Proposals for geoengineering the Earth's climate are prime examples of emerging or ‘upstream’ technologies, because many aspects of their effectiveness, cost and risks are yet to be researched, and in many cases are highly uncertain. This paper contributes to the emerging debate about the social acceptability of geoengineering technologies by presenting preliminary evidence on public responses to geoengineering from two of the very first UK studies of public perceptions and responses. The discussion draws upon two datasets: qualitative data (from an interview study conducted in 42 households in 2009), and quantitative data (from a subsequent nationwide survey (n=1822) of British public opinion). Unsurprisingly, baseline awareness of geoengineering was extremely low in both cases. The data from the survey indicate that, when briefly explained to people, carbon dioxide removal approaches were preferred to solar radiation management, while significant positive correlations were also found between concern about climate change and support for different geoengineering approaches. We discuss some of the wider considerations that are likely to shape public perceptions of geoengineering as it enters the media and public sphere, and conclude that, aside from technical considerations, public perceptions are likely to prove a key element influencing the debate over questions of the acceptability of geoengineering proposals.
1. Introduction: geoengineering and the importance of public perceptions
Many new technologies are embraced by society with very little public debate or controversy. At other times, they raise profound ethical dilemmas and concerns about the acceptability of risks, often accompanied by wider questioning of the ways in which new technologies will change society for good or for ill. Of course, every new or uncertain development in science and technology has the capacity to be viewed, initially at least, with some degree of ambivalence by people. But geoengineering—the deliberate large-scale manipulation of the planetary environment to counteract anthropogenic climate change —appears almost certain to generate very intense levels of controversy about the meaning, aims and pace of technological ‘progress’ itself. Looking back over the past 50 years, such disputes have arisen over the health and environmental impacts of chemicals in the environment , the risks of nuclear power and radioactive waste  and the impacts of industrial pollution on local communities . More recent times have seen complex debates over aspects of human and agricultural biotechnology  and nanotechnologies .
In this paper, we argue that geoengineering will raise many of the same issues that have already been observed with these other highly controversial developments. Some geoengineering proposals, if ever realized, would have potentially un-resolvable uncertainties over their global environmental impacts . Others raise profound ethical concerns [7,8] or complex trans-boundary legal and governance issues . As a result, we will need to understand how people's perceptions and responses to geoengineering are shaped by, and in turn shape, the public debate about this emerging technology. The study of public perceptions and beliefs can provide insights into value and ethical concerns not ordinarily considered by scientists and engineers, or factors such as trust in risk management, all of which are important for the governance and control of risks but which often prove more difficult to express in formal risk–benefit assessments. Perceptions also matter because beliefs (even when they turn out to be completely wrong) can have real consequences in the world. For example, the UK controversy over the triple vaccine for children—prompted by the publication of a single study (since discredited) claiming a link with autism—was the trigger for many parents in Britain to withdraw their children from the state vaccination programme .
We propose that geoengineering should be viewed as an emerging or ‘upstream’ technology. A technology can be said to be upstream if significant research and development has yet to commence, risks and benefits are highly uncertain, and entrenched attitudes or social representations have yet to be established . All of these apply to geoengineering at the present time . Most of the technology implicated in geoengineering has yet to be developed, let alone field tested, while some geoengineering proposals may yet turn out to be little more than imaginative science fiction. However, in the face of growing concerns about climate change, geoengineering is now a subject of serious consideration, and no small measure of contestation, among a small but increasingly vocal group of scientists and engineers around the world, including those of the American Meteorological Society  and the UK's Royal Society .
Dealing with the technical challenges associated with planetary climate control will involve formidable efforts in climate science, engineering and risk assessment. The Royal Society's geoengineering report  emphasized that a recurrent feature of many emerging technologies that have complex and uncertain risks is a paradox of control—sometimes referred to as the ‘Collingridge dilemma’ . Put simply, in the early phases of development, when some degree of influence can in principle be exercised over the path that the technology will take, uncertainties over potential risks and benefits mean that we are unable to fully specify what precautions and safeguards will actually be needed (e.g. through anticipatory risk regulation or ‘design in’ of desirable characteristics and safeguards; see also ). Such uncertainties can only be resolved through conducting research and development testing, but by then it may be much more difficult or expensive to significantly alter the path the technology has taken. In response, the Royal Society's report recommends early consideration of governance and regulatory mechanisms alongside any research effort, allowing also for flexibility and adjustment as new evidence of the risks, costs and potential unintended consequences of particular geoengineering proposals are obtained. Allowing for such flexibility is, however, partly contingent on the nature of the technology itself and partly on the institutional, social and material nature of its development; under such circumstances, questions such as ‘how reversible is society's commitment to the technology?’ also become paramount [15, p. 8].
The Royal Commission's study of nanotechnologies  highlights several indicators of inflexibility that require continual attention as a technology is developed. These include: long lead-in times from idea to application; capital intensity, such as investment in large or expensive equipment; major infrastructure requirements; closure or resistance to criticism; and hubris on the part of scientists and developers. An effective governance regime thus requires capability in applying indicators of inflexibility to decide when to intervene selectively if the technology represents a danger to the environment or human health [15, p. 8]. This approach is echoed in the Royal Society's geoengineering report  wherein concepts of ‘encapsulation’ and ‘reversibility’ are posited as useful in the process of characterizing governance requirements for different geoengineering methods. Significant in this process of ‘responsible governance’ is an imperative to understand how public attitudes to geoengineering are developing, alongside a programme of public engagement motivated by a recognized need for ‘the provision of ongoing opportunities for public and expert reflection and debate’  cited in the Royal Society's report .
Against this background, an important set of questions therefore emerge around the possibilities for public contestation and engagement with geoengineering technologies and proposals—this paper aims to explore these questions. What can we learn about the meaning of such responses from the study of perceptions of earlier technologies? What do we know so far about public reactions, and what are their implications? The aim here is not so much better communication of scientific information to people about an emerging technology and its potential risks and benefits (although this goal is, of course, an important one), and certainly not to seek to persuade people that any potential risk issues are unproblematic. Rather, it is to begin to map out where particular societal sensitivities lie, what ethical and value issues people consider important for geoengineering regulation and governance, and through this to contribute to the wider public debate and dialogue about such proposals.
2. Shaping public responses to geoengineering: lessons from earlier controversies
In this section, we outline some of the key factors known to shape perceptions of technical and environmental risks. The Royal Society's risk report  made clear that many so-called ‘crises of technology’ are often less about the technology per se, or the absolute level of risk involved, which in some cases might be quite trivial as judged from a natural or engineering sciences perspective. Rather, people's perceptions involve a range of concerns and value-based questions, which go beyond formal measurement of risk. Risk controversies at the interface of the environment and technology are rarely, if ever, solely about ‘risk’ alone.
It is important to recognize for purposes of subsequent discussion that, much as with biotechnology and nanotechnology, there is no single approach to geoengineering. Rather, a range of very different methods and approaches have been proposed. The Royal Society's working group  identified two distinct approaches: carbon dioxide removal (CDR) techniques, which remove CO2 from the atmosphere, and solar radiation management (SRM) techniques, which reflect a small percentage of the Sun's light and heat back into space. CDR techniques include proposals to sequestor carbon dioxide from the atmosphere by using giant chemical vents to ‘scrub’ the atmosphere (analogous to the carbon capture and storage techniques currently being developed for use on fossil-fuelled power stations) and plans to ‘fertilize’ the oceans by using particles of iron sulphate to stimulate algal blooms which absorb carbon dioxide. Although not strictly categorized by the Royal Society as an engineering intervention, global reforestation, avoidance of deforestation and better land use management could all be viewed as long-term methods for CDR. By contrast, SRM techniques include a variety of suggestions to deflect solar radiation, including the placement of trillions of tiny ‘sunshades’ in orbit around the Earth, the enhancement of marine cloud albedo using particles of sea salt, and the release of reflective aerosols into the stratosphere. Accordingly, and as we have also argued in the case of nanotechnologies , it makes better sense to signal this heterogeneity by referring to ‘geoengineering proposals’ as a plurality of approaches.
The range of very different technical approaches to geoengineering points to an immediate issue when exploring public attitudes—different methods (biochar and forestation, aerosols, weathering, cloud whitening, space reflectors, etc.) will almost inevitably prompt very different responses from people. We know that this occurs in many other areas of emerging technology, with the end-application in particular mattering a lot to people. Bauer & Gaskell  showed that people are often more positive about biotechnology for human health applications when compared with its agricultural uses. Likewise, in deliberative work discussing nanotechnologies with members of the public in both the UK and USA, energy applications were seen as relatively unproblematic when compared with health and human enhancement issues, with the latter felt to raise particular ethical and societal questions . Such different attitudes and responses across application domains of the same underlying technology can be explained using theoretical concepts from work on public perceptions of technological risk, where factors over and above probability and severity of harm are known to differentiate the acceptability of proposals.
A first observation is that people do not object to new technologies simply because they know little about them. If this were true, then greater risk acceptability would follow when people are provided with increased scientific knowledge about a technology or its risks (the so-called deficit model of science communication). While science literacy is clearly important for society, a simple link between increased knowledge of technology and greater acceptability is not borne out by the evidence [18,19]. If anything, as knowledge increases, this can either raise or lower concern, depending upon the person or individual technology. By contrast, factors that do seem to matter more include a range of qualitative characteristics of risks: including whether a risk is voluntarily undertaken (or not), is controllable and is known to science, holds the potential for catastrophic outcomes and is perceived to be (un)natural. All other things being equal, we find technologies more unacceptable if they are imposed upon us (rather than voluntarily undertaken), are relatively uncontrollable if things do go wrong, have potential for unknown side-effects, in the worst case could lead to catastrophic outcomes (however unlikely) and involve processes perceived to be ‘unnatural’ or to directly interfere with Nature .
A further important factor known to influence acceptability of a technology is the degree of trust or confidence that we have in its social oversight: that is, in the risk management or regulatory/governance processes [21,22]. If we distrust the process of risk regulation (or the organizations charged with that task), we are unlikely to feel fully comfortable with deployment of a technology. Such sensitivity is not without foundation, as studies of the preconditions to major engineering failures show clearly that technologies themselves are often neutral in creating risk, but that it is the people and organizations managing systems who make the critical mistakes that lead up to ‘technological’ disasters [23–25]. Although the Collingridge  dilemma of control means that risk management is unlikely ever to be simple or easy to effect, it suggests the need to establish transparent and trustworthy risk governance arrangements for geoengineering from a very early stage, both to effectively manage technical and social risk factors and to maintain public confidence.
The extent to which an issue prompts an emotional response (positive or negative affect) is also known to be strongly related to judgements of both risk and acceptability. If we feel a sense of unease or anxiety about an issue (however intangible that belief might be), this acts as a rule of thumb, or heuristic, warning us of the potential for risk. Recent thinking in psychology shows how a fully rational response to decision-making needs both analytic and affective components [26,27]. Again, geoengineering proposals, particularly those that interfere with living systems or have invisible and uncontained side-effects, are likely to elicit strong negative affective associations, whatever the science or risk assessments might say.
Risk perceptions are also sensitive to the history of controversy surrounding the issue or associated with technologies deemed similar as a context to people's attempts at making sense of an emerging technology. In this respect, any new risk question is likely to trigger responses based on reasoning around analogous issues or earlier technological controversies. Hence, people concerned about genetic modification of food at the height of public controversy in Britain in the late 1990s would frequently mention the regulatory failures underlying the BSE (‘mad cow’) disaster as a justification for not fully trusting this new technology . For the geoengineering case, we might expect (for example) that the idea of stratospheric aerosols will be interpreted by people through a generic ‘air pollution’ analogy, something with longstanding negative associations for many . In addition, media reporting, active campaigns by vocal interest groups both opposed to and supportive of a technology, involvement of distrusted individuals or organizations, alongside prominent accidents occurring with a new technology can all combine to shape risk perceptions. The dynamics of this process are known as social amplification of risk and research shows that it is typically a combination of several such factors that prompt extensive public concern .
More recent work, deriving primarily from sociological approaches to public engagement with upstream technologies [31,32], has concluded that people are often perfectly able to reason about and debate risk and technology issues with which they may have little day-to-day familiarity. While people will typically come into a research exercise (perhaps a focus group, interview or deliberative event) with very limited technical knowledge of the topic, many will engage enthusiastically with the subject by drawing upon a range of shared cultural narratives and resources regarding the ways science is located in (and shapes) society, and about both the promise and perils of scientific ‘progress’. Examples here include shared discourses about the morality of interfering with ‘natural’ systems [33,34], the trustworthiness and perceived interests of the social institutions promoting and regulating scientific developments [17,35] and, as noted earlier, stories about the genesis and consequences of past accidents and disasters [28,36,37].
Work on the sociology and psychology of emerging technologies has also demonstrated that people are particularly sensitive to un-resolved or long-term uncertainties (so-called unknown-unknowns). Such issues are often characterized as posing ‘trans-scientific questions’ [1,15]. ‘Trans-scientific’ refers to those problems in which questions arise that can be posed in the language of science as questions of fact, but in practice are unanswerable in those terms: for example, the question ‘is it safe?’ will always be a matter of potentially unverifiable judgements and assumptions about the future. For trans-scientific concerns, then, considerations extend beyond the (important) issues of risk and risk management to questions about the direction, application and control of innovation [1,15]. Empirical work here shows that the public(s) responses to uncertain and emerging technologies often exhibit an essential ambivalence towards the possibility of risks [34,38,39].
Finally, we have known for many years that the perceived benefits of a particular application of a technology are important [16,40], and relate to acceptability in one straightforward way (more benefit leads to greater acceptability), and one that is less obvious (an inequitable distribution of risks and benefits implies for some people less acceptability). Of course, whether something, like avoiding the risk of dangerous climate change, is judged to be a significant ‘benefit’ will vary between individuals, often depending upon other factors such as their world views, ideology or political opinions.
There is no reason to expect that the factors influencing people's evaluation of geoengineering will be any different from those identified with other emerging technologies. What will differ will be the ways in which specific factors become more or less salient for any particular geoengineering proposal. If people are persuaded of a need that cannot be met in other ways, and suitable regulatory and precautionary arrangements are in hand, they are likely to tolerate considerable initial uncertainty. Equally, geoengineering options that are perceived to be involuntary, uncontrollable, poorly regulated, inequitable, deemed to interfere significantly with the balance of Nature or otherwise tap into dystopian cultural narratives about technology and society are likely to attract significant public concern and resistance, whatever the benefits accrued. In sum, while there may be some generic beliefs that people bring to bear on the more abstract idea of planetary climate control, there is unlikely to be a single public attitude towards, or response to, ‘geoengineering’—but a range of different responses depending upon aspects of the specific proposal.
3. Empirical methodology
We currently know very little about how the public will respond to real (or imagined) geoengineering proposals. In order to begin to ask questions about initial responses to geoengineering, data were gathered from two linked research projects, a qualitative interview study and a major national quantitative British survey. Both studies were conducted at a very early point in the emerging public and policy debate about this issue in the UK. Both studies had been designed to explore multiple issues associated with energy, climate change and climate responses. Because the discussion of climate change provides the principal rationale (or framing) for geoengineering in policy and scientific discussions, embedding questions about geoengineering towards the end of these particular studies held the considerable advantage that the key rationale for geoengineering had already been raised with participants. However, and notwithstanding this, the data from both of these studies should be treated as provisional evidence of initial views, associations and responses.
(a) Study 1: qualitative interviews
Exploratory qualitative data on understandings of geoengineering were elicited as part of a series of in-house interviews, which were conducted with members of the public in both south-west England and south Wales during the summer of 2009. This meant that this study was conducted prior to the appearance of major public narratives and wider interpretive frames about geoengineering in the UK. The responses garnered are thus particularly interesting for understanding how public(s) would make sense of this sociotechnical issue in the absence of such wider discourses. A total of 53 individuals from 41 households1 took part in these interviews. These semi-structured interviews formed part of a wider project to understand how understandings of energy (both supply and demand) featured in people's everyday lives. To maximize the possibility that energy would have some salience for the participants, all lived within 8 miles of major electricity supply infrastructure: in the case of south-west England, the Hinkley Point nuclear power station; in south Wales, a large coal-fired power plant at East Aberthaw. Both locations are separated by the Bristol Channel, which has long been the focus of debates about potential tidal energy developments (e.g. the severn tidal barrage), making these two areas a particular nexus for past, present and future energy developments. Participants were selected to ensure a heterogeneous sample (table 1) and a diverse spread of opinion using a variety of demographic criteria, including specific locality, age, gender, household tenure, length of residence and personal connection with the local power station (through their own, or family, employment).
As the interview questions were also framed against the topics of climate change and energy security, the means to address climate change formed a natural part of the discussions as they developed. Towards the end of an interview section on climate change, the interviewer introduced geoengineering in order to gauge spontaneous responses to the term, and as a response to climate change. Interviewers asked whether participants had heard of the term geoengineering before and what the term brought to mind. Following this, interviewers then gave a broad definition (with two SRM examples, stratospheric aerosols/sulphates and mirrors in space, explained as ‘ideas’ rather than as existing technologies by which we might cool the Earth). Participants were then asked what they thought of such proposals and their views on the fact that such things were being considered. Geoengineering through carbon removal was not explored, because interviewers had already asked about conventional fossil fuel carbon capture and storage (creating the potential for confusion with air capture and storage). Approximately 5 min of each interview was dedicated to geoengineering. All interviews were digitally recorded and subsequently professionally transcribed.
(b) Study 2: quantitative national British survey
The second set of data reported here is drawn from questions included on a major British survey, again primarily designed to probe beliefs about energy and climate change, and conducted in early 2010  some nine months after the interview research. A nationally representative quota sample of the British population aged 15 years and older (i.e. England, Scotland and Wales; n=1822) were interviewed face-to-face by the social research organization Ipsos MORI in their own homes between 6 January and 26 March 2010. Computer-assisted personal interviews were conducted by professional interviewers, and on average the interviews took 30 min to complete. The survey was described to respondents as being conducted on behalf of Cardiff University about the environment and how our energy is supplied now and in the future. This meant that (as with the interview study) geoengineering was not mentioned at any point during recruitment.2 Interviews were conducted at 315 sample points (including Scottish and Welsh booster samples), selected randomly from a stratified sample of output areas sorted by government office and council area. Interviewers approached selected addresses within the sample points until set demographic quotas were reached. The findings from the overall British sample of 1822 are based on a core sample of 1528, to which additional booster samples from Scotland (109) and Wales (185) were added. The data were then weighted to the profile of the known British population on the basis of gender, age, working status, social grade and ethnicity. The full survey sample characteristics are shown in table 2.
The main survey questionnaire covered attitudes towards different energy sources, beliefs about energy security, detailed attitudes towards nuclear power, belief in and attitudes towards climate change and conventional mitigation responses, measures of environmental and other important social values, and standard sociodemographic questions . Towards the end of the survey, immediately after sections on climate change and on nuclear power as a response to climate change, four questions were asked about geoengineering, as follows.
(i) Question A. Awareness of geoengineering
The use of large-scale engineering projects designed specifically to combat global climate change is termed ‘geo-engineering’. How much, if at all, would you say you know about this subject? (five-point scale from ‘I know a great amount about geoengineering’, ‘I know a fair amount about geoengineering’, ‘I know just a little about geoengineering ’, ‘I have heard of geoengineering but know almost nothing about it’, ‘I have not heard of geoengineering’)
(ii) Question B. Support for geoengineering
Overall, to what extent would you support geoengineering approaches to tackling climate change? (‘strongly support’, ‘tend to support’, ‘neither support nor oppose’, ‘tend to oppose’, ‘strongly oppose’, ‘do not know’)
(iii) Question C. Support for solar radiation management and carbon dioxide removal
Scientists have made a number of geoengineering proposals to tackle climate change, although, currently, the costs, benefits and side-effects are uncertain. To what extent would you support or oppose the following?
C1. Developing technology to reduce global temperatures by reflecting sunlight back into space (‘strongly support’, ‘tend to support’, ‘neither support nor oppose’, ‘tend to oppose’, ‘strongly oppose’, ‘no opinion’, ‘do not know’).
C2. Developing technology to extract the gases that cause climate change from the air and store them (‘strongly support’, ‘tend to support’, ‘neither support nor oppose’, ‘tend to oppose’, ‘strongly oppose’, ‘no opinion’, ‘do not know’).
The main objective of these items was to complement, using a nationally representative quantitative sample, some of the findings emerging from the interview study, as well as to gauge potential responses to SRM and CDR when these were described to people. The survey questions were also intended to provide baseline measures, against which future changes in public responses to geoengineering in Britain might also be compared.
(a) Levels of awareness and first associations with the term geoengineering
Although it is relatively straightforward to investigate people's perceptions of more longstanding risk issues such as nuclear power or climate change, emerging or ‘upstream’ technologies bring particular methodological challenges for the social sciences. Few people in the upstream phase will have heard of the technology, let alone have any understanding or knowledge of potential applications on which to base a judgement. Few public representations in the media, policy or mainstream science exist either. In this context, one of the things that will anchor the development of early public understandings of geoengineering will be the set of images and concepts that people spontaneously associate with the term. We expected to find very low levels of awareness of geoengineering in both the interviews and survey, a finding also reported by the Royal Society . In the early phases of the in-house interviews (i.e. before the interviewers introduced the topic), no participant had spontaneously mentioned geoengineering as a potential means to combat climate change. The initial responses to the term, when it was introduced, showed many of the interviewees to be uncertain, with almost complete lack of awareness of the meaning of the term. Consistent associations were, for some people, with the broad category of technology, and for some with geothermal energy. Typical comments would be: ‘Really do not know to be honest. Do not know. Sounds technical, really technical’ (Adam) or ‘Is that getting heat out of the rocks?’ (Cathy). Other participants likened the term to some form of geotechnical engineering, as in ‘Geoengineering? Something to do with rocks and strata and big Earth movements and stuff?’ (Alan).
It is no surprise to find these the first responses, given the similarities between the terms geoengineering and the (relatively) more common concepts of geothermal energy and geotechnical engineering. Some people will also have heard about geothermal and ‘ground source’ in the context of renewable energy systems, and be aware that they are being promoted to help tackle carbon emissions and climate change (something which, of course, was the main focus of the interview). As Mercer et al.  point out, the existence of these associations as the basis of an initial mental model (without further information being provided) means that the use of this term when explaining geoengineering to people holds significant potential for miscommunication.
The survey results confirmed this almost total lack of awareness of the term among the national sample too. Three-quarters (75%) of the sample respondents had either ‘not heard of geoengineering’ or knew ‘almost nothing about it’, when described to them as ‘large-scale engineering projects designed to tackle climate change’. Only 6 per cent of respondents reported that they felt they knew ‘a fair amount’ and 1 per cent a great amount when geoengineering was described in this way. The latter should be regarded as upper estimates of awareness, given the interview findings, because we cannot rule out a subset of respondents associating the survey question with geothermal energy or geotechnical engineering.
(b) Support for geoengineering
Tables 3 and 4 report the raw percentages for overall support for geoengineering, and support for developing SRM and CDR approaches, respectively, also broken down by gender, socioeconomic status and education. It is important in interpreting these tables to take account of the fact that there are very high levels of the ‘do not know’ response, and that levels of the ‘do not know’ response also differ systematically for different demographic groups. In particular, more women responded ‘do not know’ than men, as did people with less formal education, as did those in manual (C2DE) compared with more professional (ABC1) occupations. Conversely, there is greater support for geoengineering in general (QB) among men, those with more formal education and those with professional occupations. A similar pattern to that with the general question is observed with CDR (QC1), whereas with SRM (QC2) there appears to be a less systematic relationship between support and the demographic variables. Overall, there is greater support for CDR (48.2% tend to or strongly support) than for SRM (40.6%)
In order to investigate whether levels of support were systematically related to demographic variables and other beliefs included on the survey (e.g. concerns about climate change, and pro-environmental beliefs), we also conducted several regression analyses. The dependent variables in these regression equations were the two support measures for SRM and CDR, respectively (QC1 and QC2). These regression analyses are shown in table 4, and control for the intercorrelations between variables. This analysis reveals that the profiles of support for SRM and CDR techniques were broadly similar, with sociodemographic variables showing little additional systematic relationship when compared with other more influential variables (in contrast to the simple analysis of means shown in tables 3 and 4). In particular, concern about climate change was positively associated with support for both variants of geoengineering (p<0.001 in both cases) as was belief in the human causes of climate change. That is, the more concerned respondents are about climate change, and the more they feel it is caused by humans rather than a natural phenomenon, the more they support developing both SRM and CDR (table 5). Self-reported awareness about geoengineering was also found to positively predict, albeit weakly, CDR support (p<0.05)—that is, the more people reported that they knew about geoengineering in general, the more likely they were to support the use of CDR techniques. By contrast, awareness was a negative predictor (p<0.05) of support for SRM—that is, the more people reported they knew, the less they supported the development of SRM techniques.
Given the very low levels of awareness (which, with the measure used here, can also be interpreted in part as self-reported knowledge), one would not necessarily expect such beliefs to remain constant as more information about geoengineering becomes available in the media, popular science and public policy domains. Note also that the effects in the regression are small, with a very limited amount of variance accounted for overall (approx. 5% of total variance explained). This means that, at the point in time of this survey, the level of ‘noise’ present in the data is far larger than the weak trends seen. As noted at several points mentioned earlier, this cautions against treating the current findings as more than provisional, exploratory measures of belief.
(c) Interviewee responses to examples of solar radiation management
The interview study also provided a broad definition of geoengineering, accompanied by descriptions of stratospheric aerosols/sulphates and mirrors in space as two examples of SRM. Participants were then asked what they thought of such proposals and their views on the fact that such things were being considered. Responses to the follow-up definition and examples were more varied than with the initial awareness question, and box 1 illustrates some of the responses obtained here.
Responses to the idea of geoengineering and SRM examples in interviews.
Responses to the idea of geoengineering and SRM examples in interviews.
We do not present a fully developed thematic analysis of all of the discourses present in these data; and quotes are provided to exemplify the broad range of responses by participants, from the curious to the incredulous/horrified. As the quotations in box 1 illustrate, some interviewees went on to liken the description given of geoengineering to science fiction, space weaponry and earlier experiments in weather modification. Note also the ethical concern of humans and scientists ‘messing’ with Nature, for which many of the participants expressed a spontaneous, affective dislike. Some also felt that, because we had already made a mess of the climate, then in effect ‘two wrongs do not make a right’, implying that pursuing geoengineering could in this way be viewed as a moral transgression. Interviewees also raised associations with earlier disasters (the 1952 Lynmouth flood in Devon), or analogous stigmatized technologies (‘chemicals’). Note also that some of the responses in box 1 pose questions about whether geoengineering would bring side-effects that cannot be anticipated or controlled, a common response found with discussion of other emerging technologies (and a marker of ambivalence and negative affect in risk discourses).
This paper has brought together background theory on risk perceptions of emerging technologies, with novel data from two studies to provide an initial view of public perceptions of geoengineering in the UK. Although both of our empirical studies are exploratory in nature, the analysis we present is strengthened because of the ability to triangulate across the two very different methodological approaches and datasets, obtained from UK participants within a nine-month period. The study therefore provides some of the very first coordinated intelligence on baseline views of geoengineering well ‘upstream’ of any possible development, and before any major public debate about the topic has taken place or wider public discourses (including in the media) have had time to develop. It is important that this research is followed up with companion studies, in order to evaluate how public views develop further, if and when geoengineering does enter more directly into the mainstream research, public policy and media domains.
A first finding from both interviews and survey is that awareness of the term ‘geoengineering’ is currently extremely low, a finding in line with other early survey studies [1,43]. This has implications for both the interpretation of the current findings and the design of future geoengineering perception studies. Under circumstances of low familiarity, we know that many people will offer responses to almost any survey question, even when they know very little. For example, people report in surveys that they are unfamiliar with nanotechnologies while in the same breath state that they believe they will contribute positively to society in the long run . Such responses are not necessarily contradictory or ‘irrational’, but reflect a bracketing of the term ‘nanotechnology’ with the more generic faith held in Western societies that many new technologies do eventually improve our own or others' lives . In theoretical terms, this process is known as ‘preference construction’  where people, faced with a survey question about a topic they are initially unfamiliar with, arrive at a response drawing upon a range of their existing beliefs and values, their instant affective responses, and a range of inferences about what the question being posed to them might mean or be analogous to. Such responses are a product of the person's attempts to make sense of an unfamiliar issue, informed by what they already know and their inferences from the wider context and cues within which a question is asked. The associations found in the interview study between geoengineering and geothermal/geotechnical engineering are a good example of this and indicate, as concluded also by Mercer et al. , that communicating the term ‘geoengineering’ at this point in time may invoke misinterpretations for some people.
Commenting upon biotechnology surveys, Fischhoff & Fischhoff  observe that, under circumstances of very low familiarity, highly structured questions inevitably leave respondents guessing about the meanings of the questions and investigators guessing about the meanings of the answers, and that, as a result, interpreting findings from such surveys must be done with considerable caution. Preference construction theory also predicts that, under conditions of low familiarity, responses will be highly sensitive to question wording (something known as framing effects). For example, the survey by Mercer et al.  found very low familiarity with the term ‘geoengineering’ but much higher recognition of the term ‘climate engineering’, which people are able to associate with the more common idea of ‘weather modification’. Such findings suggest that it will be important in future research to find ways to elicit more ‘informed’ or stable preferences for geoengineering, for example by providing sufficient balanced information as an integral part of any survey materials  or by taking individuals through more extensive deliberative exercises .
More generally, our survey responses indicate relatively positive views among the national sample for both SRM and carbon removal approaches (CRD) to geoengineering, but with high levels of ‘do not know’ responses to each question. Again, these findings have to be interpreted with very great care, and in terms of the context of the survey (climate change and energy) and the precise question wording that was used. The CDR and SRM questions were prefaced with the statement that ‘Scientists have made a number of geoengineering proposals to tackle climate change, although currently the costs, benefits and side-effects are uncertain’. We know from research on attitudes towards climate change over the past two decades that the majority of people in the UK are very concerned about climate change, but view it as a ‘psychologically distant’ problem that they themselves cannot readily influence . As a result, people express a strong desire for governments and other external actors to provide solutions to the problem [50,51]. Therefore, offering people an apparent technological ‘fix’ for climate change at this current point in time frames the question in terms of a wider set of cultural beliefs and hopes, and would indeed be expected to garner relatively positive responses among respondents concerned about this issue, irrespective of their views about the merits or otherwise of particular geoengineering interventions. This interpretation is supported by perhaps the most interesting findings from the regression analyses: that higher concern for climate change, belief in its anthropogenic origins and a pro-environmental orientation were all associated with greater support for both SRM and CDR. All of these are factors that differentiate individuals who are convinced about climate change from those who are more sceptical .
A useful parallel for understanding how beliefs about climate change and geoengineering might be related comes from work on attitudes to energy options and the environment. Here we know that, despite longstanding reservations that many people hold about nuclear power, far more support for nuclear is expressed when it is framed in surveys as a ‘solution’ to climate change [53,54]. The research also shows that such beliefs should be treated as ambivalent at best (that is, express a ‘reluctant’ rather than absolute acceptance) because the question pitches a technology with many negative associations (deadly radiation, the Chernobyl disaster, Fukushima, etc.), against the possibly greater risks posed by global climate change. This suggests a need for further detailed empirical work on the ways that different question framings will alter responses to different geoengineering proposals, and their relationship with people's prior beliefs about climate change, the environment and other values that they hold . However, this relationship may ultimately prove to be more complex than it seems on the surface (see also ). One could plausibly find some environmentalists who oppose geoengineering proposals because they believe this will interfere unduly with Nature, while other environmentalists will be prepared to accept the same technology if it can realistically offer a counter to climate change. Equally some climate sceptics might support geoengineering as an insurance policy that would allow Western high-consumption lifestyles to continue unabated (the ‘moral hazard’ argument ), while others might view it as an unnecessary expense because it addresses a problem they believe does not exist.
The survey also shows relatively greater support for CDR than for SRM. Again this finding should be interpreted with due caution, and in relation to the precise question phrasing used, but it is consistent with qualitative results from the major dialogue exercise Experiment Earth conducted in early 2010 for the UK Natural Environment Research Council . Broadly speaking, Experiment Earth participants preferred methods for CDR over SRM techniques, although views also altered as the dialogue developed. In particular, participants in the dialogue felt that CDR held the advantage of addressing the underlying problem of greenhouse gas emissions, and that some CRD approaches (e.g. biological capture) were also more in keeping with the environment's ‘natural balance’. The report of Experiment Earth also discusses how the concept of interfering with Nature was a key theme underlying participants' concerns about some SRM geoengineering approaches such as stratospheric aerosols, a finding obtained also in a subsequent deliberative study with a cross section of members of the UK general public .
The data from the current interviews also point to ‘naturalness’ as a significant theme. We observed that participants referred spontaneously to the problem of interfering with Nature when the two SRM approaches (aerosols and space reflectors) were explained to them. This in turn suggests that initial public interpretations of geoengineering—as with many emerging technologies before it—are likely to be framed by much wider concerns and discourses about the complex relationship between Nature, society and scientists' visions of technological ‘progress’ [34,58]. The issue of contradicting views of ‘naturalness’ is likely to prove particularly problematic for some geoengineering proposals. One might dismiss such an objection simply on the grounds that people have, intentionally or unintentionally, shaped the natural world for millennia, and that, philosophically speaking, geoengineering is no different in this regard. However, work on public opposition to biotechnology has demonstrated how people's unease about the unnaturalness of the technology was bound up with a wider storyline about not ‘pushing Nature beyond its limits’, fuelled by concerns that scientists could not anticipate the long-term consequences of their actions on ecosystems, human health and society  and the degree to which scientists' visions of increased technological control over both Nature and human society could ever be ethically acceptable. It is not at all difficult to see how some of the current proposals for geoengineering (e.g. iron fertilization, sulphate aerosols) might indeed have characteristics that evoke, and are subsequently interpreted, through such cultural narratives, something that may also help us in part to explain their current sensitivity among many scientists, external commentators and environmental groups. As a result, providing analytic-deliberative spaces that enable people to think through different framings of geoengineering proposals, and their wider social and environmental implications, probably sets a greater current research priority than do further surveys of public opinion .
The paper has begun to map public awareness and some of the social and ethical considerations that people bring to bear on their judgements of the acceptability of geoengineering. We reiterate that, aside from technical considerations, public perceptions are likely to prove a key element influencing the debate over questions of geoengineering and its societal acceptability.
This research was supported through grants from the Economic and Social Research Council (RES-062-23-1134), the Leverhulme Trust (F/00 407/AG), and the US National Science Foundation (co-operative agreement SES 0938099). Additional support was also provided through the Engineering and Physical Sciences Research Council (Integrated Assessment of Geoengineering Proposals project). A.S. is additionally supported by the Horizon Digital Economy Research, RCUK grant (EP/G065802/1). We thank Dan Venables and the Ipsos-MORI team for assistance with the national survey, and Naomi Vaughan for advice on the science of geoengineering. Two reviewers also provided extensive helpful comments to improve the manuscript.
One contribution of 12 to a Discussion Meeting Issue ‘Geoengineering: taking control of our planet's climate?’.
↵1 There are more participants than households. Although participants were recruited on an individual basis, on a few occasions a partner or other family member would also join in the interview.
- This journal is © 2012 The Royal Society