A strategic assessment of scientific and behavioural perspectives on 'dangerous' climate change

Irene Lorenzoni, Tom Lowe and Nick Pidgeon June 2005

Tyndall Centre for Climate Change Research

Technical Report 28

A strategic assessment of scientific and behavioural perspectives on ‘dangerous’ climate change
Tyndall Centre Technical Report No. 28 June 2005 Irene Lorenzoni, Tom Lowe and Nick Pidgeon

Centre for Environmental Risk, and Tyndall Centre for Climate Change Research, UK, Zuckerman Institute for Connective Environmental Research School of Environmental Sciences University of East Anglia, Norwich, NR4 7 TJ

This is the final report from Tyndall research project: T3.32 A strategic assessment of scientific and behavioural perspectives on ‘dangerous’ climate change.

Abstract Defining ‘dangerous’ climate change is of increasing importance for both scientific analysis and the climate policy debate. This report presents the findings of a project examining how ‘dangerous’ climate change is interpreted through scientific risk assessments, how laypeople and experts perceive the issue and the implications for managing climate change. Whereas the former build upon defining external thresholds and critical levels beyond which substantial change would occur, most people relate to climate change through direct personal experience. A review of the literature and more recent studies shows that in most developed nations, such personal experience is perceived to be limited and, although there is widespread concern about the issue, people are ambivalent about the threat of climate change and potential solutions. ‘Danger’ is context and time-specific, defined by social and political judgements. Thus, the discrepancy between expert and lay perceptions of climate change and danger may prove problematic in implementing policy options under conditions of uncertainty without consideration of public concerns over particular risks and their preferences for management. Participants to an international workshop, and in interviews undertaken using mental modelling, identified difficulties with the current policy approaches to addressing climate change, which are driven in part by international negotiations around UNFCCC article 2. Issues relating to the role of science, uncertainty, power, equity, social expectations and lifestyles were raised. To promote action on climate change, it was suggested that climate change should be situated in people’s daily lives, weaving it into related policy areas and including it within the broader scope of sustainable development.

Keywords dangerous climate change, perceptions, thresholds, risk assessment and management, lay people and experts

Section 1 - Overview of project work and outcomes Section 2 - Technical report 1 Introduction 2 Phases 1 and 2 – Interpretations of ‘danger’ in relation to climate change 3 Phase 4 – International workshop 4 Phase 3 – ‘Expert’ definitions of danger in relation to climate change 5 Overall project conclusions Page 4 7 7 8 26 29 60

BIBLIOGRAPHY APPENDICES Appendix 1 – List of project outputs to date Appendix 2 - "Perspectives on Dangerous Climate Change" participants Appendix 3 – Interview protocol Appendix 4 – Meta mental model of ‘expert’ perceptions of danger in relation to climate change


70 71 72 75


Acronyms used in this report
AIDS – Acquired Immunodeficiency Syndrome BSE – Bovine Spongiform Encephalopathy CAN – Climate Action Network CER – Centre for Environmental Risk, University of East Anglia, Norwich, UK CFCs - Chlorofluorocarbons COP – Conference of the Parties CO2 – Carbon dioxide DEFRA – Department of the Environment, Food and Rural Affairs, UK DTI – Department of Trade and Industry, UK ECF – European Climate Forum ENSO – El Niño / Southern Oscillation EORG - European Opinion Research Group GDP – Gross Domestic Product GHG – Green house gas GM – Genetically Modified HOT - Helping Operationalise article Two [research project] ICCT - International Climate Change Taskforce IPCC – Intergovernmental Panel on Climate Change MORI – Market & Opinion Research International NAO – North Atlantic Oscillation PIK – Potsdam Institute for Climate Impact Research RCEP – Royal Commission on Environmental Pollution, UK THC – Thermohaline Circulation UKCIP – UK Climate Impacts Programme UK Met Office – UK Meteorological Office UNFCCC – United Nations Framework Convention on Climate Change UNESCO – United Nations Educational, Scientific and Cultural Organisation WAIS – West Antarctic Ice Sheet WBGU – German Advisory Council on Global Change WEHAB – Water, Energy, Health, Agriculture and Biodiversity [five key focus areas proposed at the Johannesburg World Summit on Sustainable Development, 2002] WISE - Weather Impacts on Natural Social and Economic Systems [EU research]


Section 1 - Overview of project work and outcomes
Abstract Defining ‘dangerous’ climate change is of increasing importance for both scientific analysis and the climate policy debate. This report presents the findings of a project examining how ‘dangerous’ climate change is interpreted through scientific risk assessments, how laypeople and experts perceive the issue and the implications for managing climate change. Whereas the former build upon defining external thresholds and critical levels beyond which substantial change would occur, most people relate to climate change through direct personal experience. A review of the literature and more recent studies shows that in most developed nations, such personal experience is perceived to be limited and, although there is widespread concern about the issue, people are ambivalent about the threat of climate change and potential solutions. ‘Danger’ is context and timespecific, defined by social and political judgements. Thus, the discrepancy between expert and lay perceptions of climate change and danger may prove problematic in implementing policy options under conditions of uncertainty without consideration of public concerns over particular risks and their preferences for management. Participants to an international workshop, and in interviews undertaken using mental modelling, identified difficulties with the current policy approaches to addressing climate change, which are driven in part by international negotiations around UNFCCC article 2. Issues relating to the role of science, uncertainty, power, equity, social expectations and lifestyles were raised. To promote action on climate change, it was suggested that climate change should be situated in people’s daily lives, weaving it into related policy areas and including it within the broader scope of sustainable development. Objectives Dessai et al. (2004) argue that it is not possible to developing sustainable responses to climate change without recognising the central role of perceptions of danger which mould, affect, influence and shape policy and its uptake. Interaction between climate change and risk researchers has been limited. Whereas the former have focused on definitions of ‘dangerous’ climate change by experts which tend to be ‘externally’ described, little attention has been dedicated to how ‘non-experts’ view climate change, a task which lies more within the remit of risk research. The objectives of this project are to investigate experts’ and non-experts’ definitions of dangerous climate change by: 1. Reviewing and analysing data and literature in the remit of risk research to obtain indications of how non-experts relate to climate change and how ‘dangerous’ climate change may be perceived; 2. Comparing expert and non-expert perceptions and definitions; 3. Bringing together leading scholars in the climate science and socio-psychological disciplines and stakeholders to discuss (a) interpretations and perspectives relevant to their own field of expertise; (b) similarities and differences among these; (c) implications for policy-making. Underlying this project is the assumption that climate change responses cannot be adequately organised and managed if discrepancies exist between working definitions of danger by climate communities and non-experts’ perception of climate change, which equally shape policy-making. Crossovers and links between climate and risk communities to date have been rare. This strategic assessment sets out to investigate the nature of this gap and to forge a deeper understanding between communities in order to enable the search for constructive policy solutions to climate change as seen from various perspectives.


Work undertaken This project was undertaken in four phases, the first two which were effectively merged: Phases 1 and 2 - Analyses of public perceptions of climate change These two phases drew upon datasets available to CER and worldwide literature on lay perceptions of climate change. Four strands of work on lay perceptions of climate change in the UK, undertaken under the Programme for Understanding Risk by CER, were analysed and synthesised to explore the influences on perceptions of risk in relation to climate change among laypeople (i.e. a CER / MORI survey and focus groups undertaken in 2002; CER reviews of psychometric research spanning the past 20 years and postgraduate work on climate change linked to the Centre). These data were cross-referenced with similar work carried out internationally, such as the collaborations with Tony Leiserowitz (Decision Research, USA) and Timo Rusanen (University of Kuopio, Finland). Additionally, some German survey data are also comparable with the CER/MORI poll. The role of the media in shaping perceptions of climate change was also investigated based on a review of available literature. More direct input on this subject was provided by the attendance at the international workshop (see Phase 4) of scholars who had analysed these aspects in more detail with reference to the UK. Phase 3 – ‘Expert’ definitions of danger in relation to climate change The aim of this phase was to examine ‘expert’ definitions of danger within the climate science community. Dessai et al. (2004) have already provided an indication of some of the various definitions in existence. We expanded that review to a wider set of sources available in the literature, triangulated through exploratory interviews with leading exponents of the climate change research community, analysed through a mental models approach (Morgan et al., 2002). The outputs from this phase were then compared with findings of Phases 1 and 2. Phase 4 – International workshop The aim of this event was to bring together academics of a social/psychological background with natural scientists to debate the notion of ‘dangerous’ climate change from their perspectives and in terms of implications for policy-making. The workshop was held at UEA on 28 and 29 June 2004, funded jointly by the Tyndall Centre for Climate Change Research and the Leverhulme Trust. Overall, 35 attendees officially participated, including participants from North America, Oceania, continental Europe, and the UK in addition to researchers at UEA Norwich. Results Phases 1 and 2 of the project support previous findings that many laypeople are ambivalent towards climate change. Although there is widespread concern about the issue, individuals find it difficult to balance the risks and benefits (from technologies and energy sources) with the levels at which these are manifest (personal, societal), compounded also by the time component of causes and effects (present vs. future). Societies tend to define climate change through experience, which is generally discrepant with expert notions of ‘dangerous’ climate change, specifically defined by risk and vulnerability assessments. This may prove problematic if policy is implemented under conditions of uncertainty without consideration of public concerns over particular risks and their preferences for management. Participants to an international workshop and in interviews undertaken using mental modelling highlighted the difficulties inherent in current approaches to addressing climate change, driven in part by international negotiations around UNFCCC article 2. To promote action on climate change, it was suggested that climate change should be situated in people’s daily lives, weaving it into related policy areas and including it within the broader scope of sustainable development. Underlying this notion was a general disenchantment with the applicability and operationalisation of the international agreement on climate change, which is seen as generally incompatible with

context and time specific notions of danger used by laypeople. The experts in particular articulated three conceptualisations of danger, defined by different perspectives: (i) interference with the climate system (therefore related to mitigation); (ii) severity of impacts to which it is not possible to adapt; (iii) changes to current societies (e.g. costs of mitigating). These discussions invariably focused on issues of decision-making under uncertainty, the role of governments in democratic societies, power, trust and inequity. Thus, whilst the important role of wise leadership was acknowledged, the complexity and pervasiveness of climate change suggested that attention should focus on how to engage the public globally, by promoting relevant and equitable solutions. Relevance to Tyndall Centre research strategy and overall Centre objectives This project provides insights into perceptions of climate change risk by laypeople as well as ‘experts’ from different disciplines associated with climate change, and how these bear upon the various interpretations to be placed upon ‘dangerous’ climate change. This Strategic Assessment also provides theoretical insights into perception formation and behavioural change. As such, the findings of this project are relevant to the adaptation and mitigation research areas within the Tyndall Centre, through one of its foci on determinants of risk and inertia in behavioural change, as well as to climate change policy more widely. Thus, although this project was originally categorised as falling within the remit of Tyndall Theme 3 (“Adapting to climate change”), its results relate to a wider set of research within the Tyndall Centre. In particular, one of the medium term objectives of the Tyndall Centre is to “Motivate Society – promote informed and effective dialogue across society about its ability and willingness to choose our future climate”. This project has reflected on how climate change is perceived by laypeople, especially in Western societies, and includes suggestions deriving from the international workshop and data analysis on how to further public engagement with the risks posed by climate change with a view to potentially translating these into behavioural change. Potential for further work Suggestions for future research were identified throughout the international workshop (and are listed at the conclusion of section 3 of this report). Others include: - Engaging civil society: what scope is there for ‘situating’ climate change in people’s lives? - Practically examine what combinations of incentives and measures enable and encourage individuals to mitigate climate change; - Investigate the scope for multi-purpose policies, their drawbacks and potential for negative interactions or overlaps; - Explore whether there are critical thresholds that encourage an individual to translate their motivation to act into behavioural change, and whether this is affected differently by gradual vs. rapid change (in the climate); - Re-assessing definitions of danger in relation to climate change after the ratification of the Kyoto Protocol. Communication highlights • International workshop on Perspectives on Dangerous Climate Change, 28 and 29 June 2004; co-funded by the Tyndall Centre for Climate Change Research and the Leverhulme Trust. • Lorenzoni, I. and Pidgeon, N. (2004) Public views on climate change: European perspectives of a long-term risk. Paper prepared for and presented at the Workshop on Global Warming: The Psychology of Long Term Risk, Princeton University, 12 November 2004. • Lorenzoni, I. and Pidgeon, N. (2005) Closing the Gap, Defining Dangers of Climate Change and Individual Behaviour. Poster presentation at the Avoiding Dangerous Climate Change conference, Hadley Centre and Met Office, Exeter, 1-3 February 2005. • A special journal issue: Lorenzoni, I., Pidgeon, N. and O’Connor, R. (eds.) (in preparation) Perspectives on Dangerous Climate Change. Special issue of Risk Analysis.


Section 2 - Technical report
1 Introduction

The objective of the 1992 UN Framework Convention on Climate Change (FCCC), as stated in its second article, “… is to achieve … stabilization of greenhouse gas concentrations in the atmosphere at a level that would prevent dangerous anthropogenic interference with the climate system. Such a level should be achieved within a time-frame sufficient to allow ecosystems to adapt naturally to climate change, to ensure that food production is not threatened and to enable economic development to proceed in a sustainable manner”. Understanding what constitutes dangerous climate change is increasingly important for both scientific analysis and the climate policy debate. Interpretation is crucial for determining the extent and level of time and resources for managing climate change. This strategic assessment set out to further the understanding of what constitutes ‘dangerous’ climate change from climate science and risk research perspectives. Whereas the former has focused on definitions by experts, less attention has been dedicated to how ‘non-experts’ view climate change. Climate experts often focus upon notions of thresholds of danger beyond which a substantial change takes place. But definitions of danger also depend on lay judgements about the valued characteristics of ecosystems and human systems. In other words, what risks from climate change are acceptable to society? This project investigates the nature of this gap, with a view to questioning current perspectives and to enable the search for constructive solutions to climate change. Insights into perceptions and images of climate change risk (danger, tolerable change) are brought to bear upon the issue of understanding perception of dangerous climate change. This project uses a combination of methods and various datasets to obtain indications of how non-experts relate to climate change and how ‘dangerous’ climate change may be perceived; and to compare expert and non-expert perceptions and definitions. This technical report is divided into three main sections, reflecting the four phases of the project (as outlined in section 1). We report on Phases 1 and 2 initially, followed by Phase 4 (international workshop) as the latter naturally followed on from the first two. Finally, we outline the findings of the mental models approach applied to expert views of danger in relation to climate change (Phase 3).



Phases 1 and 2 – Interpretations of ‘danger’ in relation to climate change

2.1 ‘Dangerous’ climate change: scientific interpretations According to the Oxford English Dictionary (2003), danger is the “liability or exposure to harm or injury; the condition of being exposed to the chance of evil; risk, peril”. UNFCCC article 2 suggests that harm would derive to human and natural systems if certain disruptions were to occur as a consequence of excessive concentrations of greenhouse gases in the atmosphere. Three elements are key to defining such danger, as per article 2: (i) the impacts deriving from changes to the climate system, (ii) the concentrations of greenhouse gases (GHGs)1, and (iii) the rate (timing) of any change in climate parameters. By focusing on the notions of ‘threshold’ “beyond which many believe substantial climate change would occur” (Schneider, 2001:18; also Parry et al., 1996), the climate community has predominantly worked with tools at their disposal, defining danger through assessments of risk and vulnerability (Table 2.1). Table 2.1 Examples of external definitions of 2004). Danger measured through thresholds in physical vulnerability (various sources) Large scale eradication of coral reef systems dangerous climate change (source: Dessai et al.,

Danger measured through thresholds in social vulnerability (various sources) Irrigation demand exceeding 50% of annual seasonal water usage for agriculture in northern Victoria, Australia Disintegration of the West Antarctic Ice Depopulation of sovereign atoll countries Sheet (WAIS) Breakdown of the Thermohaline Additional millions of people at risk from Circulation (THC) water shortage, malaria, hunger and coastal flooding Modification of crucial climate-system Destabilisation of international order by patterns such as ENSO, NAO environmental refugees and emergence of conflicts Climate change exceeding the rate at World impacts exceeding a threshold which biomes can migrate percentage of GDP

Patwardhan et al. (2003:4) extend these notions further by arguing that a more in-depth understanding is required of what limits (i.e. levels and / or rates) of climate parameters are associated with critical impacts on key areas for human well-being, taking the WEHAB2 components as reference points. These can be conceptualised, according to the authors, using the notion of critical thresholds, of which the authors define two types: • Thresholds of type I that would entail smooth changes, which could at some point result in damages that could be considered ‘unacceptable’ by policy-makers. The authors envisage that the threshold would be defined through a socio-political process involving consideration of relative risks and benefits to particular sectors / regions. It is clearly thresholds of this type that should be the subject of societal decision-making to define what constitutes dangerous climate change and impacts thereof.

The IPCC and other government departments (such as DEFRA in the UK) emphasise that ‘stabilisation’ refers to the concentrations of GHG in the atmosphere, not CO2 alone. Technically, stabilising emissions or energy intensity will not achieve climatic stability. 2 The WEHAB framework was introduced at the World Summit on Sustainable Development held in Johannesburg in 2002 and identifies key areas for human well-being and development, namely Water, Energy, Health, Agriculture and Biodiversity.



Thresholds of type II: “natural bounds, which if exceeded, would lead to major, potentially irreversible, impacts. It is very likely that the irreversibility and scale of such changes would be considered ‘unacceptable’ by virtually all policy-makers and would thus qualify as dangerous change”. Some examples of thresholds of this type are identified in the left-hand column of Table 2.1.

Similarly, the UK Hadley Centre defines dangerous climate change on the basis of sudden change (“some components of the climate system which could change abruptly”) or thresholds beyond which changes will be inevitable and irreversible. The aim of the Centre is to quantify the risk from high-impact climate events where possible, or alternatively estimate the global and regional consequences of these events, which would provide plausible worst case climate scenarios for impacts studies and adaptation response planning (Hadley Centre, 2004). Another aspect not covered in Table 2.1 is the concept of danger as a function of adaptive capacity. Human, and up to a certain extent natural, systems are not solely passive recipients of climatic changes. In addition to contributing to these changes through GHG emissions, they can also modify their responses to reduce negative impacts or maximise positive impacts deriving from these changes. Time is the other important component of ‘danger’. If the onset of a change in the climate is faster than the capability for adaptation, an event is likely to be more ‘dangerous’ than it would have been if changes were progressive (e.g. ‘regular’ climate change). “The capacity to adapt is enhanced by foresight of upcoming changes and reduced when changes are abrupt, large and / or relatively unforeseen. Thus, abrupt changes are likely to be more dangerous than changes that are more slowly evolving and better foreseen”. “[…] The ability to adapt is itself a measure of whether any particular climate stimulus is ‘dangerous’.” (Mastrandrea and Schneider, 2001:444). Thus, a dangerous climatic outcome could be considered a function of the inter-relatedness and dependency of timing and adaptive capacity, a characteristic known as “tight coupling” in normal accident theory (Perrow, 1984:8). Dangerous climate change will therefore be defined not only by what may occur when crossing a physical threshold, but also by what it is perceived or experienced to be, and therefore by the methods enacted to counteract that danger. The other two aspects of article 2 which have been the subject of scientific analysis are the levels of GHG emissions that could stabilise the climate, in order to avoid adverse impacts on human and natural ecosystems, and the timing at which this would occur. O’Neill and Oppenheimer (2002), for instance, infer an upper limit for GHG concentrations and the timescales during which these can be achieved to prevent damage to reef systems, protect the WAIS or aver the shutdown of the THC. They estimate that a stabilisation of CO2 concentrations at 450 parts per million (ppm) by 2100 (leading to a range of warming of 1.2-2.3oC) could probably, but not certainly, avoid the collapse of the WAIS, likely prevent the THC from closing down, but incur damage to at least one vulnerable ecosystem (e.g. coral reefs). Timing is crucial: the authors estimate that meeting the Kyoto targets ten years later (in 2020 rather than 2010) could preclude us from stabilising concentrations at 450 ppm by the end of the century. The scientific uncertainties inherent in defining thresholds, or even one single threshold against which the danger of climate change can be gleaned, exemplify the difficulties in translating the notion of danger into policy-useful guidance. Yet based on scientific research some governments and organisations already appear to have made a policy judgement on what constitutes ‘dangerous’ climate change, thus transposing some external definitions into their modes of operation. For instance, the German Advisory Council on Global Change (WBGU, 2003) examined the threshold beyond which the effect on the three criteria specified in article 2 would no longer be

tolerable, estimating that “the threshold from which damage to the global natural heritage is no longer acceptable… [is] in the range of 2oC global warming relative to pre-industrial values” (WBGU, 2003:1). The WBGU recommend that “to avert dangerous climatic changes, it is essential to comply with a ‘climate guard rail’ defined by a maximum warming of 2oC relative to preindustrial values” (WBGU, 2003:2). As temperatures have already risen by 0.6oC from the times of the industrial revolution, only an increase of another 1.4oC is “tolerable”, although this should not occur at a rate faster than a 0.2oC warming per decade. The WBGU suggests the European Union (EU) should adopt a leading role in promoting the acceptance of this ‘climate window’ within the context of the UNFCCC process3. Given the uncertainties inherent in the climate system, the WBGU also recommends ambitious emissions reductions of 45-60% globally from 1990 levels by 2050 to stabilise CO2 concentrations below 450 ppm. This means that industrialised nations would need to reduce their GHG emissions by 20% by 2020. These reductions and targets are consistent with those proposed by various UK commissions (RCEP, 2000). In 2003, the UK Government announced its aspiration to reduce domestic CO2 emissions by 60% by 2050 from 1990 levels, which should result in a cut of about 65 million tonnes of CO2 (DTI and DEFRA, 2003). However, recent statistics show that at European levels, despite commitments to the Kyoto Protocol targets, GHG emissions have recently been rising (Gugele et al., 2003). This poses the question of how, realistically, GHG concentrations can be stabilised. Part of this depends upon societal views on climate change: whether people consider its manifestations dangerous and if they support climate change policies. These factors, amongst others, will influence their practical achievability. For instance, a Greenpeace report (Hare, 1998) suggests that to limit the rate and magnitude of climate change within this century to levels tolerable to human and natural systems (i.e. no significant damage) a ceiling needs to be placed on the amount of carbon dioxide released from the burning of fossil fuels. Greenpeace has calculated this ceiling or carbon budget to be around 225 billion tonnes of carbon. To meet this carbon budget and stabilise global temperature increases below 1oC in the long-term, it argues that 75% of known, economically recoverable reserves of conventional fossil fuels should remain untouched. This raises issues of whether such limits are acceptable and / or tolerable, from policy, public and climate science perspectives. It is clear, therefore, that risk assessments cannot determine what is ‘dangerous’ purely on ‘scientific’ bases without some judgement about what is acceptable (Pidgeon et al., 1992; Pidgeon, 1998), and what level of risk is tolerable, by way of a policy response to this assessment (Gerrard, 2000). Tolerability and acceptability are socially constructed. Especially in cases such as climate change, a complex and pervasive phenomenon, shrouded in uncertainty and where management stakes are high, traditional forms of science and policy making alone cannot find and deploy solutions to such an issue. It requires taking into account the context in which risk occurs and therefore the multiple views and values which will drive decisions in the face of uncertainty. Equally, decisions cannot be postponed until full understanding is available. Furthermore, decisions taken in the present will require leaving options open for the future. Within this context Funtowicz and Ravetz (1993:739) and others have argued for ‘post-normal science’ as a means to initiate and facilitate responses to societal issues and tensions, paving the
Various organisations argue that temperatures should be kept well below the 2oC average (above pre-industrial levels) as this could already result in adverse impacts to human and natural systems or rapid irreversible changes. For instance, the European Union Environment Council Decision of 1996 proposed that temperature increases should not be allowed to exceed 2oC and concluded that CO2 concentrations should be kept below 550 ppmv (parts per million volume) (EC, 1996). Others include the World Wide Fund for Nature (www.panda.org/about_wwf/what_we_do/climate_change/problems/index.cfm), the Climate Action Network (CAN, who define possible changes under temperatures 1-2 oC higher as unacceptable; www.climnet.org/pubs/CANadequacy30102002.pdf), the Alliance of Small Island States and the International Climate Change Taskforce (ICCT, 2005).


way towards a “democratization of science”. Post-normal science is an attempt to broaden the knowledge base in policy processes. In the context of climate change, this implies moving away from traditional notions of communicating science to the public towards embracing the understanding that there are multiple “publics” with different views and response patterns which need to be taken into account if policymaking on such a complex issue is to address the roots of the problem successfully. Failure to take public values and views into consideration when taking decisions on climate risk management issues will inevitably prove problematic. A number of reasons can be advanced for this. At a very basic level climate policies will require a degree of ‘buy-in’ or acceptance from those who will be affected by them if they are to be successfully implemented. Equally, where public policy and citizen frames of reference differ (e.g. regarding the balance between long-term and short-term considerations) the practice of risk communication becomes much more difficult, while policy implementation may be misunderstood, neglected or even opposed by the electorate (Pidgeon, 1998). Such considerations reflect the operational and dynamic definition of danger that Dessai et al. (2004) call for, possibly based upon elicitation and mediation of values with a view to integrating external and internal perspectives of climate change for its management. The basis for post-normal science and decision-making processes is the acknowledgement that there are multiple thresholds defined by the societal values attributed to ecosystem functions and other elements integral to human livelihoods. Crucial in these multiple assessments are therefore who / what is exposed to a certain risk, how that risk compares to others and the practical considerations of taking action. All of these issues were raised in various terms at several stakeholder workshops held world-wide as part of the project Helping Operationalise article Two (HOT): a science-based policy dialogue on fair and effective ways to avoid dangerous interference with the climate system and implications for post-Kyoto policies. For many stakeholders ‘dangerous anthropogenic interference’ was difficult to define, especially as this was context dependent and arbitrary, with no agreed definitions or accepted notions of the three impacts criteria. Others argued that article 2 did not direct enough attention to distributional and equity impacts, and the implications for human health. They also reinforced the notion that the danger of climate change is context-specific: some areas around the globe are already experiencing this form of change; while unacceptable outcomes of climate policies are multiple and defined by stakeholders’ interests, perceptions and experience (Gupta et al., 2003). To this effect, risk perception research provides an insight into individual and social considerations of climate change, and the implications of these in light of article 2. These are examined in the sections below. 2.2 Lay perceptions of climate change Dessai et al. (2004) argue that external definitions of dangerous climate change (i.e. ‘risk analysis of system characteristics of the physical or social world’) can be at odds with internal definitions of danger. Danger as insecurity or lack of safety is clearly dependent upon personal experiences, values, information, trust, cultural and institutional processes. This section provides a summary of the several consistent trends in studies of public perceptions of climate change revealed in the past two decades by quantitative and qualitative risk studies. These are mostly based on studies conducted in developed nations, as it has proven almost impossible to find documented perceptions studies conducted in developing countries. Awareness of climate change The issue of climate change has woven its way into the general public consciousness. Climate change generates awareness and concern around the globe, although these are nationally-specific (Dunlap and Scarce, 1991; Bord et al., 1998; MORI, 2002; Brechin, 2003). As with many other

environmental issues, the changing social context at any particular point in time (e.g. the activities of interest groups or the media) can serve to amplify or attenuate perceptions of climate change risk (Pidgeon et al., 2003). However, the importance of climate change tends to be secondary in relation to other environmental, personal and social issues (Witherspoon, 1994; Dunlap, 1998; Thompson and Rayner, 1998; INRA Europe, 1999; Zwick and Renn, 2002; Brechin, 2003; Lorenzoni, 2003). For instance, although 67% of respondents of a 2003 British survey maintained climate change was important or very important to them, more often their main priorities lay with health, family, safety and finances (Poortinga and Pidgeon, 2004 – see Table 2.2). According to the EuroBarometer 58.0 of 2002, the most worrying environmental risks for most European respondents were associated with nuclear power and radioactive waste, and industrial activities, pollution, natural disasters and ozone, followed by climate change (very worrying for 39% of respondents in the 15 EU Member States). The survey analysts explain these differences by relating them to media reporting: the issues that elicit more worry are linked to industrial safety and more traditional environmental problems, frequently mentioned by the media over the last 30 years (EORG, 2002:9). The lower worry about climate change, on the other hand, could be explained if it is considered an ‘old’ environmental issue which has nevertheless only very recently received more extensive media coverage. Table 2.2. Risk in context: the importance of personal (P) and social (S) issues (% responses).
Not at all important (1) 0* 1 0* 0 1 0* 1 1 0* 0* 0* 0* 3 2 1 2 9 3 2 3 5 3 8 9 17 2 0* 1 1 0* 2 1 1 2 2 1 1 1 3 3 2 2 4 3 3 5 5 5 7 8 11 Neither important nor unimportant (3) 2 4 3 4 4 6 6 9 7 7 7 8 14 14 12 18 14 20 20 29 25 25 29 33 35 4 10 10 16 18 17 23 26 24 31 33 36 34 26 31 37 36 32 36 42 32 33 39 27 26 17 Very important (5) 87 85 80 77 75 69 65 63 59 58 56 56 53 49 46 40 40 37 33 29 29 28 26 21 19 Don’t know 0* 0* 0* 0* 0* 0* 0* 1 0* 0* 0* 0* 1 0* 1 0* 1 0* 1 1 3 1 2 3 0*

Your Health (P) Partner and Family (P) Law and Order (S) Personal Safety (P) Education (S) Being Independent (P) Your Privacy (P) Terrorism (S) Environmental Protection (S) Having a Comfortable Life (P) Personal Finance (P) Social Relations/Friends (P) Radioactive Waste Animal Welfare (S) The Economy (S) Excitement/Fun (P) Work (P) Tackling World Poverty (S) Tackling Human Rights (S) Population Growth (S) Genetic Testing CLIMATE CHANGE Radiation From Mobile Phones GM Food Religion (P)

Source: UEA/MORI GM Food Survey 2003 (Weighted dataset, N =1,363) (Poortinga and Pidgeon, 2004). Note: the non-empty cells (<0.5) denoted by (*) were rounded to 0.

Such findings are, however, not always mirrored in other surveys. For instance, of the 1,508 respondents in a German survey in 2001 in the region of Baden-Württemberg, 21% felt highly threatened and 48% moderately threatened by climate change. In comparison to the other risks surveyed in this same study, climate change elicited more worry and concern than did a range of other risks (i.e. BSE, genetically-modified food, crime, nuclear power, smoking and radiation from mobile phones). Zwick (2002) found it difficult to explain this discrepancy, given the paucity of media coverage of the issue preceding the administration of the survey and a lack of direct personally-experienced weather events that may have influenced people’s views; perhaps those individuals who perceived climate change as threatening considered the potential risks associated with climate change as acutely threatening hazards to health or life. Two UK surveys undertaken in 2004 indicated that although most people have heard of global warming, and rate it as the most important environmental issue for the world today, they see terrorism and domestic issues as having a higher priority (Norton and Leaman, 2004; Kirby, 2004). These findings contrast with those of a 2001 survey, according to which Britons (33%) and Europeans (31%) rated 'the environment' as the most important global problem (MORI, 2001). In qualitative terms, laypeople often also use the terms weather and climate interchangeably, partly because short-term weather changes are more directly perceptible than long-term climate patterns (Read et al., 1994; Gowda et al., 1997; Murlis and Davies, 2001)4. Climate change is happening and will continue Many people believe the climate has already changed and would continue to do so in the future (Kempton et al., 1995 – US surveys; Dunlap, 1998 - survey of laypeople in Canada, USA, Mexico, Brazil and Portugal; Lorenzoni, 2003 – survey of laypeople and students in Italy and the UK). Whilst investigating public views of current extreme weather events and future climate change during 1997 to 1999 in UK, Italy, German, The Netherlands, Palutikof and colleagues (1999) also noted that a majority of respondents of all nationalities believed that future climate change would happen, but that it would be overall undesirable. Misunderstandings of the issue Laypeople have been shown to manifest confusion over the causes of climate change, believing it may be due to pollution in general, nuclear power, aerosol sprays or, conversely, that the burning of fossil fuels (oil, coal, and gas) is not linked to CO2 emissions (Thompson and Rayner, 1998). Research has also shown that laypeople associate climate change with ozone depletion and skin cancer (e.g. Bostrom et al., 1994 – see also Kempton et al., 1995). Similar findings from other studies have been interpreted as people confusing the two environmental issues (Stamm et al., 2000; see also below) perhaps because of the close nature of the problems and the ease of mentally depicting the ozone hole, in contrast with the difficulty of appreciating the intricacies of climate change (Ungar, 20005). However, some respondents holding knowledge of the links between ozone

Weather, according to climate scientists, indicates fluctuations of the atmosphere whereas climate is a longer-term view of weather, described statistically over periods of time, allowing the detection of variability and extremes, and enabling the calculation of averages (UK Met Office, 2002). 5 Why was the ozone hole capable of engendering some public understanding and concern, while climate change failed to do so? Ungar (2000) argues that the ozone threat encouraged the acquisition of knowledge because it was allied to, and resonated with, easy-to-understand bridging metaphors derived from popular culture, and engendered a "hot crisis" providing a sense of immediate and concrete risk with everyday relevance. It broke through the veil of ignorance, became reasonably understood, and engendered consumer boycotts and other actions. At the time of writing, climate change fails to generate such effects.



depletion and climate change should not be entirely ruled out, given recent scientific evidence on this point6. Difficulties have been observed among adults (Kempton et al., 1995) and students in the US (Gowda et al., 1997) in conceptualising the potential adverse effects of changes in average global temperatures, which scientists use to describe climate change, especially as individuals often experience daily or monthly fluctuations of more than 4oC (corresponding to the mid-upper limit of IPCC projections of average global temperature changes by the end of the 21st century; Houghton et al., 2001). Preliminary work on affective images associated with climate change amongst the British and American publics (based on surveys undertaken in summer 2002 and over the period Nov 2002 – February 2003 respectively), and analysed as part of this Strategic Assessment, indicates that both American and British respondents mostly referred to generic manifestations of the phenomenon, such as increases in temperature, to a different environmental problem (ozone), and to negative affective outcomes. Associations with personally relevant impacts of climate change, or its causes and solutions, were infrequent, suggesting that the impacts of climate change are geographically and psychologically distant for most individuals in both nations (Lorenzoni et al., submitted). The theoretical associations with the operation of the ‘affect heuristic’ (see Slovic et al., 2002) in the context of climate risks need more investigation. On a general level, then, public opinion polls do highlight a sense of moderate importance, urgency and negativity associated with climate change as an environmental, but not necessarily as a ‘domestic’, issue. However, more detailed studies on national samples serve to highlight more complex public understandings of climate change. Individuals characterise climate change in multiple terms, related to their everyday experiences and locality, distinguishing effects on different scales in space and time. It is these that we turn to next, drawing primarily upon European surveys and qualitative studies of public views. A remote risk Overall, many individuals perceive the risks of climate change as removed in space and time (Bord et al., 1998; Bickerstaff et al., 2004). In a 2004 survey, 52% of British respondents stated that climate change would have little or no effect on them personally (Kirby, 2004). Among some interviewees in a German study, climate change had “a lower cognitive presence”, being overshadowed by other events, which are more directly experienced and important to everyday life (Höhle, 2002:117). Climate change becomes evident only when it becomes immediate and personally relevant. Qualitative research in the UK (Bickerstaff et al., 2004) and in Germany (Höhle, 2002) has also found that climate change tends to be associated with a belief in higher risks for developing countries, as these are generally perceived to be more vulnerable or less adaptable to the consequences of climate change, and for future generations, an inequality identified itself by the climate community (Houghton et al., 2001). Elicitations of views on climate change relying on past events as analogues for future changes can serve as a means of stimulating critical consideration of impacts. Using this method, the WISE project compared public views of current extreme weather events and future climate change during 1997 to 1999 in the UK, Italy, Germany and The Netherlands (Palutikof et al., 1999). Asking samples of the general public to recall their experiences of extreme weather events and to
It is scientifically acknowledged that climate change and ozone depletion are not completely distinct phenomena. CFCs are important greenhouse gases as well as ozone-depleting substances. Ozone is also a greenhouse gas. As CFCs destroy ozone, the resulting reduced greenhouse effect of atmospheric ozone is partly compensated by an increased greenhouse effect of CFCs (Houghton, 1997:37). Thus, fluctuating ozone levels interact with the climate system in very complex ways (Ozone Secretariat, 2000).


extrapolate these to potential adaptation to future climate change (characterised by hotter and drier summers, milder winters), Palutikof et al. (1999) found some commonalties in opinions of climate change between Northern European countries (specifically, The Netherlands and the UK) and between more 'Southern' states (Italy and Germany). This suggested an inverse relation between existing temperatures in these countries and perceptions of future changes in the climate. Both the Dutch and the British respondents preferred warmer and drier summer weather, especially in relation to personal comfort, outdoor leisure activities and health. The Dutch perceived a warmer drier summer most favourably than other nationalities. For most of the respondents milder winters had positive connotations, except for the Germans in terms of air quality and personal moods. The study highlighted the influence of geographical and cultural elements on perceptions of climate change and how, although individuals expressed concern about future undesirable consequences of climate change impinging on the national good, this did not exclude them from considering the potential personal benefits deriving from future changes in the climate (Palutikof et al., 2004, in the UK, Galeotti et al., 2004, in Italy). Within the UK, the WISE study highlighted how English respondents perceived more negatively the effects of unusually warm summers on agriculture and air quality than did Scottish respondents (Palutikof et al., 2004), which the researchers interpreted partly as an indication of the influence of regional diversity. On a more localised level, Bickerstaff and colleagues (2004) have noted that even individuals in localities that could be considered proximal and potentially vulnerable to the effects of climate change had difficulties relating the impacts of climate change (which many were aware of) to their local area or day-to-day life. Where people did draw connections this tended to reflect issues where there was some immediate demonstration of impact (e.g. flooding, local coastal erosion). In particular, people were concerned about not being able to get house insurance because they are too low lying and thus potentially susceptible to flooding or being near the cliff edge in an area prone to erosion (such as Cromer, Norfolk, UK). This evidence suggests that, presently, climate change is salient, in perceived or experienced terms, for only a minority of individuals. Similar findings emerge from a survey of a representative sample the British population, where respondents were ambivalent about the potential for damage and catastrophe arising from climate change (i.e. the ‘dread’ factor in the classic psychometric approach developed by Slovic and colleagues, see Slovic, 2000) and about the unfair distribution of risks on particular groups in British society (Poortinga and Pidgeon, 2003a; see Table 2.3). However, respondents strongly characterised climate change as a moral issue with risks for future generations. This suggests that climate change may be made more tangible and relevant to people’s current interests by considering its impact on their children and grandchildren. Table 2.3. Climate change evaluated on various psychometric characteristics by British respondents. Psychometric characteristics Climate change Unknown consequences 4.13 (0.88) Risks to future generations 4.31 (0.73) Dread 3.06 (1.12) Well informed 2.80 (1.14) Control any risks to myself 2.48 (1.07) Unfair distribution of risks 3.00 (1.00) Moral concerns 3.44 (1.03) Note: N= 312. The scale ranges from 1 (“totally disagree”) to 5 (“totally agree”); standard deviations are in brackets. (Source: Poortinga and Pidgeon, 2003a). Some studies have shown that the perceived benefits associated with current lifestyles are sometimes felt to outweigh, on a personal level, the possible risks of climate change in the future. When the overall effects on society are considered, these benefits are still seen as considerable, but


the perceived degree of harm / risk becomes more significant (e.g. benefits from private car use in Zwick and Renn, 2002; Poortinga and Pidgeon, 2003a; see Tables 2.4 and 2.5). Table 2.4. Personal and societal benefit from private transportation (as a cause of climate change), expressed as percentage of respondents and assessment of benefit-risk balance. Level (Very) High Moderate Little / no Overall Overall benefit (%) benefit (%) benefit (%) benefit harm Individual 42 43 15 3.8 -2.7 Societal 54 40 6 4.4 -4.3 Note: N=1508; assessment of benefit-risk balance on a scale from –6 (‘harm’) to +6 (‘benefit’) (source: survey in the German region of Baden-Württemberg; Zwick and Renn, 2002).

Table 2.5. Benefits to oneself and to society from activities affecting the climate (expressed as percentage of responses), and perceived risks and benefits to oneself, to British society and combined, of climate change. Level (Very) High Moderate Little / no Overall Overall benefit (%) benefit (%) benefit (%) benefit risks Individual 33.2 33.8 33.1 3.37 (1.66) 4.83 (1.48) Societal 35.1 41.7 23.2 3.49 (1.67) 5.05 (1.37) Combined 3.43 (1.63) 4.94 (1.38) Note: risks and benefits on a scale from 1 (‘not at all’) to 7 (‘very high’), 4 being the mid-point. Standard deviations are in brackets (source: GB survey; Poortinga and Pidgeon, 2003a). Important differences in risk perceptions on a personal compared to a societal level have been observed in relation to a range of hazards (see Sjöberg, 2000). An individual’s downplaying of a certain risk to him/herself, while recognising its relevance to society overall, could be interpreted as a manifestation of a personal denial about direct effect and, more importantly, dissociation from any personal involvement in any solution. At the same time, individuals may accept that on a societal or governmental scale such a threat poses a problem and therefore that a solution should be found at that level, without impinging on, or requiring, direct personal involvement. This attitude can lead to a diffusion of responsibility, whereby others are seen as being primarily responsible for addressing the issue (Bickerstaff and Walker, 2002; P. Cox, pers. comm., Mar 2004). Zwick (2002) suggests that this discrepancy could be partly explained by the different ways in which individuals relate to the perceived risks of climate change and how these might be remedied. Thus, although society as a whole may be considered to be in danger from climate change, individuals might expect that the damage to themselves would be amortised by the compensation provided by the state’s social institutions. Furthermore, an individual’s sense of control may also exert an influence on perceptions; some people may perceive themselves to be personally able to elude risks more easily than others. Such evaluations clearly relate to subjective judgements of benefits vs. risks for individuals vs. society in general and to developed countries vs. developing nations, raising equity and moral considerations. The correlation between combined measures of risks and benefits in the British study is positive (r = 0.6) (Poortinga and Pidgeon, 2003a). Some studies have reported a negative relationship between perceived risks and perceived benefits (Fischhoff et al., 1978; McDaniels et al., 1997), which could reflect a more general (affective) evaluation of a hazard (see Finucane et al., 2000; Slovic et al., 2002). The positive relationship in the 2002 British survey suggests that individuals who associate climate change with benefits also perceive it to be a ‘risky’ issue and therefore have a more ambivalent attitude towards it (Poortinga and Pidgeon, 2003a).


Trust and responsibility Trust in institutional performance is another major influence on people’s responses to risk (see Cvetkovich and Löfstedt, 1999; Rohrmann and Renn, 2000). It reflects people’s confidence in agencies and institutions that initiate and control risk, although there are various definitions of trust (for other components of trust, see Poortinga and Pidgeon, 2003b below). Regarding the communication of environmental issues and risks, the public tend to mistrust governments, businesses, industry and sometimes experts (e.g. Marris et al., 1998; Poortinga and Pidgeon, 2003a) (Figure 2.1), although simultaneously governments are concurrently conferred a high degree of responsibility for solving these problems. This reaction is likely to stem from a combination of perceived low individual efficacy (e.g. problem of free-riders), a desire for institutional accountability (e.g. Hinchliffe, 1996) and of the degree of control of the risk (i.e. subjective ability to influence the risk). For instance, Zwick and Renn (2002) found that the more a risk is perceived to be voluntary (rather than imposed by external actors) and the higher an individual’s perceived influence on the risk, the easier the subjective control of the risk.

Figure 2.1. Percentage of British respondents who would trust a lot / a little the above organisations / people to tell the truth about climate change (% of valid responses; Poortinga and Pidgeon, 2003a).
Consumer organisations Environmental organisations University scientists Government scientists Industry scientists Environment. group scientists People from your community Friends and family Doctors EU Government Local authorities Oil companies Car companies






There is evidence in the literature that enforced risks create resentment, whereas those taken voluntarily are usually more readily accepted. The Baden-Württemberg study found that climate change was considered as caused by individuals through their own volition and therefore a risk taken voluntarily (18% of respondents) while 49% of respondents indicated they considered climate change a risk taken partly voluntarily. Only 33% felt it was imposed by external forces, therefore an enforced risk (33%) (Zwick and Renn, 2002). These findings reflect similar opinions of respondents in their perceived ability to influence the risk of climate change: 28% asserted they were not able to influence this risk, 42% thought they might be partly able to exert some influence, while a considerable proportion (30%) did consider they had the ability to influence climate change. By contrast, in the British survey, respondents tended to disagree that they could control risks of climate change to themselves (Poortinga and Pidgeon, 2003a). Similarly, barely half of respondents

(54%) in a 2004 survey believed that changes to their own personal behaviour would reduce the impact of climate change (Kirby, 2004), although 85% maintained they would be willing to alter their lifestyles for this purpose. Of those who were prepared to change their way of living, most (92% to 96%) favoured options that could be easily undertaken in the home (and that would cost almost nothing, like recycling and using less energy). Fewer would favour reducing their private transport (68% would use the car less; 62% would take fewer flights), whilst price increases were not popular (only 37% said they would pay more for petrol and 51% would pay more for flying). There is also a widespread perception that the UK's efforts on climate change will not be effective unless coupled with international action (Norton and Leaman, 2004). Of the 1007 UK citizens interviewed in 2004, 60% felt that climate change would be best addressed at a global level, 13% suggested national government, while only 9% felt climate change could be best tackled at an individual household level. Only 5% favoured the European level, which is surprising given the united front the EU tries to portray at international negotiations on climate change and the initiatives being undertaken throughout the continent (such as the EU-wide emissions trading scheme coming into force in 2005; Kirby, 2004). When Zwick and Renn (2002) asked their survey respondents to consider which institutions should be responsible for controlling risks and which would have the highest public confidence to do so, in the German region, industry and politicians were designated by about 50% of respondents to be responsible; 42% designated responsibility onto scientists. Interestingly, 27.8% of respondents maintained individuals were responsible, while 23.7% ascribed responsibility to environmental agencies, and only 3.3% to the media. British respondents in 2002 (Poortinga and Pidgeon, 2003a) did not feel that current rules and regulations in the UK were sufficient to control climate change. More involvement from organisations separate from government and industry was called for. Overall, respondents felt the public, among several other organisations and groups (see Figure 2.2) should be involved in climate change decision-making although they tended to be indifferent about their personal involvement. However, Poortinga and Pidgeon (2003b:971) posit that there may be a healthy type of distrust which they call ‘critical trust’, consisting of some form of reliance on an individual, organisation or institution tempered by some (healthy) scepticism, which allows a practical, yet limited, initiation and implementation of authority-driven regulation of an issue. This suggests that some form of government regulation or intervention with the aim of changing behaviour may be effective in the case of climate change action in the UK. An interesting contrast to these studies – that aptly demonstrates the influence of local culture - is provided by an analysis of the perceptions of climate change in Finland (Rusanen, 2004). A survey similar in format to that administered in Britain during 2002 (Poortinga and Pidgeon, 2003a) found that Finns generally “feel good” about climate change, associating it with benefits for the nation, which clearly outweigh any risks. Rusanen argues that widespread trust in the government to regulate the issue is a major influence on the acceptability of the issue amongst the Finnish population.


Figure 2.2. Percentage of British respondents who ‘strongly’ agree and ‘tend to’ agree that the following should be involved in decision-making about climate change (% of valid responses; Poortinga and Pidgeon, 2003a).
Consumer organisations Environmental organisations University scientists Government scientists Industry scientists Environment. group scientists People from your community The general public Doctors EU Government Local authorities Oil companies Car companies






2.3 Addressing climate change It is well documented that individuals constantly mediate between inner demands - shaped by beliefs and previous experience - and external signals related to social and cultural norms which invariably mould social expectations, lifestyles and behaviour (see Stoll-Kleemann et al., 2001; Langford and McDonald, 1997). To this effect, we consider the psychological understanding of how individuals make sense of the world around them, how this relates to societal interpretation of a hazard event and the implications for behavioural change implied by article 2 of the UNFCCC. 2.3.1 Internal vs. external demands and behavioural challenges Some processes can contribute to reconciling an individual’s inner demands with external signals, thus establishing a sense of harmony or consistency within the individual. A lack of agreement or consistency gives rise to cognitive dissonance (Festinger, 1957). If an individual acts inconsistently with some previous attitude, then the thought of acting in that way will be inconsistent with the thought of holding that previous attitude. People experiencing dissonance deal with it either by resolving it (e.g. change thoughts by bringing them in line with the behaviour, see Eiser, 1994), denying it or displacing it; in short: “people process new information in ways that are consistent with their existing beliefs about the world” (Greenwald, 1980, in Rachlinski, 2000). This resistance to changing belief structures can lead to counterintuitive consequences. For instance, biased assimilation is the phenomenon whereby an individual looks for and accepts evidence supporting his/her belief about the world while rejecting inconsistent evidence (also known as ‘attitudinal certainties’; Eiser, 1994). “Mixed evidence on a subject [such as uncertainty on certain aspects of climate change] about which people have strong feelings will not only fail to moderate people’s beliefs, but tend to make them more extreme” (Rachlinski, 2000).


Within this context, it is not surprising that individuals set up ‘barriers of denial’ to make sense of any dissonance that may arise, for instance, between the necessity of acting to mitigate climate change and their personal preference for a particular behaviour (e.g. lifestyle, consumption patterns etc). In the context of climate change, various mechanisms might be used by individuals to condone their limited involvement in change: • Not being willing /prepared to give up present lifestyles and comforts (‘comfort’ interpretation; Stoll-Kleemann et al., 2001:115) especially if this changes the status quo for the worse (‘choosing among losses’; Rachlinski, 2000). In other words, people will disregard moral concerns unless they are related to personal advantages; • Fear that costs of change will be borne individually and these will outweigh the benefits to society (‘tragedy of the commons’ interpretation; Stoll-Kleemann et al., 2001:115). For instance, the view that a global problem cannot be addressed by a few nations alone (a comment often made in reference to the USA’s withdrawal from the Kyoto Protocol); a perceived insignificance of individual actions; question about the perceived responsibility for the problem (e.g. climate change as an imposed risk or a side-effect of chosen actions - Lorenzoni, 2003; Bickerstaff et al., 2004); • Dismissive attitudes towards climate change as a big problem (‘tendency towards consistency in beliefs’; Rachlinski, 2000), which can also be related to distrust in science (Lorenzoni, 2003); • Faith in technological and regulatory advances to solve the problem (‘managerial fix’ interpretation; Stoll-Kleemann et al., 2001:115); • Lack of inclination to change present behaviour in the light of the government not being able to deliver its climate change mitigation commitments (‘governance distrust’ interpretation; StollKleemann et al., 2001:115). It has also been observed that people in an information-saturated society have a capacity to deny awareness (Cohen, 2000). In the case of climate change an increased scientific knowledge and understanding of the subject can be counteracted by a refusal to recognise the implications of that knowledge (Marshall, 2001) especially when it is based on scientific uncertainty. This proclaimed ‘ignorance’ can thus be used to deflect responsibility from the public onto more expert institutions or scientists (e.g. Michael, 1996). Another particular form of denial is to diffuse responsibility for the problem – in some cases, the greater the number of actors involved, the lower the probability that every single one will feel capable of taking unilateral action to deal with the issue (StollKleemann et al., 2001): “lack of visible public response is part of the self-justifying loop that creates the bystander effect” (Marshall, 2001:41). Scientific interpretations of danger associated with climate change as defined in UNFCCC article 2 focus profusely on the identification of thresholds beyond which undesirable consequences would occur. Very preliminary analysis of images associated by laypeople with climate change in the UK elicited through face-to-face interviews (Lorenzoni et al., submitted) would appear to suggest that laypeople tend to associate climate change with gradual temperature changes leading to highmagnitude consequences. This observation begs the question of whether climate scientists’ definition of ‘dangerous’ climate change would reflect more closely the current understanding of how climate change is conceptualised and perceived among laypeople. 2.3.2 Social attenuation of risk The above section has illustrated that responses to risk are shaped by the societal context in which it occurs. Various case studies have shown that events related to hazards (including information on hazards) dynamically interact with psychological, social, cultural and institutional factors resulting in amplification or attenuation of individual and social perceptions of risks. According to the Social Amplification of Risk framework (Kasperson et al., 1988; Pidgeon et al., 2003), behavioural responses to these first order events can generate ‘ripples’ or secondary reactions which have wide ranging repercussions on areas far beyond direct harm to humans and the environment, such as

liability, insurance costs and confidence in institutions. The process can help explain why some events trigger major public concern when scientists or politicians do not judge the risk as high, or public apathy rules with some issues considered of high importance by other sectors of society (Gerrard, 2000). Such outcomes can trigger even more widespread responses such as “demands for additional institutional responses and protective actions, or conversely (in the case of risk attenuation) place impediments in the path of needed protective actions” (Renn and Rohrmann, 2000:38). In this respect, climate change has been, for several decades, an attenuated or ‘hidden hazard’ (Kasperson and Kasperson, 1991:10), whose characteristics and the nature of societies in which it occurs have conspired to make it go unattended. In particular, strong societal incentives serve to conceal or downplay the nature of the risk to people. Equally, some actors have a direct interest in maintaining the management of climate change removed from society’s direct involvement, by relying on the uncertainty surrounding the issue, and through this a reduced visibility to the population at large. The Kaspersons have also emphasised the potential challenges to current economic systems, which makes the hazard unattended because it denigrates the consequences and favours associated benefits; in the Kaspersons’ words, climate change is a value-threatening and an ideological hazard. Furthermore, current political structures are often inadequate to deal with such a complex, yet pervasive and inequitable issue that is likely to have a greater negative impact on those individuals and populations which are already in a situation of marginal existence (i.e. climate change is also a globally elusive and marginal hazard; see Kasperson et al., 1999). The Kaspersons suggest that the visibility and priority of such hidden hazards may increase if environmental changes are vividly associated with particular human actions and activities, if public concern on such issues can be built internationally and sustained, and if equity problems brought to the fore by such hazards can be resolved. In as much as it is considered to provide a more holistic approach to risk perception, the Social Amplification Framework unifies the two traditions in risk perception studies (individual perception of risk characteristics and the social/cultural reactions to risk) with experiences of confidence or distrust in agencies and institutions (Rohrmann and Renn, 2000). In addition, ‘ripples’ of impacts are generated as individuals or groups select specific aspects of the event or information in question and interpret them on the basis of their perceptions and mental schemes. Interpretations are then collated into new messages which can be collected, acted upon or / and communicated to other individuals and groups, which in turn will also perform similar functions of ‘amplification stations’. Each station, depending on its role / position within society and beliefs, will inject a new, different and subjective component in the interpretation process of amplification / attenuation. This can lead us in turn to consider how individuals’ behavioural responses may reveal whether climate change is perceived as dangerous or not. Definitions of danger are culturally constructed, so dangerous climate change is not a blanket term that can be extrapolated to apply to societies world-wide. Obliquely, therefore, responses to elements or impacts associated with climate change could be an indication that certain manifestations of climate change are considered a ‘danger’ and therefore need to be addressed (R. Kasperson, pers. comm., Nov 2003). If certain changes in the climate are perceived as dangerous, they can trigger anticipatory change enhancing adaptive capacity. However, behavioural changes designed to address perceived dangers may result in negative outcomes if the measures enacted do not adequately deal with the problem (H. Osbahr, pers. comm., Aug 2003). Conversely, the evidence of individuals’ reticence to voluntarily adopt behavioural changes aimed at mitigating climate change that may cause inconveniences in daily lives (Bord et al., 1998; O’Connor et al., 1999) may suggest that laypeople conceive the danger lying in consequences of implementing measures to address climate change (associated strongly with economic, lifestyle or

political costs) rather than in the dangers of not doing adopting mitigation or adaptation measures (K. Bickerstaff, pers. comm., Nov 2003). So far, we have considered some of the characteristics of climate change and its interpretation among the scientific communities, laypeople and partly governments. The various perspectives mentioned above clearly bear an influence on what is considered acceptable both in terms of behavioural changes and effects on livelihoods. The following section explores whether there may be conceptually some means of reconciling these differences, or indeed if this were necessary, to facilitate policymaking on climate change. 2.4 Integrating internal and external perspectives: the role of the media There is a current debate over whether greater information provision interacts with the perception of a certain risk, and how this might contribute to modifying behaviours. The media have had an interesting, and arguably influential, role in communicating about climate change as important modern sources of information for individuals generally, either directly or indirectly (Stamm et al., 2000): social networks and social mechanisms are equally relevant for information exchange. For instance, in the British 2002 survey, Poortinga and Pidgeon (2003a:31) were interested in examining whether people’s views were influenced by social influences represented by friends, family and work colleagues. Most respondents felt that these were moderately concerned about climate change, which is likely to influence their views. The media’s modus operandi and interest in immediate and imminent issues (May and Pitts, 2000; Harrabin, 2000) and focus on striking and sensationalist characteristics (Smith, 2000) limit the extent to which complexity can be reported (Ungar, 2000). The media often simplify the climate change problem or proclaim to present balanced accounts of scientific views, often reporting contrasting views, rather than the most consensual view (see below). Because of this, the media have been sometimes accused of failing to promote the importance of the most scientifically accepted position, allowing space for the ‘sceptics’ to put their point across in a seemingly equally valid way, through this attenuating the importance of the problem. The media also sometimes associate climate change with other environmental issues, such as ozone (Bell, 1994, in New Zealand) which can be mistaken for suggesting causal links (as observed by Hargreaves et al., 2003, in the UK). It has also been suggested that media report climate change in different ways according to their geographical coverage and influence. Dispensa and Brulle (2003:98) argue that “society members need to realize how they are controlled and manipulated by the media” having found that the media in the USA is subservient, and even tightly controlled, by a powerful fossil fuel industry whose main interest is to confuse and persuade society that climate change is not a serious problem. Hence reporting that portrays controversies among scientists and inconclusivity of the science, especially in comparison to other media reporting in Finland and New Zealand. For example, the UK media made the most of the opportunities for 'Phew! Wot a Scorcher' stories focusing on the heat wave of summer 2003. However, coverage that year also had a harder edge than previously, with most of the media making the link with global warming (UKCIP email update, 15 Sept. 2003). The coverage of the heat wave was criticised by some environmental groups (e.g. Rising Tide) and some scientists on the grounds that beneficial implications for the UK and the scientific uncertainty, rather than international consensus, were over-emphasised (J. Turnpenny, pers. comm., Sep 2003). The issue of how the media represents complex and uncertain issues of climate science is contentious. Shanahan and Good (2000), for instance, compared temperature records in New York and Washington DC with climate change coverage patterns in The New York Times and Washington Post. Their study showed that there are some relationships between local temperature and frequency of attention to climate issues, such that journalists are more likely to discuss climate

during unusually warm periods, i.e. hotter periods are more likely to be associated with more frequent and heavier attention to the issue of global climate change. This may also lead to a tendency to associate weather issues with climate issues, in turn impacting how proposed solutions are perceived (as noted above in the section on perceptions). Which temperatures are 'normal' is an extremely subjective judgement and perhaps culturally determined phenomenon. Temperature is a part of everyone's daily life; perceptions about daily temperature abnormalities may well play a role in what we think about climate change. Local temperature can vary widely even in the context of a global increase, and very unusual local temperatures in a particular time or place might provoke attention (albeit short-lived) by journalists. Over the long term, their data showed that the physical indicator of temperature is obviously not the only major driving factor behind coverage. Other factors need to be explained such as the timing, release, interpretation of major scientific reports and the possibility of narrative interest. More likely it is political events and scientific studies that attract the majority of coverage. Over the longer term, temperature variation may provide one narrative opportunity. "People are now so used to ‘experiencing’ environmental phenomena that members of the public and those in the media may erroneously believe that a long-term global issue like climate change is manifest in a particularly hot, or dry, or otherwise unusual weather day. Such beliefs could impact not only on how the public and the media understand and ‘create’ the issue, but also on how policy makers frame their solutions" (Shanahan and Good, 2000:288). Some therefore advocate reforming the media’s reporting style to reflect more closely the type of information that different publics may find relevant to their personal situations. Hargreaves et al. (2003) suggest that current British media reporting of climate change can foster distancing from climate change. They purport that climate change should also be made more personal and immediate to people’s lives by recurrently portraying the potentially negative effects that may befall the rest of the world as a consequence of behaviours in developed nations which would stand to partially benefit from the prospected change, rather than parochially emphasising the beneficial effects to developed nations (Meyer et al., 1998). Hargreaves and colleagues (2003) maintain that simple and practical information on the most significant human contributions to climate change and how these can be limited or altered is necessary. Based on his findings from a mail survey of 673 US citizens, Leiserowitz (2003) suggests that multiple strategies may be needed to communicate about climate change. Firstly, emphasise the health implications of climate change for the American public, as this may contribute to increase public concern. Secondly, that communicators target those who confuse climate change with the hole in the ozone layer, in an attempt to disentangle and disassociate the two issues, by providing information through the official communication channels (e.g. government, media) that individuals with these beliefs have been shown to generally trust. Thirdly, the author advocates convincing sceptics that climate change is a problem. As reported above, the basis for scepticism can be quite varied. Using media and government communication channels to reach the sceptics will prove difficult, as these tend not to trust these official communicators; in fact, some interpret information from these sources in the light of their current beliefs, thus reinforcing them. Different strategies would be required to target these sub-groups via trusted information sources, with communication based upon their respective arguments and beliefs. Such diverse forms of scepticism have also been found in other studies (Lorenzoni, 2003; Bickerstaff et al.; 2004). Preliminary findings of a study in Norwich (UK) investigating the saliency of climate change and individual’s sense of efficacy in addressing the issue through images of potential impacts of climate change and mitigation measures (S. Nicholson-Cole, pers. comm., Jan 2004) suggest that only images which make the issue directly relevant issue coupled with practical ways to make a difference are likely to induce behavioural change. Combining images that promote saliency and

agency by communicating the notion that “if you do ‘x’, then you can prevent ‘y’ from happening” appear to be much more effective than using catastrophic images of potential impacts of climate change (around the world, not in the UK) as people felt distanced from the problem. Images such as famines in Africa and forest fires had become too familiar and individuals participating to the study felt powerless to remedy the situation. Images conveying a behavioural change that individuals found personally relevant were more salient. Others reject the notion that establishing cognitive links between causes and consequences of climate change requires more information provision on the mechanics of the greenhouse effect. Dunlap (1998) suggests that it may not be necessary to ensure laypeople have assimilated climate change knowledge correctly, as they generally seem to echo the view of the majority science community advocating precautionary action rather than ‘wait and see’. From the publics’ perspective, therefore, Dunlap argues that it may be more important to know who to trust and distrust in taking action rather than the intricacies of their respective arguments. These findings beg the question of whether the media are the appropriate vehicle for diffusing climate change information. The role of the media as intermediaries between scientists and the public is debatable, given their reporting requisites. Whereas some doubt scientists are able to affect public opinion other than through providing credible information, they would prefer the media to provide a sober and in-depth analysis (J. Schellnhuber, pers. comm., Sep 2003) of events and facts. Qualitative data also show that individuals themselves differ in their views on the role of the media as communicators on climate change information (Lorenzoni, 2003; Bickerstaff et al., 2004): • Media reporting can, for instance, be criticised for not representing information ‘correctly’ according to individuals’ beliefs. Thus, for instance, a participant in a focus group in the UK justified his scepticism regarding human influence on the climate by attributing the exaggerated concern about the issue to un-contextualised media coverage, which fails to set current changes within geological timescales. • Despite being ‘hyped up’ and inaccurate, media reporting and images on climate change can be considered powerful communication tools that can raise the profile of the issue by bringing climate change to the attention of the ordinary person, although this in itself may not induce any lifestyle change. • Whether the media could or should fulfil an educational role by supporting government attempts at enabling citizens to enact change. This would be possible if the media made information accessible both in format (short reports, clips of useful information) and means (tabloids and television as well as broadsheets, necessary to make information available to the wider public). The latter way of conceptualising the role of the media could be taken as an indication of diffused, not attributed, responsibility for addressing climate change. Typically, individuals refer to the selfishness of governments, the greediness of politicians, the lack of commitment by businesses and the apathy of the public. Summary This part of the report has drawn upon the literature on risk perception and behavioural psychology to describe public attitudes towards climate change as an environmental issue and raised some questions regarding the notion of ‘dangerous’ climate change. There is widespread awareness and concern, urgency and negativity associated with climate change as an environmental, but not necessarily as a ‘domestic’, issue. Most laypeople in developed nations identify climate change with an impersonal, removed and distant risk, secondary to other more

important and immediate issues, even in localities more proximal to climate change impacts (Bickerstaff et al., 2004). Climate change will occur to others or in the future. Danger is identified both with more immediate, but also potential, loss of lifestyle (as a result of mitigation and adaptation) and with disruption to ways of life in the longer-term (from the unmitigated impacts of climate change). However, experiences in other areas around the globe (such as Small Island States) would seem to suggest that for some individuals, climate change is indeed dangerous already (JC Mace, pers. comm., Feb 2004). In developed nations, individuals tend to downplay risks from climate change to themselves, while recognising its relevance to society overall. This can be a manifestation of a personal denial about direct effects and acknowledgement that a solution should be found collectively but without personal involvement. If individuals do not consider their actions contribute to the problem, or they perceive the benefits deriving from the causes greater than the risks associated with the consequences of climate change, they will be less motivated to take corrective actions on their own. This may, in turn, lead to governments taking a weaker policy stance because they perceive low support for ‘unpopular’ policies. On the other hand, governments may also be pressured politically to push through highaiming emissions reduction targets that would need an intricate system of measures, policies and incentives to enable a behavioural transition to become effective (see also McDaniels et al., 1997). Several studies mentioned above indicate there are more powerful motivators for behavioural change, such as insurance availability. Furthermore, there are also other mediators influencing perceptions of danger, such as psychological, social, moral, institutional and cultural processes affecting the desirability and acceptability of certain consequences deriving from climatic changes. Thus, danger in relation to climate change currently discriminates in space and time. ‘Danger’ cannot therefore be universally defined. Like acceptability and tolerability, it is context, scale and time dependent.



Phase 4 – International workshop

An international workshop aiming to address some of the questions raised as a result of the research carried out in phase 1 (listed in the previous paragraphs), was hosted by the Tyndall Centre and the Centre for Environmental Risk at UEA on 28 and 29 June 2004. Over 35 academics from the social and climate sciences, in addition to policy makers and researchers, attended and participated in the event. The main themes of the workshop plenary discussions are summarised below. A special journal issue of Risk Analysis (edited by Irene Lorenzoni, Nick Pidgeon and Bob O’Connor) is currently being prepared, totalling 13 papers (11 of which were authored by participants to the workshop, listed in Appendix 2) and an editorial, due for publication in early 2006. Definitions vs. action The workshop participants also acknowledged the context specificity of ‘danger’, which the literature analysed in Phase 1 of this project has also highlighted. Participants recognised that Article 2 of the UNFCCC effectively encapsulates a rich tension between what humanity as a whole should avoid and what is perceived as ‘dangerous’ by different socio-economic groups, e.g. danger may be seen as a zone of ambiguity or as an opportunity; for some climate change is a removed risk, for others the manifestations of climate change represent immediate threats. Some participants felt that there was a need to focus upon implementing action on climate change, thus shifting attention, from debate on multiple definitions of ‘danger’, to climate policies and their outcomes, with a focus on barriers to action and effective alternative approaches. Three options were discussed: (a) piggy-backing the climate change issue onto other policy areas, so that short term actions can result in long term benefits; (b) related to this, the need to focus on the 'local' in order to mobilise people. This discussion was based on understanding that people relate most effectively to things that are important to them in their everyday lives. As many of the causes of climate change are also the causes of other problems often much closer to home, climate change needs to be situated in people's everyday experiences to make it real to them; (c) addressing climate change as part of a deeper understanding enveloped within the notions of sustainable development. One participant offered more comments regarding the main points summarised in this section, urging critical thinking about the weaknesses associated with some of these suggestions: • The 'zone of ambiguity' is susceptible to being shaped by the more vociferous positions, which contribute to articulating the global climate change discourse. If the scientific and policy communities are not able to strongly emphasise the need for action and the benefits of pre-emptive action, there is a possibility that rather than a constructive zone of ambiguity article 2 may legitimate endless discussions among different positions and perspectives; • The precise nature of piggy-backing should be considered with care, such that linking local problems with global environmental change issues that have apparently less resonance can enable effective action. For instance, associating polluting particulate emissions from car traffic with CO2 could prove counter productive; this type of pollution will gradually disappear through existing legislation and decommissioning of older vehicles, whereas the CO2 emissions will persist in newer models. Beyond the ‘normal’ The primary focus of scientific interpretations of article 2 has been to identify changes that could cause major disruptions to ecological and human systems. However, slight changes to parameters that could be taken to represent ‘the norm’ may result in social discontinuity: danger resides in changes to social / technical structures (e.g. discontinuities in the Earth’s physical systems can also be represented through interference with infrastructure, such as nuclear reactors).


Utility of UNFCCC article 2 Mixed views on the usefulness of article 2 were presented at the workshop. On the one hand, defining article 2 primarily through the discourses of science (i.e. seeking to avoid large scale discontinuities causing major disruption) will enable a pragmatic approach to be taken towards managing climate change. Others argued that this may be less useful for policy making – there may be a need to make climate change more tangible for people. However, it was also suggested that article 2 is deliberately ambiguous (or constructs a “zone of ambiguity”) in that it sidesteps the issue of who loses. Many vested interests stand to gain from this ambiguity, which can be interpreted in subtle ways by different actors. Ambiguity can also be useful for policy development, allowing consideration / incorporation of understanding and values that develop over time. Another point is the wording of article 2: it explicitly focuses attention on issues of importance to some countries e.g. Small Island States (sea level rise) and East African countries (food security), but not others of priority to other countries. In international negotiations, article 2 points towards defining danger tangentially / indirectly. The general wording of article 2 can help set criteria to define the lowest common denominator (danger common to all) and may be more useful to international negotiations than trying to agree upon a new definition. Creative ways may need to be found of introducing the topic into different discourses, e.g. via notions of ‘quality of life’. In pragmatic terms though, article 2 is present in international documentation and therefore has to be worked with. Justice in relation to UNFCCC article 2 A rights-based approach to climate change focuses upon impacts. Applying this to article 2 means that everyone’s right is to avoid dangerous anthropogenic interference with the climate and its consequences. There is no one way to agree what ‘dangerous’ means. One option is to consider which stabilisation scenarios some nations opt out of as an indication of the danger they are prepared to impose on themselves and others. The Convention is seeking a balance between dangerous climate change, strategic powers and the level of danger that is acceptable. This will be defined for some, but not necessarily for all. Others argued that using the notions of rights may impede the development of policies to enact the convention. Equity and justice issues are important, but they may only be effective locally (including to enact adaptation), and less so regarding global justice. These issues tend to be debated at the backstage of international negotiations and not necessarily in the public fora, due to the difficult questions they raise. The use of danger in climate change discourse and communication A crisis narrative may not encourage proactive responses. Indeed it may foster fatalism and withdrawal. One line of action would be to promote the positive, more tangible and personalised aspects of localised actions associated with day to day problems and dangers, e.g. health, noise pollution etc (this links with the notion of piggy-backing mentioned earlier). Community networks may be of importance here due to the relevance of group understandings and concerns about risk and danger. Individual actor behaviour should also be addressed. Another important question is the transmission of information: how can messages be conveyed when politicians, and to some extent scientists, are heavily distrusted in some arenas? Media reporting The participants debated the role of the media in relation to climate change. The erratic metrics of risk in media reporting were acknowledged, identifying a need to create different kinds of register to represent different risks and other environmental issues. The media need to recognize the messy


edges around climate science, issues of scale, and the importance of making the right comparisons i.e. climate change kills more people than terrorism. The workshop also emphasised the need, in any communication exercise about climate change, for reporting to delineate how the ‘balance’ of evidence is weighted: opposing positions should be backed up by making clear which position is supported by the greater amount of scientific evidence, in order to promote a more nuanced appreciation of the debate. Suggestions for further research These were also discussed at the international workshop; the most salient are outlined below: Balance the input of expertise in establishing criteria on “dangerous” climate change, combining the knowledge available from climatology through the development of integrated assessment models with cultural and political aspects. - Undertake perception studies in developing countries (a clear gap). - Carry out long-term longitudinal social studies of perceptions, media reporting, and responses, securing funding for at least 10 year cycles, with strong theoretical underpinnings, driven to provide practical, useful and effective policy recommendations. Also, meta-analysis is needed for existing datasets (primarily qualitative) to look at longitudinal trends up to the present day in discourses about climate and danger. - Research how varying social/technical infrastructure in particular makes communities (in both developing and developed countries) more or less vulnerable to climate change. - Explore, from a social science perspective, elites (“studying up”), their ways of thinking and the important stage-door issues. There is little understanding of underlying strategies due to the delicate nature of many discussions, negotiations and agreements, which are generally poorly documented, e.g. notes from Conference of the Parties (COP) meetings provide no indication of strategies or major/sensitive issue discussions. - Research how media representations of risk and risk perception can better represent metrics of risk and differentiate between differing types of risk i.e. AIDS vs. climate change. - Investigate how to promote the link between quality of life and individual behaviour (as there is evidence that laypeople in general, e.g. even health sufferers, often do not relate these two aspects), to charge people with the responsibility to act and encourage individual behavioural change. - Study the links between risk perception, climate change and health. - Further research issues of vulnerability, justice and adaptation which become important in the context of the quality of health systems in different regions/countries as determinants of the effects of climate change. - More work on perceptions of ‘surprising events’ that do not fit people’s existing mental models e.g. global cooling. - Research on use, impacts, meaning, affective properties and interpretation of climate change images currently used by the media. − Investigate how climate scientists and policy makers interpret the images they receive (how images influence their judgments).



Phase 3 – ‘Expert’ definitions of danger in relation to climate change

4.1 Research design: applying a ‘mental models’ approach In this phase of the study, we adopted a mental models approach to elicit experts’ understandings of the notion of danger related to climate change, using the mental models approach developed at Carnegie Mellon University (Bostrom et al. 1994a, 1994b; Morgan et al., 2002) and at the University of East Anglia (Cox et al., 2003). It was felt that mental models would be a useful tool for gathering and representing the many beliefs which exist around climate change, and in particular on the use and understanding of the notion of ‘dangerous’ in relation to climate change, by eliciting the personally held views of individuals. The mental models work in this study comprised two stages: (a) undertaking a review of literature on climate change; (b) conducting semi-structured interviews with climate change ‘experts’ based mainly in the UK, with some in Europe. Subsequent to these, the information received from all the experts was combined into a ‘meta’ model representing the various perspectives portrayed by interviewees. The various phases of this process are detailed in the sections below. 4.1.1 Conceptual framework: mental model research A mental model is the representation of an individual’s conceptualisation of elements worthy of attention in relation to a specific issue, situation or phenomenon. It thus reflects his / her existing beliefs and knowledge (Morgan et al., 1992). Mental models are not ‘models’ in the more scientific sense, as they do not involve ‘strict mapping between things in the ‘real’ world and elements in the model’ (Morgan et al., 2002 p.23). Rather they provide an indication of an individual’s perceived relationship among different factors relevant to that particular issue, and can reveal how these are valued and understood. The mental models approach has traditionally been used to decipher what (usually lay) people know about a certain issue in comparison to ‘expert’ knowledge, with a view to formulating communication aids that would be meaningful to end users. Expert knowledge is commonly organised in the form of influence diagrams, which show the relationship between factors relevant to the issue on which communication is to be improved. These represent different variables and the functional relationships between them. The arrows that connect the variables indicate that the one at the head of the arrow is dependent upon the one at the arrow’s tail. Effectively, the expert influence diagram represents their mental model based on detailed knowledge about the subject in question. For the purposes of effective communication to a lay audience who may have incorrect understanding about the technicalities of the subject, the final influence diagram is developed iteratively with experts and is used as an organising device to characterise the content of lay mental models (Morgan et al., 1992:2051). Comparisons of expert and lay mental models have been used to examine relations between illness cognitions and adherence to medication (Brewer et al., 2002) and for developing various forms of risk communication (Morgan et al., 2002; Cox et al., 2003; Niewöhner et al., 2004). 4.1.2 Expert definition and selection The long-standing debate over the appropriate place of scientific and technical experts in the policy process has been exacerbated by mostly untested assumptions about experts, particularly that experts will arrive at consensus on the basis of careful analysis of the facts alone (Barke and Jenkins-Smith, 1993). It appears that no consensus exists on what constitutes an “expert” in much of the expert knowledge elicitation literature. However, it is regarded as critical to the present research that, given the high levels of uncertainty framing the climate change debate, the notion of an “expert” to be used within this study should be clearly defined through a comparative study of the literature and as basis for interviewee selection.

Experts have been defined by distinguishing them from laypeople based on the different cognitive heuristics used by them to arrive at a decision (Slovic, 2000). However, a wide range of subjective values may exist in the framing or definition of what constitutes an “expert”, from those that have “thought deeply….” about a particular subject (Nordhaus, 1994:46, Krayer von Kraus et al. 2004) to “those that have status of authority in a subject by reason of special training or knowledge” (Bray and Von Storch, 1999:441). Wright and Ayton (1987:8) note Feigenbaum’s (1979) observation that from a medical perspective, expert knowledge is “…largely heuristic knowledge, experimental, uncertain – mostly ‘good guesses’ and ‘good practice’ in lieu of facts and figures”. Thus, we can infer from this that the expert has the ability to proffer a coherent and well judged opinion of what may be (in the future), based upon a vast wealth of experience and knowledge, or in other words, a ‘best guess prophesy’. Arnell et al. (2004) suggest a number of categories into which “experts” in their study of rapid and abrupt climate change can fall. These include scientists active in the study of climate change; those reviewing the work of other scientists (through the Intergovernmental Panel on Climate Change (IPCC), for example); policymakers concerned about the implications; and, finally, political lobbyists. Experts themselves were identified as i) those leading major research programmes; ii) those who had published and iii) those who had participated in the IPCC’s Third Assessment Report (TAR) reviews of rapid and abrupt change. However, out of a total of 38 experts approached to take part in the study investigating different methods of estimating likelihoods of abrupt climate change, 21 declined to take part for reasons ranging from a lack of confidence in subjective probabilities to a lack of trust in the expert elicitation process. A study carried out by Nordhaus (1994), who investigated estimates of impacts of climatic change, used a broader and more relaxed categorization, selecting as “experts” an international network of individuals conversant with issues of the economic impacts of global warming. Utilizing the snowball sampling method, interviewees were asked to nominate other individuals who, they felt, represented a cross-section of knowledgeable opinion, eventually totalling 22 respondents. Whilst it seems possible that the process of selection by nomination could lead to an unrepresentative sample of ‘group thinkers’, ‘knowledge networks’ or ‘epistemic communities’, Nordhaus argued that it was designed to yield both diversity and informed opinion. In qualitative research methodology, it is common to use such a sampling strategy, which is oriented to providing a (purposively defined) spectrum of views on the topic (see Pidgeon and Henwood, 2004). The mental interviews reported here were intentionally circumscribed to European perspectives in order to limit the influence of other significant cultural differences e.g. US (based on observations by Bray and Von Storch, 1999). Thus, by avoiding these differences it is possible to triangulate information received in a way that would reflect a ‘consensus’ summary (Morgan et al., 2002). However, we were also interested in the views of those who may be outside the ‘consensus’ summary in order to reflect the wider range of views amongst climate change experts. A number of specific factors came to bear in the selection of our “expert” interviewees. Firstly, the high degree of uncertainty, the complexity and the wide-ranging repercussions surrounding climate change debate can create a situation in which the term ‘expert’ may be loosely defined. In addition, the multitude of disciplines contributing to research into climate change has enabled individuals with no previous background or experience in climate change to participate in the climate change debate. For the purpose of this project, we considered ‘experts’ those individuals who, having specialised in their particular area of work or research, had extensive knowledge of climate change issues demonstrated through climate change-related project involvement, publication and wider recognition as specialists in their particular field. However, it should be noted that some interviewees explicitly stated that they regarded themselves as specialists in their area rather than

experts on climate change and maintained that they did not have a clear understanding of all of the complexities and diverse views surrounding the debate (e.g. some misunderstandings about the causes and relationships with other environmental issues). 4.1.3 Sample Size Expert elicitation literature tends to give little idea as to the optimum number of respondents to use in a study such as this. Morgan et al. (2002) suggest that much of the value of conducting openended interviews can be obtained within a relatively small sample. For our research, sample size depends upon the composition of the population of expert knowledge from which the sample is selected. Morgan and Keith (1995:469), in a quantitative study of subjective judgements made by climate experts in relation to the geophysics of climate change, maintain that “because each set of expert judgements was offered as a single considered view, it is generally not appropriate to average across the results obtained from different experts”. Rather than selecting a statistically representative group, the authors sought to include at least one representative from most of the mainstream perspectives, with serious consideration of the quality of scientific credentials. Fiscal and logistical constraints limited numbers and precluded experts outside of the USA; a total of 16 experts were interviewed and asked to carry out exercises for periods as long as 5-6 hours. For the purposes of this project, following a literature review, discussions with fellow researchers and faculty at the University of East Anglia and other researchers with experience in mental modelling, it was agreed that ‘experts’ would be selected from the following: researchers working in climate modelling would be able to provide views related to the notion of human interference, whilst experts in the areas of food, water, ecosystem management, economics and law would reflect the impacts side as worded in article 2. Correspondence with Ann Bostrom [Georgia Institute of Technology, School of Public Policy, USA] suggesting that we ‘think about the chapters of the IPCC report […] and anything they say (directly or indirectly) about the structure of expertise in climate change research’ led us to use the IPCC report to guide us towards its authors and related areas of expertise. Furthermore, the pervasive nature of climate change implies that all areas of modern society will be affected, albeit on different temporal and spatial scales. Hence we expanded the potential pool of interviewees to areas not explicitly denoted by article 2, such as health and insurance, and sought to include alternative perspectives on how climate change should be addressed in relation to other issues on national and global agendas. The project was thus developed to elicit ‘expert’ understandings through a mental models approach. In total, 23 of a proposed 25 interviews were eventually carried out (see Table 4.1). This response rate was similar to work carried out by Nordhaus (1994) in which only 3 invitees declined to be interviewed (equating to a response rate of 88%). Our own high response rate is possibly due to the invitation follow-ups mailed to initial non-respondents, the short length of time required for each interview and the subjective, opinions-based questioning (rather than probabilities-based questioning utilised in some expert elicitations).


Table 4.1. Groupings of interviewees according to their expertise. Area of expertise Number of Interviewees Agriculture Climate Modelling Coastal zones Economics Ecosystems Energy Equity / Social Justice Hazards/Risk Health Insurance Policy 3 2 1 2 3 2 2 2 2 2 2

4.1.4 Interview Procedure After selection, ‘experts’ were contacted either by email, telephone or in person. Individuals were asked to participate in the study by consenting to an interview on their perspectives about danger in relation to climate change and UNFCCC article 2. Confidentiality was assured. They were informed briefly of the background to the study (Appendix 3) and provided with an example of an influence diagram (Figure A3.1 in Appendix 3) as a very basic template of a mental model, on the basis of which they could construct their own during the interview. Thus, some thinking and learning was required prior to the interview; this occurred to varying degrees among respondents. The majority of interviews were conducted by telephone. A few were carried out face-to-face, during which respondents were met at pre-arranged locations, often in their office or a meeting room. Interviews were limited to 30 minutes, unless respondents felt they could dedicate more time to discussion. It is possible that the relatively short length of the interviews may have precluded a more frank appraisal of their views surrounding the issues raised. However, given the project timescale and the time that individuals were able to commit, a longer conversation would have been limiting. The interviews were semi-structured according to a basic protocol (see Appendix 3) developed through a pilot interview process and feedback from fellow researchers and faculty within the University of East Anglia during the spring of 2004. The full interview was pre-tested four times in total. Minor modifications were carried out to the protocol following these pilots. Interviews began in August 2004. At the beginning of the interviews a more in depth introduction to the aims of the project was provided, before asking each respondent to give a brief description of their background and the work in which they were previously and currently involved. This led to a question on the respondents’ personal interpretation of the term ‘climate change’. Following this, respondents were asked to draw their own mental model of the issues they felt related to the notion of danger and its utility. At this stage, the interviewer intervened as little as possible, although he was asked a few times to clarify the concept of an influence diagram by referring to an unrelated subject (i.e. tripping on the stairs, Figure A3.1). After this the ‘expert’ was encouraged to draw his/her own diagram relating to danger and climate change. As most interviews were undertaken by telephone, interviewees were encouraged to talk through their mental model while formulating it, so that the interviewer could follow their cognitive process and associations, and the conceptualisations portrayed in their own influence diagram. Despite an initial reluctance from some respondents, this process was largely successful (only one of the 24 interviewees refusing to construct a mental model, preferring instead to list issues of

relevance in note form). Interviewees then posted their drawings to the researchers after the interview. Following the protocol, interviewees were encouraged to discuss the usefulness of the term ‘danger’ in relation to climate change and interpretations of article 2, its limitations and future prospects. Finally, respondents were asked to refer to a mental model based upon the literature on danger and climate change (Figure A3.2), which had been emailed to them upon acceptance to be interviewed. They were asked to review the diagram, making changes and adding influence arrows where they felt it was necessary. Unlike the approach utilised by Cox et al. (2003), in which each expert’s comments were combined with an analysis of the interviews in an ‘iterative development of the influence diagrams’ after each interview, our aim was to present the interviewees with an unmarked diagram containing a range of salient climate change issues. The notion underlying this approach was to compare individuals’ responses on a common template to facilitate the analysis of divergences and similarities of expert views. Our mental models approach was based upon and guided by earlier works of a similar nature, developed to suit the opportunities and constraints of our particular project. Cox et al. (2003) maintain that the use of a generic template in the construction of user diagrams (suggested by a series of headings) allows for a systematic interrogation and comparison without suppressing the diversity of views. In particular the exercise is disciplined by the normative objective of identifying communication gaps. However, following a pilot study and a series of different designs, we found the generic mental model to be too restrictive and prescriptive, instead encouraging interviewees to depict their mental model on a blank sheet of paper early on in the interview. By utilising this approach, each respondent’s starting point on their mental model diagram offered an insight into their perceptions of the most salient issues for them. As a compromise, at the end of each interview, those who were still engaged were presented with the pre-tested version of the generic literature-based diagram as explained above. Similarly to Morgan and Keith’s (1995) elicitation in which experts chose not to make significant use of the influence diagrams provided, in our study less than half of the respondents (11) felt able to complete this final stage of the interview, as some were limited by time constraints whilst others felt all relevant issues had been illustrated and discussed as part of the initial mental model diagram and subsequent interview. We also had the impression that, in a few interviews, the literature-based diagram referred to at the end of the conversation may have biased some of their comments. This diagram was available to them before the interview as it was sent to respondents with confirmation. Indeed, one mental model diagram was constructed in a similar format and using similar points to that of the literature based mental model suggesting either that the interviewer was not explicit enough in providing instructions or that the interviewee preferred to follow a set format. The influence arrows and comments inserted by the interviewees into the literature based model were of some additional guidance in deciphering their opinions; however they did not introduce new information into our analysis and consequently did not play a role in the interpretation of our findings. At the conclusion of the interview, respondents were given the opportunity to mention any issues not already covered and to suggest other possible experts that may be interested in taking part in this study. These questions tended to prompt a fairly lengthy response as the interviewees at this stage were often eager to elaborate on points of interest to them. Some found that the process of systematic thought encouraged during the interview grounded their understanding of the subject. Some even added further thoughts to their mental model influence diagram. Often, as found in

several other studies (P. Cox, pers. comm., 2004) these conversations yielded the most important perspectives. Indeed, in some cases it was demonstrable that the conversation had provoked respondents to reflect deeply upon their own position, conceptualisations and understandings of the climate change debates. As Morgan and colleagues (2001:23) state, “this modelling procedure is not a reactive one; it forces the interviewees to think more systematically about their beliefs than they might have otherwise”. 4.2 Data analysis Interviews were transcribed in full7. The transcripts were coded thematically using a coding schema developed iteratively by the research team. Coding categories were identified and refined through an interaction between concepts and themes emerging from the interviews, from mental model diagram data and from theoretical perspectives derived from the literature relating climate change to notions of danger. Respondents’ representations of their associations and interpretations were compared visually and related to their interviews. These were then synthesised to produce a metamodel representing the full range of opinions and perceptions, supplemented by the analysis of the interview recordings and understandings gained from the literature. The objective was to encapsulate the various perspectives into a comprehensive description, summarizing the pooled knowledge of a community of experts rather than the views of any one expert. 4.3 Findings and Discussion The following discussion presents issues and concepts outlined by the interviewees, with reference also to definitions of ‘dangerous’ in the relevant literature. Firstly, some observations are provided on the interviewees’ construction of their mental models and an overall description of the metamental model (see Appendix 3) derived from these. Subsequently, responses to the interview protocol are presented thematically, supported where possible by direct quotes from respondents (section 4.3.2). These are identified only numerically, to indicate their distinct source and to protect interviewees’ anonymity. 4.3.1 Building a meta-mental model Following the interviews, a meta-mental model was produced to encapsulate the various perspectives illustrated by each respondent in a manner that also captured their train of thought in the diagram construction and thus their feelings about the nature of the ‘dangerous’ discourse. Following analysis of each expert diagram by the research team, a diagram was produced which included interviewees’ perspectives. A process of distillation then followed, during which the points were discussed, grouped (by their relation to each other and position in the developing diagram) and summarized using key descriptive words which could be explained by accompanying text. The result is a clear diagram which manages to capture the feel and content articulated by all of the respondents (see Figure 4.1 – full explanation given in Appendix 4).

However, due to the poor quality of some of the recordings, three could not be transcribed in full. These were supplemented by notes taken during the interviews by the researcher to formulate a comprehensive record of the discussions.



Figure 4.1 – Meta mental model
- Influence arrow R.O.S - Role of science U - Uncertainty Danger in the Cost of Mitigation


Personal & Societal Values: Cognitive/Affective Perceptions Acceptability


Causes + Drivers •Present •Future

Energy Insecurity & Politics




∆GHG Concentrations


Climate System U

Climate Change

Primary & Secondary Impacts On Human & Natural Systems
+/- feedbacks; Degrees & Timing of Danger

‘Dangerous Climate Change’

+ve feedback

Context ‘Dangerous Anthropogenic Interference’ (2)

Ability to Adapt



For ease of interpretation, the meta-mental model (shown in Figure 4.1 on the previous page) is subdivided into three main sections: 1. Influences on the climate and impacts as a result of this change. Oval 1 encapsulates how human as well as natural causes and drivers are increasingly affecting the climate, altering atmospheric greenhouse gas concentrations (identified by some as “dangerous human interference with the climate system”, as per UNFCCC article 2). The extent to which this may cause positive feedbacks in the climate system is generally governed by a large degree of uncertainty. Notwithstanding, most respondents maintained that significant climate change impacts were likely as a result of anthropogenic interference with the climate system and would be exacerbated in the future unless major political and / or economic surprises (such as changes in energy availability) were to result in mitigation. The respondents also outlined that conversely, another conceptualisation of danger is in the potential threat to current ‘western’ lifestyles, in the form of mitigation and adaptation measures. 2. Impacts upon human and natural systems, referred to by some respondents as “dangerous climate change”. The extent to which changes in the climate will indeed be ‘dangerous’ is dependent upon the context within which these changes occur (see Oval 2). Modifications to the natural system as well as social, political and economic forces affect the ability to adapt. Thus, pre-emptive (1) or reactive (2) adaptations may temper the degree to which the effects of climate change are manifest as dangerous. According to the majority of respondents, in fact, ‘dangerous climate change’ exists only when impacts are beyond a system’s ability to adapt; a situation, it was suggested, which is likely to exacerbate conditions of existing vulnerability and inequality. 3. Finally, Oval 3 represents respondents’ view that climate change is representative of, and is affected by deeper individual, political and social drivers. The issues contained within this oval relate to the question asked by most respondents of “Danger to whom?” in framing the issue of ‘dangerous climate change’. Central to this question is the role played by human values and what is felt to be socially acceptable. Also prevalent among respondents’ comments was the question of how these values are formed, by whom and to what end. The role of science was questioned within the decision making process, as many interviewees were unsure about the extent to which research agendas and outcomes are manipulated in the service of political goals. Some respondents referred to the current energy debate to exemplify this concern, suggesting governmental concern over the unacceptably dangerous impacts of climate change functioned as a subterfuge for the desire to see the return of nuclear power as a socially acceptable option for future energy production. Some interviewees also mentioned that mitigation options could be perceived in some political spheres as leading to (unacceptable) changes to the status quo, resulting in expressed preference for adaptation. 4.3.2 Perspectives on Danger and UNFCCC Article 2 What is climate change? The second question interviewees were asked regarded their conceptualisations of climate change, in terms of causes, impacts and outcomes. All respondents maintained that natural variations in the climate system are likely to be enhanced by human-made greenhouse gasses to produce a change away from the natural or ‘regular’ cycle. It was felt that interference with natural variations, along with persistent perturbation and failure to return to an original state, defined climate change. There was a general consensus surrounding the possibility of negative impacts as a result of human-


induced climate change. However, views varied regarding the timing and extent to which climate change impacts may be experienced. It was noted that the effects of increased emissions were not confined to interference with climate systems, with other human-made pollutants also posing problems at more local levels. The message from one respondent was that if it is a problem that humans have caused, it is a problem that humans can solve. The majority of respondents maintained that climate change is symptomatic of the deeper malaise of contemporary society characterised by human and economic inequities, power struggles and issues of sustainability. To this end, climate change and its effects were socially constructed. This stance refers both to ‘at risk’ populations, relying more heavily upon marginal and thus more vulnerable resources, and what Pielke Jr. and Landsea (1998) suggest to be the result of richer people moving into harm’s way. Climate change was also defined in terms of potential impacts. Positive ones were associated with increases in economic development especially in some sectors, such as agriculture and health, through the displacement of some effects experienced currently (see quote 17 below). However, more than half of the respondents mentioned in their interviews that negative effects of climate change on human health and well-being would be unacceptable. Consequently, some envisaged that increased investment into healthcare as an adaptive response to changed conditions (see quote 3) could result in improved public health provision: (17) “If you take the UK context or northern Europe even …the probability is that our gain from reducing cold deaths, if we got warmer on average, would exceed the increase in heat deaths, and they’re the two biggest directly weather-related impacts at present”. (3) “These [climate changes] might also have positive impacts. There might be an increase in economic development, which might mean a better investment into public healthcare systems. This, on the other hand could mean a better tertiary preventative approach in terms of care and treatment.” However, on further reflection, these assertions fail to take into account the wider repercussions (environmental, political, legal) which may be associated with changes away from the ‘norm’. Indeed, as mentioned by one other respondent, the possible threat from environmental refugees fleeing resource conflicts and the destabilisation of socio-economic and cultural balances means that net improvement in healthcare provision is unlikely to be experienced for humanity overall. In addition, increased health spending may prove detrimental to spending in other social sectors: (1) “Resource scarcity is likely to cause conflict, particularly water availability and quality. On an economic basis, impacts in less-developed countries will be the key factor. [Climate change] will be very dangerous for them, and spilling over to create danger for Annex 1 nations.” The complex interactions among sectors and constituents of socio-economic systems obfuscates the assessments of the likely consequences of these impacts, thus increasing the difficulty of gauging the outcomes overall. Defining danger To assess the extent to which perceptions of ‘dangerous’ were influenced by interviewee’s knowledge and expertise, respondents were questioned on the use of the word ‘dangerous’ in article 2. From a theoretical and methodological perspective, it is interesting to note that responses varied depending on respondents’ prior knowledge and/or familiarity with the article and its wording.

Generally those more familiar with the article were more likely to analyse the underlying concepts and issues relating to its use, whereas respondents new to the article were more likely to identify issues with the actual wording and phrasing of the text, not discussing its wider ramifications. An important point, made by two of the 23 respondents, referred to the often interchangeable use of the terms “dangerous anthropogenic interference with the climate system” (as per article 2) and “dangerous climate change” 8. The two respondents, representing different specialist areas and both familiar with the wording of article 2 prior to the interview, felt that the expressions are not equivalent. One in particular argued that they refer to two different points at which the danger occurs: the former relates to the processes by which humans influence the climate system; the latter relates to effect of the impacts resulting from significant changes to the climate. As the respondents clarified: (3) “Dangerous anthropogenic interference is different from dangerous climate change. What I mean here is whether you establish 355 ppm [parts per million] or 550 ppm, or whatever you establish as a dangerous interference of human-beings, or whether you say a 2 degree centigrade increase of temperature on average represents a challenge or risk or danger to ecosystems, health, infrastructure and so on. I think these are two completely different issues.” (6) “The definition of ‘Dangerous anthropogenic interference with the climate system …’ is fine… ‘Dangerous Climate Change’ is not a synonym.” However, an appraisal of the literature suggests that the official terminology as per the UNFCCC Article 2 (i.e. “dangerous anthropogenic interference…”) is often replaced by or used interchangeably with “dangerous climate change” (e.g. Schneider, 2003; Mastrandrea and Schneider, 2004; Parry et al., 1996; Dessai et al., 2004). Respondents’ views on these two diverse ways of relating danger to changes in the climate are explored more in depth in the two subsections below. “Dangerous climate change” The vast majority of interviewees considered the physical impacts associated with climate change to be the most conspicuous elements of their conceptualisations of danger, rather than emissions of greenhouse gasses. Some in particular referred to the notion that climate change could trigger ‘surprises’, i.e. rapid, ‘big switch’ events and non-linear responses of the climate system to anthropogenic forcing, thought to occur when thresholds are crossed (see also Schneider, 2003)9. Respondents acknowledged the uncertain nature of the projections of these climate change impacts. The possible ‘surprises’ suggested by respondents included the collapse of the North Atlantic Thermohaline Circulation (THC) (see also Rahmstorf, 2000), de-glaciation of the polar ice-sheets, collapse of the West Antarctic Ice Sheet (WAIS) (O’Neill and Oppenheimer, 2002) and effects upon the Indian Monsoon, as some explained: (22) “The actual danger is the increased risk of various catastrophes, some of which we know very little about.” (9) “I would rather start [the mental model] with the number of occurrences of phenomena like hurricanes, the level of the sea and sensitive regions such as the Arctic, where we observe some tangible changes and species are being endangered or are disappearing.”
This distinction bore direct relevance to this study as reference was made to both terms throughout the project. However, as previously mentioned, the project focused upon on the article 2 definition of ‘dangerous anthropogenic interference with the climate system’ and the interviewer was careful to articulate this during discussions. 9 At a symposium held in Beijing (ECF and PIK, 2004) three concepts of danger were distinguished in relation to climate change impacts: determinative, early warning and regional.


(12) “Would have to define it [dangerous climate change] in terms of impacts rather than the climate itself; impacts to which the local community could not respond.” Dangerous anthropogenic interference was thus interpreted by some respondents in terms of impacts, which would be beyond the capabilities of most societies to cope with and which exceeded critical adaptation thresholds. In other words, several respondents related the need of a mitigation target to the degree of adaptation that would be required by agreeing a threshold. The types of thresholds mentioned were related to direct and indirect impacts exceeding an individual’s or group’s capacity to adapt; for instance, economic thresholds (beyond which livelihoods would be threatened and social upheaval would result) and, interestingly, challenges to legal relations as a consequence of some climate change impacts. This consideration is increasingly important, as infringements of legal rights become a major concern for individuals and communities in search of equitable solutions to environmental problems: (7) “Danger is when man no longer has the ability to adapt. Slow change is OK as you can adapt.” (12) “Large change – cannot adapt quickly enough, for example abrupt change, people do not have resources to adapt in time. This is dangerous.” Danger was regarded by a number of natural resource experts as the result of increased climatic variability exacerbating the risks already faced by existing vulnerable communities. It was also suggested that these effects would act to amplify existing conditions of inequality and injustice10: (23) “What you are actually trying to define is the additional level of danger, caused by additional climate change.” (6) “Climate change is either a primary cause of things or it can be a significant disruptive influence on existing problems… Climate change is itself not unjust, but it acts to exacerbate social inequity and injustice.” (15)“…at some point you have to have a threshold that gets crossed which then from a legal perspective would vault you into some world in which legal relations have changed. When you are talking about dangerous in relation to article 2 I view that as referring to the last sentence in that article [refers to ‘stabilization in a timeframe sufficient to allow ecosystems to adapt naturally, to ensure that food production is not threatened, to enable economic development to proceed in a sustainable manner’] and I define dangerous in relation to that. If ecosystems are not able to adapt naturally you get a dangerous system etc. My understanding of ‘dangerous’ is informed by the sentences surrounding it. From the legal perspective you take your victim as you find him, so dangerous is going to mean completely different things to different groups of people.” The extent to which these impacts are regarded as dangerous was strongly related to the rate at which changes were likely to occur (Perrow, 1984; Mastrandrea and Schneider, 2001; Patwardhan et al., 2003; Parry et al., 1996). Respondents also mentioned the time component as an influence on adaptation (see section below). Most interviewees considered increased variability in the

Thus, danger is not perceived as a threat to the climate system, but rather it is viewed in terms of the impacts resulting from climate change acting to exacerbate existing levels of physical and social vulnerability in areas where the ability to adapt is hindered by unforeseen impacts and evolution prevented by a lack of infrastructure and resources. To these individuals, ‘dangerous climate change’ defines their position more accurately than ‘dangerous anthropogenic interference with the climate system’.



climate system as the most significant source of potentially dangerous changes (see quote 22), with only one respondent suggesting that changes in the mean represented danger (quote 6): (22) “The variability is probably more important than the mean because of the dangers of extreme events. Climate change is really a problem of extreme events as I see it, rather than gradual rises in sea levels or gradual rises in temperatures.” (1) “We are already vulnerable to extreme events, and the response curve is generally strongly nonlinear. Climate change will increase extremes, so bound to be dangerous.” (5) “There are many dangers ahead e.g. storms, droughts, disease vectors and more precipitous events…it is entirely appropriate to call it dangerous.” Not all respondents, however, argued that rapid climatic changes resulting in climate extremes would represent ‘danger’. In fact, two respondents referred to gradual changes within ‘the norm’ leading to more dangerous circumstances which humans could find difficult to survive in: (6) “I think perhaps it’s things like persistent high temperatures in the end which really are dangerous. You can actually manage drought and increased storminess, the thing that you can’t manage is the thing that just keeps on and on – the mean shift. It has become unfashionable to think that the shift in the mean climate has a large impact but I think that might not be true. Sea-level rise is another one, it’s not really an extreme.” “Dangerous interference with the climate system” A small group of respondents made the case for the use of thresholds in the definition of danger, effectively reinforcing the notion that danger lies in human interferences with the climate. It was argued that by defining a point beyond which climate change impacts become dangerous (such as suggested by O’Neill and Oppenheimer, 2002), policy-makers would be more able to set emissions targets as the basis for consensus building and further international agreements: (2)“I suppose I see it as exceedence of some kind of critical threshold beyond which our climate system couldn’t be restored to its original previous state, or where you are exceeding some sort of critical damage threshold… I think of this perception or definition of danger and if there is some kind of threshold there it helps untangle certain differences in points of view.” (4) “The arbitrary use of ‘danger’ does not suffice. A meaningful target is needed, around which an emissions strategy can be set.” (15) “It aids policy-making, certainly. If everyone has agreed and can work towards a target, I see the benefit there. But I think it’s also misleading that you obviously pick a target that reflects this idea of achievability, which is in itself subjective, it’s not objective and you have to agree that it’s not. Once you’re already agreeing on something that everybody thinks is achievable you’ve already given-up a good bit of the game, given the enormity of this problem.” In quote 15 above, the interviewee expresses some reservations about setting a specific target for what constitutes ‘dangerous’; arguing that the establishment of a target may be accepted as a political means toward an end, but may not contribute to meaningful climate change mitigation. By its very nature, any target that is capable of being agreed will itself necessarily reflect a compromise based on what is globally achievable, rather than what may be needed to avoid what is dangerous climate change to many populations and ecosystems. Agreement on a particular emission or temperature target minimizes the fact that still-lower levels of emissions, or a still-lower temperature increase, will still have dangerous impacts on many portions of the world’s population

and ecosystems. In other words, conforming to an achievable agreement for agreement’s sake will not adequately deal with the magnitude of the problem. The respondent argues that a global definition of dangerous has the potential to be counterproductive: (15) “Probably it is achievable and it will be operationalised to the detriment of many populations. In my mind agreeing something is like anything else, picking a level, like picking Kyoto targets. Sure they are achievable and they can be operationalised and all that but at the end of the day does it help? Does it avoid what you are trying to avoid?” Acceptability of outcomes A small group of respondents from a human sciences background suggested that dangerous could be defined by the (un)acceptability of certain climate change impacts, which would be very much dependent upon the context of existing stress levels and variability already experienced by populations. This reflects the concept of type I thresholds (Patwardhan et al., 2003) defined as points which, if exceeded, result in dangers and are therefore the subject of societal decisionmaking. In addition, among these respondents there was acknowledgement that ‘acceptability’ or ‘tolerable levels’ (Hare, 1998) are highly subjective, the meaning of which may be as debatable and arbitrary as that of dangerous. As one respondent mentioned: (16) “I think possibly if we could stabilise where we are at present, we might be OK, and it really depends how much climate change you’re prepared to accept”. One interviewee highlighted the problem of attributing dangerous effects to environmental interactions. Links between climate change and physical (e.g. environmental health) effects are still poorly understood, whilst a specific number of lives are lost due to other directly attributable external factors such as road deaths, AIDS, obesity. This very clear connection between ‘cause’ and ‘effect’ enables society to label losses of this kind as unacceptable and ultimately preventable: (3) “It is extremely difficult to understand when ‘weather’ becomes dangerous. As a principle everything which kills a human life, for us, is obviously unacceptable…. Injuries from transport kill alone 80,000 people in Europe per year. So it depends upon the scale and the components used to decide what is dangerous and what is not dangerous”. The importance of context and comparison in defining dangers of climate change was recognised by the British Government’s chief scientific adviser (Sir David King, 2004) who suggested that the threats from climate change were greater than those from terrorism (a shocking yet moderate threat to life). Sir David’s remarks drew upon the reactions of a Western public to a number of high profile terrorist attacks, which had sent a clear message that such actions were not acceptable. His position has certainly raised the profile of climate change in the UK’s public and political agendas; however, the lack of directly attributable evidence, and causal links remains a barrier to the perception of climate change as an ‘unacceptable’ risk. The acceptability of climate change impacts was suggested by one respondent to be linked to feelings and attitudes towards those responsible for causing changes. The effects of others' actions on very vulnerable communities would cause outrage among ‘innocent’ victims, as one respondent phrased it: (15) “From a small island perspective I would argue that you are in a dangerous situation as soon as you’ve got folks who are basically dealing with a system that has been altered without their doing – an outside causative agent could change their status quo to their detriment.”


The unacceptability of climate change effects resulting from outside influences is being translated into legal actions. Apportioning blame and responsibility for the causes of climate change is becoming an increasingly important goal for vulnerable groups whose plight may otherwise go unnoticed by the global political process. To date, legal focus has been on emissions regulations. However, a belief among some scientists that more sophisticated computer models of climate will soon make it possible to assign blame for environmental harm stemming directly from increasing temperatures is likely to initiate moves towards litigation in relation to greenhouse-gas emissions. As indicated by quote 21 below, a number of test cases are currently underway, the outcomes of which if successful, could set a precedent for future moves to reduce emissions. During summer 2003, eight US states and New York City filed a lawsuit against five US power companies for their contribution to climate change: (21) “It will be very interesting to see what the outcome of this court case [interviewee refers to New York and New Jersey state Governments, currently involved in suing the electric utility industry], there may actually be an internal legal situation in the US which is a very litigious society where successful prosecution in this court case is going to see utilities for potential damage, and costs incurred by the sate might actually shift them away from coal.” More recently, conservationists and lawyers filed a petition with UNESCO (United Nations Educational, Scientific and Cultural Organization) asking it to rule that governments must cut back greenhouse gas emissions in order to conform to their legal obligations under the World Heritage Convention (Climate Justice Programme, 2004). New comment has also been made regarding the legal implications of the 2003 summer heat wave in France (Stott et al., 2004; Allen and Lord, 2004). Stott et al. (2004) maintained that it is very likely that human influences have at least doubled the risk of exceeding the mean summer temperature experienced during the European summer heat wave of 2003. Greater scientific evidence reinforcing such links, and an increasing use of litigation, could result in more equitable considerations and a more precautionary approach to dealing with greenhouse gas emissions. The time component of ‘dangerous’ ‘Time’ was raised separately to rate of change (section above), as an important factor in the definition of dangerous by respondents for a number of reasons. Firstly, the speed at which greenhouse gas emissions are altered has a significant influence over the expected magnitude of climate change impacts thus affecting adaptation, due to the longevity of greenhouse gasses in the atmosphere and the lag time of resulting impacts. Whilst the IPCC has called for further scientific investigation into the likely timing of climatic events, the respondents highlighted the need to generate a clearer picture: (1) “Time is an important factor and affects the ability to adapt to change.” (10) “What can happen between now and when that stabilisation takes place is quite a frightening prospect. You see the guard-rails and the tolerable windows etc. but without a doubt, in the meantime people will suffer.” (16) “They certainly want to know what timeframe you’re talking about: will this habitat be starting to have problems within the next 20 years? What sort of time are we looking at? And certainly that’s the sort of level they want to know because they want to know how much we need to be changing our policy to try and fit into that.” Secondly, concern was raised over the preponderance of scientific evidence of long-term mean climate shifts compared to near-term information on weather variability and extreme events for which, presumably, more accurate predictions can be made:

(23) “It is appalling how the scenario people have fooled the world into thinking they are predicting over the next 20 years when they actually have very little idea what the system might flip into over that time scale.” One respondent considered the impacts upon specific species in terms of changing habitats, suggesting that dangerous effects of climate change could already be observed in some areas: (16) “We’re perhaps already on the way to “dangerous anthropogenic interference”. […] Certainly on the thinner soils on south-facing slopes, beech is doing very, very badly at the moment and ash is doing better. In some senses as far as beech is concerned we’ve already had “dangerous anthropogenic interference” for that region.” Third, it was thought that if some idea could be given of the time-frame of climate impacts, a more definite link could be made with future generations and the impacts upon children and grandchildren (intergenerational equity is discussed further in section Such information could bring into sharper focus just how concerned we need to be: (4) “Timescale will affect people’s perception of what is danger and what is a risky situation.” Utility of ‘dangerous’: the role of science in relation to policy and politics Interviewees’ comments had a propensity to overlap as interviewees mentioned and considered similar aspects in their reasoning. The vast wealth of knowledge at their disposal, gained through high-level contacts with experts from other disciplines, allowed for various sides of an argument to be articulated in response to questions relating to the utility of the term ‘dangerous’ in UNFCCC article 2. This observation supports similar comments in the literature which refer to the tendency for specialists to cultivate ambitions and competence beyond their own field (Elzinga, 1997). Their ability to think ‘outside the box’ meant that whilst many were able to identify numerous faults with the wording of article 2 and the use of ‘dangerous’, from their own perspective and specialist area, individuals were equally likely to highlight the utility of such a statement. We propose that had the sample consisted of more junior individuals with less overall (cross-disciplinary) knowledge, understanding and experience, the responses may have been more clearly recognisable as coming from different fields of expertise. A wide range of views and opinions were expressed on the usefulness of the wording in article 2 and in particular on the term ‘dangerous’. Almost all interviewees accepted that a judgement of values was required in defining ‘dangerous’ and in doing so called for versions of Funtowitz and Ravetz’s (1993) post-normal science approach in an attempt to broaden the knowledge base in policy process. It was suggested by one respondent that the actual existence of the term ‘dangerous’ in a climate change policy document denoted an epistemological progression from the natural sciences toward a more social definition of what is becoming increasingly accepted as a human problem. A few interviewees pointed out that science should inform the debate whilst context-specific judgements would define the meaning of danger in relation to climate change and its implications (as the IPCC has also acknowledged, Watson et al., 2001). In the words of two respondents: (20) “You can’t say it’s inappropriate but I certainly wouldn’t like to link it to policy debates yet. I think its something that certainly scientists alone couldn’t discuss because danger is always to whom and what timescales and regional scales”. (14) “It’s a value judgement and it’s for science to inform a debate and it’s for people to decide.”

The use of ‘dangerous’, a term inevitably associated with strong connotations of negative outcomes, to describe the possible effects of anthropogenic interference with the climate system was seen by four respondents (not directly involved in policy formation) as adding a necessary boost to the importance of article 2 and the issue of climate change as a whole. They maintained that powerful and evocative words, such as ‘dangerous’, can contribute in the communication of such a long-term, global and intangible concept as climate change, by provoking action. The same group of respondents (referred to in the above paragraph) felt that article 2 was clear and simple, as it lacked ‘weasel words’, seen as common in many policy documents. Others, however, mentioned the negative effect of ‘emotional language’ upon public perceptions of climate change, particularly in developed countries, arguing that its use could lead to criticism from climate sceptics and thus be counterproductive: (9) “There is a danger that people may become de-sensitised to the use of strong words and images.” (14) “…The world is becoming rather immured to the scenes we see on television about Darfur and so on.” (16) “I probably wouldn’t have used the word ‘dangerous’, I would have used words like ‘unprecedented’, ‘you’d have substantial effects of’, but, I don’t know, dangerous is interesting.” One respondent even argued that strong language has been used with regard to environmental problems since the UN Conference on Environment and Development held in Rio de Janeiro in 1992 (also known as the “Earth Summit”) with little effect in the way of significant actions. Another set of respondents, on the other hand, argued positively for the current wording of article 2 on the grounds that its ambivalence and ‘middle of the road’ nature creates a ‘zone of constructive ambiguity’ which is necessary within the ‘big treaty’ dominant policy approach. As consensus on this is highly unlikely, the ambiguous phrasing of article 2 allows room for manoeuvre among the very diverse interests participating in the negotiations: (22) “The ambiguity is a way of bringing lots of people / countries / interest groups on board.” (21) “I think the role that the word ‘danger’ serves in article 2 is to create a zone of constructive ambiguity, which allowed the treaty to go ahead amongst a group of parties who had very different ideas about what levels might actually be appropriate to spark-off any kind of action.” (23) “The idea of ‘dangerous…’ is too vague and difficult to operationalise.” The role of science Several interviewees referred to the subservient role that science could find itself playing, serving the interests of only the most powerful. As the quotes below show, there was acknowledgement that science could be moulded to suit the requirements of the policy makers and that, at the end of the day, those with most political influence were most likely to sway the outcomes and definitions of article 2: (20) “…Depending what solutions the policy-makers have already accepted to solve global warming they will then pick out what is dangerous. Science is useful to legitimise these dangers or to make them realistic through these model forecasts. So the pressure on the modellers to come up


with these dangerous predictions is huge in my opinion…This bias which came out of the Kyoto research agenda is part of the problem and the definition should be made more neutral.” (22) “I do see the costs [of mitigation] issue as a problem of economic interest groups. This is to do with the way much research into the costs of mitigation has been funded. In the run up to the Kyoto Protocol negotiations in 1998, research was commissioned by the fossil fuel lobby to demonstrate that the US might face high costs. This is legitimate, but there is a risk of it being unbalanced and lacking independence.” Interviewees strongly shared the view that disagreements surrounding climate change and lack of trust in the decision making process were set to continue, given levels of uncertainty permeating current understanding of climate change and the tools and information which decision-makers use to take their decisions. Science can be misused or intentionally misunderstood to serve the purpose of some, especially when scientific knowledge is stripped of the uncertainty it acknowledges and spun into the language of policy: (21) “Short of the ice-caps melting at a rate far faster than anything that I think we would realistically anticipate or there really being a Thermohaline reversal, there is going to be disagreement about what constitutes danger”. (5) “It is possible that dangerous climate change could be avoided with a determined attempt at mitigation. But many key players have not been persuaded of the evidence. However, the evidence is still ambiguous or not convincing enough in many people’s eyes, particularly the policy makers. Therefore they are not prepared to take dramatic action.” (20) “It is a political definition. But the scientists say that they should be able to define it and they claim that they can put into a mathematical language, thresholds of various sorts, so there is a dispute here.” […] “I think this definition is not science; it is linguistics that we can all agree to.” “…Groups who are opposed to this definition of economy and energy security will use the science and uncertainty to gain time.” (21) “The scientific information is a resource for these social drivers that go into the definition of danger.” […] “National governments are the problem not the solution, that climate is in a sense framed by larger global policy issues rather than the other way round. Science cannot compel policy.” (16) “…Policy-makers want one line. They don’t want “maybe this, maybe that” or “it lies between these high and low and somewhere in the middle”. They just want to tell them one line to take as it where, which we’re not in the position to do.” Thus, the use of a ‘dangerous’ definition in the political domain attracts new problems associated with political rhetoric, will and power relations. Some respondents implied that political definitions of dangerous could vary widely according to political agenda or interests, noting ironically that it may be ‘dangerous’ (i.e. unacceptable) for politicians to alter the status quo by, for instance, increasing energy prices, which could restrict the range of opportunities available for mitigating climate change. It is the conundrum of potentially increasing the likelihood of some risks through decreasing others (the important question of energy provision is examined in more detail in the subsection below): (20) “I think the greatest danger comes from change in energy prices in the shorter-term, who knows what the longer-term climatic effects may be, but I think they are going to be unequal and regionalised.”

Hence, we found among a large number of respondents concerned about the possibility of dangers quite different to the physical impacts of climate change, namely those inherent in the high costs of climate change mitigation. It was argued that society would not be able to cope with the social and economic changes necessary to prevent anthropogenic interference with the climate system, with the actions of mitigation possibly creating a greater danger than climate change itself. It was generally felt that vulnerability could be increased within communities capable of adaptation by panicked measures to mitigate change, a concern which has been voiced for several years (e.g. Rayner and Malone, 1997:332). The dangers associated with mitigation for policy and decision-makers appear most closely linked to economy, energy and competitiveness in the global market. It is clear from respondents’ comments that many countries fear a capping of growth and productivity which, in the case of emergent economies such as Asia and China, is a social and economic necessity, whilst wealthy economies are reluctant to see their advantage diminish (e.g. USA and Europe). Some respondents acknowledged that moving towards a carbon-neutral society would require a large degree of social acceptance and state intervention where changes were necessary, but that the changes required may not be popular. Others maintained the costs of such change had been overestimated: (22) “The Dutch have looked at [implications of any proposed mitigation or adaptation] seriously, saying that of course it can be done technologically but the political and legal implications are very worrying. [Refers to book ‘Climate Neutral Society’] A society which doesn’t travel any more and all living in little high-rise flats, a society that anybody would want to live in. There is very little trade and very little travel. A huge amount of legislation would have to change because the state just wouldn’t have the power to do these things – on such a short time-scale.” (20) “…There is a link between what is considered as “dangerous” and the costs of abating the danger. The high cost and difficulties of action can lead to apathy and fatalism. However, my own work and other literature suggest that the costs of mitigating greenhouse gases are in fact extremely low and not overwhelming.” As the first quote above (22) outlines, the perceived danger of transition to a different set of cultural norms and priorities, can acts as a major barrier to political and behavioural change. It is preserving the status quo following the maxim ‘more of the same’ that may be used to dispel those fears of something different and potentially dangerous (i.e. disruption to our current lifestyles): (21) “…There is another set of social drivers which are those attenuating danger. You have there issues like concerns about economic competitiveness, especially the US but not only. You also have concerns about sovereignty”. These comments refer largely to discussions of the ‘possibility’ of policy change in the absence of clear evidence. Thus, many of the interviewees maintained that significant impacts associated with climate change could be the only catalysts that would succeed in mobilising political interests towards effectively addressing the problem. Similarly, the same interviewees suggested that the public could be encouraged to enact behavioural changes to mitigate their effect on the climate, following extreme events, as was warranted in the USA following the attacks on the World Trade Centre in 2001:


(14) “In order to make something politically and economically acceptable to an electorate in democracies there has to be a major incentive or a major scare. You can set-up homeland security with new laws on terrorist arrest in relation to 9/11, but you couldn’t have done it before-hand.” However, the climate change threat, expressed through an extreme event, will not be perceived in the same way globally. To follow from the US example, while the affront authorised restrictive measures in the country, the British reaction to the terror threat, compulsory identification cards, has not been received favourably. A powerful UK lobby maintains that the infringement of human rights brought about by compulsory identity cards outweighs the actual risk of terrorist attack, a perception which could possibly be altered by more ‘immediate’ threats. For instance, the heat wave in France during summer 2003 was a surprise to many who had thought it impossible for over 15,000 people to succumb to heat within a modern and developed society. However, this shock appears to have caused rapid policy changes which led to changes in key areas of infrastructure, utilising or re-directing resources already in use, the most important shift being in political will: (3) “Take the heat wave in 2003…some countries have set up focus points, there are early warning systems, there are intervention plans etc. It has become part of the budget system etc. This needed a trigger, something to stimulate this, which put at risk a policy maker such as the minister for health. If countries are able to set up something that is as effective as this in a year, it means that the technology, knowledge and infrastructure are there.” There was some disagreement among the respondents as to the rate at which infrastructural changes could or would be made, given the huge investment and commitment in developed countries towards particular types of housing, cooling, heating, drainage, public health and social systems and a greater degree of flexibility in adaptive measures. The major constraints to change still exist within current systems and infrastructure. However, as the French example has shown, developed countries are not immune to the effects of climate change. As one interviewee mentioned: (17) “I think there is a separate set of issues that I would put down which are related to the slowness of human response…we’re slow to adapt…that’s partly due to infrastructural things that are very difficult to change, like we can’t rebuild housing stock completely even if it were appropriate to do so. And there is inevitably quite a lot of inertia, that people aren’t inclined to act immediately or take steps for something that is perceived to be a distant threat and maybe not a very important threat for a lot of people.” How the 2003 heat wave affected French or European energy policies remains to be seen, given that low probability, high impact events appear unlikely to provide the impetus for far reaching energy and economic policy alterations. Energy Perhaps the most significant form of anthropogenic influence exerted upon the climate system is the provision and use of energy, which remains central to the climate change debate. Epitomising what many see as an unsustainable development trajectory, issues of energy and energy security introduce additional sources of danger associated with conflict and mass social upheaval. Alternatively, it can be argued that the energy debate is being framed somewhat by the reported harmful and dangerous effects of climate change. Some respondents argued the danger of climate change was purposefully exaggerated in order to facilitate and speed-up the transition of industrialised societies to low-carbon economies. Others linked the poor implementation of new ‘clean’ technologies to resistance from oil giants, eager to

make the most of their huge investments in oil production and supply infrastructure. It was suggested that the speed at which this transition occurs is a moral choice for those in the industrialised nations. As the diagram below demonstrates, continued commitment to fossil fuels has an increasingly deleterious effect upon marginal populations /ecosystems and those most vulnerable to the effects of climate change. Despite the existence of alternative forms of energy and innovative technologies, social, economic and political barriers exist to the apparently costly and uncomfortable transition away from a high carbon economy. In this instance, danger is defined by moral choice; the level at which society balances lifestyle change with its willingness to accept the degradation of what is perceived as the corpus of social and ecological order (see Figure 4.2). Figure 4.2. Danger: a moral choice (adapted from an interview).

Interestingly, a few respondents widened their discussions to argue that climate change should be framed within the wider requirements of energy security facing much of the western world today. One respondent implied that society may consider the risks of climate change acceptable if these are weighed against the potential destabilising effect of diversifying energy provision to address climate change: (14) “The dangers presented by climate change may be things we have to cope with, and it may be that humanity decides it’s willing to cope with them and not exacerbate the problems we have with energy security anyway.” Others referred to the danger of climate change in the context of fears over oil prices and the effectiveness of renewable energy sources in a transition to a low carbon economy. It was suggested that a convincing message about the dangers of climate change may allow for a re-appraisal and greater acceptance of other energy sources such as renewables and the nuclear option. Thus, questions are again being asked about the certainty to which climate predictions can be made and the background of those interested parties in the research and dissemination of climate change information. As some interviewees commented:

(14) “People such as Jim Lovelock at Dartington this year, and Crispin Tickell etc. are now convinced that there are now changes in the Earth’s systems that are hugely threatening and the world ought to move very quickly towards a nuclear energy policy, which for environmentalists is a major step.” (20) “Parts of the North are desperate to replace fossil fuels, not because they are running out but because of costs and political factors. So they are using, in my opinion, the global warming scare which is longer term to support the research & development shift into wind-turbines and photovoltaic…Fear of energy security…is a real issue but not everywhere in the world which is why you get this resistance to Kyoto in those areas where the climatic threats are not persuasive for one reason or another, e.g. Russia and Australia, one gets wetter the other becomes less cold, but the energy security affects mainly Western Europe.” These two comments exemplify a strand in the climate change discourse, both in the literature and among interviewees, that increasingly frames the climate change issue from within the energy debate. As stated in quote 20 above, this may represent a purely UK or Western European perspective as energy discussions are currently high on the political, environmental and economic agendas due to an increasing reliance upon fuel imports as North Sea oil and gas reserves run low and amid fears over cooling and a THC collapse. The political process When asked about the possibility of political settlement on an agreed definition of ‘dangerous anthropogenic interference’, a number of interviewees were sceptical and scathing of the political process, maintaining that such an approach was not a satisfactory, effective, equitable or fair method for approaching global greenhouse gas emissions reductions. However, the same individuals tended to be resigned to this, perhaps a reflection (referred to above) that a certain degree of pragmatism is necessary within a complicated process of debate: (15) “It’s negotiated text which is the difficulty with it but also the beauty of it. It’s a bit vague in ways that are helpful and also difficult…It’s a convention text that is already agreed, you are not going to renegotiate it, it just can’t happen. All you can really do is try to come to some common understanding of what it might mean that will never get put down in text anyway because no-one wants to renegotiate it. So it’s not going to change.” (12) “It is possible to come up with a politically acceptable agreement, Kyoto almost did. The arbitrary level could be accepted, and in many ways that is the only way it is possible.” (22) “I also think it’s very important to make the assessment in terms of achieving a global consensus, achieving agreement between the main parties involved. It is a social and political issue as well as an economic one. That is why the semi-legal route, which has been adopted, is an important one to achieve consensus and agreement.” In some cases, it was suggested, international negotiations aimed at defining dangerous for reasons of mitigation were purposefully avoided by countries unwilling to discuss an issue that is viewed as secondary in comparison with more pressing development issues: (23) “Their agenda is so full of adaptation, that to expect them to go back to very first principles and say ‘we need to have an agreed multi-attribute definition of danger’ will be seen as taking them away from (they only have a few people in these areas) what they see as the critical areas of capacity building and adaptation funding etc.”


(10) “I think it is well phrased in terms of provoking debate. But I think the issue is that the debate is constantly suppressed by governments.” In addition, some interviewees observed that despite the far-reaching implications of climate change, discussions tend to be downplayed through partial government involvement, which thus acts as a barrier to higher level or more wide ranging decision-making processes: (10) “Only environment ministers got involved in the Kyoto negotiations, it should have been broader. Had they realised how potentially significant this was for the whole of society there should have been more people there, rather than just a nice club of environment ministers.” (9) “When you move onto the implementation the tendency is to pass responsibility on to the ministry of the environment, and the ministry of the environment in the architecture of most countries is not a key ministry which doesn’t have a say in key ministerial negotiations.” Some interviewees also maintained that from a developed nation perspective, dangerous criteria are also selected depending upon existing or planned social and economic agendas: (21) “We may when it is convenient, refer to the framework convention to legitimise policies where we can claim them as successes”. It was also suggested in various ways by a large proportion of respondents and from a range of disciplines, that the climate change debate was susceptible to exploitation by normative agendas, even promoted by environmental groups and research institutions with a vested interest in the propagation of climate concerns: (23) “There are very strong statements out there and they are accepted as true, when we all know they are quite exaggerated and if you did the formal maths you would probably end up with something quite modest….I think it has given voice to this social risk exaggeration, so there is a lot of activity around “can we define danger”. And what they mean is can we communicate how dangerous we think climate change will be? As opposed to: can we define scenarios where the impacts are adaptable? No-one is doing that.” These comments suggest that to reduce the scope for normative agendas within political negotiations, subjectivity should be replaced with legitimate scientific information, thus again placing great expectation upon the sciences. Concurrently, however, despondency about the effectiveness of global agreements spurred many interviewees to support moves towards more localised policies and actions, as is being experienced in the USA (e.g. Kates and Wilbanks, 2003) (discussed further in section (3) “Dangerous for me is on the one hand CO2 increase which is also in some place, countries, cities etc. going concomitant with the increase of secondary air pollutants and other pollutants which means it’s not only a CO2 increase it’s an increase of things which leads to numerous direct and indirect health impacts.” One respondent referred in particular to perceptions of danger, mentioning that individuals tend to give consideration to the risks they perceive as relevant to themselves, which are not necessarily those implied by article 2. Thus, although the outcomes of climate change may be uncertain, risk communication should focus on representing climate change via individual concerns:

Following the ratification of the Kyoto protocol by Russia in October 2004 (and the entry into force of the Kyoto Protocol on 16 February 2005), it may be interesting to reassess respondents’ views on the effectiveness of such an international agreement.


(9) “There are many debates about whether or not there is an intrinsic risk or if they are all just perceptions. In that case I think the risk that is measured by scientists with quite high confidence in the data of a probability of real danger, this is not what is perceived by the public or the policy makers… In terms of policy response it would be better to identify some very clear impacts, in terms of public communication it would be very important to not talk about global changes as these are too big for people to cope with… People have to understand what it means for them in their daily lives.” Equity Approximately half of the interviewees maintained that the political process surrounding dangerous climate change was inequitable due to the influence of interest groups upon the trajectory of climate change research and decision-making. When initially asked for their views on climate change, many respondents suggested a metaphorical link between global issues of equity and sustainability and the plight of the global climate, maintaining that the entire meaning of ‘dangerous anthropogenic interference’ or ‘dangerous climate change’ had been hi-jacked and was missing more important points regarding development and global inequity: (20) “I would say that in the end, ‘dangerous’ will be defined by the politically powerful most influential actor in the whole system. Some of the public rhetoric i.e. “modelling predictions”, increases in weather intensity, droughts and floods etc. will be used to persuade. But I think the real driver, given our political system and the time-scale difference, will be effects upon the economy and in particular trade relationships.” (22) “Who is going to say what is dangerous, when it is affecting different parts of the world and future generations.” (6) “Climate change is itself not unjust, but it acts to exacerbate social inequity and injustice.” (21) “The discourse has to be framed within what I have in other contexts called the other components of an ordinary language definition of risk, which is the trust, liability and consent issues. In other words requirements of equity and fairness…governments … act for self interests and it would be extremely difficult to overcome that.” (10) “I think the definition which exists in article 2 is a good one; I don’t have any problem with it myself. However, people tend not to use it accurately so they look at the definition and they see the bits they want to see and they choose not to see the bits they don’t want to see, and this causes a lot of issues. […] In debating the meaning of dangerous, poor communication between the various sides leads to a lack of consensus.” This bit may go better when discussing the politics of decisionmaking (16) “A global definition [of ‘dangerous…’] which must be played out amongst major decision makers will not represent the needs of all nations in an equitable way. […] [There are] concerns over sovereignty i.e. national policy dictated by non-national considerations.” Whilst most respondents framed danger from a utilitarian anthropocentric perspective, two respondents referred specifically to the ‘right’ of natural systems to continue functioning. They argued that instead of experiencing direct physical impacts, human beings experience the dangerous secondary impacts derived as a consequence of damage to ecosystem services, loss of habitats and species. Thus, human beings have a duty to protect the wider needs of the planet and ensure that change does not make existing ecosystems less viable. The emphasis to act and the framing of these questions should therefore be moral (including spiritual and ethical stances):

(6) “Dangerous in this sense may have an ethical dimension in relation to our bad stewardship. This does not mean physical danger to ourselves but as these creations are all part of a web, so the loss of such things may adversely affect us.” (13) “In terms of biodiversity, it is a very personal view. I would have thought the most dangerous changes are those which cause species to be lost.” A minority of respondents felt that despite the need to negotiate an equitable definition of dangerous, this process could act as a barrier to effective and timely mitigation (see quote 5 below). Thus, a consequent danger suggested by one other interviewee (quote 15), lies in hasty attempts to arrive at a global solution to the major effects of climate change may ignore the fact that small islands and more vulnerable regions / communities are already being affected. International negotiations intended to prevent the THC shutdown and WAIS collapse may obscure more relevant issues of equity: (5) “Global agreement on what constitutes ‘dangerous’ may not be possible within the given timeframe.” (15) “There has been discussion of a 2 degree threshold. I am not comfortable with that because I think by 2 degrees you have already screwed-over many places and I don’t buy that. To me dangerous is something which is different to different populations, it’s the same sort of tort law concept [i.e. you take your victim as you find him]. You can’t be trading-off people in Bangladesh with the consumption habits of people elsewhere”. With regard to intra-generational equity, very few respondents explicitly considered this aspect during interviews. The limited reference to this issue may not indicate respondents felt it bore little importance in the existing climate change discourses. Rather, they may see the debate as very much centred in the present, with emphasis on action now, hence the focus on article 2. Furthermore, it is possible that, although these concerns were present at a personal level, our research tended to elicit a more professional response to climate change issues. Assessing UNFCCC article 2 Unrealistic aims? The issue of scale was most notably discussed with regard to the utility of global agreements and actions to combat climate change. Two respondents argued that the entire approach of the IPCC and article 2 was based on the assumption that there existed a global decision maker. However, the existing plurality of beliefs and priorities result in international processes which cannot achieve global targets: (23) “Analytically, if you try it on a global basis, you are actually doing something that is not done…. There is no international process that targets global, economic status in that way.” (21) “We argued that the one big treaty was probably not going to work… local authorities worldwide may have more in common on this issue than they have with their own governments.” […] “I think it is a monstrous failure of policy and of the environmental movement and of those of us who seek to advise policy makers that we have allowed Kyoto to become the only game in town”. Around half of all respondents found the diplomatic wording and implicit messages of article 2 to be unrealistic and problematic for basing serious climate change mitigation efforts upon. Without a clear background or explicit interpretations of the nature of climate impacts, for whom and where, many interviewees maintained the article was limited in its utility for policy guidance:

(23) “I have not gone through the legal text and the supporting arguments, but there is very little guidance as to what those criteria ought to be and some of the criteria implied in the earlier text were clearly unworkable – without those criteria it is very difficult to operationalise.” (17) “Whether it’s realistic to expect that whatever processes are already in train with regard to climate change can be arrested at levels such that ecosystems will be allowed to adapt naturally; that may be unrealistic.” (6) “Article 2 is put in very positive terms. ‘Preventing’ is a wrong expression as it is going to happen anyway. Thus implying we can stop any change from occurring. Should be ‘reduce and eventually eliminate’ rather than prevent which is too optimistic. Next sentence ‘…allow ecosystems to adapt naturally…’ is wildly optimistic as some ecosystems won’t, even with natural change.” (20) “What a wonderful wish list. As if we could do all of these things in advance, it’s typical diplomatic rhetoric.” However, this view was not shared by all, with several maintaining that article 2 presented a political tool towards an objective in the most simple and uncomplicated way: (15) “It’s negotiated text which is the difficulty with it but also the beauty of it. It’s a bit vague in ways that are helpful and also difficult.” […] “It’s a convention text that is already agreed, you are not going to renegotiate it, it just can’t happen. All you can really do is try to come to some common understanding of what it might mean that will never get put down in text anyway because no-one wants to renegotiate it. So it’s not going to change.” (14) “I don’t object to the word ‘dangerous’. It’s like a lot of the words produced by lawyers in policy documents e.g. critical loads, they don’t have a rigorous scientific meaning but they do have resonance with the public and they do frame the issue.” (10) “I think it’s useful. Nothing provokes debate like article 2. I think that is very useful. Whether it is universally applicable in a sense doesn’t matter because it does force people to think about it.” Finally, the majority of interviewees did view article 2 in a pragmatic way, realising its uses despite its problems: (4) “…The improvement of scientific data, evidence of change and increasing extremes will move climate change up the agenda. How article 2 is worded may not be as important as the building evidence of climate change.” (15) “To the extent that it galvanises actions at higher levels, OK fine. I just think it creates a degree of comfort that shouldn’t be there.” Improvements to Article 2 Interviewees were asked if article 2 could be improved and if so what changes they felt would be needed. Some interviewees concentrated on the wording of the article while others took a broader view. Interviewees’ suggestions for improvements were fairly widely spread, often reflecting their own areas of interest. However, a broad split can still be identified with those arguing for a more detailed definition (i.e. ‘What’ the dangers are; ‘Who’ will be most at risk; ‘Where’ will the greatest risks occur; ‘How’ will these dangers be experienced and ‘When’ to expect these dangerous effects) in contrast to positions arguing that the arbitrary nature of article 2 stimulated debate and allowed for changes to be made over time as interpretations and evidence develop:

(20) “I would take ‘dangerous’ out and replace it with a more neutral word, that the research allows you to say positive things about – perhaps a more cost benefit approach.” (4) “But, to take it forward it would be necessary to unpick some of the statements in there and the meaning of some of the statements.” (5) “If the article were re-drafted, I would be tempted to highlight what we think the main kinds of dangerous climate change are. This would make people respond by refuting it or not, and would encourage a debate and hopefully more clarity.” Redefining notions of danger: dangerous to whom and where? Often mentioned criticisms of article 2 revolved around the need to identify receptors of dangerous impacts, lack of political action resulting from the article content and the inequitable way in which groups are represented in discussions and decisions arising from article 2. The quotes listed below relate to difficulties arising from the implied meaning of ‘dangerous….’ in article 2. They suggest that the plurality of beliefs, understanding and contextual situations make climate change impacts difficult or even impossible to define. Frequently mentioned by respondents was the question of “Dangerous for whom?” noting that context is vitally important if ‘dangerous’ is to be defined, as what is dangerous for one individual or group, may not be dangerous to others. This appears to be one of the most complex questions for those in search of a well-defined notion of danger in relation to climate change and forms a major barrier to the acceptance and utility of article 2: (8) “Danger is not just an abstract phenomenon that you can just simply say there is danger. There has to be a recipient of that danger, and that is not defined in your quote ‘prevent dangerous anthropogenic interference with the climate system…’ […] The climate system is not the receptor of the danger as some kind of socially constructed value must be attached to the effects or impacts to be significant. However, there is no explanation of dangers to ‘whom’.” (4) “Dangerous means different things to different people. Climate change will be experienced in many different ways, in terms of its impacts and in terms of the vulnerability of different communities and systems to climate change.” (9) “The meaning of danger is dependent upon individual’s (group’s) perception…lack of clear / tangible impacts affects people’s perceptions.” In other words, these comments make clear arguments for a move away from a purely thresholdsbased approach to a policy which addresses who or what is exposed to a certain risk, how that risk compares to others and the practical considerations of taking action. In fact, most interviewees thought it was unlikely that negotiations on article 2 would yield a workable global consensus, particularly within a rational time-frame and in the absence of stronger scientific or physical evidence, although more recent ratification by Russia of the Kyoto Protocol on 22 October 2004 may have modified some of the respondents’ views. In effect, these opinions reflect the limitations of external definitions of danger, supporting the notion that danger is contextually, and therefore, internally defined according to how it is perceived and experienced (Dessai et al., 2004). Most respondents shared the view that scientific definitions of ‘dangerous’ based upon probabilistic estimates are too limiting, given the uncertainties inherent in understanding the climate change

system and related responses, again reflecting the distinction made by Dessai et al. (2004); as one respondent put it: (21) “If you want to talk about criteria, certainly anything which restricts the discourse to a language of probability and magnitude of consequences is too restrictive.” This argument has been reiterated in various forms in the literature; for instance, Pittock et al. (2001) maintain that repeated experiments and frequencies of measured outcomes are not compliant with socio-economic sciences and are thus unable to represent the human component of dangerous climate change. Schneider (2001) however (see also Dessai and Hulme, 2003), argues that policy makers need probability estimates to assess the seriousness of the implied impacts of climate change. In the same way, some interviewees stressed the importance of ‘possibility’ and that an indication of the probability of stochastic events was necessary to define danger, something explicitly absent from article 2. One interviewee maintained that the lack of clearly communicated evidence on the likelihood and probabilities of events such as THC shut-down or failure of annual rains diminishes the element of danger inherent in future climate change, which reflects Pittock et al.’s (2001) support of a Bayesian approach to probability estimates of future conditions on Earth. Past geological and historical events (e.g. lost civilisations) were cited by one respondent as evidence that dangerous impacts are possible and that these occurrences could act as analogues, bearing witness to similar events that may take place in the future: (22) “One thinks of previous civilizations, which have been lost or extremely damaged by climate events… we can see what has happened in Iceland or the Easter Islands and extrapolate to the planet.” Many respondents also maintained, regarding the ability of article 2 to promote actions to deal with climate change effectively, that a global definition does not suffice in a debate which is increasingly acknowledging regional impact scales and locally experienced drivers of human behaviours: (21)“Basically we are going to have to find emergent or clumsy solutions which are things which instead of everybody agreeing on the problem and agreeing on the solution, people will be motivated to do things that cumulatively help to resolve the problem but for diverse reasons”. Several interviewees maintained that the better-defined risk assessments lie at regional and national levels, based upon evaluations of underlying exposures. First, it was suggested, local impacts of climate change must be identified and considered in greater detail. These could include geographical, cultural or economic factors among others. Second, the ‘recipients’ of dangerous impacts must be identified. As one respondent explained, coastal communities may experience a higher level of danger than those inland even within a fairly small area (such as, for instance, occurred around the Indian Ocean rim following the tsunami on 26 December 2004). Third, the community affected should be defined in terms of institutional capacity, social capital etc., which define its ability to adapt. One respondent proposed the use of a ‘standard catalogue’ in terms of outcomes of vulnerability. This would not be exhaustive but would relate to key factors such as lives lost, disease burden, livelihoods, economic systems under threat, welfare, amenities, perceptions and social stress. The interviewee explained that these data gathered through consultations and elicitations may reveal

quite different reasons for concern or elements of danger if provided by different sectors and groups in geographically diverse areas. Thus, a common theme shared by the respondents was the need to make climate change more pertinent to people’s lives by communicating a more tangible set of dangers at a regional or local level. The majority of interviewees argued that global consensus agreements had had their day and were generally ineffective, referring transversely to the Kyoto Protocol to exemplify their point: (14) “The judgements on what actions are to be taken in response to that should be taken nation by nation.” (21) “Basically we are going to have to find emergent or clumsy solutions which are things which instead of everybody agreeing on the problem and agreeing on the solution, people will be motivated to do things that cumulatively help to resolve the problem but for diverse reasons.” The cultural shift required to seriously incorporate climate change considerations in all areas of decision-making were highlighted by some interviewees. Thus, for instance, one mentioned the potential role of interested investors, whilst another lamented the compartmentalisation of discussions in international fora: (1) “We should consider responsibility in ownership of money…pension funds are coming on-board now. Every company is owned by people and the chain of control is fuzzy with very little consultation between owners and managers in decision making. There is potential for change here…. The investment world is important…they have done a lot to promote the issue of climate change among stakeholders and will be a very powerful influence in the future.” (10) “Nobody really talks about it; everybody sits in their own little compartment. You go to these international meetings where the Saudis are saying the same thing as they have always said and the Americans and the Chinese and the Brits and the EU all have their positions and they all completely fail to talk to one another, so I guess you could say that there is a failure to have a dialogue.” Summary discussion and conclusions Methodological observations A major finding from this mental models interview procedure was that interviewees were encouraged to think systematically about their beliefs. For those who had not considered the issues in detail before the interview and those (the majority) who had not previously constructed mental model diagrams, unfamiliarity was manifested in an initial reluctance and hesitancy. However, once the first few points were on paper many found it increasingly easy to pursue an in-depth discussion. As mentioned earlier, interviewees chose not to make significant use of the literature based influence diagrams provided, preferring to enter information into their own mental model diagram. In terms of the final sample size (23 respondents), we believe this number to be adequate in eliciting and representing the majority of issues and beliefs existent among our target group (western European perspective) of climate change interviewees. Indeed, the construction of a meta-mental model (a synthesis of all the mental model diagrams produced by the respondents) revealed an overlapping of opinion and repetition of salient issues within the breadth of interests and perspectives (e.g. agriculture, policy, coastal zones, energy). This study could be extended to include interviewees with representatives of other areas of expertise, following suggestions from interviewees, such as annuities/investments and climate modelling.


Telephone interviews may have aided or hindered the process, depending on the context and perspective of the respondent. In some instances the lack of personal contact with the interviewees may have adversely affected the depth and breadth of discussion, particularly during the visual mental model diagram exercise. In other instances, respondents seem to have preferred the lone telephone interview style. Although the interview information sent to all respondents as part of the official invitation to participate in the project included the wording of UNFCCC article 2, it became apparent that familiarity with the terminology of the article influenced interviewee’s responses. Those who had considered the article previously in their work discussed the underlying concepts and issues relating to its use. Other respondents tended to focus on the wording and phrasing of the text. Most interviewees exhibited a wide knowledge extending well beyond their specialist areas of interest which became evident as the discussions touched upon similar issues and lines of reasoning. We would argue that if the interviewed sample had consisted of more junior individuals (i.e. specialists in their fields, without that great degree of overall knowledge, understanding and experience of the complexities and cross-overs among different aspects of climate change), the mental models and discussions may have been more clearly linked to particular areas or disciplines. However, some interviewees explicitly stated that they regarded themselves as specialists in their area rather than experts in climate change and maintained that they did not have a clear understanding of all the complexities and diverse views surrounding the debate (e.g. some misunderstandings about the causes and relationships with other environmental issues). Reflections on the mental model discussions The exact terminology used to define ‘dangerous’ in relation to climate change was not an issue to many respondents. Most preferred to use “dangerous anthropogenic interference with the climate system” (as per article 2) and “dangerous climate change” interchangeably. However, two interviewees explicitly distinguished them, maintaining that the former refers to the process of avoiding such interference (often quantified in terms of atmospheric CO2 concentrations), the latter focusing on the impacts that would arise from certain changes in the climate. ‘Danger’ in relation to climate change was conceived by some respondents in terms of critical thresholds of irreversibility, either (a) in the climate system or (b) in terms of impacts. The former interpretation clearly has implications for mitigation, whilst the latter relates closely to adaptation. The scales at which these responses are instigated vary greatly: internationally and nationally in the former, mainly regional and local levels, geographically contextualised, with regards to adaptation. Most respondents did, however, specify that the degree of danger related to impacts of climate change would be defined by various inter-related aspects, including: • • The rate of change (i.e. severity and timing) in the climate system: abrupt or rapid events vs. gradual changes, progressively deviating from the ‘norm’; The context (geographical and social) of changes in the climate system, related to the ability to adapt to these changes. Most respondents argued that ‘danger’ in climatic terms means reaching a state that is irreversible and potentially associated with disastrous outcomes (i.e. impossible to respond to), in whatever form they may be defined. These in turn would amplify existing social conditions of inequality, injustice and vulnerability.

One respondent widened the discourse beyond mitigation to adaptation, maintaining that it could also act as a catalyst by varying the amount of danger perceived or realised. Adaptation itself is clearly dependent on (and therefore reflective of) political, social and economic factors and unlike mitigation, which may be instigated through choice, adaptation may be an unavoidable necessity.

The extent to which adaptation is successful will be mediated by the availability of finances and resources. Focus on the agents involved in defining ‘danger’ exemplified the complexities of addressing climate change successfully: the role of science in providing information for decision-making was mentioned, although its subservient role to the political powers was also explicitly raised as a concern. Uncertainty was also mentioned by many respondents as a fundamental characteristic of scientific research and findings, sometimes subject to misuse in the political decision-making process. Furthermore, the issue of acceptability of change and measures to address this change were raised. The contrast between the interest in public support defined by maintaining the status quo and recognition of the dangers identified by climate change science were debated by respondents (i.e. variation between scientific and political definitions of danger). Several respondents argued that the focus of article 2 was predominantly on the physical changes resulting from dangerous climate change not, however, taking into account the social contexts which dictate the varied responses to impacts through adaptation. Thus, two approaches were described: • Official Approach: International implementation of article 2 focuses upon mitigation. Establishing a range of thresholds, which dictate the target for a future climate regime. At the international level, therefore, this process develops into discussing the acceptability of a range of impacts. Because the latter are very difficult to define with scientific precision, the international aim has been to reach definable mitigation targets (based on thresholds), as these set a ‘numerical responsibility’ according to which strategies can be defined (a few respondents referred to negotiations surrounding the second phase of the Kyoto Protocol). Equity Argument: Some respondents suggested that no matter what the target set, there would be winners and losers. The ways in which targets were negotiated did not represent all interests in an equitable way and the issue of a global mitigation target is missing the point in terms of wider issues of sustainability.

The disenchantment expressed by several interviewees with the international process of negotiations partly reflected the weariness of the long-standing US reticence in ratifying the Kyoto Protocol thus enabling it to enter into force. Ratification by Russia on 22 October 2004, after the interviews were completed, may have altered some views expressed to this regard and summarised in this report. Interview discussions and mental model diagrams focussed to a great extent upon climate change policy at national and international level, with only a small proportion (3) of respondents making specific reference to public involvement in climate change issues. This may not reflect the importance given to public involvement in decision-making but rather the nature, design and trajectory of the interview. Of the respondents who mentioned the influence of public on climate change, the necessity was noted of engaging the public through communication programmes designed to increase their involvement and contribution to mitigation. One respondent argued that communication is essential to keep the issue of climate change in people’s minds with a view to behavioural change. To date, focus on extremes potentially associated with climate change has kept public interest in the issue. However, communication campaigns have not been able to successfully convey the message of the long-term dangers of climate change. The need to work more on public associations and communication methods was identified, given that uncertainty could cloud the message the UK government is interested in spreading.


A number of respondents also mentioned the need for climate change to be made more relevant to publics, who will ultimately influence the uptake and success of any actions to address climate change. A few respondents suggested this would occur by encouraging common actions for different reasons, communicating climate change within the context of other threats and dangers that individuals are able to relate to personally (i.e. situating climate change in people’s daily lives). Thus, proposed improvements to UNFCCC article 2 included moving away from debating notions of ‘danger’ to practically implementing mitigation, by ‘localising’. This term refers to two possible approaches: (a) talking directly to people about how climate change will affect their daily lives, and (b) representing climate change in the form of other issues through the use of ‘multi-purpose’ policies. Our finding from this study and the 2004 Dangerous Climate Change workshop was that making climate change relevant to people by linking it with more immediate issues does not require agreement at international levels - a notoriously difficult goal to achieve. Examples of localised activities at community levels that have achieved great mitigation targets were mentioned. A couple of respondents explicitly mentioned the differences in climate change risk perceptions between expert and lay publics. The corollary that followed was that within a democratic system, whilst the voting public remains unconvinced and unwilling to make changes for a global cause, it is unlikely that all but the strongest and most motivated of political leaders will take serious action on climate change. A few interviewees argued that for this to occur, a strong drive needs to be created so that available channels can be used for enforcing such messages and employing them to effect change (e.g. investor power and interaction with stakeholders). Several respondents widened the scope of the interview arguing that in fact climate change should be framed, as is being framed by some political establishments, within a wider energy debate. This emphasis could be partly related to the intense debate widely covered in the press about the nuclear power as a ‘carbon neutral’ energy source which could substantially contribute to meeting the ambitious UK 60% CO2 reduction target by 2050 (Caldeira et al., 2003). However, more recent events such as the admission by the UK government in December 2004 that the target for reducing CO2 emissions by 20 per cent by 2010 was likely to be missed (DEFRA, 2004; Wintour, 2004), could have reinforced some of these opinions. This situation could thus reflect the current governmental quandary, mentioned by some interviewees, relating to sustained public and environmental lobby resistance to the nuclear option and pressure from energy providers and industry. Political rhetoric surrounding this issue was seen by a larger proportion to be driven by modelling predictions (e.g. increases in weather intensity, droughts and floods) and scientific enquiry, which are used to persuade the wider public. However, it was suggested that the real driver of energy transition, given our political system, will be the impacts of climate change upon the economy and in particular trade relationships.



Overall Project Conclusions

The four phases of the study reinforce the notion that danger is context and time specific. By focusing on the notions of ‘threshold’, the climate science community has predominantly defined danger through assessments of risk and vulnerability. These have been translated into greenhouse gas emissions and temperature limits, setting ambitious policy targets. These ‘external’ interpretations are discrepant with those defined internally, according to personal experience, beliefs and knowledge (corroborating Dessai et al.’s 2004 distinctions). Among publics, especially in Western nations, the perception – action gap is still wide. Although climate change may be perceived through the senses via weather anomalies, it is not an immediately intuitive phenomenon and not considered as important as other more pressing personal or social issues. Individuals still have difficulty relating climate change to their everyday lives and behaviour. It persists as an ‘unsituated’ risk: that is, individuals tend to psychologically distance themselves from the issue, relinquishing it to other spaces and future times. Yet some people accept that at a societal level it is a moral and ethical concern. Interestingly, recent research has shown that ‘western’ publics identify extreme and catastrophic events, highlighted by the scientific community, with climate change (such as for instance, a collapse of the West Antarctic Ice Sheet and halting of the Thermohaline Circulation). From this angle, expert and lay perspectives of climate change are not too dissimilar, both referring to thresholds. It is likely that laypeople’s associations with these events will have been influenced by media reporting of scientific findings. However, there are more subtle differences between the two groups related to the way in which societies experience climate change: (a) sense of urgency: in the UK and rest of Europe generally, science and some political leaders have identified climate change as a serious threat to be addressed, whilst many members of the lay public perceive climate change as impersonal and removed from their daily lives; (b) difficulties in associating climate change with the correct causes and solutions: research shows that individuals interested in modifying their behaviour often need to be given guidance on the correct actions; and (c) diffused sense of responsibility: most individuals are ambivalent towards climate change as they attempt to balance perceived risks and benefits (of technology and energy sources) at different levels (individual, societal) on various timescales (present vs. future). Furthermore, workshop participants acknowledged that the perceived need to act does not necessarily lead to action. Fatalism and withdrawal are common reactions to a crisis narrative. The workshop also highlighted that communication of climate change presents a challenge for the media, accustomed to represent balance of opinions (which may actually act to distort the scientific message). The media is not set up to represent long-term, subtle and uncertain processes. Transforming the media function will be the challenge, in order to present readers with a more realistic understanding of the difficulties involved in defining policy on climate change. Managing climate change means contending with this diversity, by judging what risks are ‘acceptable’ to society, something which has obvious implications for institutional adaptation and behavioural responses to climate change. A common theme throughout the interviews undertaken as part of this project is the acknowledgement that climate change is still a discourse defined by western / developed nations, which invariably also influences the conceptualisation of solutions to the problem, primarily because the global community is still subject to the dominant westernnorthern hemisphere centric view of climate change issues. The manner in which climate change is addressed at international meetings raises issues of equity and inequality, as the importance of climate change in relation to other issues faced by communities worldwide is questioned. Some argue that the focus should shift to addressing the more immediately relevant issues, integrating climate change considerations within a wider sustainability discourse. Others questioned whether climate change was a smokescreen for more pressing considerations which could be exacerbating

the impacts of climate change, such as for instance energy security. Furthermore, some also argued that the discourse on climate change should be broadened to acknowledge the complexities of climate change, that is, how positive impacts may be balanced against those which may affect negatively other parts of the globe. Underlying all these considerations is the awareness that discourses of danger and hazard are politically and socially generated, thus entwined strongly with issues of power, equity and morality. As the ‘expert’ interviews highlighted ‘danger’ associated with climate change can be schematically defined in three ways: the degree of human interference with the climate; the impacts of climate change and the current costs from pre-emptive mitigation and adaptation. Danger fundamentally resides in the threat of not being able to cope with a certain degree of change, be it in the climate system or to current lifestyles. The timing and rate of change was considered fundamental: rapid climate change could be dangerous to all, whereas gradual changes are generally manageable, although dangerous to some. Most interviewees maintained that defining danger is a political debate and gives rise to inequality. Setting a target based on a thresholds approach, as article 2 would suggest, may indeed still endanger ecosystems and human populations who will be affected below a prescribed (international) threshold. In his most recent environment speech at the beginning of September, the British Prime Minister acknowledged that addressing climate change cannot be based on radical revisions of current lifestyles, which are currently unfeasible in practical terms as well as being politically unacceptable. Thus, part of the role he envisaged for those within government and outside it is “telling people what they can do that would make a difference” (Blair, 2004). The British Prime Minister recognised the difficulties inherent in bridging the perception-action gap, although as risk perceptions research has highlighted many times, there is more to changing behaviour than just providing information (see Fischhoff, 1995). Public views are affected by many factors including scales, the balance of benefits and costs, values and cultural influences, and trust in societal and individual actors. In so far as tangible benefits loom larger than losses, people are unlikely to undergo significant change to their daily routine unless pushed by external forces. Few will take action based on a moral imperative. The question is whether individuals will want to act given that climate change is a psychologically, spatially and temporally distant phenomenon. People are not likely to support initiatives addressing climate change unless they consider the issue a serious problem affecting them personally. It thus becomes evident that it is impossible to address environmental degradation without properly taking account of “local as well as global initiatives and of citizen-oriented as well as state-led programmes” (Irwin, 1995:6). Appreciating how environmental issues are interpreted at the local level through cultural practices and social world-views is becoming increasingly valued. There is an increasing recognition at various levels that situating climate change in the locality will provide the driver to initiate behavioural change (e.g. Rayner and Malone, 1997; Kates and Wilbanks, 2003 in the USA; K. Jones, pers. comm., 2004 in the UK) perhaps even through multi-purpose policies, aiming to implement mitigation and adaptation measures by integrating climate change considerations into other policy areas focusing on issues that are more relevant to people’s everyday experiences. We must be careful, however, in considering this as the panacea for increasing engagement with the issue, as there is evidence that many laypeople tend not to relate their quality of life to their individual behaviour.


Thus, there is a need to investigate how to promote that link, charge people with the responsibility to act and encourage individual behavioural change (as discussed at the workshop on dangerous climate change held at the University of East Anglia in June 2004) by, inter alia, localising climate change through: − Embedding climate change in other policy areas; − Addressing the diversity of interpretations of danger via tailored communication, via powerful icons for different groups or focusing on specific consequences for particular groups; − Enabling individuals to effect behavioural transitions, focusing on actions that individuals can easily adopt and that result in tangible benefits (see also Nicholson-Cole, 2004): people must feel the threat but also have the power and responsibility to act. This project supports previous findings which indicate that discrepancies exist between working definitions of danger by climate communities and non-experts’ perception of climate change. Both have the capacity to shape policy-making, although science is still in a privileged position. Any decision taken on managing climate change will involve making choices under conditions of uncertainty with far reaching consequences. Publics’ risk perceptions can inform this process by presenting the concerns that people associate with particular risks and their preferences for management. The future of climate change effectively rests on moral and ethical judgements, on which citizens may be called upon to decide and will be expected to take action. The findings outlined in this report reinforce the notion that climate change will not be successfully addressed by dealing with it separately from other issues that affect individuals in their day-to-day lives. Wise and proactive leadership is essential in providing guidance to remain within ‘safe’ limits whilst enabling societies worldwide to effect the cultural and behavioural transition needed to slow down the threat to survival on Earth.

Acknowledgements This project was funded by the Tyndall Centre for Climate Change Research. Additional support for this research was provided by the Leverhulme Trust as part of the Programme on Understanding Risk (RSK990021). Thank you also to all the interviewees who shared their views and mental models on danger in relation to climate change, and to Pat Cox and Ann Bostrom who provided guidance and advice on the application of the mental models approach.


BIBLIOGRAPHY Allen, M. R. and Lord, R. (2004) The blame game. Nature 432: 551-552. Arnell, N.W., Tompkins, E.L. and Adger, W.N. (2004) Eliciting information on the likelihood of rapid climate change. Draft paper presented at the International Workshop ‘Perspectives on Dangerous Climate Change’, University of East Anglia, Norwich 28-29 June 2004. Barke, R.P. and Jenkins-Smith, H.C. (1993) Politics and scientific expertise: scientists, risk perception and nuclear waste policy. Risk Analysis, 13(4):425-439. Bickerstaff, K. and Walker, G. (2002) Risk, responsibility and blame: an analysis of vocabularies of motive in air-pollution(ing) discourses. Environment and Planning A, 34:2175-2192. Bickerstaff, K., Simmons, P. and Pidgeon, N. (2004) Public perceptions of risk, science and governance: Main findings of a qualitative study of five risk cases. Unpublished working paper, Centre for Environmental Risk, University of East Anglia, Norwich. Blair, T. (2004) PM Speech on Climate Change, Tuesday, 14 September 2004, at: http://www.number10.gov.uk/output/page6333.asp (accessed 15 October 2004). Blake, J. (1999) Overcoming the ‘Value-Action Gap’ in environmental policy: tensions between national policy and local experience. Local Environment, 4(3):257-279. Bord, R.J., Fisher, A. and O’Connor, R.E. (1998) Public perceptions of global warming: United States and international perspectives. Climate Research, 11:75-84. Bord, R.J., O’Connor, R.E. and Fisher, A. (2000) In what sense does the public need to understand global climate change? Public Understanding of Science, 9:205-218. Bostrom, A., Morgan, M.G., Fischhoff, B. and Read, D. (1994) What do people know about global climate change? 1. Mental models. Risk Analysis, 14(6):959-970. Bray, D. and Von Storch, H. (1999) Climate science: an empirical example of postnormal science. Bulletin of the American Meteorological Society 80 (3):439-455. Brechin, S.R. (2003) Comparative public opinion and knowledge on global climatic change and the Kyoto Protocol: the US versus the rest of the World? International Journal of Sociology and Social Policy, 23(10):106-134. Brewer, N.T., Chapman, G.B., Brownlee, S. and Leventhal, E.A. (2002) Cholesterol control, medication adherence and illness cognition. British Journal of Health Psychology, 7(4): 433-447. Caldeira, K., Jain, A. K. and Hoffert, M.I. (2003) Climate sensitivity uncertainty and the need for energy without CO2 emission. Science, 299:2052-2054. Climate Justice Programme (2004) UNESCO danger – listing petitions presented. Legal moves to protect world heritage glaciers and coral reefs from climate change. At http://www.climatelaw.org/media/UNESCO.petitions.release (Released 17 November 2004; accessed 30 November 2004). Cohen, S. (2000) States of denial: knowing about atrocities & suffering. Polity, Cambridge. Cox, P., Niewöhner, J., Pidgeon, N., Gerrard, S., Fischhoff, B. and Riley, D. (2004) The use of mental models in chemical risk protection: developing a generic workplace methodology. Risk Analysis, 23(2):311-324. Cvetkovich, G. and Löfstedt, R. (1999) Social trust and the management of risk. Earthscan, London. DEFRA (2004) UK Climate Change Programme review: consultation launch. DEFRA News Release 504/04, available at: http://www.defra.gov.uk/news/2004/041208b.htm (12 December 2004). Dessai, S. and Hulme, M. (2003) Does climate policy need probabilities? Working Paper 34, Tyndall Centre for Climate Change Research, Norwich. Dessai, S., Adger, W.N., Hulme, M., Turnpenny, J., Köhler, J. and Warren, R. (2004) Defining and experiencing dangerous climate change - an editorial essay. Climatic Change, 64 (1-2): 11-25.

Dispensa, J.M. and Brulle, R.J. (2003) Media’s social construction of environmental issues: focus on global warming – a comparative study. International Journal of Sociology and Social Policy, 23(10):74-105. DTI and DEFRA (2003) Our energy future – creating a low carbon economy, TSO, Norwich, UK. Available at: http://www.dti.gov.uk/energy/whitepaper/ourenergyfuture.pdf (24 November 2004). Dunlap, R.E. (1998) Lay perceptions of global risk. International Sociology, 13(4):473-498. Dunlap, R.E. and Scarce, R. (1991) The polls- poll trends: environmental problems and protection. Public Opinion Quarterly, 55:651-672. ECF and PIK (2004) What is dangerous climate change? Initial results of a symposium on key vulnerable regions, climate change and article 2 of the UNFCCC. COP10 event at the EU pavilion, Buenos Aires, 14 December 2004. Eiser, J.R. (1994) Attitudes, chaos and the connectionist mind (see especially Chapter 1). Blackwell Publishers, Oxford. Elzinga, A. (1997) From Arrhenius to megascience: interplay between science and public decision making. Ambio 26: 72-80. European Community (EC) (1996) Climate Change - Council Conclusions 8518/96. European Opinion Research Group (EORG) (2002) The attitudes of Europeans towards the environment. Eurobarometer 58.0. For Directorate-General Environment; survey managed and organised by Directorate-General Press and Communication “Public Opinion Analysis”. Available at: http://europa.eu.int/comm/public_opinion/archives/ebs/ebs_180_en.pdf (20 October 2004) [carried out between 1 September and 7 October 2002; with 16067 respondents in the 15 EU Member States]. Feigenbaum, E.A. (1979) Themes and case studies in knowledge engineering, in: D. Michie (ed.) expert systems in the micro-electronic age, Edinburgh University Press, Edinburgh. Festinger, L. (1957) A theory of cognitive dissonance. Stanford University Press, Stanford. Finucane, M. L., Alhakami, A., Slovic, P., and Johnson, S. M. (2000) The affect heuristic in judgements of risks and benefits. Journal of Behavioral Decision Making, 13:1-17. Fischhoff, B. (1995) Risk perception and communication unplugged: twenty years of process. Risk Analysis, 15: 137-145. Fischhoff, B., Lichtenstein, S., Slovic, P., Derby, S.L. and Keeney, R.L. (1981) Acceptable risk. Cambridge University Press, Cambridge. Fischhoff, B., Slovic, P., Lichtenstein, S., Read, S. and Combs, B. (1978) How safe is safe enough? A psychometric study of attitudes towards technological risks and benefits. Policy Sciences, 9:127-152. Funtowicz, S.O. and Ravetz, J.R. (1993) Science for the post-normal age. Futures, 25(7):739-755. Galeotti, M., Goria, A., Mombrini, P. and Spantidaki, E. (2004) Weather impacts on natural, social and economic systems (WISE) part II: individual perceptions of climate extremes in Italy. Nota di lavoro 32.2004, FEEM, Italy. Gerrard, S. (2000) Environmental risk management. Chapter 17 (pp. 435-468) in: O’Riordan, T. (ed.) Environmental Science for Environmental Management. Second edition. Prentice Hall, Pearson Education Ltd., Harlow. Gowda, M.V.R., Fox, J.C. and Magelky, R.D. (1997) Students' understanding of climate change: insights for scientists and educators. Bulletin of the American Meteorological Society, 78(10): 2232-40. Greenwald, A.G. (1980) The totalitarian ego: fabrication and revision of personal history. American Psychologist, 35:603-618. Gugele, B., Huttunen, K. and Ritter, M. (2003) Annual European Community greenhouse gas inventory 1990-2001 and inventory report 2003. Submission to the UNFCCC Secretariat. European Environment Agency, Copenhagen, Denmark. Available at: http://reports.eea.eu.int/technical_report_2003_95/en/tech_95.pdf (19 February 2004).

Gupta, J., Berk, M. and van Asselt, H. (2003) Defining dangerous: report of the Annex 1 workshop on Article 2 of the Climate Convention (HOT WD1), Report number W-03/35. IVM – Institute for Environmental Studies, Amsterdam. Hadley Centre Met Office and DEFRA (2004) Uncertainty, risk and dangerous climate change. Hadley Centre and Met Office, Exeter. Hare, B. (1998) Fossil fuels and climate protection: the Carbon Logic. Greenpeace, London. Hargreaves, I., Lewis, J. and Speers, T. (2003) Towards a better map: science, public and the media. Report for the ESRC, Swindon. Hinchliffe, S. (1996) Helping the earth begins at home. Global Environmental Change, 6(1):53-62. Höhle, E. (2002) Global climate change as perceived by the public. Chapter 5 (pp. 115-130) in Zwick, M.M. and Renn, O. (eds.) Perception and Evaluation of Risks. Findings of the “Baden-Württemberg Risk Survey 2001”. Joint working report by the Centre of Technology Assessment in Baden-Württemberg and the University of Stuttgart, Sociology of Technologies and Environment, Germany. Houghton, J.T., Ding, Y., Griggs, D.J., Noguer, M., van der Linden, P.J., Dai, X., Maskell, K. and Johnson, C.A. (eds.) (2001) Climate Change 2001: The Scientific Basis. Contribution of Working Group 1 to the Third Assessment Report of the Intergovernmental Panel on Climate Change. Cambridge University Press, Cambridge and N.Y. ICCT (International Climate Change Taskforce) (2005) Meeting the climate challenge. Recommendations of the International Climate Change Taskforce. International Climate Change Taskforce, London, Washington and Canberra. Available at: http://www.tai.org.au/ (26 January 2005). INRA Europe – European Coordination Office (eds.) (1999) Les Europeens et l’environnement. Enquete realisee dans le cadre de l’Eurobarometre 51.1. Produced for the European Commission, DG Environment, Nuclear Safety and Civil Protection. Office for Official Publications of the European Communities, Luxembourg. Irwin, A. (1995) Citizen science. A study of people, expertise and sustainable development. Routledge, London, UK. Kasperson, R. E., and Kasperson, J. X. (1991) Hidden hazards, Chapter 1, pp. 9-28 in D. G. Mayo and R. D. Hollander (eds.) Acceptable evidence: science and values in risk management. Oxford University Press, N.Y. Kasperson, R. E., Kasperson, J. X. and Turner, B. L. (1999) Risk and criticality: trajectories of regional environmental degradation. Ambio, 28(6): 562-568. Kasperson, R. E., Renn, O., Slovic, P., Brown, H. S., Emel, J., Goble, R., Kasperson, J. X., and Ratick, S. J. (1988) The social amplification of risk: a conceptual framework. Risk Analysis, 8(2): 178-187. Kates, R.W. and Wilbanks, T.J. (2003) Making the global local: Responding to climate change concerns from the ground up. Environment, 45 (3): 12-23. Kempton, W., Boster, J.S. and Hartley, J.A. (1995) Environmental values in American culture. The MIT Press, Cambridge, Massachusetts, USA, and London, UK. King, D.A. (2004) Climate change science: adapt, mitigate or ignore? Science, 303:176-177. Kirby, A. (2004) Britons unsure of climate costs. BBC News Online, at: http://news.bbc.co.uk/1/hi/sci/tech/3934363.stm. Poll results available at http://news.bbc.co.uk/nol/shared/bsp/hi/pdfs/28_07_04_climatepoll.pdf (3 October 2004). Krayer von Kraus, M.P., Casman, E.A. and Small, M.J. (2004) Elicitation of expert judgements of uncertainty in the risk assessment of herbicide-tolerant oilseed crops. Risk Analysis, 24(6): 1515-1527. Langford, I.H. and McDonald A-L. (1997) Risk perception, health and environmental change: a multidimensional model. CSERGE Working Paper GEC 97-14, Centre for Social and Economic Research on the Global Environment, University College London and University of East Anglia, Norwich.


Leiserowitz, A. (2003) Affective imagery and risk perceptions of global climate change. Paper presented at the Society for Risk Analysis Annual Meeting, Baltimore 7-10 December 2003. Lorenzoni, I. (2003) Present Choices, Future Climates: A cross-cultural study of perceptions in Italy and in the UK. Doctoral Thesis, School of Environmental Sciences, University of East Anglia, Norwich. Lorenzoni, I., Doria, M. F., Leiserowitz, A, Poortinga, W. and Pidgeon, N. (submitted) Crossnational comparisons of image associations with 'global warming' and 'climate change' among laypeople in the United States of America and Great Britain. Submitted to Journal of Risk Research. Marris, C., Langford, I., Saunderson, T. and O’Riordan, T. (1998) A quantitative test of the cultural theory of risk perception: comparison with the psychometric paradigm. Risk Analysis, 18(2): 635-647. Marshall, G. (2001) The psychology of denial. Pp. 40-41 in Retallack, S. (ed.) Climate change. Time to act. The Ecologist Report 2001. The Ecologist Special Issues Office, Think Publishing Ltd., London. Mastrandrea, M.D. and Schneider, S.H. (2001) Integrated assessment of abrupt climatic changes. Climate Policy, 1:433-449. Mastrandrea, M.D. and Schneider, S.H. (2004) Probabilistic integrated assessment of “dangerous” climate change. Science, 304:571-575. May, R.M. and Pitts, R. (2000) Communicating the science behind global environmental change issues. Chapter 2 (pp. 15-25) in: Smith, J. (ed.) The Daily Globe: environmental change, the public and the media. Earthscan Publications Ltd., London. McDaniels, T.L., Axelrod, L.J., Cavanagh, N.S. and Slovic, P. (1997) Perception of ecological risk to water environments. Risk Analysis, 17(3):341-352. Meyer, W.B., Butzer, K.W., Downing, T.E., Turner II, B.L., Wenzel, G.W. and Wescoat, J.L. (1998) Reasoning by analogy. Chapter 4 (pp. 217-290) in: Rayner, S. and Malone, E.L. (eds.), Human choice and climate change. Volume 3: tools for policy analysis. Battelle Memorial Institute, Columbus, Ohio. Michael, M. (1996) Ignoring science: discourses of ignorance in the public understanding of science. Chapter 5 (pp. 107-125) in: Irwin, A. and Wynne, B. (eds.) Misunderstanding science? The public reconstruction of science and technology. Cambridge University Press, Cambridge. Morgan M.G. and Keith D. W. (1995) Subjective judgements by climate experts. Environmental Science and Technology, 29(10):468-476. Morgan, M.G., Fischhoff, B., Bostrom, A., Lave, L. and Atman, C.J. (1992) Communicating risk to the public. Environ. Sci. Technol., 26(11):2048-2056. Morgan. M.G., Fischhoff, B., Bostrom, A. and Atman, C.J. (2002) Risk communication. A mental models approach. Cambridge University Press, Cambridge. MORI (2001) What’s worrying Britain? 2001. A UN sponsored European-wide survey on social concerns. Survey undertaken by MORI, commissioned by the UN Population Fund (UNFPA), at: http://www.mori.com/polls/2001/pdf/unfpa.pdf (20 September 2004). MORI (2002) Public uncertainty over environmental issues. MORI Survey dated 13 June 2002. Undertaken for Scientific Alliance. Press release of survey results available at: http://www.mori.com/polls/2002/scientificalliance.shtml (28 September 2004). Murlis, J. and Davies, G. (2001) Public perception of the health impacts of climate change. Chapter 2 (pp. 50-54) in: Health effects of climate change in the UK. Department of Health, London. Nicholson-Cole, S. (2004) Imag(in)ing climate change: explore people’s visual imagery, issue salience and personal efficacy. Doctoral Thesis, School of Environmental Sciences, University of East Anglia, Norwich.


Niewöhner, J., Cox, P., Gerrard, S. and Pidgeon, N. (2004) Evaluating the efficiency of a mental models approach for improving occupational chemical risk protection. Risk Analysis, 24 (2):349-361. Nordhaus, W.D. (1994) Expert opinion on climatic change. American Scientist, 82: 45-51. Norton, A. and Leaman, J. (2004) The day after tomorrow: public opinion on climate change. MORI Social Research Institute, London. O’Connor, R., Bord, R.J., Fisher, A., Staneva, M., Kozhouharova-Zhivkova, V. and Dobreva, S. (1999) Determinants of support for climate change policies in Bulgaria and the USA. Risk Decision and Policy, 4(3):255-269. O’Neill, B.C. and Oppenheimer, M. (2002) Dangerous climate impacts and the Kyoto Protocol. Science, 296:1971-1972. Oxford English Dictionary (2003) Oxford English Dictionary. Oxford. Palutikof, J.P., Agnew, M.D., and Hoar, M.R. (2004) Public perceptions of unusually warm weather in the UK: impacts, responses and adaptations. Climate Research, 26:43-59. Palutikof, J.P., Agnew, M.D., Subak, S. and Holt, T. (1999) Weather Impacts on natural, social and economic systems (WISE): Report of the Climatic Research Unit, University of East Anglia. Pp. 79-206 in: Weather Impacts on natural, social and economic systems (WISE). Final Report for the European Commission Directorate-Research General under contract ENV4-CT97-0448. Parry, M.L., Carter, T.R. and Hulme, M. (1996) What is a dangerous climate change? Global Environmental Change, 6(1):1-6. Patwardhan, A., Schneider, S. and Semenov, S.M. (2003) Key vulnerabilities including issues relating to UNFCCC article 2. A concept paper requested by the First Scoping Meeting of the IPCC Fourth Assessment Report. Draft dated 4 Sept 2003, revisited after Second Scoping Meeting at Potsdam, Germany, 1-4 September 2003. Perrow, C. (1984) Normal Accidents. Living with high-risk technologies. Basic Books, Inc., Publishers, N.Y. Pidgeon, N. (1998) Risk assessment, risk values and the social science programme: why we do need risk perception research. Reliability Engineering and System Safety, 59:5-15. Pidgeon, N.F. and Henwood, K.L (2004) Grounded theory. Pp. 625-648 in M. Hardy and A. Bryman (eds.) Handbook of Data Analysis. Sage, London. Pidgeon, N.F. and Beattie, J. (1998) The psychology of risk and uncertainty. Pp. 289-318 in P. Calow (ed.) Handbook of Environmental Risk Assessment and Management, Blackwell Science, Oxford. Pidgeon, N.F., Hood, C., Jones, D., Turner, B. and Gibson, R. (1992) Risk perception. Ch 5 (pp. 89-134) in: Risk Analysis, Perception and Management: Report of a Royal Society Study Group, The Royal Society, London. Pidgeon, N.F., Kasperson, R.K. and Slovic, P. (2003) The social amplification of risk. Cambridge University Press, Cambridge. Pielke, R.A., Jr. and Landsea, C.W. (1998) Normalized hurricane damages in the United States, 1925-97. Weather Forecasting, 13: 351-61 Pittock, A.B., Jones, R.N. and Mitchell, C.D. (2001) Probabilities will help us plan for climate change. Nature, 413:249. Poortinga, W. and Pidgeon, N.F. (2003a) Public perceptions of risk, science and governance. Main findings of a British survey on five risk cases. Technical Report. Centre for Environmental Risk, University of East Anglia, Norwich. Poortinga, W. and Pidgeon, N. (2003b) Exploring the dimensionality of trust in risk regulation. Risk Analysis, 23: 961-972. Poortinga, W. and Pidgeon, N.F. (2004) public perceptions of genetically modified food and crops, and the GM Nation? public debate on the commercialisation of agricultural biotechnology in the UK. Technical Report of a British survey. Centre for Environmental Risk, University of East Anglia, Norwich.

Rachlinski, J.J. (2000) The psychology of global climate change. symposium: innovations in environmental policy, University of Illinois Law Review, 1:299-319. Rahmstorf, S. (1999) The thermohaline ocean circulation: a system with dangerous thresholds? Editorial Comment, Climatic Change, 46:247-256. Rayner, S. and Malone, E. (1997) Zen and the art of climate maintenance. Nature, 390:332-334. Read, D., Bostrom, A., Morgan, M.G., Fischhoff, B. and Smurs, T. (1994) What do people know about global climate change? Part 2: survey studies of educated laypeople. Risk Analysis, 14(6):971-982. Renn, O. and Rohrmann, B. (eds.) (2000) Cross-cultural risk perception. A survey of empirical studies. Technology, risk and society: an international series in Risk Analysis. Volume 13, Mumpower, J. and Renn, O. (eds.). Kluwer Academic Publishers, The Netherlands. Rohrmann, B. and Renn, O. (2000) Risk Perception Research – An Introduction. Chapter 1 (pp. 1154) in: Renn, O. and Rohrmann, B. (eds.) Cross-cultural risk perception. A survey of empirical studies. Technology, risk and society: an international series in Risk Analysis. Volume 13, Mumpower, J. and Renn, O. (eds.). Kluwer Academic Publishers, The Netherlands. Royal Commission on Environmental Pollution (RCEP) (2000) Energy - The changing climate. Summary of the Royal Commission on Environmental Pollution’s Report. Printed for The Stationery Office, CP5902, UK. Rusanen, T. (2004) Perceptions of Environmental and Health Risk Issues in Finland. Paper presented at the international workshop on “Perspectives of Dangerous Climate Change”, Norwich 28-29 June 2004. Schneider, S.H. (2001) What is ‘dangerous’ climate change? Nature, 411:17-19. Schneider, S.H. (2003) Dangers and thresholds in climate change and the implications for justice. Paper presented at “Justice in Adaptation to Climatic Change” International Seminar organised by Tyndall Centre for Climate Change Research, FIELD, IIED & CSERGE, University of East Anglia, Norwich, September 2003. Shanahan, J. and Good, J. (2000) Heat and hot air: influence of local temperature on journalists’ coverage of global warming. Public Understanding of Science, 9:285-295. Sjöberg, L. (2000) Factors in risk perception. Risk Analysis, 20(1): 1-11. Slovic, P. (2000) The perception of risk. Earthscan, London. Slovic, P., Finucane, M., Peters, E. and MacGregor, D. (2002) The affect heuristic. In T. Gilovich, D. Griffing and D. Kahneman (Eds.) Heuristics and biases: the psychology of intuitive judgement. Cambridge University Press, Cambridge. Smith, J. (ed.) (2000) The Daily Globe: environmental change, the public and the media. Earthscan Publications Ltd., London. Stamm, K.R., Clark, F. and Eblacas, P.R. (2000) Mass communication and public understanding of environmental problems: the case of global warming. Public Understanding of Science, 9:219-237. Stoll-Kleemann, S., O’Riordan, T. and Jaeger, C.C. (2001) The psychology of denial concerning climate mitigation measures: evidence from Swiss focus groups. Journal of Global Environmental Change, 11:107-117. Stott, P.A., Stone, D.A. and Allen, M.R. (2004) Human Contribution to the Heat wave of 2003. Nature, 432: 610-614. Thompson, M. and Rayner, S. (1998) Cultural discourses. Chapter 4 (pp. 265-343) in: Rayner, S. and Malone, E.L. (eds.), Human choice and climate change. Volume 1: the societal framework. Battelle Memorial Institute, Columbus, Ohio. UK Met Office (2002) The climate system. Available at: http://www.meto.gov.uk/research/hadleycentre/models/climate_system.html (19 August 2003). UNFCCC (1992) United Nations Framework Convention on Climate Change. Published for the Climate Change Secretariat by UNEP / IUC, Châtelaine, Switzerland.

Ungar, S. (2000) Knowledge, ignorance and the popular culture: climate change versus the ozone hole. Public Understanding of Science, 9:297-312. Watson, R.T. and Core Writing Team (eds.) (2001) Climate change 2001: synthesis report. Summary for policymakers. Contributions of Working Groups I, II and III to the Third Assessment Report of the Intergovernmental Panel on Climate Change. Cambridge University Press, Cambridge. Available at: http://www.ipcc.ch/pub/syreng.htm (16 January 2004). WBGU (German Advisory Council on Global Change) (2003) Climate Protection Strategies for the 21st Century: Kyoto and beyond. WBGU, Berlin, Germany. Available at: http://www.wbgu.de/wbgu_home_engl.html (17 December 2003). Wintour, P. (2004) Climate change policy review reflects failure on emissions. The Guardian online, http://www.guardian.co.uk/guardianpolitics/story/0,,1368621,00.html (12 December 2004). Witherspoon, S. (1994) The greening of Britain: romance and rationality. In: British social attitudes. The 11th report. Jowell, R., Curtice, J., Brook, L. and Ahrendt, D. (eds.). Social and Community Planning Research, Dartmouth. Wright, G. and Ayton, P. (1987) Eliciting and modelling expert knowledge. Decision Support Systems, 3: 13-26. Zwick, M.M. (2002) Descriptive findings of the Baden--Württemberg risk survey 2001. Chapter 2 (pp. 7-32) in Zwick, M.M. and Renn, O. (eds.) Perception and Evaluation of Risks. Findings of the “Baden-Württemberg Risk Survey 2001”. Joint working report by the Centre of Technology Assessment in Baden-Württemberg and the University of Stuttgart, Sociology of Technologies and Environment, Germany. Zwick, M.M. and Renn, O. (eds.) (2002) Perception and Evaluation of Risks. Findings of the “Baden-Wüttemberg Risk Survey 2001”. Joint working report by the Centre of Technology Assessment in Baden-Wüttemberg and the University of Stuttgart, Sociology of Technologies and Environment, Germany.


Appendix 1 – List of project outputs to date Conference presentations and papers Lorenzoni, I. and Pidgeon, N. (2004) Perspectives on ‘dangerous’ climate change: implications for adaptation. Presentation at the 8th Biennial Scientific Conference, International Society for Ecological Economics, Montreal, Canada, 11-14 July 2004. Lorenzoni, I. and Pidgeon, N. (2004) Interpreting ‘dangerous’ climate change: perceptions as inputs to UK risk management? Draft discussion paper for the International Workshop on ‘Dangerous’ Climate Change University of East Anglia, Norwich, 28-29 June 2004. Lorenzoni, I. and Pidgeon, N. (2004) Public views on climate change: European perspectives of a long-term risk. Paper prepared for and presented at the Workshop on Global Warming: The Psychology of Long Term Risk, Princeton University, 12 November 2004. Lorenzoni, I. and Pidgeon, N. (2005) Closing the Gap, Defining Dangers of Climate Change and Individual Behaviour. Poster presentation at the Avoiding Dangerous Climate Change conference, Hadley Centre and Met Office, Exeter, 1-3 February 2005. Lorenzoni, I., Leiserowitz, A., de Franca Doria, M.F., Poortinga, W. and Pidgeon, N. (submitted) Cross-national comparisons of image associations with 'global warming' and 'climate change' among laypeople in the United States of America and Great Britain. Presented at the Society for Risk Analysis – Europe conference (Paris, 15-17 November 2004) and submitted to Journal of Risk Research. Lorenzoni, I. and Pidgeon, N. (in prep) Public views on climate change: European perspectives of a long-term risk. For submission to Climatic Change.

International conference International Conference: Perspectives on Dangerous Climate Change, 28 and 29 June 2004; cofunded by the Tyndall Centre for Climate Change Research and the Leverhulme Trust.

Special journal issue Lorenzoni, I., Pidgeon, N. and O’Connor, R. (eds.) (in prep) Perspectives on Dangerous Climate Change. Special issue of Risk Analysis.

Report Lowe, T., Lorenzoni, I. and Pidgeon, N. (2004) Project T3.32 A Strategic Assessment of Scientific and Behavioural Perspectives on ‘Dangerous’ Climate Change. Report on Phase 3: ‘Expert’ mental models of ‘dangerous’ climate change. CER and Tyndall Centre, School of Environmental Sciences, University of East Anglia, Norwich, UK.


Appendix 2 - "Perspectives on Dangerous Climate Change" Participants Neil Adger, Tyndall Centre for Climate Change Research and CSERGE, University of East Anglia, Norwich, UK Stephanie Baldwin, Parliamentary Office of Science and Technology, Houses of Parliament, London, UK Russell Blong, Risk Frontiers / ELS, Macquarie University, Sydney, Australia Jacquie Burgess, Department of Geography, University College London, London, UK Anabela Simões de Carvalho, Departamento de Ciẽncias da Comunicação, Univerisdade do Minho, Braga, Portugal Suraje Dessai, Tyndall Centre for Climate Change Research, UEA, Norwich, UK Simon Gerrard, CRed, UEA, Norwich, UK John Handmer, Centre for Risk and Community Safety, RMIT, Melbourne, Australia Peter Höppe, GeoRisks Research Department, Munich Reinsurance Company, Germany Mike Hulme, Tyndall Centre for Climate Change Research, UEA, Norwich, UK Branden Johnson, New Jersey Department of Environmental Protection, NJ, USA Sari Kovats, Centre on Global Change and Health, London School of Hygiene and Tropical Medicine, University of London, UK Anthony Leiserowitz, Decision Research, Oregon, USA Cinzia Losenno, Global Atmosphere Division, Department for Environment, Food and Rural Affairs (DEFRA), London, UK Franziska Matthies, Tyndall Centre for Climate Change Research, UEA, Norwich, UK Cees Midden, Department of Human-Technology Interaction (MTI), Eindhoven University of Technology, The Netherlands Robert O’Connor, Decision, Risk and Management Sciences Program, Division of Social and Economic Sciences, National Science Foundation, VA, USA Saffron O’Neill, Tyndall Centre for Climate Change Research, UEA, Norwich, UK Tim O’Riordan, CER, Tyndall Centre and CSERGE, Tyndall Centre for Climate Change Research, UEA, Norwich, UK Michael Oppenheimer, Woodrow Wilson School, Princeton University, USA Judith Petts, Centre for Environmental Research and Training, School of Geography, Earth and Environmental Sciences, The University of Birmingham, UK Marc Poumadère, Inst. Symlog/Ecole Normale Supérieure, Cachan, France Steve Rayner, ESRC Science in Society Programme, University of Oxford, UK James Risbey, School of Mathematical Sciences, Monash University, Timo Rusanen, Department of Environmental Sciences, University of Kuopio, Finland John Schellnhuber, Tyndall Centre for Climate Change Research, UEA, Norwich, UK Joe Smith, Geography Discipline, Faculty of Social Sciences, The Open University, UK Linda Steg, Department of Psychology, University of Groningen, The Netherlands Emma Tompkins, Tyndall Centre for Climate Change Research, UEA, Norwich, UK Peter van der Werff, Institute for Environmental Studies (IVM), Vrije Universiteit, Amsterdam, The Netherlands Charles A.J. Vlek, Department of Psychology, University of Groningen, The Netherlands Elizabeth Wright, Tyndall Centre for Climate Change Research, UEA, Norwich, UK Farhana Yamin, Institute of Development Studies at the University of Sussex, Brighton, UK


Appendix 3 – Interview protocol: expert perceptions of ‘dangerous’ climate change [Interview length: ca. 30-40 minutes.] The aim of this project is to provide a strategic assessment of scientific and behavioural perspectives on ‘dangerous’ climate change. It comes in response to article 2, which calls for comprehensive and integrated investigations among the natural, technical and social sciences and is the UN Framework Convention on Climate Change (FCCC) major stated objective: “…. stabilization of greenhouse gas concentrations in the atmosphere at a level that would prevent dangerous anthropogenic interference with the climate system. Such a level should be achieved within a time-frame sufficient to allow ecosystems to adapt naturally to climate change, to ensure that food production is not threatened and to enable economic development to proceed in a sustainable manner”. The interview will be conducted as follows: • Ask permission to record interview • Briefly describe nature of study • Introductory questions – background / field of expertise etc. • Construction of influence diagram illustrating personal feelings about “dangerous” definitions and its influences (using Figure A3.1 as an example). • Further questions relating to “dangerous” climate change • Finally, the respondent will be shown an influence diagram, already completed by the research team, based upon relevant literature (see Figure A3.2). They will then be asked to make comments and changes as they see necessary. Questions: - Can you first tell me a little about your work and how this relates to climate change? - How would you define ‘climate change’? - Is it appropriate to discuss danger in the context of climate change? • How do you interpret ‘dangerous interference with the climate system’ (UNFCCC article 2 - above)? - Are there any limitations / difficulties in your opinion on the current framing of art 2? How would you suggest this is framed? [for our purposes, keep in mind the criteria listed below] • What criteria should any definition discourse be based upon e.g. effects/ impacts; benefits/costs; vulnerability; intra and inter-generational equity; scientific uncertainty? - Do you think a single definition of dangerous climate change is achievable / possible? o (If so) how could the current one be improved? (alternative word?) o What implications would this bring to bear on policy making? o How would you see it played out among the major decision-makers in the international / regional /local arena? - How do you view the future prospects with regard to art 2 responses i.e. definitions, and its usefulness in creating policy? - Would you like to mention anything else we might not have had the chance to talk about? - Is there someone else you feel we should contact?


Floor covering Remodel the house Height and width Of steps Lighting Use the stairs Railing

Trip on stairs
Discipline the children Sleeping habits of cat

Fall on stairs

Children’s behaviour

Toys on the floor


Figure A3.1. Example of influence diagram provided to respondents in advance of the interview: risk of tripping and falling on the stairs (adapted from Morgan et al., 2002). Ovals represent nodes (i.e. uncertain circumstances or states); the boxes indicate risk choices made by a decision-maker; the arrows indicate influence (the node at the end of an arrow’s tail has influence over the node at the arrow’s head, e.g. sunshine has influence over temperature).



AS DEFINED BY ARTICLE 2 Ecosystem Food Economy



NATURAL VARIABILITY Sun spots, volcanic eruptions, Earth / Sun orbit, etc

HUMAN INDUCED Increasing greenhouse gases (CO2; methane, NOx, etc).

PRIMARY Flood Drought Warming vs. cooling Rapid vs. incremental

WEHAB Ecological well-being Economic sustainability Equity

SECONDARY Insurance Conflict Security

Trust Control Knowledge Context Judgement


Acceptability / Tolerability POLICY


Figure A3.2. Literature based mental model on danger and climate change without influence arrows.


Appendix 4 – Meta mental model of ‘expert’ perceptions of danger in relation to climate change Part of each expert interview included the construction of a ‘mental model’ diagram (see Figure 4.1 in the text of the main report). On completion of the interviews, the research team constructed a ‘meta-model’ in an effort to represent all of the respondents’ views in a single diagram. This process enabled us to triangulate our interpretations of the experts’ interview responses with their own representative diagrams of the major issues influencing a definition of ‘Dangerous’ climate change. During the analysis phase a number of diagrams were constructed in order to simplify and distillate the information provided respondents whilst maintaining the sequence of influences suggested. Respondent diagrams were considered individually, along-side written notes taken during the interviews. The views of all interviewees are represented and, whilst many of the issues represented in the meta-model were raised by the majority of respondents, some ‘outlier’ issues are also shown. The strong linear progression of cause, effect and impact in the final diagram is influenced both by the wording of UNFCCC article 2 and the commonly held perception of human-induced climate change. Explanation of the main elements of the meta-mental model Oval 1: Cause and drivers of climate change 1. Causes and drivers of change e.g. in energy consumption, at present or in the future.
Causes + Drivers •Present


This oval considers respondents’ reference to human over-consumption and over-population going beyond certain limits. Reference was made to major economies (e.g. USA) and the future growth of emergent economies such as China. This related to changes in quantity and patterns of energy consumption. Also mentioned were; geography and land use, changes in agricultural practices and the effects that this has had and will have upon ecosystems. In particular, drivers were strongly linked to global economic, social and political inequities and the role of international competition in trade and energy. 2. Political or economic shocks arising from unsustainable progress. Shocks Some respondents believed that unforeseen economic or political shocks, unrelated to climate events (e.g. changes in energy price, availability or trade relations) were likely to lead to changes in policy or political rhetoric which could act to mitigate climate change. Thus, it was suggested that such actions would be motivated, for instance, by fears over energy security (in which case those with the largest stake i.e. politically most powerful), would be likely to dictate change; or for reasons of international harmony or relations.


3. Mitigation: measures to reduce anthropogenic interference with the climate system. Mitigation Mitigation was described by respondents in terms of: introduction of low carbon products and processes, carbon taxes, renewable energies, energy efficiency and international agreements i.e. Kyoto Protocol. The degree to which mitigation occurs is mediated by a complex set of contextual factors including; social understanding, perception and acceptance of climate change impacts; political factors (as seen above in ‘shocks’), power, sovereignty, competitiveness etc. These are mentioned in greater detail in oval 3. 4. Is it wise to pursue costly mitigation targets when adaptation will suffice?

Danger Inherent in the Cost of Mitigation

It was argued by a minority of respondents that society would not be able to cope with the social and economic changes necessary to limit anthropogenic interference with the climate system, with the actions of mitigation creating a greater danger than that of climate change itself. Some maintained that vulnerability could be increased within communities capable of adaptation by panicked measures to mitigate change. The dangers associated with mitigation for the policy and decision-makers appear most closely linked to economy, energy and competitiveness in the global market. It is clear from respondents’ comments that many countries fear a capping of growth and productivity, which, in the case of emergent economies such as Asia and China, is a social and economic necessity, whilst wealthy economies are reluctant to see their advantage diminish (e.g. USA).


5. Changes in atmospheric GHG concentrations ∆GHG Concentration

5.1 ‘Dangerous Anthropogenic Interference with the climate system’
‘Dangerous Anthropogenic Interference’

The vast majority of respondents viewed an increase in concentrations of greenhouse gasses from human activities as the main cause of ‘Dangerous Anthropogenic Interference with the Climate System’. However, opinions differed regarding the terminologies referring to the different points at which danger occurs: ‘Dangerous Anthropogenic Interference’ relates to the process of human influence on the climate system (defined in terms of GHG concentration thresholds); whilst ‘Dangerous Climate Change’ relates to the effect of the impacts resulting from significant changes to the climate and, as will be discussed, changes beyond the capabilities of human adaptation. Thus, it is possible to identify in the diagram the point at which ‘Dangerous Anthropogenic Interference’ is perceived by some respondents to occur.

6. The Climate System

6.1 Climate Change

6.2 Positive feedbacks in the climate system

Climate System

Climate Change

The climate system was seen by most respondents as the receptor of man-made greenhouse gasses. This interference, exacerbating natural variations, along with persistent perturbation and failure to return to an original state, defined climate change. There was a general consensus surrounding the possibility of negative impacts as a result of human-induced climate change, also articulated in terms of positive feedbacks or surprises in the system, irreversible changes and larger than expected changes. However, views varied regarding the timing and extent to which climate change impacts may be experienced. 7. Primary and secondary impacts resulting from climate change upon human and natural systems

Primary & Secondary Impacts On Human & Natural Systems

The majority of respondents defined ‘dangerous climate change’ in terms of significant (primary) physical impacts causing irreversible damage to Earth systems, e.g. halting of the Thermohaline Circulation (THC), collapse of the West Antarctic Ice Sheet (WAIS) and failure of Indian Monsoons. Also mentioned were the wider (secondary) environmental, political and legal implications, which may be associated with changes away from the ‘norm’. In particular, the


possible threat from environmental refugees fleeing resource conflicts and the destabilisation of socio-economic and cultural balances. 8. Cost and benefit feedbacks resulting from climate change impacts +/- feedback

Danger was also regarded by a number of natural resource experts as the result of increased climatic variability exacerbating the risks already faced by existing vulnerable communities. It was also suggested that these effects would act to amplify existing conditions of inequality and injustice. Impacts were also viewed in terms of benefits and opportunities for some.

Oval 2: Contextual influences upon our ability to adapt to climate change impacts 9. Ability to adapt to climate change Ability to Adapt 9.1 Pre-emptive adaptation 9.2 Reactive adaptation



The ability of humans and natural systems to adapt to climate change was linked by most respondents to the speed or rate at which changes occur. In the case of human beings this includes the ability to pre-empt change and plan for impacts which can greatly reduce potential risks Thus, based upon respondents’ comments we suggest the ability to adapt can be represented in two parts: (1) ability to adapt to potential threats, and (2) the ability to adapt to or respond to actual impacts. A range of socio-economic and political factors including knowledge, attitude and economic status will mediate both of these. These points are described further in ‘Context’. ‘Ability to adapt’ and ‘Context’ are shown separately to ‘Dangerous Climate Change’ in order to reflect respondent’s views that climate change impacts become dangerous only when they are beyond the capabilities of human adaptation. 10. Context: Factors which respondents believed affect the degree to which climate change is ‘dangerous’. Context Respondents mentioned a large number of factors as acting to mediate the degree to which individuals or communities are likely to be affected by climate change impacts and thus the degree to which they experience climate change as ‘dangerous’. In order to simplify our model we have grouped these factors into ‘context’. The factors mentioned most frequently are detailed below: • Ability to adapt: Influenced by knowledge and experience in dealing with social and environmental change, and range of change (amount that change ‘beyond the norm’). • Vulnerability: (also linked to resilience and adaptive capacity). Elements which have an influence upon vulnerability were suggested as; o Geography o Social Issues o Health • Relevance and influence of Climate Change in relation to other issues e.g. poverty, health and other, more immediate environmental risks (although these can all be linked to the effects of climate change).

• •

Scale: Regional definitions of danger (based on thresholds) may differ strongly on a regional basis. Inequity: e.g. less developed countries may be less resilient to the effect of Climate Change due to the nature of their economies. When initially asked for their views on climate change, many respondents suggested a metaphorical link between global issues of equity and sustainability and the plight of the global climate

Oval 3: Social and Political Influences Oval 3 articulates the importance given by respondents to the social and political influences upon a definition of ‘dangerous’ climate change. The majority of respondents maintained that climate change is symptomatic of the deeper malaise of contemporary society characterised by human and economic inequities, power struggles and issues of sustainability. For many, these factors govern both the causes of climate change and the minimisation of resultant dangerous impacts; however, the factors discussed in this oval appear to be the most subjective and open to debate. 1. Personal and societal values relating to climate change impacts.
Personal & Societal Values: Cognitive/Affective Perceptions

Personal and societal values were suggested to be import mediators of perceptions of risk relating to climate change and the ability to adapt to change: • Information and awareness • Observed changes (tangible, immediate, people care about) • Quality of life • Costs – relates to adaptation • Economic 2. The Acceptability of climate change impacts.

A small group of respondents from a human sciences background suggested that dangerous could be defined by the (un)acceptability of certain climate change impacts, which would be very much dependent upon the context of existing stress levels and variability already experienced by populations (see box 1 above). This reflects the concept of type I thresholds (Patwardhan et al., 2003; see Background section 2) defined as points which, if exceeded, would result in dangers and are therefore the subject of societal decision-making. 3. Responsibility for climate change impacts.

A question relating closely to the issue of mitigation was suggested as individual, national and international feelings of responsibility. This was described in terms of moral responsibility or obligation and from a legal point of view i.e. the ability to apportion blame for climate change impacts and to take necessary legal actions. Globally, it seems, moral obligation to act on climate change has been weak (as the Kyoto protocol has shown), with issues of international equity e.g. economic growth, trade and production suggested as barriers to mitigation action.


4. The role of energy security and politics in the mitigation of climate change.
Energy Insecurity & Politics

Epitomising what many see as an unsustainable development trajectory, issues of energy and energy security introduce additional sources of danger associated with conflict and mass social upheaval. Alternatively, it can be argued that the energy debate is being framed somewhat by the reported harmful and dangerous effects of climate change. Some interviewees argued the danger of climate change was purposefully exaggerated in order to facilitate and speed-up the transition of industrialised societies to low-carbon economies. Others linked the poor implementation of new ‘clean’ technologies to resistance from oil giants, eager to make the most of their huge investments in oil production and supply infrastructure.

5. The Role of Science in communicating and understanding climate change

Some interviewees maintained that ‘Dangerous’ climate change was defined through the impacts outlined by the IPCC. However, several interviewees referred to the subservient role that science could be subjugated into; serving the interests of the most powerful. Science could be moulded to suit the requirements of the policy makers and, that at the end of the day, those with most political influence were most likely to sway the outcomes and definitions of article 2. The role of science box is placed in two areas in which interviewees maintained that science plays a significant role in informing debate and influencing decisions on climate change. The first (oval 1), relates to the possible effects of increased greenhouse emissions upon the climate system, also denoted by an element of uncertainty. The second box (oval 2) refers to the role of science in informing both public perceptions of the dangers posed by climate change and policy makers in social, political and economic decisions. It was suggested that science is likely to play a greater role as accuracy is increased.

6. The Role of Uncertainty in communicating and understanding climate change.

The issue of uncertainty was seen as relevant in almost all areas of respondent mental models with many wishing to articulate an underlying element of uncertainty throughout the entire ‘dangerous’ debate. Whilst the pervasive nature of uncertainty surrounding climate change and its’ possible natural and human outcomes is accepted, we have chosen to represent in the meta model, areas in which respondents indicated a particular relevance to the issues concerned. Our diagram indicates that uncertainty is greatest with regard to scientific predictions and forecasts of future climate patterns and the impacts thereof. Respondents accepted that despite improvements in climate modelling, the remaining uncertainty poses a significant problem to policy makers and the public in basing decisions upon uncertain information.


The inter-disciplinary Tyndall Centre for Climate Change Research undertakes integrated research into the long-term consequences of climate change for society and into the development of sustainable responses that governments, business-leaders and decisionmakers can evaluate and implement. Achieving these objectives brings together UK climate scientists, social scientists, engineers and economists in a unique collaborative research effort. Research at the Tyndall Centre is organised into four research themes that collectively contribute to all aspects of the climate change issue: Integrating Frameworks; Decarbonising Modern Societies; Adapting to Climate Change; and Sustaining the Coastal Zone. All thematic fields address a clear problem posed to society by climate change, and will generate results to guide the strategic development of climate change mitigation and adaptation policies at local, national and global scales. The Tyndall Centre is named after the 19th century UK scientist John Tyndall, who was the first to prove the Earth’s natural greenhouse effect and suggested that slight changes in atmospheric composition could bring about climate variations. In addition, he was committed to improving the quality of science education and knowledge. The Tyndall Centre is a partnership of the following institutions: University of East Anglia UMIST Southampton Oceanography Centre University of Southampton University of Cambridge Centre for Ecology and Hydrology SPRU – Science and Technology Policy Research (University of Sussex) Institute for Transport Studies (University of Leeds) Complex Systems Management Centre (Cranfield University) Energy Research Unit (CLRC Rutherford Appleton Laboratory) The Centre is core funded by the following organisations: Natural Environmental Research Council (NERC) Economic and Social Research Council (ESRC) Engineering and Physical Sciences Research Council (EPSRC) UK Government Department of Trade and Industry (DTI) For more information, visit the Tyndall Centre Web site (www.tyndall.ac.uk) or contact: External Communications Manager Tyndall Centre for Climate Change Research University of East Anglia, Norwich NR4 7TJ, UK Phone: +44 (0) 1603 59 3906; Fax: +44 (0) 1603 59 3901 Email: tyndall@uea.ac.uk

Recent Tyndall Centre Technical Reports Tyndall Centre Technical Reports are available online at http://www.tyndall.ac.uk/publications/tech_reports/tech_reports.shtml Warren, R. (2002). A blueprint for integrated assessment of climate change, Tyndall Centre Technical Report 1. Gough, C., Shackley, S., Cannell, M.G.R. (2002). Evaluating the options for carbon sequestration, Tyndall Centre Technical Report 2. Köhler, J.H. (2002). Modelling technological change, Tyndall Centre Technical Report 3. Goodess, C.M. Osborn, T. J. and Hulme, M. (2003) The identification and evaluation of suitable scenario development methods for the estimation of future probabilities of extreme weather events, Tyndall Centre Technical Report 4. Steemers, K. (2003) Establishing research directions in sustainable building design. Tyndall Centre Technical Report 5. Macmillan, S. and Köhler, J.H., (2004) Modelling energy use in the global building stock: a pilot survey to identify available data, Tyndall Centre Technical Report 6. Adger W. N., Brooks, N., Kelly, M., Bentham, S. and Eriksen, S. (2004) New indicators of vulnerability and adaptive capacity, Tyndall Centre Technical Report 7. Skinner, I., Fergusson, M., Kröger, K., Kelly, C. and Bristow, A. (2004) Critical Issues in Decarbonising Transport, Tyndall Centre Technical Report 8 Gill, J., Watkinson, A. and Côté, I (2004). Linking sea level rise, coastal biodiversity and economic activity in Caribbean island states: towards the development of a coastal island simulator, Tyndall Centre Technical Report 9. M. N. Tsimplis (2003). Towards a vulnerability assessment for the UK coastline, Tyndall Centre Technical Report 10. Berkhout, F., Hertin, J. and Arnell, N. (2003). Business and Climate Change: Measuring and Enhancing Adaptive Capacity, Tyndall Centre Technical Report 11. Palutikof, J. and Hanson, C. (2004) Integrated assessment of the potential for change in storm activity over Europe: Implications for insurance and forestry, Tyndall Centre Technical Report 12 Jenkins, N., Strbac G. and Watson J. (2004) Connecting new and renewable energy sources to the UK electricity system, Tyndall Centre Technical Report 13 Levermore, G, Chow, D., Jones, P. and Lister, D. (2004) Accuracy of modelled extremes of temperature and climate change and its implications for the built environment in the UK, Tyndall Centre Technical Report 14 Levermore, Bristow, A., Pridmore, A., Tight, M., May, T., Berkhout, F. and Harris, M. (2004) How can we reduce carbon emissions from transport? Tyndall Centre Technical Report 15 Brown, K., Boyd, E., Corbera-Elizalde, E Adger, W. N. and Shackley, S (2004) How do CDM projects contribute to sustainable development? Tyndall Centre Technical Report 16 Few, R. (2005) Health and flood risk: A strategic assessment of adaptation processes and policies, Tyndall Centre Technical Report 17

Dutton, A. G., Bristow, A. L., Page, M. W., Kelly, C. E., Watson, J. and Tetteh, A. (2005) The Hydrogen energy economy: its long term role in greenhouse gas reduction, Tyndall Centre Technical Report 18 Shackley, S., Bray, D. and Bleda, M., (2005) Developing discourse coalitions to incorporate stakeholder perceptions and responses within the Tyndall Integrated Assessment, Tyndall Centre Technical Report 19 Dlugolecki, A. and Mansley, M. (2005) Asset management and climate change, Tyndall Centre Technical Report 20 Shepherd, D., Jickells, T., Andrews, J., Cave, R., Ledoux, L, Turner, R., Watkinson, A., Aldridge, J. Malcolm, S, Parker, R., Young, E., Nedwell, D. (2005) Integrated modelling of an estuarine environment: an assessment of managed realignment options, Tyndall Centre Technical Report 21 Abu-Sharkh, S., Li, R., Markvart, T., Ross, N., Wilson, P., Yao, R., Steemers, K., Kohler, J. and Arnold, R. (2005) Microgrids: distributed on-site generation, Tyndall Centre Technical Report 22 Anderson, D., Barker, T., Ekins, P., Green, K., Köhler, J., Warren, R., Agnolucci, P., Dewick, P., Foxon, T., Pan, H. and Winne, S. (2005) ETech+: Technology policy and technical change, a dynamic global and UK approach, Tyndall Centre Technical Report 23

Brown, K., Few, R., Tompkins, E.L., Tsimplis, M. and Sortti, (2005) Responding to climate change: inclusive and integrated coastal analysis, Tyndall Centre Technical Report 24 Timms, P., Kelly, C., and Hodgson, F., (2005) World transport scenarios project, Tyndall Centre Technical Report 25 Dearing, J.A., Plater, A.J., Richmond, N., Prandle, D. and Wolf , J. (2005) Towards a high resolution cellular model for coastal simulation (CEMCOS), Tyndall Centre Technical Report 26 Dearing, J.A., Plater, A.J., Richmond, N., Prandle, D. and Wolf , J. (2005) Towards a high resolution cellular model for coastal simulation (CEMCOS), Tyndall Centre Technical Report 26 Lorenzoni, I., Lowe, T. and Pidgeon, N. (2005) A strategic assessment of scientific and behavioural perspectives on ‘dangerous’ climate change , Tyndall Centre Technical Report 28