; MacCallum, 2008), and development studies (Mahanty, 2002).  Only one self-identified environmental sociologist draws on the tools of ANT to study collaborative processes in environmental management, although it is in a study of the assessment of large dams within the environmental impact assessment process (Lockie, 2007). Nevertheless, ANT holds great potential for the study of INRM, including, as I will explain below, a particular sociological contribution.  In particular, ANT promises to provide a way of investigating innovation – with respect to complex new relationships among humans and nonhumans – in ways that are not available in other approaches  6 Some environmental sociologists, such as Irwin (2001) do speak about “co- construction” but do not specifically align themselves with actor-network theory.  45 (Steins, 2001).7  ANT investigates innovation by systematically refusing any and all presuppositions, and instead attending to things in their specificity to learn how new relations, patterns, and beings emerge.  In particular, ANT scholars avoid common assumptions that reality is ordered in a series of dualisms between society and nature, micro and macro, subject and object, agency and structure, and reality and representation. Spheres such as these, according to ANT scholars, do not exist as given features of reality, but are produced through processes – largely performed by academic disciplines but also in common venues such as daily newspapers – that “purify” the underlying heterogeneity of the networks on which they rest (Latour, 1993).8  Indeed, according to actor-network theorists, nothing exists in itself.  Rather, “everything in the social and natural worlds” is produced or constructed – and is thus the effect of the performances of the network that sustains them. One consequence of the claim that nothing exists in itself, but is constructed by networks of humans and nonhumans, is that nothing can be said to have an essence, nor be reducible to any other essential thing.  As Law (1992, p. 7) argues, ANT is “a semiotic machine for waging war on essential differences.”  Anything that is assumed to have an essential nature, whether it is a scientific fact, a piece of technology, a human being or an organization, can in practice be analyzed for the things that constitute it.  The solidity and stability of things is merely provisional, as can be witnessed when the networks that  7 Indeed, Latour (2005: 9) even considered re-labelling actor-network theory the “Sociology of Innovation.” 8 Latour’s (1993) use of the term “purify” refers to the ways in which some elements of heterogeneous networks are systematically ignored or conceptually excluded.  For example, when placing a “fact” in “nature,” the human work involved in its production becomes hidden from view.  In contrast, when representing a “value” as “social,” the nonhuman means by which it possesses durability become hidden from view.  46 support such things as “organisations or systems which we had always taken for granted – the Union of Soviet Socialist Republics, or Continental Illinois – are swallowed up” (Law, 1992, p. 1).  Nor is stability secured by virtue of a more foundational reality behind inessential appearances: phenomena cannot be explained as mere manifestations of an underlying cause or reality, whether it is presumed to be “social,” such as religion, or “natural,” such as physical laws of nature.  As Latour (1988, p. 163) argues, “nothing can be reduced to anything else, nothing can be deduced from anything else, everything may be allied to everything else.” Critics have interpreted this principle of irreduction (Latour, 1988) to entail a flattening of differences in which everything is the same as everything else.  For example, the world depicted by ANT, according to Shapin (1988, p. 547),  is the world of the seamless web, a world in which everything is connected to everything else, in which even the discrete existence of things and the categorization of processes cannot be used to interpret or to explain the actions of those who are said to produce them.  If everything is connected to everything else, if nothing can be reduced to anything else, and if nothing can be explained as an outcome of pre-existing causes, then everything is the same and there are no differences.  As Laurier and Philo (1999) put it, ANT produces the “problem of installing a great indifference between the countless things of the world…which arises when they end up being portrayed as potentially all the same” (p. 1016). However, ANT theorists “do not deny differences; [rather, they] refuse to consider them a priori and to hierarchize them once and for all” (Callon & Latour, 1992,  47 p. 356).  That is, differences emerge through processes of network formation and thus cannot be assumed to pre-exist such ongoing activity.  Moreover, they do not suggest that everything is connected to everything else: “not everything comes together, not everything is connected” (Latour, 1996a, p.152).  As Harman (2009) explains, “a philosophy of networks does not require that the network be devoid of separate parts.  If everything were already linked, translation would not be such a pressing issue for Latour” (p. 47).  Indeed, notwithstanding various critiques of the use of the term (Haraway, 1994; Strathern, 1996), Latour (2005) suggests that the network metaphor has the advantage of reminding us that not everything in the world is connected, since “nets, networks and ‘worknets’ leave everything they don’t connect simply unconnected” (p. 242).  Therefore, ANT provides neither a framework that would specify in advance the kinds of entities, structures, and forces that need to be considered in sociological research, nor a holistic theory in which everything is one.  Instead, ANT researchers propose that specific realities are produced through specific networks that can be traced by the analyst, thereby providing a practical way of studying innovation. ANT-inspired analyses of INRM can draw on the principle of irreduction – the idea that nothing can be reduced to anything else and that nothing has an essence – to investigate the specific and contingent details involved in the production of new institutions, management regimes, groups, relations, knowledges, economies, and so on. As such, these kinds of analyses can avoid both the application of lists of criteria to the evaluation of outcomes, and attempts to explain outcomes as resulting from given social and political forces, as reviewed above.  Moreover, an ANT analysis allows the researcher to take into account the sheer diversity and complexity of human and  48 nonhuman elements that go into the construction process.  For these reasons alone, ANT is well suited to the study of resource conflict and integrated management: according to Woods (1997),  given that rural conflicts are influenced by (and influence) local, national and global actors; that they are contested by coalitions of actors; and that they could conceivably be described as involving both human and non-human actors, many of the features of actor network theory […] would appear to be relevant. (p. 322)  Moreover, the basis that ANT provides to study the complex and contingent ways in which innovative relationships “between” humans and nonhumans are produced, also supports an analysis of power and hierarchy, science and politics, and the relationship between nature and society – all of which, as explained above, are essential areas of investigation for INRM.  2.6.1  ANT, INRM and Power    Integrated resource management is premised on the devolution of power and authority from central decision-makers (the Government and industry) to a large range of stakeholders. Yet, this is not a simple or straightforward process.  As described above, devolution of power is not just about sharing power between government and community, as if these two things are monolithic; rather, communities are complex and internally fractured.  Since power is distributed across fractured coalitions and networks,  49 rather than “held” by any particular group, ANT provides a useful analytic approach. According to Woods (1997):  In positing a new conceptualisation of power the focus is shifted from the power attributed to any one actor to the construction of networks which allow an outcome to be achieved. These 'tactics of translation' provide a framework for understanding how coalitions are constructed and structured. (p. 323)   The concept of translation refers to the process of (re)defining the properties of entities – in particular, an entity’s identity and interests – as they become connected to other entities.  The sociology of translation, according to Callon (1986), amounts to a “new approach to the study of power” (p. 196).  According to Callon’s (1986) classic formulation, the process of translation has four overlapping “moments.”  During problematization, the proponent of an idea or plan of action defines the problem to be solved, identifies the entities that are required to solve it, and defines the entities in such a way that the plan becomes an “obligatory point of passage,” necessary for them to pass through in order to meet their (re)defined needs.  This moment thus involves the first stage in linking a number of entities in a network and defining them relationally.  As Callon (1986) explains, “problematization describes a system of alliances, or associations, between entities, thereby defining the entity and what they ‘want’” (p. 8). The second moment involves attempts to “interest” these entities in the program of action so defined.  Callon draws on the etymological meaning of the term “interest” as being in between (inter-esse) to come up with the neologism, interessment, to describe this second moment.  Entities of course are already supplied with interests and identities defined by the networks in which they are implicated.  This second moment thus involves building  50 “devices which can be placed between them and all the other entities who want to define their identities otherwise” (ibid., p. 9).  These “devices” are infinitely varied, ranging from physical devices to verbal and written arguments. Severing entities from “enemy forces” (ibid., p. 11) requires the consent and participation of the entities themselves, resulting in a series of negations defining the third moment, enrolment.  Entities may demand any number of conditions, concessions, or other transformations of the plan in exchange for their agreement to participate, and enrolment can take a number of forms, including “physical violence […], seduction, transaction, consent without discussion” (ibid., p. 12). Finally, enrolled entities are mobilized – the fourth moment of translation – by others who represent and speak on behalf of the assembled network.  Entities that were at first dispersed, unconnected (but connected to other entities) and differently defined and interested, become progressively displaced and translated into a form that allows them to be easily represented, transported and combined (Latour, 1987, p. 227).  In Callon’s study, for example, scallops that once rested on the sea floor have been 1) (re)defined as entities interested in accepting shelter that allows them to proliferate and survive, 2) severed from enemies such as currents, predators and greedy fishermen via the interessment device of a net and towline, 3) anchored (by themselves) on towlines on condition that they were at the right level and the right sort of substances for anchorage were used, and 4) mobilized in the form of diagrams and tables presented by biologists, their spokespersons, at a far away conference.  According to Callon (1986), “chains of intermediaries which result in a sole and ultimate spokesman can be described as the progressive mobilization of actors who render […] propositions credible and indisputable by forming alliances and acting as a  51 unit of force” (p. 14).  In other words, the scientists who spoke on behalf of the scallops (and fishermen, consumers and scientific colleagues) were able to claim factual knowledge about scallops on the basis of the network assembled for that purpose.  The fact that their claims rested on a network became apparent when the network began to fall apart: scallops refused to anchor, fishermen refused to withhold from fishing, colleagues became sceptical. Several studies of resource management have adopted this approach to study how particular conservation visions become dominant and “how certain actants/networks are able to impose their views over those of others” (Hillier, 1998, p. 84; see Selman & Wragg, 1999; Burgess et al., 2000; Kitchen, 2000; Davies, 2002; MacCallum, 2008).  For example, Kitchen’s (2000) study of environmental policy in the UK examines how “social actors struggle to ensure that their representations of the area prevail” by “striving to build networks and alliances designed to enrol others to their interests” (p. 135). Burgess et al.’s (2000) study of farmers’ identities relies on the understanding that “translation is thus about attempting to gain rights of representation, to speak for others and to impose particular definitions and roles on them” (Burgess et al., 2000, p. 123). Selman and Wragg’s (1999b) study of countryside planning in England draws on an understanding of ANT as  a theory of `translation’, explaining how an innovation `translates’ spatially and temporally from its origin to multiple destinations and is translated into the lingua franca of a cluster of spokespeople, so that a particular perspective gains dominance and displaces competitor theories and practices. (p. 653)   52 While this is a useful approach, the view of translation as a process of “heterogeneous engineering” (Law, 1987) has been heavily critiqued within science and technology studies for being too centred and even “Machiavellian.” ANT researchers in STS have tended to focus on “great men” such as Pasteur, examining how they convene, direct, and order networks in order to achieve their aims.  Thus, while the “greatness” of these figures in ANT accounts is distributed somewhat among the various elements that are required for them to achieve their goals, they are nevertheless at the centre of the action, enrolling others as a means to their ends (Star, 1989; Fuller, 2000, p. 20).  In these terms, ANT “sees only attempts to dominate, strategies for winning battles, means of attack, trials of strength, and other forms of violence” (Amsterdamska, 1990, p. 496). Indeed, Latour has been prone to use language associated with warfare – such as “battle,” “allies,” “trials of strength,” and the like (Fujimara, 1992).  In these terms, ANT’s “actors work out their impulses to grow, to transform themselves from 'micro-actors' to 'macro- actors'.  This they do vampire fashion: by subduing others, by insinuating themselves into others' bodies and by turning them into agents of their own volition” (Shapin, 1988, p. 534).  The language of warfare has especially been picked up by feminist science studies scholars: as Haraway (1994, p. 59) questions, “must technoscience – with all its parts, actors and actants, human and not – be described relentlessly as an array of interlocking agonistic fields, where practice is modelled as military combat, sexual domination, security maintenance, and market strategy?”  This kind of language centres the protagonist – generally male – and depicts him as in control of processes of network assemblage.  But is the centre really a centre and is “he” really in control? As Knorr- Cetina (1985, p. 583) asks, “where, in Latour's account of Pasteur's success, is the  53 reference to the accidents, mistakes, breakdowns, circumstantial factors and disinterested attributions which presumably helped in the constitution of the phenomenon 'Pasteur', and which occurred perhaps against the will of contributing agents?” Moreover, critics charge that ANT’s focus on “great” heterogeneous engineers leads to a neglect of marginalized voices, while its participation in an elitist discourse that serves to maintain the status quo favours certain social interests and supports continued discrimination against disadvantaged groups.  As Star (1991) argues, ANT accounts focus on network builders, but not on those who occupy a marginal position in relation to the networks of others.  Because it is not as easy or as interesting to hear stories from the perspective of groups that are not able to assemble networks, ANT tends to follow the stories of heroes or failed heroes and thus describe networks from the perspective of the actors who represent them (Sismondo, 2009, p. 89).  ANT thus participates in an elitist discourse.  As a result, ANT “colonizes” the “other” and incorporates them into “the same” while producing accounts of dominant networks (Lee & Brown, 1994, p. 779; Hetherington, 2000). Others charge that ANT supports the status quo by not critically engaging with or judging the social structures and forces that produce social inequality in the first place, relying on “an implicit but clear distinction between describing a state of affairs and judging them” (Berg, 1996, p. 256).  Critics argue that ANT researchers do not study “economic, political, cultural, historical, and ideological factors” (Scott, 1991, p. 12) or embed their studies in a “larger historical or cultural context” (Martin, 1998, p. 27), and thus fail to recognize how science reflects large social forces (Schaffer, 1991, p. 189; Fuller, 2000, p. 27), particularly those associated with “vested interests, strategic  54 competition and traditional political forces” (Scott, 1991, p. 13).  Several critics argue that ANT researchers neglect to account for the ways in which “the powerful” direct scientific research – theirs included – the result being unwitting complicity in relations of oppression (Etzkowitz, 1987, p. 696; Winner, 1993; Fuller, 2000).  For example, Fuller (2000) suggests that ANT is “mode 2” research, or client-driven research that is shaped by collaboration with state and industry.  More forcefully, Fuller suggests that ANT has been “captured” by the interests of “the powerful,” and is providing the latter with the means to oppress even in new situations where “top-down” authority is questioned, while simultaneously showing the indispensability of ANT research for policy-makers (Fuller, 2000, p. 12-17).  Moreover, this flattened ontology entails that everyone has equal power, thus reinforcing capitalist ideology (Fuller, 2000, p. 11; Mirowski & Nik-Khah, 2004; Boltanski & Chiapelllo, 2005).  Consequently, the researched remain pawns in a power struggle and ANT researchers fail to either judge the current order or to recommend a radical new order (Fuller goes so far as to accuse ANT researchers of “totalitarianism”).  The twin charges of Machiavellianism and total neglect of power structures ought to cancel each other out.  On the one hand, critics claim that ANT treats the production of knowledge and technology as a purely political contest, thus taking away any ability of the oppressed to “speak truth to power” (Sokal & Bricmont, 1999).  On the other hand, ANT is accused of being indifferent to inequalities and power struggles, and thus “content only to connive with those in power” (Latour, 2005, p. 251).  Both of these critiques rest on two assumptions: (1) that power resides in the social whereas knowledge and truth reside in nature, and (2) that power is something that some groups “have” and others fail to “have” on the basis of their position within social structures.  On this basis,  55 both criticisms make sense: ANT extends politics to where it should not be – nature (unless it is to debunk ideology) – and withdraws it from where it ought to be contested – society (unless it is to use truth against power).  But, ANT does not make these kinds of distinctions; nor does it consider some characteristics, such as power, to be essential features of some ontological categories, such as society.  Instead, for ANT theorists, power is produced and reproduced through networks of humans and nonhumans.  As Latour (2005) puts it:  power, like society, is the final result of a process and not a reservoir, a stock, or a capital that will automatically provide an explanation. Power and domination have to be produced, made up, composed.  Asymmetries exist, yes, but where do they come from and what are they made out of? (p. 64)  Conventional sociologists assume that asymmetries are produced by the structures of “society.”  This, according to Latour (Strum & Latour, 1987), is an “ostensive” definition of society: it assumes that actions are contained “within,” and explainable with reference to, a social realm defined by particular principles, rules and structures.  In contrast, ANT explanations rely on a “perfomative” definition of society.  Actors create their social worlds (and their principles and structures), by linking heterogeneous materials – economic, biological, psychological and so on – together.  Society is not what glues people together, but is what is glued together.  Moreover, the process of “gluing” must be kept up: as soon as it stops, the reality which it performs ceases to be. Conventional sociology, argues Latour (2005, p. 100), simply replaces the explanandum with the explanans: “society” whether taking the form of actors, norms, structures, or interests, cannot explain but instead needs to be explained.  As Latour (1987) writes,  56  Analysts who use groups endowed with interests in order to explain how an idea spreads, a theory is accepted, or a machine rejected, are not aware [sic] that the very groups, the very interests that they use as causes in their explanations are the consequence of an artificial extraction and purification of a handful of links from these ideas, theories or machines (p. 141).   When adopting an ANT approach for the study of INRM, the analyst must neither apply a Machiavellian model – in which some actors assemble, enrol and colonize others in order to ensure that their interests prevail and dominate – nor a social structures model – in which pre-existing social forces determine individual outcomes.  As Davies (2002) reminds us, the value of ANT is that “it enables analysis of partnerships to move beyond the simple dichotomy of structures and agents to include people, institutions and the non- human realm and to explicitly examine resources, arguments and flows of knowledge between them” (p. 190).  The analysis must focus on the assemblage of multiple entities such as these while avoiding assumptions that any one entity is ultimately in control of such assemblage, or that it will serve to realize any one set of interests.  Rather, the focus must be on how identities, interests, and visions emerge and become transformed as networks are built.  ANT should help us understand how “the nature of a program followed changes as new actors are enrolled and as the program responds to the 'anti- programs' of opponents. This can contribute to understanding how the nature of a political campaign or issue can change in relation to shifting contexts” (Woods, 1997, p. 324; see also Lockie, 2007).  Additionally, this approach can help us understand how social change is possible. If giant forces are assumed to exist and condition all outcomes, then there is little room for analysing how groups who have little influence with those occupying powerful  57 positions are able to generate influence.  Thus, instead of analyzing cases as examples of well-known types (Latour, 2005, p. 22), ANT learns from the actors they study how power is produced in particular instances.  This is perhaps the best way to make a difference, since “if there is no way to inspect and decompose the contents of social forces, if they remain unexplained or overpowering, then there is not much that can be done” (Latour, 2005, p. 252).  By contrast, according to Latour (2005):  action is possible only in a territory that has been opened up, flattened down, and cut down to size in a place where formats, structures, globalization, and totalities circulate inside tiny conduits, and where for each of their applications they need to rely on masses of hidden potentialities. If this is not possible, then there is no politics. (p. 252)  2.6.2  ANT, INRM and the Relationship Between Science and Politics    INRM represents a move away from top-down, expert-led resource management and toward the inclusion of multiple voices, perspectives, and knowledge (Selin & Chavaz, 1995, p. 189; Ludwig, 2001; on the rise of civic science for sustainability, see Lee, 1993; Bäkstrand, 2004; Cortner, 2000; Plummer, 2006).  As such, it contains an implicit critique of the relationship between science and politics.  The traditional model of science and politics assigns different roles to experts and policy-makers.  Experts are tasked with discovering the incontrovertible facts.  By contrast, policy-makers work to balance multiple and conflicting interests and positions while they make decisions.  This model presupposes two distinct spheres of reality: the world of mute, nonhuman reality, and the world of cacophonous, human politics.  The goal of conservation advocates is to  58 use science to bring the true, incontrovertible facts to the political sphere in order to quell debate and provide the foundation for decision-making.  This model of science and politics is thus an apolitical ecology, one in which scientists break away from the human world of politics to access the truth of nonhuman nature, only to break away from the world of nonhuman nature to return to the world of politics with their truths and tell people what they should do.  This “double rupture,” as Latour (2004, p. 11) refers to it, allows experts to enter politics without appearing to be political: “they can make the mute world speak, tell the truth without being challenged, put an end to the interminable arguments through an incontestable form of authority that would stem from the things themselves” (Latour, 2004, p. 14).  However, the fortress conservation approach that would maintain a clear distinction between nonhuman reality and human politics has been replaced with the recognition that conservation science and politics are inseparable.  As Selman and Wragg (1999b, p. 667) note, “it is likely that the generally democratic spirit of participatory fora is conducive to classically trained scientists becoming more amenable to the legitimacy of lay knowledge.”  In Latour’s terms (2004, pp. 22-24), “matters of fact” have been replaced with “matters of concern.”  That is, the idea that experts can determine the essential requirements of ecosystems, that those requirements are well-defined, and that they entail exclusion of human use, has been replaced with recognition of uncertainty, instability, and disturbance, the involvement of multiple groups in defining and valuing ecosystems, and the inclusion of humanity as a component of ecosystems.  Science remains fundamental to resource management, but its position vis-à-vis politics takes on a new aspect.  As argued by Lockie (2004):  59  Even though the “natural sciences” have established themselves as “obligatory points of passage” (Callon, 1986), or “centres of calculation,” in the networks of natural resource management, environmental disputes are seldom straightforward conflicts between technocentric science and romantic environmentalism. Rather, they are conflicts over whose science, and the ends to which it is to be applied, are to prevail (Beck, 1992). The expression of agency in such situations is highly dependent on the ability to open the black box of ‘science’ and to enrol its actants in one’s own networks. Environmental controversies are often, as much as anything else, conflicts over who may speak on behalf of non-humans (p. 51).  ANT is particularly adept at opening the “black box” of science to study its matters of concern, because it was invented for just that purpose.  Instead of assuming two separate spheres – nonhuman reality and human politics – ANT investigates how the world is made up of a variety of heterogeneous elements.  Politicians and their constituents, economists and their markets, scientists and their ecosystem integrity, NGOs and their moral concern, communities and their economic needs, species and their habitats – all of these elements go into the construction of a conservation project.  The assemblage of such networks is simultaneously a scientific and political process.  This is not to say that politics envelopes science, projecting and defining nonhumans purely on the basis of political gain.  Rather, as Shapin (1990: 538) notes, “Latour's erosion of the conventional boundaries separating politics from science is predicated upon the insistence that objects and non-human entities as well as people are political beings. Things belong to the study of political order as much as human agents.” Nevertheless, ANT’s erosion of the boundary between science and politics has been heavily criticised.  In particular, this view has been critiqued for being overly relativistic, or for suggesting that there is no reality beyond what is “socially constructed” (e.g., Amsterdamska, 1990; Gross & Levitt, 1994, p. 58).  In this critique, ANT is  60 characterized as entailing that reality is solely a product of human practices of representation, and that there is no observer-independent reality to which representation corresponds.  For example, Amsterdamska (1990) argues that Latour’s position entails that:  we are unable to make any distinctions between things and their representations: Things are what we collectively represent them to be, nothing more and nothing less.  Accordingly, a true statement is a statement we express as true, nature is what we collectively represent as nature, a fact is what we collectively express as a factual proposition. (p. 497)  In Amsterdamska’s view, ANT confines the analyst to the sphere of human representation, whereas one ought to be able to specify a link between representation and that which is represented.  Similar critiques have been levelled against social constructionist perspectives in environmental sociology (Benton 1994; Dunlap & Catton 1994; Martell 1994; Murphy 1994; Dickens 1996).  These debates have been particularly heated, since realists believe that constructionist arguments ultimately entail that there are no real environmental issues and threats.  However, while social constructionist environmental sociologists respond to these charges by emphasizing that they believe in an external biophysical reality but prefer for methodological reasons to restrict analysis to epistemological issues while withholding ontological statements (Burningham & Cooper, 1999), ANT takes a very different approach.  Indeed, Amsterdamska’s critique misrepresents ANT.  The misrepresentation becomes apparent when one takes into account another, symmetrical critique of ANT, one that is in line with a social constructionist approach and accuses ANT of not being relativistic enough.  61 For example, Bloor (1999) suggests that Latour’s position is a step backward from his own approach to science and technology, known as the “Strong Program” in the sociology of knowledge.  Bloor argues that where Latour naively takes account of nonhumans in the production of knowledge, the Strong Program rightly restricts analysis to the assumption that “systems of belief, that is, shared and institutionalized forms of knowledge, are the medium through which people co-ordinate their shared interactions with non-social nature” (Bloor, 1999, p. 88). According to Bloor (1999) “the important point is to separate the world from the actor’s description of the world.  It is the description that is the topic of inquiry” (p. 93).  Moreover, this entails a clear restriction to the sphere of representation and meaning: “only by sustaining the distinction between the subject and object, and by driving a wedge between nature itself and the descriptions of it provided by the knowing subject can we highlight the problematic character of those descriptions” (ibid., p. 94).  Similarly, Collins and Yearly (1992) argue that Latour is forced to uncritically defer to claims made by scientists about nonhuman nature, since he has no particular scientific expertise himself.  Consequently, he neglects an appropriately sociological analysis of scientific knowledge which involves methodological relativism, or “the ability to switch between different frames of reference” (Collins & Yearly, 1992, p. 301).  Thus, Latour (and colleagues whom he represents) is alternately accused of giving too much to representations and not enough to nonhuman nature and too much to nonhuman nature and not enough to human representations.  The accusation rests on the assumption that there is a big gap between reality and representation – but this is another one of the dualisms that ANT has attempted to dissolve.  According to this assumption, the epistemological question concerns how to  62 ascertain the extent to which representations “correspond” to reality, otherwise known as the “correspondence theory of truth” (Latour, 1999a, p. 24). Yet, this assumes two distinct ontological domains – language and nature – that are separated by a giant gap. By contrast, Latour attempts to describe how scientists cross multiple small gaps as they “pack the world into words” through what he calls “circulating reference” (ibid, p. 24). In tracing how scientific knowledge is produced he shows that there is a chain that involves translating material reality into new forms through the application of instruments and categories (or “inscription devices”), which are further transformed by other instruments and categories.  Thus, in Latour’s formulation, representation and reality are not related as two domains joined by a bridge of “correspondence,” but are conjoined in a single process that loads things into words – things are real because they are constructed (Latour, 2005, p. 89).  This view goes well beyond either social constructivism or realism by demonstrating how both humans and nonhumans produce realities. An ANT approach to the study of INRM thus must neither treat science as a black box and matters of fact as given, nor retreat to a purely social constructionist account of nonhuman reality.  It must focus on circulating reference, the manner in which nonhumans are “loaded into discourse.”  It must study the innovation of resource management as an ongoing process of network building that is simultaneously scientific and political.  Nature and society are not two separate spheres: there is only one process to follow – the process of association.  Science and knowledge are the products of a “circulatory system” (Latour, 1999a, p. 80) that links multiple human and nonhuman groups.  Similarly, society is conceived of as a “circulating entity” (Latour, p. 17).  63  2.6.3  ANT, INRM and the Relationship Between Humans and Nonhumans    INRM asks us to see that nature and society are not separate from one another, but are intertwined.  The challenge is to see people and nonhumans as co-inhabitants in ecosystems.  As noted above, the understanding of how nature and society are related in the conservation field has varied over time (Salafsky & Wollenberg, 2000).  In traditional approaches to conservation, nature and society are seen as oppositional.  Some of the early integrated conservation and development projects specified indirect linkages, such as the buffer zones around biosphere reserves that were meant to provide alternative livelihood options for local populations.  More recent approaches specify direct linkages, in which conservation and development are dependent upon one another.  Despite this trend, approaches to the study of INRM have not evolved to the point of taking into account direct linkages.  Of the approaches reviewed above, all but one restrict analysis to social processes, leaving the analysis of nonhumans to natural scientists.  Only resilience perspectives promise to take into account humans and nonhumans in the same framework.  However, as mentioned above, this perspective tends to be too theory-driven and thus unable to account for the nuances and surprises involved in the construction of new worlds.  Environmental sociology should be well placed to contribute a theory of how nature and society are linked, since this question forms the core of the sub-discipline. However, environmental sociology has generally maintained a boundary between the two  64 assumed spheres, even when developing frameworks to understand how they are related (Murdoch, 2001, p. 115).  In the original formulation of environmental sociology as the “study of societal-environmental interactions” (Catton & Dunlop, 1989), society and nature are seen as simultaneously related and separate.  Human societies are seen to be embedded in and dependent upon natural systems, but nevertheless defined by unique features and dynamics that are not shared with nature.9  The various theoretical frameworks in environmental sociology that have arisen since – such as the treadmill of production (Schnaiberg, 1980), ecological modernization theory (Spaargaren & Mol, 1992, ecological Marxism (Benton, 1993), risk society (Beck, 1992), and social constructionism (Hannigan, 1995) – have either maintained this view or modified it by saying that our understanding of nature and society is inextricably determined by social processes, while assuming that there is a real, nonhuman reality outside of such constructions.  In this context, ANT stands to make a significant contribution both to the study of IRNM and environmental sociology.  Murdoch (2001) argues that ANT is well-placed to respond to the “ecological challenge to sociology,” that is, to core ecological tenets such as the notion that humans and nonhumans are connected and interdependent, the rejection of nature and society as distinct ontological categories, the rejection of dualistic thinking, and a focus on a unified vision (p. 112).  ANT, according to Murdoch (2001),   9 Dunlap and Catton emphasized early on that they did not deny that humans were unique with respect to other species.  To avoid misunderstanding, they changed the name they applied to mainstream sociology’s (and mainstream society’s) view of nature from the “Human Exceptionalist Paradigm” to the “Human Exemptionalist Paradigm.”  65 is ecological in the sense that it seeks to overcome any underlying distinctions between natural and social entities, thereby extending agency to non-humans as well as humans.  In this fashion the theory shifts our attention away from humans and the social to collectives and complex ecologies. (p. 120)  Indeed, ANT fundamentally reconfigures our understanding of the “social.” According to ANT theorists, in order to understand society, we need to understand the nonhumans with which we are bound.  On one hand, nonhumans have a social history just as much as humans do, traceable in laboratory studies (Latour, 1988, p. 262).  On the other hand, human intentions would not be achievable without the many nonhumans that render them “durable” (Latour, 1991).  Humans delegate activities to nonhumans: cumbersome hotel room keys replace the need for verbal or written directions to do with leaving your key at the desk (Latour, 1991, p. 104), speed bumps replace the need for police men to direct people to slow their cars (Callon and Latour, 1992), seat belts ensure the morality of safety and precaution, and automatic door openers replace the need for human door openers (Latour, 1992).  The translation of human goals and intentions into objects, texts, devices and so on, enables people to act at a distance, that is, to structure outcomes far removed from face-to-face interactions (Latour, 1987, p. 219).  At the same time, nonhumans pass along some of their traits to humans.  For example, the harnessing of electric power by Edison helped him build corporate power (Hughes, 1983; Latour, 1999, p. 204), the domestication of microbes by Pasteur transformed him into a great scientist (Latour, 1988): in short, machines are used as models of social organization (Latour, 1999, p. 206).  Other writers have similarly noted the ways in which humans are transformed by nonhumans, speaking of, for example, the domestication of people by microbes or animals (Haraway, 2008).  66 In this view, nature and society are not considered to be pre-existing spheres that are brought into some kind of relationship.  According to ANT, for any outcome to be produced – whether it is a piece of knowledge, a form of technology, an identity, or an institution – a network must be assembled.  Once the network is in place, the outcome that it performs can be “purified” of its messy, heterogeneous elements and placed in discrete categories, such as nature and society. As Murdoch (1997) puts it, “it is only when the networks have been established, and roles and identities distributed within them, that a clear-cut difference emerges between ‘things out there’ and ‘humans in here’” (p. 744). ANT provides tools to study both the ways in which heterogeneous entities become associated with one another and translated into one another’s terms and the processes in which these complex networks are purified of their attachments.  It is only through such processes that the categories of nature and society are (re)produced. Latour (1993) refers to the dual process of translation and purification as the “modern constitution” (p. 10).  This is a key concept that I will come back to repeatedly in my analysis of the case. This approach has far reaching consequences that challenge the humanist position underlying much of sociology. According to ANT scholars, the attribution of special properties to human beings – such as intentionality and agency – purifies the underlying heterogeneous networks that give rise to these attributes. Since there is no specific realm that might be called “society,” it is inaccurate, according to ANT theorists, to attribute specific properties to it.  Properties such as agency, intentionality, and interests, generally thought to be exclusively human, are shared and distributed across human-nonhuman networks.  Nonhumans participate in the course of actions, for example, by facilitating  67 “action at a distance.”  This is not to say that they merely reflect, embody, or transport human intentions and desires, but they help constitute intentions and desires and thereby translate and transform them.  For example, in a study of the ecological restoration of a river, Eden et al. (2000) describe how the project proponents needed to rely on the contributions of clay to help cut a sinuous, meandering bend.  However, the clay did not perform as required, resulting in an alteration to the plan and thus the ultimate goals and intentions “behind” it.  ANT theorists prefer to speak of agents as “actants” because they do not restrict agency to humans.  However, they do not mean to imply that nonhumans “have” agency just as humans do; rather the idea of agency is reconfigured as an effect of networks of heterogeneous relations.  Agency, whether attributed to humans or nonhumans, is not given, but emerges.  As Pickering (1993) notes in relation to nonhuman agency in scientific accounts, “the contours of material agency are never decisively known in advance, scientists continually have to explore them in their work, problems always arise and have to be solved in the development of, say, new machines” (p. 564).  As Latour (1999c) explains, “actantiality” is what “provides actants with their actions, with their subjectivity, with their intentionality, with their morality” (p. 18). Nonhumans provide and are provided with agency just as much as humans. Some have taken ANT’s claim that the social is comprised of nonhumans and that agency is not a solely human capacity to entail that people are mere objects.  For example, Fuller (2000) claims that “in its proliferation of agency, actor-network theory dehumanizes humans” resulting in the “treatment of humans as cogs in the wheels of a machine” (p. 21).  Similarly, Bloor (1999) likens Latour’s argument that no a priori distinctions ought to be made between humans and nonhumans to the view that “minds  68 are brains and that brains are computers” and that “humans such as Pasteur and Millikan are just like microbes, in being collections of electrons and other basic particles” (p. 96). For this reason, according to Amsterdamska (1990), there is not much to be gained from breaking down distinctions between humans and nonhumans: humans and nonhumans are “hardly comparable, and the elimination of differences among them leads only to confusion” (p.  501). However, this position arises from a misunderstanding of ANT. ANT does not intend to reduce humans to the status of nonhumans, if the latter are understood as a given ontological reality.  Materialistic monism is only possible with a prior definition of materialism.  For one, irreduction, as mentioned above, is a basic ANT tenet.  Moreover, ANT’s understanding of nonhumans is very different from what was crafted in opposition to society.  As Latour (1999b) writes, “nothing could be further from our definition of actants.  They are not in nature, nor in society (nor in language)” (p. 125).  Rather, actants are networks of heterogeneous elements. Nevertheless, ANT has been critiqued for taking an anti-humanist position.  The primary claim is that ANT’s dissolution of the boundary between humans and nonhumans fails to adequately take into account what is distinctive about people – particularly, their intentionality and capacity to have interests.  According to Pickering (1993), this is the real “sticking point.”  As he writes, “we humans differ from nonhumans precisely in that our actions have intentions behind them, whereas the performances (behaviours) of quarks, microbes, and machine tools do not” (ibid., p. 565). Moreover, intentions are generally thought to be based on underlying interests, such as people’s interests “in encouraging or enlisting in courses of action which promise to give scope and value to their skills and routine” (Shapin, 1991, p. 546).  Finally, interests are  69 thought to be shaped by one’s position in wider social structures.  Hence, according to this position, the proper role of the sociologist is to approach their topic of study as “social realists” who take the social world as the “foundation of reality” (Collins & Yearley, 1992, p. 308).  But, again, ANT writers do not deny the idea that people are associated with intentionality and interests, only the assumption that intentionality or interests are unique, essential traits or capacities.  Intentions and interests have to be produced, as does everything else.  Latour (2005), for example uses the metaphor of “plug-ins” to suggest that putatively unique and essential characteristics of humans exist only when they are constructed out of materials that come from without, such as information labels on consumer products to which people can subscribe to become rational consumers.  With this understanding, according to Latour (2005):  You don’t have to imagine a “wholesale” human having intentionality, making rational calculations, feeling responsible for his sins, or agonizing over his mortal soul. Rather, you realize that to obtain “complete” human actors, you have to compose them out of many successive layers, each of which is empirically distinct from the next (p. 207).  Thus, ANT scholars do not deny the existence of important traits such as intentions, desires, responsibility, and agency, but they attempt to show the heterogeneous ways in which these characteristics are produced, rather than assuming them a priori as fundamental features of human beings.  Accordingly, “rather than utilizing Nature and Society as explanatory resources ANT authors believe these two great domains have to be examined as outcomes” (Murdoch, 1997: 743).  Indeed, Latour and others argue that as a result of science and technology, relationships between humans and animals, and the ecological crisis, attempts to purify heterogeneous networks into  70 “nature” and “society” are becoming more difficult, and less politically desirable (Murdoch, 1997, pp. 731-732; Braun & Castree, 1998; Whatmore, 2002, p. 1; Latour, 2007, p. 3; Haraway, 2008; White & Wilbert, 2009).  As Latour (2004, pp. 20-22) argues, environmental problems have progressively rendered explicit all of the links and attachments between people and nonhumans that were previously taken for granted.  As a result, “the ecological crisis has forced us to abandon the nature and society collectors” (Latour, 2007, p. 5).  The question is, what can replace the nature/society framework? According to Latour, this cannot be determined in advance; rather it is an open question that must be examined empirically. At a minimum, the new conception can be referred to as a “collective,” or a “procedure for collecting associations of humans and nonhumans” (Latour, 2004, p. 238).  The role of the analyst is to examine the formation of networks and the ways in which a variety of elements come together to form a coherent, common world. This perspective is particularly applicable to the study of INRM.  INRM projects consist of multiple groups and the attempt to integrate humans and nonhumans in a coherent order.  Rather than assuming that the network so assembled can be split into experts and lay people, scientists and politicians, and that the network is a “society” that engages with an external “nature,” ANT researchers should study how multiple groups in the network – scientists and their ecosystems, environmental non-governmental organizations (ENGOs) and their charismatic megafauna, politicians and their constituents, species and their conservation biologists, forestry companies and their customers, First Nations and their rights and title – connect with one another, how their interests and identities are mutually defined, how they go about determining which  71 elements to bring into their world and how those elements may be understood, and how they form those elements into one good common world (Latour, 2004).  The analysis should not assume that certain properties – such as agency and intentionality – belong to certain beings and not others, but examine how these features, and others, emerge through the process of network formation.  Indeed, a key focus will be on whether the actors themselves seek to purify their network into “nature” and “society” or – of central importance for IRNM – whether they develop an alternative identity for the collective that avoids the society/nature dualism.  2.7  Summary   The Great Bear Rainforest agreement highlights issues of collaboration among ENGOs, forestry companies, and First Nations, the application of new “ecosystem- based” forms of forestry management, and the reconciliation of conservation and economy in the coastal forests.  As I detailed in this chapter, the GBR thus represents a case of integrated natural resource management.  INRM seeks to avoid the problems that came with the fortress conservation approach to wilderness conservation, such as injustices meted out to local populations and species loss through ecological fragmentation.  To address these problems, a number of overlapping initiatives have been introduced, including community-based natural resource management, ecosystem-based management, and integrated conservation and development projects.  On the basis of these initiatives, INRM works to include multiple stakeholders in resource management,  72 to manage resources on the basis of unstable ecosystems that include people as a key component, and to directly link conservation and development. However, research has highlighted particular problems with INRM.  In particular, INRM projects take place within and can exacerbate existing relations of power, must attend to challenges to scientific authority, and often involve an uncertain understanding of the conceptual link “between” society and nature.  In this chapter, I discussed how a number of different research traditions have addressed these issues, including common property theory, social learning, political ecology, social constructionism, and resilience theories.  While these perspectives have taught us much about INRM, they are not always well adapted for taking account of the complex and contingent nature of INRM projects.  Moreover, with the exception of resilience theories, existing research tends to restrict its focus to social processes without taking into account the role of both humans and nonhumans in resource management projects.  To address this need for a research approach that can investigate INRM on its own terms as a contingent process that associates humans and nonhumans into innovative formations, I introduced a novel set of conceptual tools and resources, collectively referred to as actor-network theory.  Although it has only been drawn on by a handful of researchers of INRM-type projects, ANT promises to make a strong contribution to the study of INRM.  ANT researchers refuse to make a priori assumptions about the worlds that they study, thus allowing them to take into account the complex and contingent processes through which actors to create new worlds.  Studying these worlds as the outcomes of processes that assemble heterogeneous networks, ANT researchers provide unique insights into issues of power, science and politics, and the relationship “between”  73 nature and society.  This is not to say that ANT has not gone unchallenged, however. Researchers within and outside of STS have challenged the approach for simultaneously being too relativistic and not relativistic enough, anti-humanist and anti-science, apolitical and extending politics everywhere, ignorant of the marginalized and willing to give voice to the “missing masses” of nonhumans.  The symmetry of these criticisms indicate that, often, critics mischaracterize the position of ANT.  Nevertheless, since ANT does challenge important elements of sociological thought, such as agency, intentionality, power, knowledge, and the relationship between society and nature, it is important to put it to the test to see if it stands up to its promises.  2.8  Outline of the Analysis    While ANT has been tested in numerous fields, it has hardly had a hearing in environmental sociology despite the fact that it promises to make a key contribution to the sub-discipline’s preoccupation with the relationship “between” society and nature.  In this dissertation, I test concepts and perspectives associated with ANT as I examine the GBR agreement as a case of INRM.  In the following analysis, I examine the GBR as the outcome of a process of network building among a variety of actors.  I look at how a variety of elements – human and nonhuman – were drawn into the project, how their identities and interests were produced in the encounter, and how the overall focus of the project shifted and changed as elements were enrolled in support of it.  I roughly adopt the schema of translation developed by Callon (1986) and its overlapping four  74 “moments.”  In Chapter 4, I examine the processes of network-construction engaged in by environmentalists as they sought to “problematize” the region and “interest” other groups in saving it.  In Chapter 5, I examine environmentalists’ practices of “interessement,” or how they sought to interpose themselves between their potential allies and other groups that would define their interests differently. I look at how some environmentalists attempted to enrol the wider BC wilderness movement into their project by directing the movement’s focus away from individual, fragmented campaigns and toward a large, comprehensive campaign.  I look at how the concepts and practices of conservation biology positioned bears as the representatives of the coastal temperate rainforest.  I examine in detail the strategies of environmentalists to position themselves between forestry companies and their retail customers.  Finally, I note how, in their campaigns, First Nations resisted the attempts of environmentalists to draw them into their campaigns and to speak on their behalf.  In Chapter 6, I examine the negotiations that took part in the “enrolment” of the various actors.  In particular, I look at the exchange of properties between environmentalists and forestry companies as they formed a new group between them, the Joint Solutions Project (JSP).  Not only did forestry companies take on a new identity as they joined the network, but so too did environmentalists.  Noting the resistance of multiple groups during this stage – among the “public,” environmentalists, forestry companies, workers, and First Nations – I look specifically at the efforts of First Nations to alter the network for their purposes.  Finally, the last chapter before the conclusion (Chapter 7) looks at how the various actors in the network were “mobilized” into a connected and reconciled network of interests via ecosystem-based management and the conservation economy.  In particular, I look at  75 how these mechanisms collect multiple human and nonhuman interests, commensurate them, and put them in a format that can be represented in an agreement.  The purpose of this analysis is to bring to light issues of power, science and politics, and society and nature that arise in the course of the emergence of an innovative approach to natural resource management.  I examine how the process of network construction shifted relations of power among the actors involved.  In particular, the focus is on the network interventions engaged in by environmentalists to alter the network of power that was given by the prevailing approach to collaborative resource management.  I analyse how ENGOs generated power through the enrolment of nonhumans in their campaigns and intervention in the commodity chain linking coastal forests to international markets.  I also consider how First Nations generated power through their counter-enrolment of the network.  Additionally, I consider some actors that felt that they were not represented and were left out of the agreement.  Second, I examine the process of network formation as a simultaneously scientific and political process.  In this analysis, I try to detail the chain linking words and world, rather than posit mechanisms to leap over the giant gap between representation and reality.  In other words, I do not counterpoise “social constructions” to “biophysical reality.”  I look at how both experts and lay researchers (or “citizen scientists”) simultaneously drew on material and discursive elements as they worked to enrol humans and nonhumans into the emerging network.  This analysis is applied to the problematization of the region as a “coastal temperate rainforest,” to the designation of the “Great Bear” as its representative, and to mechanisms of reconciliation between ecosystem integrity and human well-being. Finally, I analyse how actors in the GBR case worked to resolve the relationship among  76 the elements they assembled together in a network.  In particular, I examine attempts to purify the network into “nature” and “society.”  However, my primary focus is on examining the extent to which actors eventually replaced practices of purification with attempts to construct a “collective.”  That is, while early chapters show how the actors both assembled heterogeneous networks and attempted to purify them, the last chapter focuses on the mechanisms invented by the actors to reconcile the interests of multiple humans and nonhumans into a new configuration that cannot be understood as either nature or society.  In the Conclusion, I consider the contributions of this analysis for research on INRM.  I suggest that ANT provides a means of studying INRM as an innovation in ways that other perspectives are unable to offer.  In particular, I highlight that ANT provides unique insight into the issues of power, science and politics, and the relationship “between” nature and society.  Moreover, I offer this study as a contribution to the field of environmental sociology, which has recognized the existence and potential of an ANT approach but which has few empirical examples (and none in the field of INRM).  I emphasize that ANT provides a new way of conceptualizing not only nature, but also society, replacing both terms with empirical explorations of the “collective.”  I emphasize that this study highlights the role of nonhumans in network formation but does not specifically analyse nonhuman agency in the form of resistance to enrolment. Nevertheless, I suggest that there is no reason why this could not be done in principle.  77 3  A Method and not a Theory   3.1  Methodological Strategies    According to Latour (1999), ANT is “a method and not a theory” (p. 20).  That is, ANT researchers do not try to explain the world that they study with reference to pre- determined conceptual models and frameworks.  This approach would be problematic for ANT scholars, since it would rely on an understanding of what the world is made of – particular social structures, forces, and so on – rather than on learning from the actors themselves how they go about constructing their worlds.  Many of the categories that conventional sociologists take for granted, including the groups to be studied, who and what has agency, the status of objects, and the taken-for-granted reality of natural “facts,” are all empirical questions for ANT scholars (Latour, 2005b).  Since “sociologists of associations” cannot start with assumptions such as these but must start with a “clean slate” (Law, 1992, p. 2), they have no recourse but to “follow the actors themselves” (Latour, 2005, p. 68).  Actors know very well what they are doing and leave evidence of their activities that is traceable by sociologists.  The key is to not explain away the actors’ activities with recourse to social structures and forces, but to maintain a light theoretical repertoire or “infralanguage” (Latour, 2005, p. 30) that is capable of deploying controversies around things such as groups, interests, agency, objects, and facts, thereby discovering the traces by which such things are linked and constructed.  As Latour (1999) puts it:  78  Actors know what they do and we have to learn from them not only what they do, but how and why they do it.  It is us the social scientists, who lack knowledge of what they do, and not they who are missing the explanation of why they are unwittingly manipulated by forces exterior to themselves and known to the social scientist’s powerful gaze and methods. (p. 19, emphasis in original)  For this reason, ANT researchers “follow the actors themselves” to trace their world- building activities.  A network in this sense is not a theory of what the world looks like, but a method to be deployed in order to trace how worlds are constructed: “a tool to help describe something, not what is being described” (Latour, 2005, p. 131). A theory consists, at its most basic level, “of two concepts joined by a proposed relationship” (Maxwell, 2005, p. 42).  ANT’s network methodology enables the researcher to study how the actors themselves establish relationships and, thus, how they explain their worlds in the process of building them.  Ontology and epistemology are folded together and the distinction between description and explanation becomes meaningless: a full description provides an explanation (Latour, 2005, p. 137).  This close focus on actors’ world-building activities lends itself to the detailed, qualitative studies of particular cases.  According to Creswell (1998, p. 17-18), qualitative research describes how a state of affairs came into being (rather than trying to explain why); explores a topic in which variables and explanatory theories are not easily identified or available; presents a detailed view of the topic; studies individuals in their natural setting through interviews, participant observation and other data collection; writes the report in a narrative, literary style; spends a long time collecting and analyzing extensive data; and tells the story from the participants’ view rather than from an expert’s judgments.  79 My study of the GBR involves a qualitative, in-depth study of a process.  It therefore conforms to a case study approach.  According to Creswell (1998), “a case study is an exploration of a ‘bounded system’ or a case (or multiple cases) over time through detailed, in-depth data collection involving multiple sources of information rich in context” (p. 61).  The purpose of a case study is to focus intently on a particular process or event to explore a problem (Lincoln et al., 1985).  In this dissertation, I investigate the GBR to explore issues of power and hierarchy, science and politics, and nature and society in integrated natural resource management from a network perspective.  Creswell (1998) suggests that cases can be identified by being “bounded” in space and time (p. 37).  However, from an ANT perspective, neither place nor time can be assumed a priori since they are the outcomes of networks of associations.  As with other elements under consideration, the methodological approach is to “follow the actors themselves.”  Accordingly, I treat the “place” of the GBR as a controversy that I deploy in Chapter 4 where I examine the heterogeneous processes involved in environmentalists’ determination of the area as the “Great Bear Rainforest.” I demonstrate the convergence of efforts to define the area in maps, which have come to form the spatial bounds of the case.  With respect to time, I take the first attempts to redefine the central coasts in the early 1990s and the 2006 agreement as the rough temporal bounds of the case. Case studies involve in-depth explorations, and thus use extensive and multiple sources of data, including observations, interviews, audio-visual material, and documents and reports (Creswell, 1998, p. 61).  My data collection proceeded in two steps.  First, I collected textual and audiovisual data about the GBR from websites and online databases. Websites included environmental organizations (individual organizations and the  80 Rainforest Solutions Project coalition), forest company groups (Coast Forest Conservation Initiative), First Nations (Coastal First Nations – Turning Point), Provincial Government (Integrated Land Management Bureau), and mainstream and grassroots online media sources such as the CBC and The Dominion and Raven’s Eye.  Databases included Canadian Newsstand and a public provincial government file-sharing (ftp) site. From these sources, I collected agreements (LRMPs, government-First Nations, ENGOs- industry, etc.), terms of reference, work plans, legal orders, presentations, reports (technical, organizational, workshops), scientific articles, ecological data, newsletters, press releases, news stories, public and customer information materials, campaign materials, histories and timelines, minutes and agendas of meetings.  With respect to audio-visual material, I collected maps, videos, films, and radio interviews. Second, I conducted semi-structured face-to-face interviews with key individuals purposefully selected from involved environmental organizations, forest companies, First Nations organizations, the Provincial Government, the local government, and consultants.10  To do this, I created a sample frame of nearly 300 individuals from a variety of sources, including: participants in the Central and North Coast Land and Resource Management Plans (LRMPs), members of the scientific body set up to provide information on ecosystem-based management (Coast Information Team), participants in institutions designed to implement the agreement (Ecosystem-Based Management Working Group [EBMWG], Land and Resource Forums [LRF], Plan Implementation and Monitoring Committees [PIMC]), and key names identified in news stories, press releases and campaign materials.  This sample frame was stratified by group, including  10 See Appendix for copies of letters of introduction, consent forms, and interview guides.  81 environmentalists, forestry industry, First Nations, local government, provincial government, stakeholders, and consultants.  Each grouping was subdivided into subgroups and membership in coalitions was identified (see Table 1). As mentioned above, from an ANT perspective, groups should not be assumed before research commences.  I became aware of groupings used in my sample frame from engaging with the material collected in the first step.  These groups were identified “by the actors themselves” in press releases, campaign materials and in official processes such as the LRMPs.  I found, in putting the frame together, that the groups were not entirely distinct, with some individuals fitting into more than one group.  Indeed, group formation in the GBR was a fluid and contested process with many overlaps and hybridizations.  For these reasons, identification of these groups was a useful procedure (1) to identify the groups that had formed, (2) to identify individuals who were (and are) associated with the GBR project, (3) to gain a sense that the groups were not static but underwent a process of formation, and (4) to learn that individuals moved between groups and were at times in more than one group at a time.  The latter point is crucial since, as Latour (2005) suggests, “actors are made to fit in a group – often in more than one” (p. 28).   82  Group Subgroup Coalitions # Indi- viduals # Letters Sent # Res- ponding # Inter- viewed Environ- mentalists (E) Mainstream and grassroots RSP 42 11 6 7 Forestry Industry (F) Major and small forestry, labour CFCI 29 13 9 8 First Nations (FN) Bands/Nations CFN, NC 74 18 4 1 Local Govern- ment (LG) City, town and regional district  26 3 2 2 Provincial Govern- ment (PG) Ministries  25 5 5 5 Stake- holders Sectors  33 1 0 0 Consultan ts (C) Consultants and Academics  67 9 7 11 TOTALS   296 60 33 34 Table 1: Interviews by Group   Throughout the dissertation, I identify interviewees’ “location” within the emerging networks using the abbreviations given in the left-hand column of Table 1.  In order to ensure confidentiality, I have used a coding scheme that hides the interviewee’s identity. Identification follows this format: position – coded identity: page number of transcript. As an example, this identification scheme will look like the following: E – NT: 156. Next, I reduced the number of individuals in my sample by consulting with a key participant who was (and remains) deeply involved with the project, both from provincial government and First Nations groupings.  This person helped me identify key individuals  83 on the basis of degree of involvement, influence, representation of wider constituencies, and time period.  I subsequently shared my reduced list with the individual, who then stratified each grouping into “top” and “bottom” lists.  To deal with the possibility of bias that could occur from using the help of one of the individuals in the process, I asked all of my interviewees who they thought would be important people to talk to.  I found that the original list that I developed was reinforced by interviewee responses. Expecting a response rate of approximately 60% and with a goal of conducting approximately 35 interviews, I pared the “top” list down to approximately 60 names and sent out 60 letters.  Of these, 33 people responded, yielding a response rate of 55%.  Of these, 6 were environmentalists, 9 were forestry industry, 4 were First Nations, 2 were local government, 5 were provincial government, and 7 were consultants.  However, I was able to talk to four individuals who did not reply, but whom I met at meetings which I attended (see below) and at which they agreed to be interviewed.  In addition, I was able to add 5 more interviewees through referrals and in-person requests.  By contrast, there were 6 people who responded to letters with whom I did not conduct an interview.  In one case, logistical reasons prevented an interview, in two cases we decided that the interview was not necessary, in two other cases, the contact went “cold,” and in a final case, I was unable to obtain band council approval to conduct the interview. In total, I conducted interviews with 34 people, made up of 7 environmentalists, 8 forestry industry, 1 First Nations, 7 local and provincial government, and 11 consultants. Response rates for First Nations was very low, with only 4 out of 18 responding to my letter, and with only 1 First Nation person interviewed.  However, 4 of the consultants interviewed worked exclusively for First Nations organizations, providing them with  84 technical support and representing their interests in land use planning forums and negotiations with other parties.  This brings the number of individuals interviewed within the First Nations category to 5.  Moreover, there are two main First Nations organizations that together represent the interests of almost all First Nations on the coast: Coastal First Nations and the Nanwakolas Council.  I interviewed key individuals who represent these organizations.  Stakeholder sectors such as mining and tourism were not included in the scope of the research for logistical reasons and for the reason that the actors consulted did not identify these sectors as key players in contrast with ENGOs, forestry, First Nations, communities (represented by local government), and the Provincial Government.  One individual among these stakeholders was identified as a key individual, but did not respond to my letter requesting his participation in my study. Interviews were conducted face-to-face at the organizations where the individuals worked, or at a place of convenience such as a coffee shop.  In a few cases where it was not possible to meet face-to-face, interviews were conducted by telephone.  I designed an interview guide template around 13 topics, including description of the area; history of the case; key groups; the GBR project; key actions; environmental campaigns; group negotiations; LRMP history; the LRMP process; First Nations group formation; the government-to-government process; ecosystem-based management; and the conservation economy.  The full guide was not used in individual interviews, however.  Some topics were more relevant to some interviewees than others, such as First Nations group formation to First Nations and the environmental campaigns to environmentalists. Accordingly, I created interview guides that were tailored to each group.  Next, since each individual had a different history in the process and engaged in it with different  85 skills, knowledges, and interests, I designed an interview guide for each individual before the interview.  I based these guides on the individual’s grouping and what I knew of their history and position in the process.  For example, if I knew that they sat as an environmental representative on the LRMP process, I would ask them specific questions about the LRMP process.  By contrast, if they belonged to the environmental grouping but did not engage with the LRMP process, I asked only general questions about it. Some texts on research methods advocate that researchers ask the same questions of all interviewees.  Such an approach allows the researcher to (1) systematically compare responses, and (2) relate differences among responses to other observed differences.  Such a structured approach to research assumes that there is a given reality – either pointed out by interviewees’ responses or in the responses themselves – that can be explained by the researcher.  This approach tends toward a “sociology of the social” (Latour, 2005b) which invokes social forces and structures to explain the behaviour of the subjects of a study. ANT, on the other hand, engages in a “sociology of associations,” following the actors themselves.  Interviewees’ responses to questions are analyzed as traces of processes of network formation and are not analyzed to find underlying forces. Thus, since the goal is not to systematically compare responses to analyze differences and offer an explanation, the same questions do not need to be asked of every respondent. Instead, I took an unstructured approach in my research design.  According to Maxwell (2005), unstructured approaches,  allow you to focus on the particular phenomena being studied, which may differ from others and require individually tailored methods.  They trade generalizability and comparability for internal validity and contextual understanding, and are  86 particularly useful in revealing the processes that led to specific outcomes.  (p. 80, emphasis in original)  From the ANT perspective, I recognized that interviewees were located in unique positions with respect to the actor networks that they were in part constructing.  Thus, the focus was on learning about these networks from their perspective.  The interview guide served to direct conversation towards certain topics, such as EBM and the conservation economy.  However, the interviewees were free to discuss processes and issues that they considered to be important.  Indeed, I found that most of the interviewees were so knowledgeable and so invested in the process that I would only need to provide an introduction before they would take they lead and launch into impassioned accounts of the processes and events involved in the GBR.  Interviews generally lasted for approximately 1 hour.  They were conducted between September 2007 and January 2008 in Vancouver, Victoria, Nanaimo, Campbell River, Port McNeil, and Port Alice. In addition to collecting textual and audio-visual material and conducting interviews with key participants, I engaged in a limited degree of observation.  This consisted of attending two meetings of the Ecosystem-Based Management Working Group in Vancouver and Victoria, and attending a conference at which actors presented the GBR in Vancouver.  While limited, the observations that resulted gave me a sense of how some of the main actors relate with one another in group settings and how decisions are made.  Also, I collected textual materials while in the field doing interviews. Generally, the interviewees offered these to me.   87 3.2  Analytic Procedures   Ideally, analysis of materials commences as soon as collection begins (Maxwell, 2005, p. 95).  According to Maxwell (2005, p. 96), there are three main strategies for analysis: memos, categorizing strategies, and connecting strategies.  Memos are researcher notes that are used as a means of reflecting on and analyzing research materials, theory, methods, and observations.  They are recorded at all stages of the research and organized in a format that allows for easy retrieval.  “Categorizing strategies” are ways of reducing data into manageable units by grouping information into categories.  These may take the form of: organizational categories, in which data is sorted into broad areas or issues; substantive categories, in which data are organized into descriptions of different topics, concepts, events, and so forth; and theoretical categories, in which data is placed in a more general or abstract framework (Maxwell, 2005, pp. 96- 97).  Connecting strategies are “attempts to understand the data … in context, using various methods to identify the relationships among the different elements in the text.” The point is to “look for relationships that connect statements and events within a context into a coherent whole” (Maxwell, p. 98). I applied these procedures at each stage in data collection and analysis.  As noted above, I collected a great deal of textual and audio-visual material.  During this process, I created organizational categories in which to place the materials.  For example, I organized texts into “grouping,” “process” and “topic.”  Each category includes subcategories.  For example, the topic category is subdivided into “agreements,” “campaigns,” the “conservation economy,” “ecosystem-based management,” and the  88 “spirit bear.”  Each of these categories has further subdivisions and so on.  These organizational categories served the practical purpose of organizing a vast quantity of material into a manageable system, facilitating retrieval of specific types of data. Additionally, it served the analytic purpose of learning from the actors themselves what the main pieces and elements of the GBR project consist of. As I collected and engaged with this data, I wrote memos on individual topics, key issues and themes, conceptualization, actors, actions, research focus, and research design.  These memos helped me understand the features of the GBR project and how I was engaging with it.  Moreover, I began to draw maps of relationships among the elements of the GBR.  These sketches served as a connecting strategy, which is the key strategy for actor-network theory.  They helped me get an initial sense of how elements became connected over time and what arose as a result of some of the transformations. All of these memos and sketches became resources on which I drew to follow the controversies involved in the GBR, to trace the chains of translations involved in the settlement of these controversies through the obligatory passage point of EBM, and to represent the “collective” that the GBR actors were together forming.  Furthermore, since my research project is also an actor that has become linked to the GBR collective, these memos also serve as sources of data that I will draw on in the following chapters. I also wrote while I was in the field conducting interviews.  First, this involved adapting the interview schedule for each interviewee.  As mentioned above, this process was guided by my evolving understanding of the elements of the GBR project and how they were associated, and by my knowledge of the individual’s position in the network. Second, I wrote notes as interviewees were talking.  Third, I wrote memos reflecting on  89 the interviews, the interview process, and broader ideas and relationship mapping that were prompted by the interviews or other reflections.  I collected these notes and memos in three notebooks. I used the qualitative analysis software Atlas.ti to code the interviews.  Since I did not come to the field with predefined theories or conceptual frameworks, I chose an “open coding” process (Strauss et al., 1990) in which I read the interviews and developed codes from the text as I went along.  Codes were primarily organizational and descriptive, with primary codes including: actors, agreements, EBM, ecology, economy, governance, groups, ideas, knowledge, law, land use planning, people/nature, relationships, and “vision.”  Each code included sub-codes.  For example, “ecology” encompassed: conservation management, ecological integrity, old growth, operating areas, protection, and risk.   Some of these sub-codes were further subdivided.  For example, “protection” encompassed: conservancies, moratoria, and precaution.  With the large number of codes, sub-codes, sub-sub codes, and so on, I developed over 260 codes in total.  However, there are only 15 first order organizational and substantive codes.  One first order code – “ANT codes” – functioned as a theory code, seeing as it included 15 concepts related to ANT. Since the purpose of the coding exercise was not to rearrange the data into categories in order to facilitate comparisons between things in the same category but to group information into descriptive categories so as to learn how elements are connected, a finer degree of precision and a larger-than-normal number of codes is warranted.  3185 codings were applied to 1885 quotations. I used Atlas.ti’s “network view” function to visually arrange and connect the descriptive codes with one another.  This function is primarily designed as a theory-  90 building device, in which codes (representing concepts) are connected with one another. However, as my codes are primarily descriptive, the linking function allows me to visually represent elements that are linked in the field.  These network views complemented my sketches of relationships, allowing me to trace the formation of the GBR network and, in Atlas.ti, to link that formation to quotations that talked about it. Finally, I also engaged in writing as I coded.  First I attached comments to quotations.  The comments contextualized the quote and highlighted issues that I believed to be important about the quote.  Second, I wrote five (self-defined) categories of memos: application, commentary, method, queries, and theory.  In the first category, I wrote memos that applied ANT concepts to the empirical material.  The “commentary” memos commented on particular aspects of the case, such as the market campaign and protected areas.  “Method” memos focused on issues like the work plan, things I needed to do, thoughts about how I might structure the report, reflections on the coding process, and so forth.  “Queries” included memos on questions that I had about the material and case, and things that I needed further information on.  Finally, “theory” memos reflected on ANT concepts, often with the empirical material serving as a source of prompts and examples. As for the memos written during the collection of textual and audio-visual material, they will also serve as sources of data.      91 3.3  Validity   The concept of validity refers to “the correctness or credibility of a description, conclusion, explanation, interpretation, or other sort of account” (Maxwell, 2005, p. 106). The goal is not so much being “right” (as in accurately reflecting an observer- independent reality), but adequately dealing with the possibility of being “wrong.”  That is, the research study must deal with alternative explanations and other sources of validity threats, not by attempting to make them disappear, but by describing how they may influence the conclusions.  A key threat to validity in qualitative research is researcher bias (Maxwell, 2005). First, bias in the form of conclusions that conform to the researcher’s preconceptions needs to be addressed.  It is thus important to explain any possible biases and how they may influence the research directly.  My interest in the GBR derives from my long interest in and commitment to environmental issues, as well as my connection to British Columbia as a place.  In addition, I am influenced by social constructionist approaches to environmental issues that consider the society/nature dualism to be a key cause (one of the key roots) of environmental problems.  As such, I am committed to exploring cases in which “nature” and “society” are constructed as categories.  However, I also believe that social constructionist approaches are insufficient because they do not consider the role of nonhumans, and am desirous to uncover cases in which the assumed society/nature dualism is questioned and transgressed.  I have chosen the GBR to study for precisely this reason.  The changing nature of environmentalists’ strategies, conceptions of nature, relations with First Nations, and models of resource management  92 disrupt, I believe, the society/nature dualism and associated dualisms.  In addition, this is the reason why I draw on actor-network theory to inform my methodology. With respect to analysis, I expected that the case would reveal ways in which the “modern constitution” (Latour, 1993) does not hold.  However, I was open to the possibility that the conflict revolves around conventional distinctions between society and nature.  Indeed, as I show, I found negative instances (Creswell, 1998, p. 196) where “society” and “nature” were employed as conceptual categories.  However, I did find evidence that early attempts to purify heterogeneous networks into the categories of nature and society were replaced with attempts to assemble humans and nonhumans in a collective.  In addition, I am aware that other perspectives on the conflict – such as purely political, economic, or cultural ones – are just as plausible.  Thus, I do not make the claim that the “collective dimensions” are the real or true ones, or that they are more important than other perspectives.  By contrast, I offer a network approach to the study of INRM as a complement to other studies and suggest that it provides a possible, and useful, way of understanding new associations of humans and nonhumans in such projects. Moreover, the validity of my descriptions is bolstered through “triangulation,” or the collection of “data that are detailed and varied enough that they provide a full and revealing picture of what is going on” (Maxwell, 2005, p. 110).  I engage in “thick, rich description” (Creswell, 1998, p. 196) to convey this picture, giving the reader enough information to decide whether my presentation and analysis are plausible.  Finally, by “following the actors themselves,” I reduce the influence of researcher bias.  By withholding presuppositions about the field and, instead, deploying controversies over actors and action, I have provided a venue in which the story of the GBR can be told.  93 However, I am aware that I am the one telling the story, and I do not want to convey the impression that I am trying to transparently reflect an observer-independent reality.  My perspective on the GBR is just that: one perspective.  The plausibility of my perspective, on the basis of the validity checks detailed above, is up to the reader to decide.  94 4  Problematizing the Coast: Shifting the Terrain of BC’s Wilderness Conservation Movement   4.1  Finding British Columbia’s “Forgotten Coast”    When I arrive on the mid-coast in the early 1990s (here, in this report, in 2009), the first thing that I notice is an absence.  This is the place that would become the “Great Bear Rainforest” but there are no maps, no books, no news releases, no websites to tell me so.  There isn’t even an agreement – or a failure to reach an agreement.  As far as environmentalists go, no one seems to be around.  Environmentalists don’t seem to be interested in this place at all.  Where is everybody?  I could ask Wayne McCrory, because he’s here.  Unfortunately, a decade and a half later, I never got a chance to talk to him.  We tried to connect, but there were four months of work between us and a potential telephone interview, as well as a crowd of “post-graduate students studying the process and outcome of the coastal land-use issues in B.C.” vying for his time (personal communication, August 29, 2007).  But I know some things about what he was up to on the coast.  McCrory had been active on the coast since the mid-1980s trying to encounter – and then protect – a white, black bear.  In an email communication, he told me,   95 Along with Dr. Stephen Herrero and Ralph Archibald, a provincial bear biologist we flew into a valley called the Khutzeymateen in a late fall storm in October in about 1985 after I/Valhalla Society received an anonymous envelope about a famous bear valley about to fall to the chain-saw and therein the story begins (personal communication, August 29, 2007).  This story, like all stories, is multiple and mostly untold, so I can only scratch at a part of it (but that’s all I really need in order to tell my story).  One of McCrory’s stories has to do with his efforts to create Canada’s first and only grizzly bear sanctuary, as well as coastal BC’s first protected estuary (Khutzeymateen Grizzly Bear Sanctuary was created in 1994 on BC’s north coast).  But this is just the lead-in to another of his stories, which involves his and fellow “ursaphiles’” (McCrory, 2003) encounter, in another part of the coast, lower down in Kitasoo and Gitga’at territories, of a white bear.  A press release in which McCrory’s environmental group, the Valhalla Wilderness Society (2006), applauded the announcement of the 2006 agreement, stated:  Eighteen years ago McCrory and a few of his colleagues were awe-struck by their first sighting of a white bear on Princess Royal Island.  “We saw bears and salmon in every big and little valley, cathedral groves of giant Sitka spruce, and wolves on the beaches,” says McCrory. “It was UNLOGGED and about as close to a wild bear heaven you could ever find on this earth.”  Even earlier than McCrory, another non-native person took up an interest in the mid-coast’s white bears.  In 1905, W. T. Hornady, a naturalist from the New York Zoological Society, described a white bear that had been spotted in a range spanning from River’s Inlet in the South, to the Nass Valley in the North, and east up the Skeena River to Hazelton, but principally concentrated on the islands and adjacent mainland on the north central coast. Hornady named the bear after Francis Kermode, an assistant to  96 the director of the BC Museum of Natural History, calling it Ursus americanus kermodei, or Kermode American Black bear.  Hornady considered the bear to be a distinct species, but in 1928 it was reclassified as a subspecies of the black bear that contains a unique double recessive gene endowing what would otherwise be black bears with white fur. In the mid-1980s, McCrory and his colleagues searched for white Kermodes. After the “awe-inspiring” encounter in 1987, they developed a proposal for a conservancy, calling for the protection of 262,000 hectares on Princess Royal Island, smaller islands, and several valleys on adjacent mainland areas.  They termed the proposed protected area the “Spirit Bear Conservancy,” coming up with a name and image that would have a large impact on the future development of the GBR and the Coastal Land Use Agreement (CLUA).  According to McCrory (2003), “The spirit bear became the international poster icon of the whole Great Bear Rainforest Campaign.” McCrory and his “spirit bears” are an important story about the mid-coast in the early 1990s.  At the time, the area was still not called the “Great Bear Rainforest.”  Why do I refer to the region as the “mid-coast”?  I’m not sure if McCrory called it that: he was primarily focused on calling one area the “Khutzeymateen Grizzly Bear Sanctuary” (which is north of the mid-coast) and another area the “Spirit Bear Conservancy.”  I’m calling the region the “mid-coast” because that’s how it was known by somebody else who was interested in it: the BC Ministry of Forests (MoF).    MoF called it the “Mid- Coast Timber Supply Area.”  According to MoF (2002), a Timber Supply Area is “an area of Crown land designated by the minister of forests in accordance with the Forest Act and managed for a range of objectives including timber production.”  Thus, in the early 1990s, the region was connected to forestry legislation and the Ministry of Forests’  97 authority to determine how it was managed, including the issue of timber production. This is important because the management of the area, and its timber production, became significant issues.  Merran Smith (2006) indicated the importance of the name “Mid Coast Timber Supply Area” and its contested status in her speech at the 2006 CLUA announcement,  In 1996, when we started on this work, this area was known as the mid coast timber supply area, and the only value of this rainforest was dollars per cubic meter.  Today, the Great Bear Rainforest is valued as an ecological legacy.  This is another important story.  How did the region shift from a network of relations which were measured in monetary terms to one which is measured as an “ecological legacy”?  Of course, this is also a story of whether forestry business and ecological legacies are compatible with one another and whether their conflict is the only story to tell.  So why don’t I use any of the many First Nations’ terms for the areas making up the central and north coasts?  To tell you the truth, I don’t really know any of them.  But the reason for that is that they didn’t seem to come up.  Maybe they would have if I had the chance to talk to some of the First Nations people who live in the area.  I did send out letters to 18 First Nations people who were identified as connected to the issue, but I only got 4 letters back and ended up only interviewing one person who is actually First Nations.  I suspect that one reason for a poor response is that I sent letters rather than personally visiting the communities (a stipulation of the Behavioural Research Ethic Board).  Another reason may be that those folks I sent letters to simply were not interested in talking to me.  This would not be surprising, since First Nations have not always had the best relationship with the research community.  Indeed, communication  98 with one tribal council displayed quite a degree of distrust of my intentions, even though I received consent by one of its members for an interview and even though I sent it all of my ethical review materials and completed its research registration form.  As a result, I never gained council approval to interview the member I had been in contact with.  In some cases, contacts went cold, while, in yet other cases, I suspect that individuals never received a letter. Nevertheless, First Nations did not use any of their own names publicly, aside from using the phrase “traditional territories,” which is interesting in its own right.  For example, in the congratulatory words accompanying the CLUA announcement, KNT First Nations president Dallas Smith noted, “now our people have a more active role in how and where business is done in our traditional territories” (as cited in Government of British Columbia, 2006).  Similarly, Heiltsuk Chief Ross Wilson anticipated that “completion of the government-to-government land use agreements will ensure the well- being of the lands, waters and peoples within our Traditional Territories” (as cited in Government of British Columbia, 2006).  It makes sense that First Nations kept their own names to themselves, since ownership and the right to speak in one’s own name was also an issue.  Additionally, First Nations’ voices had been appropriated in the past. So here we are on the “mid-coast” in the early 1990s.  The Ministry of Forests is here, with its “timber supply area,” McCrory is here with his “Spirit Bear Conservancy” proposal, and First Nations are here with their “traditional territories.”  Otherwise, it is pretty quiet.  There are some outdoor enthusiasts having fun paddling around in a couple of recreation areas that were established in the mid-1980s.  99 Fiordland Recreation Area and Hakai Protected Area had been established on the central coast in 1987 as a result of Premier Bennett’s Wilderness Advisory Committee (The committee had been assigned to deal with the growing war in the woods – controversies over particular areas such as Mears Island, South Moresby Island, and the Stein Valley).  And then there was the Kitlope Valley.  Ecotrust came all the way up from Portland in the early 1990s to help the Heiltsuk oppose West Fraser Timber Co.’s plans to log the Kitlope Valley.  Some of BC’s wilderness preservation groups became involved, including the Sierra Club of Western Canada, the Western Canada Wilderness Committee, and Valhalla Wilderness Society (McCrory’s group).  They even managed to get it designated as a conservancy, to be co-managed by the province and the Heiltsuk, in 1994. Beyond these areas, there was really only one other small conflict in the region – the Koeye River Valley, adjacent to the Hakai Pass, in 1990.  However, this one got Ian and Karen McAllister interested, which is another important story.  MacMillan Bloedel was planning on logging the valley and a developer was planning on building a resort at the mouth of the river.  Ian’s dad, Peter McAllister, a former director of the Sierra Club of Western Canada, organized a sailing trip to the Koeye River, inviting bear biologists, photographers, journalists, environmentalists and his son, Ian.  According to Ian (McAllister et al., 1997), in the evocative prose nestled between the gorgeous photographs of his 1997 coffee table book promoting the “Great Bear Rainforest”:  On the return journey through Queen Charlotte Sound, everyone on board fell silent as the obvious question moved through us like electricity.  If the Koeye River could be so spectacular and yet so unrecognized, what about the eighty or  100 ninety other river valleys on the mainland coast that were still intact and unprotected? (p. 13)  This is a pretty good framing device to get us thinking about the coast as a whole.  And that’s what McAllister started doing, he says.  Moreover, that’s what his coffee table book is about – it lets us see, through photographs, maps and a narrative of his journey up to the BC-Alaska border and back down, “Canada’s Forgotten Coast” (the subtitle of his 1997 book).  Before he wrote it, BC’s central and north coasts were not a single thing – the “Great Bear Rainforest.”  Rather, there was just the “spirit bear,” the “mid-coast timber supply area,” First Nations’ “traditional territories,” and a few protected areas. These elements would be rearranged and given new meaning through the production of a thing that was to become the object of an environmental campaign.  4.2  Constructing the Object of Environmental Politics   What is the “object” of environmental politics?  Is it a material reality “out there” that needs to be brought “in here” and placed on the political agenda?  Is it a set of beliefs that remains forever “in here”?  The view pursued in this dissertation is that the object of environmental politics is a “thing” in the etymological sense of “a gathering or assemblage” (Latour, 2005a).  That is, the “thing” that comes to form the centre of environmental politics is produced when material, discursive, and collective elements are gathered in a format that provides this gathering some form of representation.  This representation is itself heterogeneous, at once scientific, discursive, and political.  As I  101 will show in this chapter, practices that mixed science and politics produced representatives of the central and north coasts in the form of a “coastal temperate rainforest” (representing a globally rare forest type), the “profound symbol” of the grizzly bear (representing a place of rugged beauty and grandeur), and as a satellite map that lets people see “how much of the forest was gone and what remained” (Sierra, 2008).  In other words, in this chapter I look at how environmentalists (re)defined, or “problematized” (Callon, 1986), the central and north coasts through material and symbolic means.  How did environmentalists define the coastal forests in a way that interested other groups enough to want to usher in a campaign to save them? Just as the “object” of environmental politics cannot be taken for granted, neither can its “subject.”  As noted above, there was no pre-existing group of environmentalists with interests in the central and north coasts.  This group and their interests had to be constructed to the same extent that the “Great Bear Rainforest” had to be constructed.  In the following, I trace processes that produced both the GBR as an object of environmental politics and the group – discussed more fully in the next chapter – that emerged to protect it.  These processes involved a “scientific” ecosystem mapping project, a “discursive” project to collect stories and images of the coast, as well as a “collective” project, woven through the other two, to shift the focus and interests of environmentalists.  I do not consider these processes to be attached to different spheres of reality.  Rather, I show that all processes include material, discursive, and collective elements. In this chapter, I show how these processes are linked together through what Latour (1999) terms “circulating reference.”  There is no huge gap for the central and  102 north coasts between the way they exist in themselves and their scientific and discursive representations.  Rather, what I show is that the central and north coasts are progressively transformed into the representation of the “Great Bear Rainforest” through a series of intermediary steps, each of which involves the application of a frame of reference.  I show how frames and formats are applied to mobilize the central and north coasts, transforming them along the way.  This is not a case of applying socially constructed ideas to a material reality, since the frames and formats consist of devices and data as well as metaphors and aesthetics.  Moreover, the material itself is heterogeneous, since it includes such things as land cover data, photographs, and maps.  Rather than split up this heterogeneous process into science, discourse, and politics, I trace a single chain that is simultaneously scientific, discursive, and political.  At each stage, the “material” reality of the central and north coasts is transformed by the application of a new frame, thus producing a new hybrid reality that becomes material for the next frame. This process does not achieve a transparent representation of an external reality, nor a social construction having more to do with ideas and values than materiality, but what Latour (2005b) refers to as a “panorama” (p. 183).  In other words, I show how the variety of elements produced in relation to the central and north coasts are assembled in such a way as to present a total view of the region.  This view presents the object of interest as a whole, showing what it fits into and what fits into it.  It is “Canada’s Rainforest,” a “coastal temperate rainforest” that is globally rare.  It contains beautiful creatures and it is connected to future generations.  These connections between the region and other places, times, and people are all staged in a single representation that can then be made to circulate to other places and times.  The difference between my account of  103 this representation and a social constructionist account is that I show how it is produced through simultaneously material and discursive means and, in the next chapter, how the representation circulates in socionatural networks.  4.2  Translation 1: The Coastal Temperate Rainforest   One of the central features of the GBR campaign, as we will see in the next chapter, was the claim that “one of the largest remaining tracts of ancient coastal temperate rainforest in the world is found in the Great Bear Rainforest on B.C.’s Mainland Coast” (Thomas et al., 1998). Where did this claim come from? Little evidence of origin accompanies this claim, since it was presented as a statement of fact (c.f. Latour, 1987). Did it arise through scientists’ discovery of a previously unrecognized forest type lying along BC’s coast? Or, was this a socially constructed representation projected onto the forest for political purposes?  Was it a case of science or of politics?  If we look at the practices involved in the production of the “temperate coastal rainforest,” we find that these questions involve false dichotomies between reality and representation and between science and politics. What we find, instead, are activist scientists who (1) constructed a reality that did not exist before these interventions, and (2) quite purposefully mixed science and politics. In the early 1990s, Ecotrust and Conservation International engaged in a “Coastal Temperate Rain Forest Mapping Project” that sought to define the “coastal temperate rainforest” and place it in its “global context” (Weigand et al., 1992).  As they write in the introduction of their 1992 report on the project,  104  This paper proposes a new biome, a subdivision of the previously acknowledged temperate rain forest type, the coastal temperate rain forest […] The decision to formally define this forest type grew out of an interest in placing the distribution and status of coastal temperate rain forests in a global context.  Like the tropical rain forests which have rightly received so much attention, these forests are an important part of our global heritage. (p. 1)  As this passage makes clear, this project was simultaneously scientific and political.  On the one hand, the project involved the identification and specification of a real thing out there in the world, the “coastal temperate rainforest.”  While it represents a “new biome,” the authors did not invent the coastal temperate rainforest; rather, it was “proposed” as a “subdivision” of a “previously acknowledged” reality: the “temperate rain forest type.” Nevertheless, the rainforest is a hybrid of concepts and substance: simultaneously form – the subdivision of a category – and matter – a previously acknowledged reality.  At the same time, the identification of the coastal temperate rainforest was the consequence of a “decision.”  The authors chose to “formally define” this forest type in order to link it to tropical rainforests which, they note, have “rightly received so much attention.”  The political traction of such a linkage is substantial; as noted by one interviewee:  Vicky Husband tells this fabulous story of, um. . .I think this actually happened in Rio when she was down there some time in the late ‘80s or one of the follow up, something like that.  And Michael Apsey, Mike Apsey is his name.  He was the head of COFI at the time, the Council of Forest Industries, and they were all over the tropical rainforest protection stuff, they though this was great.  COFI was up there saying, “We’ve got to protect the tropical rainforest, one is the planet, blah, blah, blah.”  Well, along with the crappy things that are here right?  So he was down there, she runs into him in the hallway somewhere and she made a comment to him, something to the effect of, you know, Canada has a rainforest as well, and one of these days, people are going to figure that out.  And she said, “he just went white because he knew what that meant.” [Laughter] [E – MN: 82]   105 Thus, the goal of defining the coastal temperate rainforest was both scientific and political: it involved the subdivision of an acknowledged reality that it was then possible to link to a politically charged issue.  However, this was not a simple project: the authors could neither simply point to the coastal temperate rainforest that was out there waiting to be discovered, nor invent it out of the blue.  Many steps were needed to progressively “load the world into discourse” (Latour, 1999, p. 96).  That is, the authors worked to translate the central and north coasts into the “coastal temperate rainforest” through a number of frames and formats.  Latour’s expression is useful here since it keeps us from thinking that the authors projected a representation onto material reality; by contrast, material reality was mobilized, enrolled, and shaped into new forms through definitions and devices, thus creating a materially-dense representation – one that can be traced back to its source. First, the authors (Weigand et al., 1992) needed to decide which elements would be used to define the coastal temperate rainforest, a difficult task since “within the scientific community there remain[ed] some discussion regarding a global definition of coastal temperate rain forest” (p. 3).  Indeed,  Forest classification schemes within the coastal temperate rain forest zone vary considerably from country to country.  Chile and Argentina, for example, classify their forests by geoclimatic parameters at the regional level, and by microsite and habitat at the community level. Tasmanian ecologists distinguish forest types biogeographically, according to altitude and climatic variables. (p. 3)  Given this level of uncertainty and even controversy over the features that define the coastal temperate rainforest, the authors were forced to make a decision.  Weigand (1992) of Ecotrust proposed the following “working definition” of the coastal temperate  106 rainforest: “areas between 32 and 60 degrees latitude, with the presence of vegetation (if not currently, then originally in a forested condition), with at least 2000 mm (80 in) of annual rainfall” (p. 4).  This definition frames and orders data associated with the coastal forests.  However, the frame does not simply consist of words, ideas, beliefs, and meanings that “socially construct” the forests.  Rather, it is a hybrid frame made up of heterogeneous materials: a political goal of inducing interest in a particular region, a working definition bringing provisional settlement to a scientific controversy spanning several countries, and geographic, botanical, and meteorological specifications. The frame enrols and assembles rain, space, and plant life – which had already taken the form or frame of tables and maps providing information about precipitation, vegetation, forest, and land-use. The result is the translation of particular areas of the earth into a forest type – the coastal temperate rainforest – which were then assembled onto a map to depict their global distribution. In Weigand’s judgment, approximately 30 million hectares of coastal temperate rainforest existed before major human alterations,11 an estimation that he depicts visually on a map (see Figure 2).   11 Subsequently, Ecotrust/Conservation International digitized the maps employed by Weigand (as well as other maps), using a Geographic Information System.  These maps were then overlain with digitized annual precipitation maps published by UNESCO. Their computer-based analysis arrived at 41 million hectares as the original extent of coastal temperate rainforests world wide.  The discrepancy derived from differences in precision between the two methods, the use of different maps at different scales, and the use of informants and ground, truthing in the manual but not computer method.  107  Figure 2: Original Global Distribution of Coastal Temperate Rain Forests Source: Wolfe et al. (1995).  Reprinted with permission.   Not only does this map situate the original locations and extent of coastal temperate rainforest in their “global context,” but by contrasting the vast white spaces with the small specks of green, it visually depicts the global rarity of the coastal temperate rainforest type. Against this backdrop, the authors could highlight the threatened nature of this forest. While they admit that “the total area of remaining coastal temperate rain forest is unknown,” they refer to anonymous “researchers [who] believe that 17.3 million hectares (42.7 million acres) or 56% of the total has been logged and converted to non- forest use”12 (Weigand et al., 1992, p. 5).  12 A later development helped to place the region in a “global context.”  In March 1997, the World Resources Institute (WRI) released a report entitled “Last Frontier Forests: Ecosystems and Economies on the Edge” (Bryant et al., 1997).  The report claims that only 20% of the world’s original forest cover remains, that only 3% of these forests lie in the temperate zone, and that temperate frontier forests are the most endangered of all. The report also notes that half of the world’s temperate forests are gone.   108  The report and its attempt to formally define a new biome as a subdivision of the previously acknowledged temperate rainforest type was the outcome of practices that mixed (among other elements) working definitions, precipitation, rhetorical devices (maps), and satellite imagery for the purpose of engaging in public debates about land use.  Is the coastal temperate rainforest real or constructed?  Is it science or is it politics? If we accept the first perspective, then we grant scientific representation a privileged access to reality.  According to this view, scientists go out into the world (or the laboratory) to “discover” facts about the world that always existed, independent of the work of scientists to uncover them.  Knowledge of natural processes only enter the political sphere after the fact, so to speak, through the work of scientific popularizers and activists who work to put scientific claims on the political agenda (Hannigan, 2006; Kranjc, 2002). By contrast, if we accept the second perspective, then we’re suggesting that people can only know the world through social processes of interpretation, and that science – rather than escaping these processes to directly access the world as it is in-itself – actively contributes to the production of frameworks of meaning (Irwin, 2001).  Despite claims of critics (Dunlap et al., 1994; Murphy, 1994), the latter perspective generally does not hold that there is no reality behind or beyond social constructions, but withholds ontological claims altogether in order to focus on epistemological questions (Burningham et al., 1999).  Thus, notwithstanding their differences, both perspectives entail that mixtures of science and politics occlude access to reality: the first as a form of bias; the second as an inevitable effect of epistemology.  109 However, a third possibility exists: reality is both constructed and real; indeed, real because it is constructed (Latour, 1999a, p. 127).  In this view, there is no huge gap between representation and reality that is either bridged by an accurate correspondence between word and world, or filled with social constructions.  Rather, the connection between word and world is made through movement across multiple gaps, or what Latour refers to as “circulating reference” (Latour, 1999a):  At every stage, each element belongs to matter by its origin and to form by its destination […] We never detect the rupture between things and signs, and we never face the imposition of arbitrary and discrete signs on shapeless and continuous matter.  We see only an unbroken series of well-nested elements, each of which plays the role of sign for the previous one and of thing for the succeeding one. (p. 56)  In Latour’s terms, the classification of the coastal temperate rainforest as a global forest type does indeed refer to BC’s coastal forests, but not through a correspondence between the concept “coastal temperate rainforest” and the coastal forests.  If the authors walked (or, more likely, boated) into BC’s coastal forests, they would not have seen, smelled, or touched a coastal temperate rainforest.  However, this does not entail that the authors are horribly biased, merely projecting an unrecognized social representation onto the forests.  Rather, as described above, there are a number of intermediary steps linking the word and the world, thus “loading” the coastal forests into the report on the coastal temperate rainforest. At each step – coming up with a provisional definition, collecting data on precipitation, vegetation and land cover, calculating an estimation of original coverage, translating that estimation into a visual depiction on a map – there is a gap that must be  110 crossed.  “Previously acknowledged” definitions of temperate rainforests must be sifted through and translated into a provisional definition of the coastal temperate rainforest; this definition must be translated into data; data must be translated into an estimate; the estimate must be translated into a map; the map must be translated into a call to action. At each step, the thing undergoes a transformation but nevertheless refers back to its former self: the map can be turned back into an estimate, an estimate back into raw data, and so on.  At the extremes are BC’s coastal forests and the concept of the coastal temperate rainforest.  However, it is the chain of translations between that links them together, the circulation in both directions across the multiple gaps of reference.  It is the entire chain – one constructed out of provisional definitions, rhetoric, data, and satellites – that provides the reality of the coastal temperate rainforest and its need to be protected. Particularly with respect to the latter feature, the reality of the coastal temperate rainforest attests to the fact that it came to serve as the unquestioned basis for other claims.  Its process of construction, the controversies at the heart of the provisional definition became hidden from view in a black box (Latour, 1987) as the following key claims were made:  1) Coastal temperate rainforests are globally rare: “Coastal temperate rain forests constitute a relatively rare forest type, originally covering 30 to 40 million hectares, less than 1/5 of 1% of the earth's land surface” (Weigand et al., 1992, p. 1) 2) Globally, coastal temperate rainforests are threatened: “over half of the forests in this ecological zone logged or converted to non-forest use” (Weigand et al., 1992, p. 13) 3) BC’s coastal temperate rainforests make up part of the world’s original largest contiguous rainforest zone: “The Northern Hemisphere harbors the "Amazon Basin" of temperate rain forests. Here, in North America, the largest contiguous, coastal temperate rain forest zone on Earth ranges from the Alaskan Peninsula,  111 south through British Columbia and Washington state to Oregon's Siuslaw River” (Weigand et al., 1992, p. 5). 4) BC has the largest expanse of undeveloped coastal temperate rainforest in the world: “The largest areas of undeveloped coastal temperate rain forests in the more productive zone of this biome exist in British Columbia” (Weigand et al., 1992, p. 1)  4.3  Translation 2: Unprotected Watersheds   The frame that Ecotrust and Conservation International applied to the coastal forests translated the forests into a new format.  However, their work could not stop there. If their simultaneously scientific and political project was to be successful, they needed to enrol more elements than rain, vegetation, and geography: they also needed activists. Indeed, “coastal temperate rainforest” was not intended to be a neutral description, but a source of inspiration, a call to arms.  It was meant to circulate to get people interested in the region in a way that they had not been before.  But, if this was to take place, then another translation was required: one that would translate the new “biome” into terms that would interest activists.  The problem now facing the scientists was not scientific controversy over how to define temperate rainforests, but the environmental activists who were too busy focusing on individual, unconnected valleys in the southern portion of the province to take notice of a giant coastal temperate rainforest to the north (Stansbury, 2000; Wilson, 1998).  Activists were too busy chaining themselves to trees to see or save the rainforest.  The task facing Ecotrust and Conservation International was to shift the focus of the environmental movement away from reactive battles focused on the protection of individual valleys in the south, and toward a proactive, comprehensive  112 conservation vision for the largest remaining intact coastal temperate rainforest to be found anywhere in the world. This call took the form of a report commissioned by Earthlife Canada Foundation and Ecotrust/Conservation International, An Inventory of Watersheds in the Coastal Temperate Forests of British Columbia (Moore, 1991).  In the preface to the report, John Broadhead – a well-known BC environmentalist – writes:  To date, British Columbia has confronted the issue [of wilderness conservation] on a piecemeal basis … watershed by watershed … jobs versus the environment and “the last unlogged watershed.”  This paper is provided for those who are calling for a different approach – for a comprehensive land use strategy […](Moore, 1991, p. 2)  To help facilitate a new direction and focus for the BC wilderness movement, the author of the report, Keith Moore, translated the coastal temperate rainforest into terms likely to interest BC’s environmental movement.  As Moore (1991) notes in the introduction to the report, “in recent years, many of the conservation efforts in the coastal temperate forests of British Columbia have focused on the need to preserve entire, intact watersheds” (p. 4).  This statement at once recognizes the efforts of the BC conservation movement and the focus of this movement on valleys or watersheds as primary units for conservation, and places these efforts and watersheds in the context of “coastal temperate forests.” Thus, Moore translated the coastal temperate forest into terms that environmentalists are familiar with – watersheds – while at the same time shifting environmentalists’ focus “away from the ‘last unlogged watershed’ syndrome” (Moore, 1991, p. 4) and toward a comprehensive focus on an entire forest type. Toward this dual end, Moore’s report sought to answer a number of questions:  113  • How many undeveloped watersheds remain in the coastal temperate forest of BC? • Where are they? • What size are they? • How many watersheds are presently protected and where and how large are they? • What ecological units do these protected watersheds, and the remaining undeveloped watersheds, represent?  Like Weigand, Moore was forced to begin with a definition of the coastal temperate rainforest.  In a manner that aligns his definition with Weigand’s but is more adapted to the specific context in BC, Moore defined the temperate rainforest as a forest existing within the “Coastal Western Hemlock (CWH) and Coastal Douglas-Fir (CDF) biogeoclimatic zones, as described by the Biogeoclimatic Ecosystem Classification system” (p. 7).  The Biogeoclimatic Ecosystem Classification system was developed in the Department of Botany at the University of British Columbia in the 1960s and 1970s and adopted by the Ministry of Forests in 1976. Next, having defined the “areal extent of the coastal temperate forest” (Moore, 1992, p. 8), Moore proceeded to mobilize the forest from within, translating it into watersheds.  Moore strengthened the case for using the watershed as “the most logical unit to consider for the conservation of representative ecological units” (Moore, p. 4) by invoking other actors, including public conservation groups, parks system planners, researchers in biological processes, and conservation professionals.  These groups, he notes, argue that watersheds are representative of their ecosystems, are large enough to prevent fragmentation of wildlife habitat, and can preserve recreational and wilderness values (Moore, p. 4).  He then defined a watershed as including “all the land area draining into a stream system that has its terminus in salt water.  It is a complete drainage  114 area from salt water to height of land, including all the tributary drainage areas of the main stream” (Moore, p. 5).  To identify watersheds, Moore used the BC Ministry of Environment’s “Watershed Coding System,” as well as existing maps that identified watersheds on Vancouver Island and the Queen Charlotte Islands. The goal was to identify watersheds over 5,000 hectares in size, a number somewhat arbitrarily chosen but matching with the Wilderness Advisory Committee’s suggestion that 5,000 ha is “the minimum appropriate size for an area to be considered wilderness” (Moore, 1992, p. 4). Thus, instead of beginning with the “global context” as did Ecotrust, Moore begins with the local by drawing on systems and classifications made in BC, identifying watersheds, and working in the larger context. Through these methods, Moore identified over 600 watersheds that possibly exceeded 5,000 ha. By drawing them on maps and measuring their area with a planimeter – an instrument that measures area of an arbitrary two-dimensional shape – he identified 354 watersheds over 5,000 ha in size (a small number were determined from other sources). Moore then crosscut these watersheds with information about development status and protected status. With respect to development status, Moore conducted interviews with Ministry of Forests staff to determine whether and to what extent logging activity had taken place within each watershed. He then verified these interviews with air photos and detailed forest cover maps. Where evidence of limited industrial activity (logging roads, powerlines, pipelines, mining, settlements) was found, Moore measured the extent on the photos and maps. With this information, he defined watersheds as “pristine” (having less virtually no, i.e. less than 5 ha, evidence of industrial activities), “modified”  115 (having less than 2% of the total area affected by industrial activities), and “developed” (having more than 2% of the total area affected by industrial activities).  With respect to protected status, Moore drew the boundaries of protected areas on his maps to determine whether and the extent to which watersheds were encompassed by protected area boundaries. The result was a report, map, and table that identified watersheds larger than 5,000 ha and included information about development status and protected status (see Figures 3 and 4).  The report translated the coastal forest into 354 watersheds, 236 (67%) of which were developed, and 118 (33%) of which were either pristine (72 or 20%) or modified (46 or 13%).  Of the 354 watersheds, only 9 were fully protected and only 6 of those were pristine (3 were modified).  In other words (if modified areas can be considered conservation opportunities due to the limited extent of industrial impact) Moore’s report translated the coastal forests into over 100 conservation opportunities that could be identified as dots on a map.  Most of the conservation opportunities existed north of Vancouver Island, where, in comparison with the south coast, “many more undeveloped primary watersheds remain” (Moore, 1992, p. 19).  Of 174 watersheds in the southern portion of BC, only 14 were pristine or modified.  By contrast, of 180 watersheds in the central and northern coasts, 104 were pristine or modified.  116  Figure 3: Coastal Watersheds, South Coast Source: Moore (1991).  Reprinted with permission.  117  Figure 4: Coastal Watersheds, Mid and North Coasts Source: Moore (1991).  Reprinted with permission.   118 These maps presented the BC environmental movement – used to fighting over individual watersheds – with a veritable smorgasbord of conservation opportunities.  By comparing the south coast map with the mid and north coast map, environmentalists would be able to see at a glance that, by far, greater conservation opportunities existed to the north than to the south.  Moreover, a paired essay by Ecotrust and Conservation International (Beebe et al., 1991) links these individual conservation opportunities back into the “global context” of the coastal temperate rainforest.  As the authors write:  While the biota and productivity values in these watersheds are unique and in some cases exceptional, their ecosystem characteristics are nonetheless similar to the coastal forests of Chile, southern Norway, and Tasmania.  (Beebe et al., p. 37)  These ecosystem characteristics are, of course, those that qualify BC’s coastal watersheds as belonging to the “coastal temperate rainforest.”  Their essay, entitled, “The Coastal Temperate Rain Forest: An Ecosystem Management Perspective,” offers descriptions of this forest type’s rarity, threatened status, and remaining extent.  On this basis, the authors conclude, “British Columbia occupies a position of central, indeed global importance” (Beebe et al., p. 37).  Here the coastal temperate forest zone blankets 6.5 million hectares over the full length of the coastline.  It contains a wide variety of local forest types within the Coastal Western Hemlock and Coastal Douglas fir biogeoclimatic zones.  And in marked contrast to the US, significant opportunities still remain to protect large, unlogged and highly productive coastal watersheds […] These [watersheds] represent an important conservation opportunity – not just for BC and Canada, but for the world as a whole.  (Beebe et al., p. 37)   119 A great deal of work went into translating the coastal temperate rainforest into terms that might interest activists.  Biogeoclimatic classification schemes, a planimeter, interviews with MoF personnel, air photographs, and maps – all of these forms were applied to the central and north coasts to translate them into irresistible dots on a map. Was it successful? The Western Canada Wilderness Committee (WCWC) first indicated interest in this report.  The Committee used one of its “educational reports” (WCWC, 1992) to promote the idea of the BC coastal “temperate rainforest,” circulating the political potential of the coastal temperate rainforest and its watersheds to a wider audience.  When most people think about rainforests, they imagine steamy hot tropical jungles of South and Central America, Asia and Africa. But rainforests—lush forests that grow where precipitation is at least 2000 mm (over 6 feet) and is spread out relatively evenly over most of the year, are also found in temperate regions of the world. Temperate rainforests grow along a thin band of land where moist ocean air collides with coastal mountains. (WCWC, 1992)  The report repeats many of the same points made in Beebe and Wolfe’s (1991) essay, and provides details on Moore’s report, including advice on how to read his tables.  The report notes a few characteristics of the temperate rainforest, its rarity as a global forest type, that temperate rainforests have been greatly diminished in extent, and that BC provides the best conservation opportunity for temperate rainforests: “Only one third of BC’s primary temperate watersheds are still wild. We have a responsibility to all inhabitants of this planet, present and future, to set aside self-sustaining areas of temperate rainforest as wilderness, forever” (WCWC, 1992).  120 Within this wide-angle zoom, the report then moves on to provide details on opportunities to preserve particular coastal watersheds.  The report notes “in the southern-most regions of the coast, we have already lost the chance to protect whole, undeveloped watersheds over 5,000 ha in size. Only fragmented watersheds remain” (WCWC, 1992).   By contrast,  most of the remaining undeveloped watersheds in coastal BC are located in the North Coast region. Here there is the opportunity to create a huge protected area which extends from the coastal divide, and Tweedsmuir Park, all the way south to Fiordland Recreation Area (which must be upgraded to class A park) and the ocean. (WCWC, 1992)   Thus, the chain constructing the scientific and political reality of the “coastal temperate rainforest” became a little longer.  Through the mediation of a watershed inventory, a local, grass-roots environmental group induced its members to shift their focus from the southern part of the province, where “only fragmented watersheds remain,” to the north coast region where there exists an “opportunity to create a huge protected area.”  As they note, the impetus to create this protected area derives from a “responsibility to all inhabitants of this planet.”  This goal brings together international environmental organizations (Conservation International and Ecotrust) – with their focus on a “global” forest type – and a local grassroots environmental group – with its focus on specific, local watersheds, which, nevertheless, are of interest to the entire “planet.”  Is the effort global or local?  Neither; rather, there is a single chain connecting all of the actors.  If you look at any one link in the chain – Moore’s planimeter, Weigand’s working definition, the Ministry of Forestry’s biogeoclimatic classification system, Broadhead’s appeal, or WCWC’s educational report – you do not find global or local, you just find a  121 link in a chain.  Start removing links, and the “global context” of the coastal temperate rainforest becomes just an idea that is not connected to anybody, while WCWC’s members continue to focus on the “last unlogged watersheds” of the southern coast.  But link them together and “local” things in many locations start to become connected with one another.  4.4  Translation 3: Stories and Images    If one of the central claims circulated about BC’s central and north coasts was that together they comprised a coastal temperate rainforest, another was that this forest was a place full of life and beauty.  It was not simply a stockpile of resources – a “timber supply area” – but contained “lush rainforest valleys [that are] are home to some of the oldest and biggest trees on earth and provide critical refuge for grizzly bears, salmon and a rare snow-white variation of the black bear called the 'Kermode' or 'Spirit' bear” (Greenpeace, 1997d).  As noted by one environmentalist, one of the most significant victories was getting people to recognize the central and north coasts as the “Great Bear Rainforest”: “In 1996, when we started on this work, this area was known as the mid coast timber supply area, and the only value of this rainforest was dollars per cubic meter. Today, the Great Bear Rainforest is valued as an ecological legacy” (Merran Smith, 2006).  Where did this claim come from?  There is a tendency among sociologists to treat scientific and cultural claims separately (Callon, 1986).  Associated with this tendency is the view that science accesses the facts of nature – there whether we like them or not – while culture freely  122 constructs meanings and symbols.  While sociological analyses of scientific claims are thereby generally restricted to examining their social context – analyzing how economic, political, and ideological forces influence choice of research topics, for example, while leaving the content of the research unexamined – no such constraints exist for the study of culture.  As such, sociologists are free to study the “cultural logic,” or the “sets of institutionalized beliefs, practices and mythologies” (Rossiter, 2004, p. 141) behind environmentalists’ claims about nature.  According to this perspective, we could analyze Ian and Karen McAllister’s 1997 coffee table book, The Great Bear Rainforest, laced as it is with gorgeous photographs and evocative prose, for the ways in which it visually and discursively constructs the “Great Bear Rainforest” as a set of meanings that are imposed onto the landscape.  For example, the McAllisters (1997) write:  We never tire of watching [grizzly bears], because each bear has a unique personality and because their relationship with the forest is so uncanny.  At first the bears’ massive bulk and heavy armament seem out of place in an environment so soft and spongelike, but the grace with which the huge creatures disport themselves among all this fragile complexity is a virtuoso performance that we can’t stop applauding.  Sometimes on busy bear trails we find clumps of untouched wildflowers we swear they must be stepping around deliberately. Elsewhere, bears searching for root plants have ripped up estuary soils like bulldozers – which couldn’t be better for the estuary.  It is this multi-faceted relationship between the bear and the forest that we have found our most rewarding study, and if we dwell on it, it is because we find it the most profound symbol of what this ancient ecosystem is all about (pp. 25-26).  We do not have to search for very long to find the primary symbol that the authors are constructing: they explicitly point it out for us on Page 26.  Grizzly bears represent the “ancient ecosystem.”  They are ideal symbols, since we can relate to them (they are full of personality), we can admire them (they are beautiful, graceful performers), and we can  123 respect them (they carry out important roles in the ecosystem).  Moreover, the bears are represented as managers and stewards of the rainforest’s ecological integrity.  Strong and powerful, the bears are nevertheless gentle when they need to protect the rainforest’s “fragile complexity,” going so far (perhaps) as “deliberately” stepping around “untouched wildflowers.”  Delicate and graceful, the bears nevertheless unleash their massive power where appropriate, as when they contribute to the health of estuaries by digging them up “like bulldozers.”  The “Great Bear,” loaded with symbols and meaning, is put forward by the McAllisters as the ideal representative of the “Rainforest.”  In this representation, the word seems quite removed from the world.  It appears that the gap is crossed by a projection, an interpretation wherein the forests are socially constructed as the “Great Bear Rainforest.”  Yet, how is this representation produced?  Is it merely the product of visual and discursive rhetoric?  Where did these pictures and words come from?  Immediately below the McAllisters’ identification of bears as the “most profound” symbol of the rainforest, we find an excerpt from Ian’s field notes, accompanied by a full-page picture of a bear gazing into the eyes of the reader:  IAN’S JOURNAL:  I should have realized that the sudden flurry of gurglings and throaty cracks from the ravens above me meant that there were more life forms about than just me and the birds.  Suddenly the devil’s club and salmonberry bushes began to shake and I knew that within seconds huge claws would be digging into the mud of the well-worn bear trail where my gumboots were currently planted.  I backed off to the side about twenty feet, trying to decide whether to run, yell, play dead or pray to God, and finally chose, out of confusion mixed with curiosity and fascination, to do nothing.   I sank deep into the moss of an old spruce stump – bear spray in hand – and just watched as the big bear lumbered down the trail, nose up, and stopped in mid-stride right in front of me. We stared at each other across the sword ferns.  Salmon blood stained his mouth and he seemed well fed.  He did not seem alarmed at my presence.  The look in his eyes when they met mine was one of gentleness, almost sentience.  […]  Then the 225-kilogram bear lowered his head and passed on without even snapping a  124 twig, as beautiful as anything I have seen. (McAllister et al., 1997, p. 26)    Figure 5: McAllister's "Great Bear" Source: McAllister & McAllister (1997).  Reprinted with permission.    This excerpt places the reader on the edge of their seat.  The excitement of immanent danger – foreshadowed by images of shaking bushes, huge claws, and the blood stained mouth of a big grizzly bear – serves to draw the reader into the scene and its visceral  125 experience.  It also serves to highlight the bravery and humanity of McAllister.  He willingly puts himself into dangerous situations in which, like us if we were in his position, he does not know “whether to run, yell, play dead or pray to God.”  He does this to bring us the stories and experiences of a place that we would otherwise never see, a place that he selflessly is working so hard to protect.  The tension set up in McAllister’s narrative is resolved in a pleasantly unexpected manner.  The grizzly bear is not fierce after all, but gentle, calm and beautiful.  McAllister has nothing to fear: the bear moves on, gracefully as ever, a gentle giant who won’t even snap a twig, let alone wantonly disembowel a human.  Moreover, the encounter is not one of violence but of connection; a meeting of eyes and, perhaps, of minds – a meeting that the reader is invited to make by gazing into the eyes of the photographed bear who gazes back.  All of these discursive effects of McAllister’s narrative and accompanying photograph are worth analyzing.  These elements construct a particular set of meanings and symbolic connections between bears and us who, in turn, are constructed as the gentle giant representatives of the rainforest.  However, it is worth noting another feature of the text: it is a journal entry.  It thereby refers back to a different time and place. Similarly, the photograph obviously came from somewhere.  If you look closely, you can notice that the background in the photo consists of water, not forest.  Additionally, given that McAllister was too busy squeezing himself into a stump in fear for his life to snap a photo of his would-be killer, this shot is obviously of a different bear.  Yet, it is assembled in the text in support of Ian’s story.  Moreover, other elements, from different times and places, also help construct the “Great Bear” as a representative of the “Rainforest.”  For example, the McAllisters invoke claims from the science of  126 conservation biology that grizzly bears serve as the indicator species of ecosystems.  As they write,  We can look at healthy grizzly populations and have confidence that the integrity of the coastal ecosystem is intact and that the 230 bird species, 68 mammals, and thousands of insects and microorganisms that make their home in the old-growth forests are also healthy.  If the grizzly numbers start to go down, we can be sure that those other less visible values are declining too.  This is reason enough to focus worldwide attention on these bears.  (McAllister et al., 1997, p. 25)  The coffee table book is comprised of images, metaphors, and symbols, but these elements are assembled from different times and places in order to construct these representations.  Where did these images and stories come from?13  If we trace the history of these objects and the representations that they create, we can fill in the links in the chain connecting the coastal forests with the representation “Great Bear Rainforest” – just as I did for the “coastal temperate rainforest.”  These two representations – one scientific and the other cultural – do not have to be analyzed in separate ways, but may be considered symmetrically (Callon, 1986).  They are both black boxes that can be opened up if we attend to the practices through which they were assembled.  In fact, when we attend to practice, we find that the scientific chain (which was made up of a diverse mixture of data, provisional definitions, instruments, and rhetoric) is directly connected to the cultural chain (which, as I will describe below, is similarly made up of a mixture of science and politics).  The McAllisters’ interest in the central and north  13 Here, I will focus on the McAllisters’ photos and stories, rather than their mobilization of science, since I have dealt with science more fully in the preceding sections and since this is the main focus of the authors themselves, who write, “we have tried to include the basic information required to understand this vast and complex area, but our main purpose has been to express our own appreciation of it” (p. 16)  127 coasts, just like the WCWC’s, was prompted by Moore’s watershed inventory.  While a lot of the Vancouver Island stuff was going on in 1989, Keith Moore was contacted by Conservation International and Earthlife Canada and, I think, Ecotrust, to do a watershed inventory of the entire BC Coast.  He published that report about them, and it basically was the catalyst for our work up on the Central and North Coast of BC because it showed there was maybe half a dozen intact primary rainforest river valleys over 5,000 hectares in size on the Island, and yet, on the Central/North Coast it was just massive clusters of dots on the map showing many, many intact river systems [but] nobody was working up there at that time.  [McAllister – 020]  The McAllisters took up the relay offered by Conservation International and Ecotrust’s “coastal temperate rainforest” by translating Moore’s watersheds into stories and images. This was not simply a matter of inventing symbols and metaphors but – just as with the coastal temperate rainforest representation – it involved the hard work of loading the world into discourse (Latour, 1999).  Indeed, this work involved a fair amount of lay science, or what the McAllisters refer to as “inventory and research.” As they recount in their book, the McAllisters (1997) began to collect “as much information as we could about every dot on the map, scouting provincial and federal government offices and libraries.  The information barely filled a shoebox” (p. 15).  In response, they decided that they had to go out and collect information directly from the coastal forests.  In an interview, Ian McAllister recounts:  We didn’t know what was up there and it took a number of years just to do the basic baseline inventory and research.  And I mean basic – like, we were just running from valley to valley, looking at estuaries, trying to understand the status of salmon and bears, trying to understand the basic forest cover, just having the first real look from the outside world of these river systems and it’s a massive coastline.  [McAllister – 021]   128 It is one thing to take the coast in at a glance by referring to a map produced by Weigand or Moore, but to go back to the territory to which these maps refer and fill in the dots is another.  The same remoteness and ruggedness of the area that had thus far prevented large-scale forestry operations made it difficult to canvass the coast for conservation reasons.  The shear size of the coastline led to technical innovations on the part of the McAllisters, who first accessed the region by boat:  We’re talking over 10,000 miles of coastline.  [We were] going up and down these fiords and whatnot and your boat might only go 7, 8 knots and you can only cover so much ground in a day, so I spent a lot of time with volunteer pilots, flying up and down valleys.  Like old World War II fighter pilots, like Mike Humphries […]  [W]e would spend, literally, weeks and weeks in the air.  Day after day, landing in Bella Coola, and Prince Rupert, and Kitamaat, and all these places, with a video camera on the wing.  And I had the door off taking pictures. [We were] going up and down the river systems and then going back and cataloguing it and beginning to put together a piece.  But the more and more we did that, the more we realized that how extensive this coastline was. [McAllister – 022]  When the McAllisters visited the coast physically rather than through mediation on Moore’s map, the “massive” and “extensive coastline” was impossible to see synoptically.  If they were to take in more than a tiny fragment of the forest at a time, the McAllister’s had to work to displace their perspective (Latour, 1999, p. 66).  To do so, they enlisted the skills of a pilot trained for a war, the capabilities of a small aircraft, and the technology of video and still photographs.   This displacement is not different in kind from the dislocation of perspective achieved by Conservation International and Ecotrust via their global information system technology, or by Moore via his planimeter.  Rather, all three parties were forced to rely on devices to gain new perspectives on the coast which, in their absence, takes on the aspect of a “delightfully tangled up territory”  129 (McAllister et al., 1997, p. 44).  For their part, the McAllisters were able to mobilize the coastline into a transportable format that could be reviewed during any time away from the coastline itself.  By means of photographs, they could associate dots on Moore’s map with visual depictions of individual valleys.  By means of video, they could fast-forward, rewind, and pause on certain images of the coast as they catalogued its watersheds. The McAllisters spent the next five years sailing up and down the coast, “ground- truthing” [McAllister – 030] the rough information they collected from the air.  The couple populated Moore’s watersheds with waterways, salmon streams, estuaries, plants and trees, and wolf and bear populations.  Perhaps even more importantly, they collected stories and photographs (such as those reproduced above).  Their coffee table book is full of these stories and images, which take the form of narrative, journal excerpts, and glossy full-page colour photographs.  By the time that readers reach the end of the book, having virtually travelled with the McAllisters along the entire coast, they encounter a map listing the coast’s “endangered intact watersheds,” one not much different than Moore’s. However, in this case, each dot has been filled in with stories and gorgeous photographs. Having taken over the relay offered by Conservation International and Ecotrust, the McAllisters (and their environmental group, the Raincoast Conservation Society) worked to interest the BC environmental movement in the region.  However, before their book was published – during their period of “research and photography” (McAllister et al., 1997, p. 16) – most wilderness advocates were engaged in a battle to save the “last untouched watersheds” of Vancouver Island’s Clayoquot Sound.  According to McAllister,   130 It was almost surreal for us to be sailing quietly up on the Central and North Coast [along] these vast intact, unprotected and threatened river valleys day after day, and then reading or listening to the news [about] the massive blockades and the huge displays of civil disobedience happening in Clayoquot Sound.  In the back of our minds we’re wondering, “when are people going to pay attention to this Coast?”  It took a few years, for sure. [McAllister – 036]  Thus, while the coffee table book was directed towards a general audience who might be induced to support the aims of an environmental campaign, the McAllister’s first task was to help “usher in” such a campaign.  Largely, this involved deploying the same images and stories that would be compiled in the book.  As Ian recounts: “a lot of our earlier work was really just as a messenger, getting those tools [out] – those video and still images, and stories, and introducing people to First Nations, and just trying to usher in a campaign” [McAllister – 038].  They brought the information that they collected about the area to local, national, and international environmental organizations: “we were going to Europe each winter, and we were travelling around Canada and the United States, banging on doors and trying to get people engaged in the issues and the campaigns” [McAllister – 036]  4.5  Assembling a Panorama of the Coastal Forests    Three important translations of BC’s coastal forests took place between 1990 and 1996.  The place was scaled up into a coastal temperate rainforest located within a global distribution, drilled down into an inventory of unprotected watersheds, and filled out with experiences, stories and symbols.  While some aspects of these translations appear to be  131 purely scientific and others to be purely discursive, all of the translations blended material, discursive, and collective elements.  Moreover, they are connected together in a continuous chain.  Rather than by nonhuman reality “out there” which scientists discover or which environmentalists misrepresent, the coastal forests were mobilized by a variety of devices and techniques in order to influence how people connect with them.  Each translation created a particular depiction of the forests, connecting the latter through a single chain.  Yet, these depictions were not assembled into one synoptic, total view of the forests until the Sierra Club produced a composite map of the region. In 1996, the Sierra Club released a satellite map of coastal BC, entitled “Canada’s Rainforest – Worth Saving” (see Figure 6). The map is an aesthetically striking work of art, as captivating, in its own way, as McAllister’s photographs.  Richly saturated aquamarines, forest greens, and un-earthy yellows draw the viewer into the glossy illustration, offering a rare and unique view on a rare and unique place.  This is “Canada’s Rainforest”: large enough to extend along the entire west coast, but tiny – and therefore precious – when considered in relation to the vast size of the continent, as can be seen in the box to the right.  A small panel of complementary photographs tell the larger story depicted on the larger map.  A cedar tree so large the we can only see its trunk, salmon eggs viewed so closely that we cannot fail to recognize their importance, a grizzly bear so large that it fills the frame, and a First Nations carving so old that it appears part of nature itself – these representatives suggest what is “worth saving” in this rainforest.  There are no people in these photographs, only plants and animals.  132  Figure 6: Canada's Rainforest – Worth Saving Source: Sierra Club of BC (1996).  Reprinted with permission.  133   The presence of people in the rainforest is only hinted at by the old, weathered carving – the handicraft of invisible First Nations as ancient and premodern as the giant cedar tree. The first time we see the depiction of a person it is of a logger, his back turned to us, bent toward the task of cutting down a giant tree – perhaps the tree we see in the first picture. Pull back the perspective in the next photo and we can see the larger consequences of his work: decimation, disaster – a clearcut.  Whom does this impact?  The innocent future generations, as the last picture suggests.  The little girl, noticeably blond and white, has an obvious connection to nature and the forests, smiling innocently as she peaks out at us from within the very heart of an old tree.  These elements of the map thus represent the coastal forests in a particular way and I have given a particularly social constructionist reading of this representation.  But this is a hybrid map.  Other elements are present that resist reading it as a pure human construction.  In the text describing the “rare, unique, and threatened” status of the rainforest, we once more encounter the claim produced by Conservation International and Ecotrust’s rainforest mapping project: the rainforest covers “just a fifth of one percent of the earth’s land surface” with BC’s portion representing “almost one quarter of all that is left in the world” (Sierra Club of BC, 1996).  This scientific-political “factish,” to use Latour’s (1999a, p. 274) term for facts that have gone through traceable processes of fabrication, does its intended work by entailing that “here we have one of the best chances to conserve these wild and ancient rainforests, along with the grizzly bear, salmon and countless species that depend on old growth for their survival” (Sierra Club of BC, 1996), the latter of which we see in the photographs on the left.  134  Other features of the map refer to things that exceed a purely discursive reading. The most important features, of course, are the yellow and green areas.  They are aesthetically striking but they also refer to things beyond the map itself.  The key to this reference is given in the pie chart to the right.  Yellow areas refer to areas that have been “logged,” while green areas refer to the “remaining ancient forest."  The text accompanying the chart tells us “more than half [53.1%] of B.C.’s coastal rainforest is gone” (Sierra Club of BC, 1996).  If we conclude that the situation is good because we still have about 50% of the remaining forest, our feeling of comfort is quickly taken from us with the knowledge that only a “thin sliver” (see pie chart to right) “is protected for our children” (see photo to left), while what remains faces an unrelenting, violent onslaught, since “virtually all pristine valleys will have logging roads punched into them in the next two decades” (Sierra Club of BC, 1996).  Were these facts and figures simply conjured up for rhetorical purposes?  Are they simply transparent depictions of realities discovered by scientists?  What about the view from space?  Surely, this seemingly objective view is not what we would see if we were aboard the orbiting satellite. This map is the product of a chain of circulating reference, just like Conservation International and Ecotrust’s “coastal temperate rainforest” and the McAllisters’ photographs and stories.  Each element assembled in the map, from the colours, to the photographs, to the pie chart, has a history.14  Indeed, the mapping project itself has a history: it is the extension of an earlier mapping project covering Vancouver Island.  As  14 Or “historicity” in Latour’s (1999: 149-150) terms.  Latour uses this term to contrast the notion of history associated with a correspondence theory of truth – in which facts, if they exist, have always been there, outside history while history is reserved for humans – with a recognition of the history of things – involving changes in the series of transformations constituting circulating reference.  135 recounted by Braun (2002, p. 215), the Sierra Club produced two maps of Vancouver Island in the early 1990s which compared the extent of forest cover in 1954 to 1990.  The maps were presented as satellite images, seemingly objective snapshots of the island in two different time periods.  The purpose of the maps was to visually demonstrate the extent of “the disappearing forest” on Vancouver Island (reproduced in Braun, 2002, p. 215). Yet, according to Braun (2002), the images were not simple snapshots from space: they were the outcome of a great deal of work.  The images were computer-generated, with Landsat imagery (satellite photographs) forming only one source of information for the final product.  Other resources included forest inventories produced by the Ministry of Forests, “vegetation zone” categories produced by biogeographers and represented in maps of biogeoclimatic units, and colour schemes introduced by cartographers.  These sources were digitally combined with one another to produce the final image.  Thus, according to Braun (2002):  The images combined and translated material from multiple sources (satellite photographs, biogeoclimatic maps, forest inventories, air photographs), mixed these with the skills of technicians (photographers, computer programmers, cartographers), relied on the competencies of various instruments (computers, software, satellite technologies, cameras, printers), and drew on a set of guiding metaphors and concepts from sciences such as ecology. (p. 223)  The result of this hard work was a new actor: “reproduced in pamphlets, hung on walls, shown at rallies, and reproduced in the pages of newspapers and magazines, it [the map] helped to fuel a global campaign to save the ‘ancient rainforests’ of Vancouver Island” (Braun, 2002, p. 222).   Similarly, the Sierra Club’s “Canada’s Rainforest” map became an important actor, inasmuch as it helped “usher in” a campaign for BC’s central  136 and north coasts.  The map served to draw environmentalists’ attention to the coastal forests lying to the north of their most recent battles.  As the map made apparent, wilderness advocates had previously focused their efforts on protecting tiny pockets of green in an expanding sea of yellow in the southern half of the province.  Their successes were thus best viewed as small green wins in a wider yellow failure.  By contrast, the map directed environmentalists’ attention to an area where it would be more effective: the north.  According to the Club, the map:  Showed the extent of rainforest destruction on Vancouver Island, along B.C.’s south coast and its gradual extension northwards up the coast. It also highlighted the extensive intact areas that could still be saved in the Central and North Coast. For the first time, British Columbians could clearly see how much of the rainforest was gone, and what remained.  (Sierra, 2008)  Indeed, this was the first time that viewers could “see” the rainforest and its extent in coastal BC.  The map presented a panoramic view of the central and north coast, one that situated the coasts in relation to Canada, the world, future generations, forestry, animals, and the concerned viewer.  But, as described above, this was not an unmediated, objective vision; nor was it merely a representation with no connection to the reality beyond the text.  The visual depiction was an achievement produced through the assemblage – through chains of translations – of many different types of things, thus offering viewers a synoptic “God’s eye view” of forest cover data, ecological classifications, historical and contemporary logging practices, future generations, “ancient” metaphors, trees, bears, and salmon, and geography.  Much work went into the production of a map reproducible on a single, and thus highly transportable, piece of  137 paper.  This composite but single actor-network helped environmentalists “see” where they should focus their efforts.  A panorama is a kind of projection, a representation of a world or state of affairs. It provides a total view.  It does not provide a transparent representation of the world as it is in itself.  But this is not to say that it is merely socially constructed.  Rather, as I have detailed in this chapter, the construction of a panorama is a material undertaking as much as it is a discursive one.  Environmental politics applies to the assemblage of heterogeneous networks such as these.  Only if we ignore all the hard work in constructing this panorama can we conceive of a material reality “out there” over which politics is engaged in “in here.”  Rather, environmental politics involves the chaining of elements such as rain, vegetation, watersheds, experiences, stories, and people into a new quasi-object – “Canada’s Rainforest” – and a new quasi-subject15 – the viewer who believes that it is “worth saving.”   Moreover, the point of constructing such panoramas is not to transparently depict the world as it is in itself, but to induce others to act in particular ways.  It becomes a practical means of creating new associations.  Thus, care must be taken in the representation of the panoramas themselves – if they are to induce interest, they must be made interesting.  Sierra’s satellite map is visually compelling, yet, what viewers could see as obvious was still rather formless: “It became obvious there was this huge green blob.  In fact, we used to call it the ‘Big Green Blob’ before it was called the ‘Great Bear Rainforest.’  I mean, we didn’t call it that publicly, but that’s what we called it in meetings” [E – NT: 86].  The “Big Green Blob” was open-ended and ill defined.  15 I refer to these entities with the prefix “quasi” because they are the provisional result of networks made up of humans and nonhumans.  138 Satellites, land use data, mapping technologies, conservation biology, attempts to influence the focus of the environmental movement – all of these things came together to create an open-ended thing.  In Latour’s (2004, p. 247) terms, the “blob” was little more than a “proposition,” or an association of humans and nonhumans before it becomes recognized and instituted in a collective.16  At this stage, the possibilities for what this ill- defined thing was were still open-ended. However, the forms of thought that make up the modern constitution were at play and, while environmentalists and others worked to create this new hybrid network, they sought to prematurely purify it into nonhuman nature without proper debate, or what Latour refers to as “due process.”  Ecotrust and Conservation International worked to produce a new forest type – not to connect the coastal forests more intimately with people, but to protect them from people.  The McAllisters worked to populate the coastal watersheds with images and stories to argue for their preservation.  WCWC popularized Moore’s watershed inventory and the concept of the coastal temperate rainforest to promote the idea of creating a giant park off limits to development.  The Sierra Club produced its map to help people see that Canada has a rainforest and that it is “worth saving.”  Indeed, while the processes involved in the above endeavours associated heterogeneous materials in novel formations, these realities were purified to present a wilderness that had to be protected from people. These practices of purification are evident in the first name given for this “Big Green Blob”:  McAllister referred to it as the “Great Bear Wilderness.”  The term “wilderness,” as numerous scholars have pointed out (e.g. Cronon, 1996b), presents a  16 I will discuss the concept of proposition more fully in Chapter 3 in connection with “ecosystem-based management.”  139 view of nature as nonhuman.  However, this purification was resisted even at this stage. As McAllister notes,  If you look at our original conservation [proposal] for the Coast, it was called the “Great Bear Wilderness,” but we actually got a fair amount of push back from First Nations.  You know, you read the Webster’s Dictionary and its definition of “wilderness” is actually quite clear, it says that it’s void of humans, right?  […] [McAllister – 56-60]  First Nations, as we will see in Chapter 5, resisted definitions of “wilderness” that excluded their cultural, jurisdictional, and economic interests, “so it [the name] got changed to the ‘Great Bear Rainforest.’”  The name-change took place at a meeting between the McAllisters and a Greenpeace representative in a restaurant in San Francisco in 1996:  I still have distinct memories of Karen [McAllister], Ian McAllister and myself [Tzeporah Berman] sitting in a restaurant in San Francisco in 1996 writing on a paper tablecloth – Great Bear Wilderness? Raincoast Wilderness? Northern Rainforest? Coastal Rainforest Wilderness? And the moment when we wrote Great Bear Rainforest. We all knew immediately that was it. And we were damn sure that Great Bear Rainforest was going to solicit more concern that the “mid coast timber supply area,” which is what the region was known to us up until we launched the campaign.  (Berman, 2006a)  The name “Great Bear Rainforest” is, as with any name, a representation.  Its “words” refer to the “world.”  Yet, as we have seen above, the world is “loaded into” discourse (Latour, 1999a, p. 24).  In one sense, the name was “dreamed up” by three people sitting in a restaurant.  However, the words chosen were not empty signifiers that could be freely loaded with a “cultural logic.”  Rather, the words were already dense with signification.  140 The term “Great Bear” refers to the “profound symbol” of the grizzly bear, which, as the McAllisters (1997) note in their book, is a natural representative of the forests due both to its gentle yet industrious role in their functioning, and to its ability to function as an umbrella species within the science of conservation biology.  However, the “Great Bear” serves as an umbrella of its own, encompassing not only grizzlies but the Kermode or “spirit bear,” for whom the most significant conservation proposal in the region (the “Spirit Bear Conservancy”) existed at the time of the campaign.  Both bears were loaded into the name – not only as discursive symbols, but also as representatives of ecosystem integrity and as an existing conservation proposal.  Nonetheless, as “charismatic megafauna” (Leader-Williams et al., 2000), they served well as “poster icons” (McCrory, 2003) for the campaign.  The term “Rainforest,” as analyzed above, is simultaneously scientific and political: a subdivision of a previously recognized forest type that links up with existing international concern over tropical rainforests.  Linking these two terms together, the name “Great Bear Rainforest” is at once rich with meaning, dense with reference, and the product of relations of power. This gave the region a particular kind of figuration, which gave it a personality in its own right (Latour, 2005b).  As one newspaper commentator noted, “a nondescript and emotionally neutral region of British Columbia known as the central and north coast timber supply area entered the public's imagination as a personality worthy of ecological recognition” (Gigg, 2006). The GBR was not already out there waiting to be discovered by scientists, but was merely “nondescript.”  Neither was it a blank screen waiting for a discursive representation to be projected onto it; it was merely “emotionally neutral.” But afterwards, through the processes of circulating reference that involved working  141 definitions, devices such as planimeters, machines such as satellites, metaphors such as “rainforest,” stories, and photographs, the coasts came to possess a “personality worthy of ecological recognition.”  The next step was to (1) articulate the GBR in a manner that would enable it to be “recognized” in a specifically “ecological” way, and (2) introduce this “personality” into wider networks in order to solicit concern and enrol other groups in the project of recognizing the GBR.    142 5  Generating Power: Spokespeople, Interessment and the Market Campaign   In the previous chapter, I examined how environmentalists problematized the central and north coasts, transforming them into a reality that could be made to interest others. Specifically, I traced chains of associations that translated the coastal forests into a new form – the “Great Bear Rainforest.” While the last stage of this process involved the production of a new name, this process was not simply about the power of naming, but about simultaneously material, discursive, and collective practices that sought to transform the forests into an evocative, scientifically-defensible conservation opportunity that would be of interest to environmentalists. The original name given to this new network – the “Great Bear Wilderness” – attempted to purify and prematurely close off debate on the area by representing it as nonhuman nature. However, First Nations contested this representation, indicating that they would have a role in defining what the area would become. The revised name – the “Great Bear Rainforest” – presents a panoramic vision of the coastal forests, a total view that connects the forests to the globe, to animals, to future generations, and to viewers who agree that it is “worth saving.” It presents a vision of the future wherein these relations would be realized; thus, while it does not transparently depict the world as it is in itself, it usefully previews the world to come (Latour, 2005b, p. 189). The protection of “Canada’s Rainforest” cannot happen in the panorama itself since it remains only a picture of what is to be achieved. More work – and more translations – was required to achieve this end.  143  In this chapter, I examine the ways in which environmentalists worked to define the identities and interests of groups that they needed to achieve their vision, and the “interessment” devices environmentalists deployed to enrol allies into their network. This analysis looks at environmentalists’ attempts to enrol the BC wilderness preservation movement, First Nations, nonhumans, and forestry companies into their vision of the forest. I look at how the wilderness movement was redefined as a coalition interested in comprehensive, rather than valley-by-valley, conservation through the mediation of personal growth and healing. Specifically, I look at the formation of the Canadian Rainforest Network (CRN) subsequent to a retreat designed to address activism burnout. The analysis of attempts to enrol First Nations is more implicit. I show how First Nations’ rejection of environmentalists’ attempts to act on their behalf involved an implicit definition of First Nations’ interests as synonymous with environmentalists’ interest in environmental protection. The interests of environmentalists and First Nations did indeed overlap, but were by no means synonymous. Moreover, I look at environmentalists’ attempt to enrol First Nations in their campaigns through standing together with First Nations in blockades in Nuxalk territory and blockades in Kitasoo territory. Again, I find that this attempt at interessment was uneven, with some First Nations travelling to Europe with environmental activists but many others rejecting environmentalists as modern day colonialists. I also investigate how environmentalists defined nonhumans and the means by which they attempted to enrol them into the emerging network. In particular, I look at how the “problematized” panoramic vision of the central and north coasts was given voice through the designation of an authorized spokesperson for the forests and their  144 inhabitants. Who better to represent the Great Bear Rainforest than the Great Bear itself? But how can a bear “speak” for others? Nonhumans obviously do not speak in the ways that humans speak, and the reader will note that I do not include any direct quotations from bears. However, in this dissertation, I do not consider speech to be a capacity that is “held” by some types of beings. Rather, I consider speech to be a process or event in which multiple types of beings participate. Specifically, I look at how the bear was produced as a hybrid actor in relation to the science of conservation biology. In particular, I look at how conservation biologists applied the concepts of the course filter and umbrella species to authorize the bear to speak for others in the emerging network.  I analyze how this bear was made to circulate along with environmental activists during their campaign to protect the GBR. The point of this analysis is to examine in detail how environmentalists placed themselves and the nonhuman inhabitants of the central and north coasts in between forestry companies and their main allies – their customers – in order to take away their power and draw them into environmentalists’ network. The study of this process of interessment is a study of power. Just as with the concept of speech, I do not consider power to be “held” by any particular group. Rather, I study how it is an effect of its distribution across networks. I am aided in this analysis by environmental activists themselves who worked to trace the networks making up the forestry industry. While analysts of BC forestry politics have described the economic power held by forestry companies, suggesting that it can only be counteracted with similarly large powers such as public concern (Marchak, 1983; Wilson, 1998; Hayter, 2000), environmental activists rendered the networks constituting forestry industry’s power explicit by tracing the commodity chain linking them to their retail customers. In  145 Latour’s (2005b) terms, environmentalists traced an “oligopticon” (p. 175). Latour uses this term to contrast it with Foucault’s (1977) idea of the “panopticon” or a form of power in which everything is seen from nowhere. By contrast, the oligopticon is not everywhere at once, but located in a specific place, such as an office or a business. The power connected to oligoptica is generated by their multiple connections, rather than by their possession of large social forces. Moreover, since it is the networks that generate power for oligoptica, intervening in networks can challenge this power. As I detail in this chapter, this is exactly what environmentalists did.  In both cases – creating a spokesperson and deploying the spokesperson to enrol allies – I note the workings of the modern constitution. As they developed spokespersons and traced the commodity chain, environmentalists and conservation biologists worked together to create new hybrid networks. However, they also attempted to sever these networks into nature and society. It is only in Chapter 6, when I discuss how ENGOs began to negotiate with forestry companies – that we see a shift in approach from the modern constitution to the collective. Nevertheless, just as with First Nations’ “push back” noted in Chapter 4, this chapter notes some elements that were put in place to help prompt the shift.  Before I move on to examine the production of spokespersons and the tracing of oligoptica, I will describe the new coalition that was convened to do the work – the Canadian Rainforest Network. This coalition is noteworthy for two reasons: first, it represents a victory for those who attempted to convene a new social group to save the GBR, even though this came at the cost of a translation. In particular, while the original goal was to work for the establishment of a giant protected area covering the central and  146 north coasts, the coalition included elements that focused on the plight of people and economies as much as nonhumans. While this element did not come out until later when the coalition exploded due to internal conflict, it was significant that it was in there from the beginning. Second, the coalition rejected a government-sponsored land and resource management planning process that putatively took into account the interests of multiple stakeholders. It was therefore instrumental in paving the way for alternative approaches to reconciling interests in the coastal forests, processes investigated in Chapters 6 and 7.  5.1  The Canadian Rainforest Network     By 1996, several environmental organizations had become interested in BC’s central and north coasts: the McAllisters’ Raincoast Conservation Society, the BC Chapter of the Sierra Club, Greenpeace (which had recently established its Ancient Forests Campaign headquarters in Vancouver), the Western Canada Wilderness Committee, and McCrory’s Valhalla Wilderness Society. Other groups also became interested in the region. For example, the Forest Action Network (FAN) – a grassroots, direct action group – established an office in Bella Coola, the heart of the central coast, to support the Nuxalk Nation in its 1995 blockades of the BC logging company International Forest Products (Interfor) on nearby King Island. However, if a concerted campaign to protect the GBR was to take place, then these different groups needed to be brought together.  Interestingly enough, these groups were united not only by their desire to protect the central and north coasts but also by the mediation of personal growth and spirituality.  147 In 1996, BC Wild convened an “activist training program” consisting of a number of workshops at the Hollyhock Centre focused on helping activists in the wilderness preservation movement deal with issues of “burn out.” According to an environmentalist involved in convening these workshops [E - KI1: 032], the Hollyhock Centre is “a personal development retreat centre, basically, like that kind of what it does. It’s all about building personal mastery and personal consciousness and it’s spiritual.”  While the goal of the workshops was to help activists who were “just flaming and burning [out] all over the place,” they originally produced a “huge culture clash,” since, in the view of this environmentalist, most activists were “basically about ‘get out there in the world and fucking save the planet and we could care less about crystal fucking, that’s not what it’s about.”” However, the workshops were eventually able to induce a transformation in the activists: “we’re kind of doing this meshing of the spiritual level with the activism piece and gradually it’s getting less and less hard to sell that message internally in the movement.”  The message was one of “compassion, and [focused] on the idea of a larger unifying force and all of those kinds of things.” Moreover, the workshops provided a venue in which the different groups could develop a common approach to their coastal work: “the group of people who were working on this particular campaign are not only coming here for training, but then they’re also having strategy sessions here, and they’re starting to do a bunch of work here, just around idea development” [E - KI1: 032]. A coalition formed out of these meetings, termed the Canadian Rainforest Network. The alliance involved a mixture of twenty small and large, Canadian and American, and radical and moderate groups. The groups differed in their goals, with  148 some calling for outright protection of the entire area and others focused on protecting some areas while recognizing that a forest economy would continue:  The Canadian Rainforest Network was the “who’s who” of everybody and it was like sixteen or eighteen groups of people who were all working up and down the Coast. And it was hell, it was like every megalomaniac you’ve ever met or understood in the world, they were all in the room, and they all had their own ideas about what had to happen, and the spectrum went from the complete idealist who said, “Reject it all, not a single other tree can come down,” to the people who were going, “Okay we’ve got to phase out clear cutting gradually and now we’ll protect some of it.”  And, you know - the incrementalists vs. the idealists – and that tension was just palpable in the room, and we spent most of our time fighting. [E - KI1: 004]  Like WCWC’s suggestion in their 1992 Educational Report, some argued, “the whole thing should be protected” (Hamilton, 1996). However, others suggested that economic well-being would have to be considered: “We would like to see as much of this area protected as possible,” said Vicky Husband, the Sierra Club of B.C.'s moderate leader, “but we recognize there have to be economic opportunities” (as cited in Luke, 1996). Indeed, a new approach that went beyond simple protection seems to have been the official focus of the group. As quoted by Gay (2001), the CRN was convened to promote “‘fundamental changes in the way British Columbians approach both forestry and forest protection,’ developing conservation strategies that ‘will ensure the continued ecological integrity of BC’s rainforest and the communities that depend on it’”   (p. 136, emphasis added). Thus, the original goal of Conservation International and Ecotrust to shift the focus of BC’s wilderness preservation movement away from the valley-by-valley  149 struggles characteristic of the “last unlogged watershed syndrome” was largely successful. As noted by the CRN coordinator, Jill Thomas:  The idea of doing a valley-by-valley fight on the mainland coast is impossible to contemplate […] There are 60 valleys. But it's time to move beyond that anyway. When we fight valley-by-valley we end up with fragmented valleys here, there and everywhere. This is quite a major paradigm shift for the environmental movement. (Hamilton, 1996)  Yet, this shift came at the price of another shift. There was at this point a slight shift from the goal of creating a huge protected area to an alternative conservation vision. Latour (1991) notes, “a statement…is in the hands of others” (p. 105). In other words, a statement travels through a chain of speakers, each of whom transforms the statement in some way before passing it along.17 Conservation International and Ecotrust said coastal temperate rainforest, Moore said over 100 unprotected watersheds, WCWC said huge park, the McAllister’s and Berman said Great Bear Rainforest, Hollyhock said compassion, and now the CRN was saying paradigm change and ecological protection, forestry practices, First Nations’ rights and economic development. The original statement (which, as I described in the last chapter, was derived from multiple material and discursive practices) was passed along but changed form as more “hands” touched it. The broader emphasis that it gained – something that I will explore more fully in the next chapter – is already seen in the CRN list of overarching goals (Econews, 1996):  • Protecting critical ecological areas  17 A statement, according to Latour (1991, p. 151), does not refer only to language; rather a statement can refer to “a word, sometimes to a sentence, sometimes to an object, sometimes to an apparatus, and sometimes to an institution.”  150 • Stopping all clearcutting in coastal temperate rainforests • Limiting road construction in pristine areas • Supporting First Nations struggles to protect traditional territories, and • Promoting sustainable community economic development  This list of goals was not uncontested and internal disagreements eventually led to the demise of the CRN. However, while it held together, the alliance worked very hard to make these goals a reality. Obviously, the CRN was not initially in a position where it could dictate land use policy for the Province: a list of goals was not enough. According to one environmentalist  you need to generate power […] if you don’t have power, you will just be considered some side input and they will, you know, monkey-wrench around the edges to shift, to try to appease you, but if you don’t have power, you can’t fundamentally alter the system. [E – NT: 318]  How did ENGOs generate power in order to fundamentally alter the system? According to writers such as Marchak (1983), Wilson (1998) and Hayter (2000, 2003), economic and political structures combined to consolidate and centralize control over BC’s forests in the hands of a “development coalition” or “government-forest industry compact” (Wilson, 1998, p. 81). In their analyses, this power could only be countered by an equal or greater power, such as the “powerful imperatives [of] neoliberalism, aboriginalism, and environmentalism” (Hayter, 2003, p. 707). Similarly, in Wilson’s (1998) terms, industry’s economic power could only be countered if environmentalists were able to “mobilize sufficient political resources to neutralize these advantages and push their issues onto the agenda.”   In other words, only the power of social structures can combat the power of social structures.  151 However, as Latour (2005b) argues, “if there is no way to inspect and decompose the contents of social forces, if they remain unexplained or overpowering, then there is not much that can be done” (p. 252). In other words, if the power of the forestry industry is explained simply as the product of economic forces such as capitalism, the best that environmentalists can do is hope that other existing gigantic forces – such as Science or social movements – can be invoked in opposition. However, if you choose to isolate one of the features of the network and then abstract it as the power of the market, of science, of economic resources, or of images, then not only will you have an impoverished explanation, but a tautological one. By contrast, a more pragmatic approach to power is to recognize that “only a skein of weak ties, of constructed, artificial, assignable, and surprising connections is the only way to begin contemplating any kind of fight” (Latour, 2005, p. 252). Or, in the words of John Law (1992):  If we want to understand the mechanics of power and organization it is important not to start out assuming whatever we wish to explain. For instance, it is a good idea not to take it for granted that there is a macrosocial system on the one hand, and bits and pieces of derivative microsocial detail on the other. (p. 2)   In contrast to the idea that power is centred and total, in the following I analyze power as a product of networks. In the case of the environmentalists’ campaign, this took the form of their ability to use their status as representative of one network (the environment) to attack and intervene in another (the commodity chain). As one interviewee notes, the   152 thing that the companies had to grapple with was that, you know, you needed to understand who had power and why they had it. And you needed to understand, look it, the environmental groups have power. Why do they have it?  Well, they have it because they represent the environment. [E – EK: 018]  But, what does it mean to “represent the environment”? Do environmentalists speak for the environment because it cannot speak for itself?  Who has the authority to speak on behalf of the environment? In the following, I will suggest that environmentalists did not simply start speaking on behalf of the environment, but that they worked to designate a spokesperson for the rainforest – the grizzly bear – with whom they could share speech (Latour, 2005, p. 64). That is, they enrolled the science of conservation biology to create an intermediary position between environmentalists who spoke on behalf of the environment and the environment speaking on its own behalf. Second, ENGOs generated power through conflict: “the Government employees all said yeah, if there wasn’t a conflict – and the conflict is what led to us having power – then they wouldn’t have heard us”  [E – NT: 310]. On the one hand, building power involved a process of representing the interests of thousands of species and ecological processes of the environment in a single spokesperson. However, on the other hand, it involved processes of exposing and attacking the network constituting a seemingly single actor, the forestry industry. This, as we will see, was largely done through the commodity chain linking forestry companies to their retail customers in Europe and the US. As one environmentalist noted, “we have to get power from somewhere, and so the market campaigns can target the companies, we can get the companies to agree to stop logging” [E – NT: 090].  153 In the following, I show how environmentalists and the Great Bear acted together as spokespersons for the rainforest, intervening in the network of associations that connected the coastal forests to forestry companies and the retail customers of BC forest products in Europe and the US. Additionally, I show how industry and the market were traced and mapped out as oligoptica – that is, as well-connected but local sites that could be attacked at a variety of points. Finally, I show how the actors worked to create new networks of power, on the one hand, and to purify these networks into nature and society on the other.  5.2  Conservation Biology and Spokespersons for the Rainforest     According to one environmentalist, the impetus to shift attention to the north was aided by conservation biology. As she notes, two studies came out simultaneously with the publication of the Sierra Club’s map.  One was a study of all the National parks in Canada, which said that in every park they were losing species, except for one, and that park had been created the year before, so they couldn’t monitor any species loss. The second one was the study of Western North America18 and it determined that every – so, these were conservation biology studies – that every park was loosing species except for the Banff-Jasper: it goes into the States, and you know, it was the only one that was big enough to actually maintain species. [E - NT: 86]  These studies, along with the Sierra Club’s map, prompted some people in the environmental community to reassess their strategies: “a number of us in the  18 Newmark, W. D. (1995). Extinction of mammal populations in Western North American national parks. Conservation Biology, 9(3), 512-526.  154 environmental community sat back and said, ‘Okay, well we’re failing. Creating parks is not actually winning, it’s failing’” [E – NT: 86]. The ‘Big Green Blob’ on the map allowed environmentalists to see an entirely new kind of conservation opportunity for BC, one built on the science of conservation biology:  We had an opportunity to do it right here – it’s this huge landscape level, we can take a whole landscape conservation biology approach, it’s not like we’re just trying to protect the remnants that are left, like in Southern BC, we could do it right from the beginning. [E – NT: 88]   Thus, the science of conservation biology entered the scene as an important actor in the crafting of a new set of relations for coastal BC. Not only was the science telling environmentalists what they were doing wrong, it was telling them how to get it right. The recommendations coming from conservation biologists aligned nicely with efforts to demonstrate that th