UC Berkeley

This dissertation investigates the ways in which societies are coming to know and govern solar geoengineering. The question at the heart of this dissertation is not whether solar geoengineering will succeed, or even whether it should, but rather what makes it --- and its governance --- imaginable. To this end, the bulk of this dissertation aimed to analyze the co-production of the evidence --- and governance assumptions --- for a sociotechnical system that does not yet exist. To do so, I draw on work in science and technology studies (STS) and political science to elucidate and analyze the political and scientific claims underpinning expert attempts to capture the public imagination and put solar geoengineering on mainstream public policy agendas. I argue that the ability to put an emerging technology on the public agenda constitutes an exercise of power, determined neither by social structures nor entrepreneurial social actors alone, and entails its own, oft-neglected, evidentiary politics.

Decades of scholarship in the interpretive social sciences demonstrates that framing and producing technoscience requires imaginative as much as technical work. Sheila Jasanoff's concept of `sociotechnical imaginaries' offers a useful point of entry into these dynamics. Sociotechnical imaginaries describe ``collectively held, institutionally stabilized, and publicly performed visions of desirable futures'' co-produced with advances in science and technology. As a theoretical concept, imaginaries help to explain why some visions of scientific and social order are co-produced, while others are not. Coupling this work with responsible research and innovation (RRI), which is concerned with the responsible steering of technoscientific developments, draws attention to the ways these imaginaries may play a vital role in the development, assessment, and governance of emerging technologies in the present, making scrutiny of their content and prospects for institutionalization urgent and timely.

Any social scientific study of solar-geoengineering-in-the-making presents challenges for the analyst, some of which are shared across `emerging technologies,' and some of which are unique to this topic, at least at this stage. For one, the supply of research on solar geoengineering --- social scientific and otherwise --- has outpaced any demand function. It is not yet a topic of research in the private sector, nor is it entangled in broader imaginaries of national identity or competitiveness, though this may change. As Steve Rayner has pointed out, solar geoengineering is at a research impasse. Moreover, the primacy of models as an evidentiary basis for contemplating solar geoengineering has contributed to its stabilization as an object of governance before we know much about what it is likely to become, or even whether it is doable at all. This has contributed to a set of early assumptions about solar geoengineering (for example, as cheap and easy, or likely to make things better or worse for specific people in specific places) that need to be revisited. In this supply-driven context, the visions of a relatively narrow set of actors --- and narrow kinds of evidence --- are forming the foundation for future policy regimes.

In Evoking equity as a rationale for solar geoengineering research? Scrutinizing emerging expert visions of equity, I examine the scientization of debates about the equity implications of solar geoengineering research. In so doing, I identify three sets of equity-related arguments advanced by sociotechnical vanguards advocating for more solar geoengineering research. The first is a call for more research as a means to shed light on the distributional outcomes of envisioned futures with and without solar geoengineering. This includes a call to reduce uncertainties inherent in scientific models examining distributional outcomes of potential deployment of solar geoengineering. Accompanying such calls is a discernible shift in the content of science itself, from more extreme to more `realistic' modeled scenarios of deployment, and from consideration of global to regional effects. The second equity-related rationale for more research is a call for comparative risk-risk assessment, underpinned by the claim that equity demands that potential risks and benefits of solar geoengineering be compared to the risks of climate change itself, especially for vulnerable populations. The third equity-related rationale for more solar geoengineering research is the invocation of the 1.5 degree aspirational goal of the Paris Agreement as requiring research on solar geoengineering, out of concern for the global poor and those most vulnerable to the consequences of climate change.

My research suggests reveals several implications of this expert-driven, outcome-oriented, and risk-based understanding of equity. First, it may suggest that more research on solar geoengineering is the only rational choice, since many of the relevant equity concerns are empirical matters, amenable to resolution through the provision of more science. Second, it sidesteps the question of whether and how diverse non-experts should have a say in whether and how such research moves forward --- even if it is to occur on their behalf, in part by assuming that climate-related preferences are knowable and quantifiable. Third, the focus on predicting the outcomes of any future deployment at this stage represents an exercise in speculative ethics, and risks ignoring alternative ways of thinking about equity and responsibility in the context of technological innovation. Finally, I suggest that further analysis should be directed toward whether the vanguard visions I explore reflect a broader shift in operationalizing equity within multilateral climate politics, with those bearing the greatest responsibility now recast as `risk managers' on behalf of the global poor and the vulnerable. I argue that those characterized as `the vulnerable' in expert discourses should regain their status as agential subjects, rather than remain undifferentiated objects in expert discourse. Empirical research suggests that publics have a set of concerns not captured in the approach to equity I analyze in this dissertation, including issues around moral responsibility, historical global injustices, the ability to be included in, and benefit from, technological development, and concerns around lack of agency and self-determination in shaping innovation pathways.

In The Politics of Climate Models for Solar Geoengineering Research, I argue that there is an oft-neglected politics of evidence around attempts to put emerging topics on the formal public agenda, which has the potential to shape future policy regimes. In this chapter, I analyze the mutual construction of solar geoengineering modeling and policy framing. Climate models have been understood as important nodes at the interface of climate science and policy, and as capable of shaping societies' understanding of, and responses to, climate change. As other scholars have pointed out, less has been said about the development of this relationship over time, which can help explain how it is that the intersection of modeling and politics takes on the form that it does.

There are at least two issues around uncertainty and representation in the use of climate models for knowledge about solar geoengineering, which raise questions at the intersection of modeling and politics. The first is that models are being used to represent technologies which do not yet exist, black-boxing the engineering in geoengineering ideas. As one interviewee stated, ``In the model, you can just make geoengineering work. You can just assume that the oceans have a higher albedo because of ocean bubbles, whether it's possible or not.'' This results in the management of the representation of a technology in models, rather than managing the development of the technology itself, eliding important near-term questions around the complexities of technology development and the structure of responsible research programs, and stabilizing solar geoengineering as an object of governance in potentially problematic ways. Secondly, there is significant debate about whether these models can usefully predict outcomes at all; uncertainties that may be less relevant to models of and for climate science and mitigation policy may become `matters of concern' when it comes to predicting or promising regarding the effects of geoengineering.

I argue that imaginaries of solar geoengineering technologies --- despite not serving current regulatory demands, and despite the non-existence of the technologies themselves (perhaps because of it) --- are engaging directly with policy needs (both current and predicted). With regard to current needs, the focus on models as proxies for actual deployment of these imagined technologies has the effect of making it seem as though societies `know' more about whether and how to develop these techniques than they do, which is resulting in debates about the management of the representation of a technology which does not yet exist. This has contributed to the current research impasse, in which ``technologists await a green light from social scientists before proceeding with research, while social scientists are limited to commenting on highly speculative ideas about how geoengineering might turn out in practice.'' In this context, policymakers are avoiding decisions regarding the advisability of a research program aimed at answering societally-relevant questions about technology development, and are content to fund indoor modeling studies. Alternatively, one might argue that the existing settlement, at least in the US, between governments, non-governmental organizations (NGOs), and scientists, in which governments seem willing to fund indoor modeling studies but accept an informal moratorium on everything else, may itself be a kind of clumsy solution, the stability of which depends on its non-articulation.

There is a broader question around displacement in the realm of climate policy raised by this research. Several scholars and commentators have raised questions about the role of imagined technologies in the present, especially since the 2015 Paris climate agreement. As Steve Rayner has pointed out, the agreement maintains the belief that global temperature targets are achievable via the inclusion of imaginary technologies, which represents a kind of `magical thinking.' Noting that the line between ambition and delusion is not always sharp, Rayner argues that the reality seems to be that the world is already likely to exceed the temperature limit agreed to absent some form of geoengineering. Despite this reality, the inclusion of climate engineering technologies in modeled scenarios has the effect of making political targets seem achievable. This is true even without any instrumental action --- and potential near-term political costs --- to policymakers when it comes to actually funding research and development on these imagined technologies, and assessing their impacts and implications.

Finally, in Climate Researchers' Views of Solar Geoengineering: Benefits, Risks, and Governance, I present the results of the first survey of climate change researchers' views of solar geoengineering research and its appropriate oversight. I argue that definitions of `expert' in emerging domains is itself a contested political category, and far from straightforward, particularly when the technologies under consideration do not yet exist. Respondents in this survey, much like surveys of general publics, report concern about the moral hazard operating at the level of political decision-making. Nevertheless, respondents generally support research on solar geoengineering, including small-scale outdoor studies --- despite both a general concern that research may result in lock-in and slippery slopes to deployment, and skepticism about the advisability of ever deploying these techniques. I find strong support for some form of novel or supplementary governance arrangement(s) for research, and a belief that scientific self-regulation is insufficient to manage risks. There seems to be less agreement, however, on particular governance approaches; I find mixed responses regarding the desirability of a `physical thresholds' approach to governing geoengineering experiments, for example.

Despite the fact that most respondents express skepticism about the desirability of future deployment, respondents tend to support more research into these techniques, both indoor and, to a lesser extent, outdoors. This might be explained by a view that research will reveal reasons not to move forward, or because of a belief that concerns about slippery slopes are overstated (although this seems less likely, given that most respondents report concern that research may result in lock-in and slippery slopes to deployment). Alternatively, a substantial number of researchers surveyed here may have an interest in scientific research moving forward in general, irrespective of its strategic aims. Respondents express skepticism about prediction and controllability when it comes to solar geoengineering deployment. It remains an open question whether a desirable future world with solar geoengineering would depend upon predicting such outcomes, although most respondents do report a belief that uncertainty in our understanding of the climate system means we should never deploy solar geoengineering.

Given low awareness of solar geoengineering, participation by a narrow set of actors --- including scientists, but also those who claim to represent the views of civil society --- can close down discussion of this imaginary technology, rather than open it up. In this way, the views of relevant but disempowered publics are assumed before most people have even heard of these ideas. It remains to be seen whether and how early visions of solar geoengineering will cohere or acquire collective stability, or whether they will be radically disrupted. My hope is that the data and analysis in this dissertation may prove useful in tracing the evolution of solar geongineering and its governance over time.