NOTICE: this is the author’s version of a work that was accepted for publication in Environmental Science & Policy Changes may have been made to this work since it was submitted for publication. A definitive version was subsequently published as: Leith, P., K. O’Toole, M. Haward, B. Coffey, C. Rees, and E. Ogier. 2014. “Analysis of Operating Environments: A Diagnostic Model for Linking Science, Society and Policy for Sustainability.” Environmental Science & Policy 39: 162–71. doi:10.1016/j.envsci.2014.01.001.

Analysis of operating environments: A diagnostic model for linking science, society and policy for sustainability Peat Leith a,b,*, Kevin O’Toole c, Marcus Haward a, Brian Coffey d, Chris Rees a, Emily Ogier a

a Institute for Marine and Antarctic Studies, University of Tasmania, Hobart 7001, Tasmania, Australia b Tasmanian Institute of Agriculture, University of Tasmania, Private Bag 98, Hobart, Tasmania, Australia c School of Humanities and Social Sciences, Deakin University, Warrnambool, Victoria, Australia d Alfred Deakin Research Institute, Deakin University, Burwood, Victoria, Australia

a r t i c l e i n f o

Article history: Received 5 October 2013 Received in revised form 6 January 2014 Accepted 6 January 2014 Available online 2 February 2014

Keywords: Science policy Research outcomes Environmental governance Coastal zone management Values Problem structuring

a b s t r a c t

Through analysis of the dynamics between science and decision-making, we argue that diagnosing fit-for purpose approaches to linking science and decision-making may be possible. Such diagnosis should enable identification of appropriate processes, institutions, objects (e.g. tools, information products) and relationships that can facilitate outcomes. We begin the paper by unsettling the traditional constructions that science must distance itself from debates about values and what is at stake, and so from policy making. Then, drawing from mixed methods case studies in coastal South-eastern Australia, we describe how scientific research has had a bearing on decisions affecting society and the environment. These analyses suggest that the willingness and capacity of research organisations, pro- grammes or projects to actively reflect on and participate in the evolution of the ‘operating environment’ for their research is integral to their ability to inform outcomes through science.

1. Introduction

A fundamental challenge for sustainability stems from long- standing tensions between the domains in which knowledge is made and applied in contemporary society. Jasanoff (2003: 235) sums the challenge up well: ‘‘how to institutionalize polycentric, interactive, and multipartite processes of knowl- edge making within institutions that have worked for decades at keeping expert knowledge away from the vagaries of

populism and politics’’. Traditional narratives in science and policy organisations tend to treat science, policy and politics as three separate spheres. Yet empirical research on the demarcation of roles and responsibilities across these domains indicates that their boundaries are blurred and continually renegotiated (Jasanoff, 1987; Wynne, 1994; Guston, 2000). Approaches to addressing the interactions between science and decision-making have tended to be normative rather than diagnostic. For example, boundary organisa- tions that operate between science and decision-making

* Corresponding author at: Tasmanian Institute of Agriculture, University of Tasmania, Private Bag 98, Hobart, Tasmania, Australia.

Tel.: +61 3 6226 2650; fax: +61 3 6226 7444. E-mail addresses: Peat.Leith@utas.edu.au (P. Leith), kevin.otoole@deakin.edu.au (K. O’Toole), Marcus.Haward@utas.edu.au

(M. Haward), brian.coffey@deakin.edu.au (B. Coffey), Chris.Rees@utas.edu.au (C. Rees), Emily.Ogier@utas.edu.au (E. Ogier). 1462-9011/$ – see front matter # 2014 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.envsci.2014.01.001

163

organisations and are accountable to both spheres (Guston, 2001) tend to be presented as a generic institutional and structural means of mediating and translating science for decision-making (Cash et al., 2003; McNie, 2007). Alternately, a focus on improving information products has drawn on psychological research (e.g. Kahneman and Tversky, 1996) to tactically refine scientific messages to improve ‘science impact’.

In this paper we draw on relevant literature and a synopsis of five case studies of science integration for South-eastern Australia. We argue that this empirical and theoretical work provides grounds for a diagnostic approach to developing context appropriate interventions into the interactions between science and decision-making, in what we term the operating environment. We define the operating environment for sciences as the dynamic and cumulative interactions between actors, values, stakes, and the institutions, processes, discourses and objects that mediate such interactions. Our diagnostic model is proposed as an approach for analysing and re-configuring operating environments for science in specific problem contexts.

2. Trends in making useful science for environmental management

Empirical research on the effectiveness of science in environ- mental governance demonstrates that the linear model of science and its application to decisions (Wynne, 1994) is rarely effective in creating outcomes (McNie, 2007; Nelson et al., 2008; Leith, 2011). Analyses of practice have detailed how scientists and policy-makers frame problems in partial ways, between diverse values and interests, and negotiate the credibility and meaning of knowledge in relation to such framing (Jasanoff, 1987; Wynne, 1994; Guston, 2000, 2001). A more public concern for science has been the associations between sciences and specific interests, undermining the legitimacy of scientists, science agencies, or the scientific enterprise as a whole (Oreskes, 2004). Claims by scientists may also be perceived as illegitimate when they are made in isolation from local knowledge, through their apparent finality and purported authority (Wynne, 1992a,b).

The metaphor of boundaries has become an influential framing of the interactions between science and decision- makers. Boundary work has moved beyond its origins as a methodology for analysing ‘credibility contests’ among scien- tists (Gieryn, 1983; Gilbert and Mulkay, 1984). Scholars have adapted the concept of boundaries to analyse a variety of situations where science intersects with lay and policy domains. Jasanoff’s (1987) seminal work on science-policy boundaries highlighted the complex negotiation of boundaries around responsibility and authority. Star and Griesemer (1989) detailed how ‘boundary objects’ (such as graphs, maps, and report cards) can be used to mediate knowledge between actors. Guston (2001) suggested that the creation of science- for-policy and policy-for science are two sides of a principal- agent problem that can be resolved through setting up ‘boundary organisations’ which sit between science and decision-makers and are accountable to both. For Cash et al. (2003) boundary spanning includes processes of

convening, translating and mediating to create knowledge for decision-making in which synergies and trade-offs between the salience, credibility and legitimacy of that knowledge are negotiated. In all these forms of boundary spanning ‘what we know’ and ‘who we are’ are linked together (Jasanoff, 2004). Knowledge production and governance become a single system (Whatmore, 2009) in which the supply of scientific information meets the well-developed demands of decision-makers (McNie, 2007; Sarewitz and Pielke, 2007). There are multiple dimensions to problem definition that could help to guide focussed application of boundary span- ning. However, two frames of reference – systems uncertainty and stake – have consistently been at the centre of debates about how to link science with decision-making (Funtowicz and Ravetz, 1993). Below we draw on two recent framings of these broad issues.

Firstly, Allenby and Sarewitz (2011) suggest that sciences and technologies can be thought of as having effects on three levels depending on the system complexity and uncertainty. The direct ‘level one’ effects are the usual goals of the application of a technology. ‘Level two’ interactions are the less immediate consequences that emerge from the interac- tion of human and biophysical systems at a local and regional level. At ‘level three’, these second level interactions are extended through interactions with global drivers to create even less predictable or manageable consequences. These three levels of interactions are reflected in inter- and trans- disciplinary research as means to better understand complex systems problems (Holling et al., 1998). Such research under- pins adaptive governance, where policy creates experiments and research plays a key role in their evaluation (e.g. Holling et al., 1998; Innes and Booher, 2004; Cash and Buizer, 2005; Nelson et al., 2008; Hallegatte, 2009). The three levels of complexity and system uncertainty provide a useful lens for considering and defining system boundaries and therefore contexts for learning, policy review, and the scope of enabling research.

Secondly, stakes and the politics that arise from them are crucial to the structuring of problems. By stakes we mean the degree of interest or concern that individuals or groups have regarding particular issues, and the degree of associated value consensus or divergence. Stakes may be based on pecuniary, instrumental, non-instrumental, intrinsic, or any other values. Authors such as Hoppe (2011) and Turnhout et al. (2008) argue that politics structures problem situations in different ways depending on what is at stake for whom. Turnhout et al. (2008) develop a typology of problem structuring across which the role of scientists, policy processes and the forms of useable knowledge all vary substantially (Table 1). First, well-structured problems are those for which converging values and/or low stakes make a problem amenable to direct application of technical informa- tion. Second, moderately structured problems exist where there is a possibility of a majority reaching agreed goals, and relatively high certainty about science. Third, poorly struc- tured problems are typified by dilemmas such that an outcome that is considered positive will create another that is considered negative, often depending on the divergent values and stakes. Finally, for unstructured problems, divergent perspectives exist about what the issue actually is, and

164

Table 1 – A typology of problem structures and appropriate processes, roles and forms of knowledge.

Well structured

Problem structure

Moderately Poorly structured structured

Unstructured

Policy process Role of scientist Type of research Useable forms of knowledge

Linear Solve problem Disciplinary Data

Negotiation Compromise Policy options Accommodate Inter-disciplinary Trans-disciplinary Contextualised Conceptual knowledge, information/argument linked with values

and interests

Learning Signal issues Partial analysis Options and perspectives

Source: Adapted from Turnhout et al. (2008).

therefore agreement on goals associated with the issue is difficult to accomplish. Pielke (2007) provides exemplars of how stakes and thus politics structure actions situations in which science has a potential role – the cases of what he calls ‘tornado politics’ and ‘abortion politics’ (cf. Scho n and Rein, 1994). In extreme weather events such as tornados, consistent societal stakes (e.g. risks to life and property) leave predictive science uncontested. These are well-structured (albeit com- plicated) problems. In abortion politics, strongly held societal values diverge and agreement about the basis of societal debate is absent – a hallmark of unstructured problems. For such unstructured problems, science tends to have little traction and can lead to intensification of controversy (Sarewitz, 2004).

In addressing well-structured problems, the issue, desired outcomes and stakes are generally either uncontested or inconsequential. In moderately structured problems there is some interaction amongst interests and identifying alterna- tive options, but science or information can play neatly into informing options and their negotiation. In trying to resolve poorly structured problems, decision-makers will tend to use compromise to trade-off and ‘balance’ between opposing outcomes or values in attempts to define a majority view. Unstructured problems give rise to divergent claims, interests and values. These differences are likely to foster adversarial politics in which various players appear to be talking about different issues. In such settings of substantial distrust among stakeholders the salience and credibility of science itself may be threatened (Clark et al., 2011).

In the central question for understanding problem struc- turing – ‘what is at stake, for whom?’ – the latter part implies some analysis of power and knowledge relations. Schattsche- neider (1960: 106) famously emphasised the bias in political decision-making towards elites: ‘‘whoever decides what the game is about decides also who can get into the game’’. The ways that problems are signalled and represented, and through which options are generated, selected and enacted, tend to be entrenched within existing institutions and incumbent power/knowledge structures, relations and cul- tures (Foucault, 1980; Hoppe, 2011). While it is beyond our scope to fully review the literature on power in decision- making, it is useful to briefly highlight the shifting power relations that are reflected in the cases that follow. In particular, the global north has experienced shifts to more distributed and networked forms of knowledge production and governance, especially in complex resource management

issues (Innes and Booher, 2004). Driven in large part by information and communications technologies (ICTs), citizens have rapidly developed new ways of influencing markets, policy processes, knowledge and politics. This has created new imperatives, including the development of a ‘social license to operate’ beyond regulatory compliance (Gunning- ham et al., 2004). Such distributed forms of power and knowledge production blur boundaries between traditional roles of science, policy and politics as well as between the public and private spheres. Inclusion and engagement, however, can also be subsumed (often subtly) by pre-existing knowledge/power relations to achieve existing organisational goals (Cooke and Kothari, 2001).

3. Science input into coastal zone management in South-eastern Australia

Coasts are particularly useful sites for examining the roles of sciences in managing complex issues because diverse inter- ests interact with complex biophysical and socio-political processes. In this section we summarise five case studies from the southeast coast of Australia. These case summaries stem from mixed methods studies reported elsewhere (see Table 2) including qualitative semi-structured interviews, document analysis, participatory workshops, focus groups and partici- pant observation. An important commonality across the cases was the use of forms of discourse analysis. Participants were asked to describe and reflect on how processes and interac- tions among stakeholders had, among other things, created or undermined legitimacy, credibility and salience of science, built capacity to use scientific results, and precipitated community involvement or dissent in relation to the relevant science. Cases used a variety of largely qualitative approaches as detailed in relevant citations.

3.1. Planning for sea-level rise in Clarence and Kingborough Councils, Tasmania

Assessing the present and future risk of sea level rise (SLR) in Australia falls particularly to local governments. Trade-offs exist between the protection of private assets of substantial value (e.g. beachfront houses) and public assets of substantial amenity and ecosystem value (e.g. beaches). The science of local impacts is uncertain, and events such as storm surge create a public debate that turns on and off, literally with the

165

Table 2 – An overview of case studies summarised.

Name of case Region case Data collection Number of participants Participantsa Citations

Estuary entrance Western Victoria Key informant 42 interviewees SG, LG, I, SO, O’Toole et al. management interviews; 30 workshop participants NGO, CO, CA (2013) and

Workshops; Keneley et al. Document analysis (2013)

Marine protected Victoria Key informant 42 interviewees SG, LG, I, SO, Coffey and areas interviews; 30 workshop participants NGO, CO, CA O’Toole (2012)

Workshops; Literature review

Derwent Estuary SE Tasmania Key-informant 10 interviewees SG, LG, I, SO Leith et al. Program interviews; (2012a,b)

Document analysis Salmon aquaculture SE Tasmania Key informant 18 interviewees SG, I, SO, NGO, Leith et al.

interviews; 12 workshop participants CO (2014) Workshops

Coastal council SE Tasmania Interviews; 6 Interviewees (Clarence); LG, PC Leith et al. sea-level rise focus group 13 Focus group participants (2012b) planning (Kingborough)

a Participants from different sectors indicated as follows: state government (SG), local government (LG), industry (I), private consultants (PC), science organisations (SO), non-government organisations (NGO) community organisations (CO) community activists (CA).

weather. Clarence and Kingborough Councils in South-eastern Tasmania represent two proximate but vastly different situations of SLR planning.

Clarence City Council on the Eastern shore of the Derwent Estuary has 191 km of coastline of which many areas are soft sediment. Much of the coastline remains undeveloped, but there are a number of small to medium settlements that are erosion or inundation prone. Community concern about these impacts led Clarence to apply for early Australian Government financial support in April 2009. The ensuing project provided an integrated scientific assessment of climate change risks on public infrastructure and private property for 18 coastal locations in the municipality. In contrast to Clarence, the Kingborough Council, has 336 km of coastline with fewer low- lying settlements on erodible substrates. Kingborough pro- vides an informative contrast to Clarence. By virtue of their geography, topography and aspect, none of the Kingborough coastal settlements have been historically exposed to sub- stantial erosion or inundation. Most notably, storm surge activity in Kingborough is limited by the presence of a large offshore island. The Clarence population had been sensitised to the prospect of coastal hazards by frequent erosion events including a 9 m recession of beach backed by residential properties over the 8 years to July 2011. As a result of these geographical differences, public interest and participation across relevant projects was very low in the Kingborough and high in Clarence.

In Clarence, the synergy between key players’ capacities and experience and timely scientific research, storm surge events and related political developments and opportunities was critical to the success of SLR planning. Participants described early science reports as incomprehensible to most people, and reported that inundation maps triggered a strong fear reaction in Council. The lead consultant was able to effectively translate high quality science into locally relevant forms, and work with relevant Council staff to empower the Council to be transparent and effectively engage the commu- nity. The Council’s communication plan committed them to

actively involving, informing and being an advocate for the community, and respecting, listening and responding to the community’s concerns around the project. Council was praised by community members for such transparency and became comfortable with releasing highly sensitive informa- tion to the wider public about coastal inundation and erosion risks. Appreciating the uncertainties in the science, short- term, ‘experimental’ works for hazard management were developed, alongside ongoing public consultation and com- munication.

Despite the extent of serendipity involved, the ‘Clarence process’ has been promoted as a model for other councils’ SLR planning. It demonstrated the effectiveness of open commu- nity engagement with science and policy directions. The experience of Kingborough demonstrated that, despite com- parable processes, low political interest followed an absence of public interest, which in turn was driven by low perceived stakes as a result of a geography which limited the impact of storm surges on residents. Thus, scientific reports were left ‘on the shelf’.

3.2. Marine protected area planning,

Victoria

The consideration of marine protected areas (MPAs) attracts significant and sustained input from a diverse and changing range of stakeholders. These stakeholders include conserva- tion groups (national, state, and local groups), fishing organisations (commercial and recreational), other interest groups (tourism operators and diving groups), political parties, public servants and members of the scientific community. Central elements in ongoing debates about MPAs include how marine and coastal areas should be managed, the role of MPAs in resource management and conservation, the nature of MPAs (‘no take’ versus ‘multiple use’), the boundaries of any protected areas established, and the compensation arrange- ments for those impacted.

Victoria’s Environment Conservation Council (ECC) worked through these issues by applying strategic environmental

166

assessment process. Between 1991 and 2000, the ECC and its predecessor organisation conducted a Marine, Coastal and Estuarine Inquiry (ECC, 2000). ECC investigation processes involved five main steps: (1) initiation of investigations; (2) preparation of descriptive reports; (3) preparation of draft options; (4) preparation of final recommendations; and (5) preparation of a government response (with steps 1 and 5 undertaken by the government and not the Council) (Coffey et al., 2011: 309). In the Marine, Coastal and Estuarine Investigation, there were six formal periods for public comment, ongoing consultation with a wide range of stakeholders, and technical support provided by an advisory group. 2500 written submissions were received following the release of draft recommendations (ECC, 2000). Following the release of its final report, the focus of attention shifted to the Victorian Parliamentary processes. Protracted negotiations led to the development of ‘‘a substantial compensation package for people adversely affected’’ (Wescott, 2006: 910) and subsequent passage of the MPA legislation.

Throughout this process, many and varied forms of knowledge were drawn upon. First, available technical information on Victoria’s marine, coastal, and estuarine environments was assembled in various background reports. Information on the socio-economic values derived from making use of Victoria’s marine, coastal, and estuarine environments were collated from a range of sources, including the fishing industry, and the socio-economic impacts of recommendations considered. More informal forms of knowl- edge (lay and indigenous) were also considered through the widespread consultation processes, including public meetings where ECC members met with members of the community, of a separate consultation process ‘‘to facilitate and coordinate the input of aboriginal people’’ (ECC, 2000: xii). In summary, extensive opportunities for written and verbal input were made available, which provided for the consideration of various forms of knowledge. While the underlying foundation for the ECC is rational and managerial and centred on the compilation of technical information on the biophysical environment, it had clear requirements for consideration of economic, social, and environmental objectives and thus provided a forum within which diverse viewpoints could be aired.

Stakeholders held markedly different visions of the value of marine, coastal, and estuarine environments and how they should be used and managed. The presence of the ECC as an independent, transparent, structured, and respected process (Coffey et al., 2011) enabled the many frames and related controversies to be included and thoroughly considered, even if they were not necessarily conclusively resolved.

3.3. Estuary entrance management,

Victoria

Estuaries provide habitat for fish, birds and other species, as well as being important sites for recreation, agriculture, fishing and urban and industrial development. Estuaries are also affected by numerous local and upstream human activities. In Victoria many estuary entrances close periodi- cally resulting in raised water levels, which can result in flooding of agricultural land, buildings, roads and structures, such as jetties and boat ramps (Sherwood et al., 2008). In the

past these effects have been addressed by artificial river mouth openings, usually with little reference to environ- mental impacts, or broader social and economic implications. The ecological risks of artificially opening estuary entrances include impacts on water bird habitat in fringing wetlands and fish kills through the lowering of dissolved oxygen levels in water following estuary opening, as well as degradation of estuarine catchments (Sherwood et al., 2008).

An estuary entrance management support system (EEMSS) was developed by local catchment management agencies. Key features of the EEMSS process were the inclusion of stakeholder input and the integration of their concerns with the views of scientific experts to develop a workable solution to a serious environmental problem. The EEMSS was designed to facilitate the management of estuary openings in a manner that was acceptable to stakeholders whilst minimising environmental repercussions.

Preliminary assessment of the EEMSS process suggested a positive acceptance of this approach by local stakeholders (Keneley et al., 2013). Other positive outcomes have included a greater understanding by local stakeholders of the complex issues and processes involved in estuary management and a reduction in complaints from those local landowners involved in the consultation process.

Whilst benefits have accrued from the EEMSS process, incorporating this type of grass roots initiative into broader integrated coastal strategies and processes remains a con- siderable challenge. In particular, ‘scaling up’ the EEMSS process across all relevant estuaries is challenging in an environment of time-bound interventions as state-funded projects (O’Toole et al., 2013). Improving estuary entrance management therefore is not merely a matter of using a ‘cookie-cutter’ to scale up adoption of new approaches. Instead, scale and context are central to the design of stakeholder engagement that underpins success. There is a need to consider the higher order rules (and policy processes), which ensure participatory mechanisms are built into policies, programmes and plans, to enable a context derived response to occur (see, for example, Jentoft, 2000).

3.4. Salmon aquaculture in South East Tasmania

Salmon aquaculture is a substantial and growing industry in Tasmania, and now constitutes Australia’s most economically lucrative seafood sector. Three main companies lease areas of public waterways and are regulated under legislation. In recent years, the industry has been the focus of some controversy and concern among sectors of the community, including environmental non-governmental organisations, local community groups and recreational fishers. There is currently a substantial effort in the sector to improve sustainability, variously supported by government, industry and the community. However, sustainability, and reporting on what it means, relies on science and science agencies that effectively target research and communicate to diverse audiences – government regulators, local communities, consumers, and industry decision-makers.

A pervasive set of narratives from participants in the community, science and industry, highlight an erosion of trust between some vocal community groups, the government

167

regulator and the industry organisations. These narratives often include concerns about the transparency, efficacy and accessibility of environmental monitoring information. Recent changes appeared to have tempered these tension including: the industry leader becoming a publicly listed company and needing to undertake triple bottom line accounting; the role of individuals in precipitating more transparent and participatory forms of engagement, interac- tions and programmes; the industry agenda to expand and the necessity of political support for this expansion; the recogni- tion that community groups and NGOs can affect consumer perceptions and thus markets; a recognition that the adversarial approach that typified other environmental con- flicts and ‘salmon wars’ elsewhere in the world had not resulted in desirable social, economic or environmental outcomes for any parties.

Social license is constituted by different players as critical to understanding the current and future role of environmental science and of science policy in enabling this role. In general, social license is described by participants as being founded in mutual understanding of the effects of Salmon aquaculture on communities and public waterways and capacity to engage in legitimate decision-making processes around these issues. Some participants’ concerns are indicative of a deficit of trust in public representation of environmental monitoring and report- ing by government and industry. Lack of access to scientific interpretations that are perceived as legitimate (and the data on which they are based) has created a highly politicised operating environment for science in which no intermediary is currently effective. The majority of effort in communicating science has occurred between scientists, industry and government, and these communication efforts have themselves not always been easy. Communication and engagement with community members and environmental NGOs has generally been poorly resourced and not seen as a priority in an operating environ- ment in which science has traditionally been viewed in relation to its regulatory application.

Most participants suggested that the industry is opening to public view, becoming more transparent and accountable to the community, not only through the regulatory processes but through commitment to transparent communication. As one industry participant stated; the industry and NGOs had to ‘‘stop meeting in the media’’ where the imperative was to argue rather than negotiate or deliberate. Another participant suggested ‘‘you can only challenge each other when you are connected and communicating. If it is adversarial it is logical for the adversary to shut down or fight.’’

Traditional, top-down regulatory systems tend to involve small technical audiences regularly engaged with scientific analysis, interpretation, risks, uncertainties and how these align with specific policy options. In a more distributed knowledge system, in which social license is a core concern, the legitimacy and credibility of scientific interpretation to diverse audiences can be crucial to mediating mutual under- standing. This requires effective intermediary organisations and/or individuals that can legitimately translate rigorous science for diverse audiences. Many participants in this case appeared to have developed this understanding alongside an evolutionary appreciation of social license as a critical underpinning of ongoing business viability.

3.5. The Derwent Estuary Program, Tasmania

The Derwent Estuary Programme (DEP) in South East Tasmania is a regional partnership comprising the Tasmanian State Government and Local Councils and commercial and industrial enterprises bordering the Derwent River. It also engages closely with research organisations, and community- based groups. Established in 1999, the DEP develops, coordi- nates and implements framework agreements and practical initiatives aimed at the reduction of water pollution, con- servation of habitat and species, the monitoring of river health and enhancing the use of the Derwent foreshore areas. Taking a long term strategic approach to managing challenges in the Derwent by providing its partners and stakeholders with a strong science base, the DEP coordinates targeted projects and ongoing monitoring programmes, producing an annual report card and a State of the Derwent Report every five years.

The DEP has positioned itself carefully on multiple boundaries: between state and local government, industry, science agencies, and the broader community. It serves particular goals of each of these groups in a way that allows for synergy and delimits political controversy and risk. The DEP’s successful navigation of political issues by maintaining a science focus, contrasts with its inception in a bid to manage controversy associated with a legacy of heavy metal pollution in the Derwent River. This controversy was itself precipitated by highly credible research that identified dangerous zinc and cadmium levels in shellfish in the Derwent Estuary, following a poisoning incident in 1970.

Through early wins, the DEP cemented a partnership between key industries and relevant levels of government which resulted in the development of a relatively stable entity with a strong science-oriented programme of work. The science focus and the explicit avoidance of political issues have created what one participant referred to as a ‘‘a safe space’’ in which science and management appear to be kept separate. Yet the DEP also maintains a dialogue between these separate endeavours enabling a broader awareness of activity across the estuary in both science and management. Such a dialogue allows individuals and groups to view and critique the modus operandi of others, but also has enabled a gradual shift in consensus from a central focus on heavy metals and water quality to a more systems oriented programme focussed on estuarine health.

Although the pathway to the current stability of the DEP is broadly acknowledged as being relatively smooth, there have been key flashpoints in which hard decisions, complex negotiations and diverging values occurred. These appear to have been largely well managed to create opportunities for learning across member organisations and a resulting increase in their level of commitment to the DEP. For example, when a favoured recreational fish in the upper estuary, Bream, were found to be containing high levels of heavy metals, the DEP mediated discussions between parties on how this information might be communicated. Ongoing negotiation of the framing of information appears to be a core component of the social and political work undertaken through (not by) the DEP by creating an ethos of collaboration on science among its partners. The DEP has built substantial legitimacy among its constituents by coordinating and writing reports

168

and other outputs for all partners while maintaining a reputation in science circles for producing credible and relevant outputs.

4. Towards diagnosis and intervention in the ‘operating environment’ for sciences

The above cases highlight a variety of ways in which interactions between science and decision-making are con- sistently structured by recurring characteristics. Among these, the stakes associated with issues are pervasive. Issues, such as the loss of public amenity of private property in our SLR cases, can precipitate high stakes situations depending on the associated values. If the values across stakeholder groups converge around the importance of private property over public values then decisions may be relatively straightforward – for instance, priority is given to private property without regard for the public amenity of beach users or the loss of shore-nesting bird habitat. In the cases presented here, however, there was substantial perceived trade-offs between public and private benefits associated with intervention. The problems were mostly structured by relatively high stakes and substantial values divergence, and mostly accompanied by entangled historical knowledge/power relations.

Following Cash et al. (2003), Gieryn (1983), Guston (2001), Jasanoff (1987) and others we refer to the mediation of issues and stakes as ‘boundary spanning’ whether this is undertaken actively or otherwise. The term embodies a diverse range of possible practices. Done well, boundary-spanning links issues and stakes and thereby defines a useful role for scientists or science communicators to contribute to useable knowledge. When done poorly, especially where stakes are high and diverse and stakeholders numerous and/or powerful, bound- ary spanning can rapidly result in the politicisation of science and undermine the perceived credibility and legitimacy of science organisations or scientists.

The way ‘stakes’ and ‘issues’ were co-created in our cases did not solely depend on underlying human values (and their divergence, or convergence). How these stakes and issues were mediated was crucial to outcomes. In the Victorian MPA process, time and effort were given to iterative public consultation in parallel to scientific assessment (including socio-economic) work, which enable these issues and stakes to be co-created in a constructive rather than destructive form.

Scientific assessment was responsive to public concerns, and as a result publics were responsive to science (Wynne, 1992a,b). In the more rapid Clarence SLR case an individual actor drove much of the comparable boundary spanning activity. In the Tasmanian Salmon aquaculture case, there was an apparent shift through relationships and products to articulate the component issues of a relatively unstructured problem. With an eye to social license, the industry appeared to be moving from an adversarial position in which the mainstream media and marine farm planning appeals processes were the primary sites of boundary spanning, to a more engaged footing in which dialogue is seen as integral to future operations.

Science was perceived to have a strong bearing on decision- making where it was linked effectively to the interests of stakeholders; their stakes and values. These cases were the DEP, the ‘Clarence process’ for SLR planning, the EEMSS process, and the Victorian MPA planning process. Effective- ness was not achieved through a single means but through a combination of ‘design elements’ with different foci. These elements represent five different but linked boundary span- ning activities and are presented in Table 3.

These elements can be mutually reinforcing. Science communication with a simple product focus, or a ‘loading dock’ approach (Cash et al., 2006), was not apparent in our cases. Rather, relationship, actor, network or organisation foci were used to ground science in specific decision-making or learning contexts. The actor and network focus was successful where it was able to link values, policy objectives and science into coherent narratives and leave a legacy within relevant organisations to continue do the same. Actors and networks appeared to have key roles in reconfiguring the way existing organisations and their members understood relevant science and were able to articulate it with diverse problem frames of stakeholders. The relationship focus presented a less formal or explicit means of developing mutually agreeable narratives that can bring science and values together through a focus on specific issues. Relationships underpinned by products enabled issues to be discussed separately rather than conflated, thereby structuring problems more cohesively. The less common organisation focus was undergirded by other design elements.

The boundary spanning function with an organisation focus was apparent in two of the cases that participants considered successful, the DEP and the Victorian MPA planning process.

Table 3 – The key design elements of boundary spanning. Element and focus Explanation Science communication (product focus) Informal linkages (relationship focus)

Brokering/intermediary (actor focus)

Temporary organisation (structure/network focus – e.g. reference groups)

Boundary organisation (organisation focus)

Development of boundary objects. Where problems are poorly structured or unstructured building informal linkages among key stakeholder groups can begin to create mutual understanding of stakes and values across groups, thereby allowing clearer definition of issues. The building of capacity within organisations that manage problems in which science and community values are both important. Temporary organisations or projects used to address complex issues and/or short-term imperatives. Long-lived, persistent ‘wicked’ problem, managing complex conditions, often within multiple organisations.

169

Cases Product focus

Relationship focus

Actor focus

Structure/ network

focus

Organisation focus

Sea level rise, Clarence and Kingborough

Marine protected areas Estuary entrance management support system

Salmon aquaculture Derwent Estuary Program

Table 4 – Presence (denoted by dark shading, lighter shading indicates element emerging) of different ‘boundary spanning

foci’ across cases.

These were clearly not the only cases in which ‘success’ was apparent. They were both long-term efforts with sustained interests, associated with diverse management objectives, and organisational/statutory responsibilities, as well as the complex politics associated with divergent values, stakes and attendant knowledge/power relations. Different ‘boundary spanning elements’ were distributed across our case studies as illustrated in Table 4. These five key design elements have many potential forms that need to be explored in context, rather than as blueprints. While the cases represent some of this diversity, it is beyond the scope of this paper to unpack the many forms these elements can take and the ways they might interact.

5. From cases to model: the operating environment

Issues, stakes and boundary spanning can be considered to constitute the ‘operating environment’ that affects whether and how sciences can have an impact on decision-making. An operating environment is an emergent property of elements as diverse as an advertising campaign, a well-networked policy entrepreneur and a storm event that threatens coastal homes. It is neither deterministic, nor fully tractable to an analyst. Any analysis of an operating environment will be partial. Among stakeholders there will be diverse interpretations of operating environments, and much understanding will be tacit, vaguely articulated, or contested. For level two and three problems (Allenby and Sarewitz, 2011), a key challenge will be in problem definition as system boundaries extend. In this context, the aim of the analyst and facilitator is not to achieve consensus, but to develop a coherent, credible and legitimate account of the operating environment, highlighting points of tension, diver- gence and consensus. ‘Naming up’ such elements can help to categorise the operating environment in terms of its problem structure. Problem structures vary continuously so it may be useful to use exemplars as anchor points for each to compare a specific problem context to. For instance, is the operating environment more like climate change (unstructured) or bookbinding (well-structured)?

The above constraints do not undermine the goal of analysing operating environments which is to explicitly

articulate the linkages between issues and the stakes of relevant actors in a manner which is credible and legitimate to those actors. Thus, definition of category is not crucial to analysis. Rather, deliberation on the critical question of what is at stake for whom provides avenues into a nuanced understanding of problem structure. While robust social research methods are needed, analysis of operating environ- ments must be legitimate in order to comprise a useful conceptualisation of the problem. Legitimacy requires effec- tive mediation and facilitation, especially in the context of entrenched power dynamics or divergent values and stakes (Innes and Booher, 2004). As depicted in Fig. 1, the goal of such analysis is ultimately the practical development and applica- tion of context appropriate boundary design elements.

Application of the model is, in the first instance, achieved through cycling iteratively through a three stage process (Fig. 1). The first stage involves iteratively building up an understanding of the operating environment through facili- tated dialogue. Within workshops or other forums, dialogue can be used to understand issues and related stakes, and how these are mediated by existing boundary spanning elements. At this stage, biases can be limited through inclusion of diverse relevant stakeholders or their representatives (Innes and Booher, 2004). Workshop design is oriented to mutual understanding of the constellation of issues (e.g. beach activities that constitute public amenity) that are potentially at stake (e.g. what elements of ‘lifestyle’ are at risk) in relation to a particular issue (e.g. managing the impacts of storm surge). Looping between issue and stake requires open dialogue and reflection that should gradually develop greater focus on how existing boundary spanning activities exacer- bate and/or ameliorate the relationships between issues and stakes.

The refinement or redesign of boundary spanning, via specific design elements, we suggest, can effectively articulate concerns and knowledge about them, so each part of the problem can be seen in the context of a wider constellation of issues and stakes. We would suggest that this process can enable the maturation of demand for science, and thereby inform the supply of science in a manner that enables intervention (McNie, 2007; Sarewitz and Pielke, 2007). Segre- gating the components of unstructured problems and dealing

171

Fig. 1 – Schematic process for diagnosing and intervening in the operating environment for sciences.

with them in isolation may be necessary; but components will eventually need to be re-articulated.

6. Conclusion

Considering interactions between science and decision-mak- ing in terms of boundary spanning design elements provides a means to better link science and decision-making, and thereby better reconciles the supply of and demand for science (McNie, 2007; Sarewitz and Pielke, 2007). Through a series of cases, and drawing on diverse literature, we have argued that such reconciliation can lead to outcomes where well-targeted science is deployed via appropriate boundary spanning design elements. Such design elements can expli- citly bring scientific information into debates about issues and stakes. Analysis of interactions between issues and stakes via boundary spanning elements, allows the analyst and partici- pants to potentially diagnose missing or ineffective boundary spanning elements that might enable more effective, efficient, equitable and legitimate processes for informing decision- making.

The approach of analysing the operating environment with the explicit intention to refine boundary design elements represents a substantial step beyond applications of generic principles. The approach outlined here begins to address constraints to the application of science, but will not allow all problems to become tractable. Where problems are unstruc- tured it may well be that the role of research becomes one of mediating, and documenting debates over values. We do argue, however, that the role of scientific information in varied decision-making contexts is mediated by the relative ade- quacy of boundary spanning design elements and the interactions between them. These design elements for boundary spanning provide a potentially fruitful focus for ongoing empirical, practical and theoretical work concerned with linking science, society and policy for sustainability.

Acknowledgement

This research was supported by the CSIRO Coastal Collabora- tion Cluster with funding from the CSIRO Flagship Collaboration Fund. We thank all participants in the case study research.

r e f e r e n c e s

Allenby, B.R., Sarewitz, D., 2011. The Techno-Human Condition. MIT Press, Boston, MA.

Cash, D.W., Borck, J.C., Patt, A.G., 2006. Countering the ‘‘Loading Dock’’ approach to linking science and decision-making: a comparative analysis of ENSO forecasting systems. Science, Technology and Human Values 31 (4) 465–494.

Cash, D.W., Buizer, J., 2005. Knowledge-action systems for seasonal to interannual climate forecasting. In: Roundtable on Science and Technology for Sustainability Policy and Global Affairs. National Research Council, Washington, DC.

Cash, D.W., Clark, W.C., Alcock, F., Dickson, N.M., Eckley, N., Guston, D.H., Jager, J., Mitchell, R.B., 2003. Knowledge systems for sustainable development. Proceedings of the National Academy of Sciences of the United States of America 100 (14) 8086–8091.

Clark, W.C., Tomich, T.P., van Noordwijk, M., Guston, D., Catacutan, D., Dickson, N.M., McNie, E., 2011. Boundary work for sustainable development: Natural resource management at the Consultative Group on International Agricultural Research (CGIAR). Proceedings of the National Academy of Sciences of the United States of America, http://dx.doi.org/ 10.1073/pnas.0900231108.

Coffey, B., Fitzsimons, J.A., Gormly, R., 2011. Strategic public land use assessment and planning in Victoria, Australia: four decades of trailblazing but where to from here? Land Use Policy 28 (1) 306–313.

Coffey, B., O’Toole, K., 2012. Understanding coastal knowledge dynamics: the potential of knowledge systems. Conservation and Society 10 (4) 318–329.

170

Cooke, B., Kothari, U., 2001. Participation: The New Tyranny?

Zed Books, London. Foucault, M., 1980. Power/Knowledge: Selected Interviews and

Other Writings, 1972–1977. Harvester, Brighton. Funtowicz, S.O., Ravetz, J.R., 1993. Science for a post-normal

age. Futures 25 (7) 739–755. Gieryn, T.F., 1983. Boundary-work and the demarcation of

science from non-science: strains and interests in professional ideologies of scientists. American Sociological Review 48, 781–795.

Gilbert, G.N., Mulkay, M.J., 1984. Opening Pandora’s Box. Cambridge University Press, Cambridge, UK.

Gunningham, N., Kagan, R.A., Thornton, D., 2004. Social license and environmental protection: why business go beyond compliance. Law & Social Inquiry 29, 307–341.

Guston, D.H., 2000. Between Politics and Science: Assuring the Integrity and Productivity of Research. University of Chicago Press, Chicago.

Guston, D.H., 2001. Boundary organisations in environmental policy and science: an introduction. Science, Technology and Human Values 26 (1) 399–408.

Hallegatte, S., 2009. Strategies to adapt to an uncertain climate change. Global Environmental Change 19 (2) 240–247.

Holling, C.S., Berkes, F., Folke, C., 1998. Science, sustainability and resource management. In: Berkes, F., Folke, C. (Eds.), Linking Social and Ecological Systems: Management Practices and Social Mechanisms for Building Resilience. Cambridge University Press, London, pp. 342–362.

Hoppe, R., 2011. The Governing of Problems: Puzzling, Power and Participation. Policy Press, Portland, OR.

Innes, J., Booher, D., 2004. Reframing public participation: strategies for the 21st century. Planning Theory & Practice 5 (4) 419–436.

Jasanoff, S., 1987. Contested boundaries in policy-relevant science. Social Studies of Science 17 (2) 195–230.

Jasanoff, S., 2003. Technologies of humility: citizen participation in governing science. Minerva 41, 223–244.

Jasanoff, S. (Ed.), 2004. States of Knowledge: The Co-production of Science and Social Order. International Library of Sociology. Routledge, London/New York.

Jentoft, S., 2000. Legitimacy and disappointment in fisheries management. Marine Policy 20, 141–148.

Kahneman, D., Tversky, A., 1996. On the reality of cognitive illusions. Psychological Review 103 (3) 582–591.

Keneley, M., OToole, K., Coffey, B., MacGarvey, A., 2013. Stakeholder participation in estuary management: the development of Victoria’s Estuary Entrance Management Support System (EEMSS). Australasian Journal of Environmental Management 20 (1) 49–62.

Leith, P.B., 2011. Public engagement with climate adaptation: an imperative for (and driver of) institutional reform? In: Whitmarsh, L., O’Neill, S., Lorenzoni, I. (Eds.), Engaging the Public with Climate Change: Behaviour Change and Communication. Earthscan, London, pp. 100–119.

Leith, P., Coffey, B., Haward, M., O’Toole, K., Allen, S., 2012a. Improving science uptake in coastal zone management: principles for science engagement and their application in South-eastern Tasmania. In: Kenchington, R., Stocker, L., Wood, D. (Eds.), Sustainable Coastal Management and Climate Adaptation: Lessons from Regional Approaches in Australia. CSIRO Publishing, Collingwood, pp. 135–155.

Leith, P., Ogier, E., Rees, C., Haward, M., 2012b. Progressing analysis of the structure and function of knowledge-action systems for coastal zone management. In: Presented at Coast to Coast 2012 Conference, Brisbane 17–21 September 2012, http://www.coast2coastaustralia.com/2012- presentations-a-l/.

Leith, P., Ogier, E., Haward, M., 2014. Science and social license: defining environmental sustainability of Atlantic Salmon

aquaculture in South-eastern Tasmania, Australia. Social Epistemology (in press).

McNie, E.C., 2007. Reconciling the supply of scientific information with user demands: an analysis of the problem and review of the literature. Environmental Science & Policy 10 (1) 17–38.

Nelson, R.A., Howden, M., Stafford Smith, M., 2008. Using adaptive governance to rethink the way science supports Australian drought policy. Environmental Science & Policy 7, 588–601.

Oreskes, N., 2004. Science and public policy: what’s proof got to do with it? Environmental Science & Policy 7, 369–383.

O’Toole, K., Keneley, M., Coffey, B., 2013. Participatory logic and coastal management under the project state: the case of the Estuary Entrance Management Support System (EEMSS) in Victoria, Australia. Environmental Science and Policy 27, 206–214.

Pielke Jr., R.A., 2007. The Honest Broker: Making Sense of Science in Policy and Politics. Cambridge University Press, New York.

Sarewitz, D., 2004. How science makes environmental controversies worse. Environmental Science & Policy 7, 385–403.

Sarewitz, D., Pielke Jr., R.A., 2007. The neglected heart of science policy: reconciling supply and demand for science. Environmental Science & Policy 10, 5–16.

Schattscheneider, E.E., 1960. The Semisovereign People: A Realist’s View of Democracy in America. Holt, Rinehart and Winston, New York.

Scho n, D.A., Rein, M., 1994. Frame Reflection: Toward the Resolution of Intractable Policy Controversies. Basic Books, New York.

Sherwood, J., Mondon, J., Fenton, C., 2008. Classification and management issues of estuaries in western Victoria, Australia. Proceedings of the Royal Society of Victoria 120 (1) 257–276.

Star, S.L., Griesemer, J.R., 1989. Institutional ecology, ‘‘translation’’ and boundary objects: amateurs and professionals in Berkeley’s Museum of Vertebrate Zoology. Social Studies of Science 19 (3) 387–420.

Turnhout, E., Hisschemo ller, M., Eijsackers, H., 2008. Science in Wadden Sea policy: from accommodation to advocacy. Environmental Science & Policy 11 (22) 227–239.

Whatmore, S.J., 2009. Mapping knowledge controversies: science, democracy and the redistribution of expertise. Progress in Human Geography 33 (5) 587–598.

Wynne, B., 1992a. Misunderstood misunderstandings: social identities and the public uptake of science. Public Understanding of Science 1, 281–304.

Wynne, B., 1992b. Uncertainty and environmental learning: reconceiving science and policy in the preventive paradigm. Global Environmental Change 3.

Wynne, B., 1994. Scientific knowledge and the global environment. In: Redclift, M., Benton, T. (Eds.), Social Theory and the Global Environment. Routledge, London, pp. 169–189.

Peat Leith is a research fellow at the Tasmanian Institute of Agriculture, University of Tasmania (UTAS).

Kevin O’Toole is associate professor in politics and policy at Deakin University.

Marcus Haward is professor at the Institute of Marine and Ant- arctic Studies, UTAS.

Brian Coffey is a post-doctoral research fellow at Deakin University.

Chris Rees is research fellow at the Institute of Marine and Ant- arctic Studies, UTAS.

Emily Ogier is research fellow at the Institute of Marine and Antarctic Studies, UTAS.