Version classiqueVersion mobile

Ethical Aspects of Climate Engineering

Gregor Betz
Sebastian Cacean

1. Introduction

Texte intégral

1.1. Geoengineering – Climate Engineering

The term “climate engineering” (CE) refers to large-scale technical interventions in the climate system with the objective of offsetting anthropogenic climate change. One distinguishes roughly between solar radiation management (SRM) and carbon dioxide removal (CDR) technologies. The main questions in dispute are whether CE technologies should (a) be researched into and (b) be deployed where appropriate.

1Man is changing the climate. Although both the exact extent and the details of anthropogenic climate change are still uncertain, the fact that we massively affect the climate system, in particular by increasing the concentration of atmospheric greenhouse gases, is not seriously argued anymore (IPCC WGI 2007). Recent observations even indicate that the IPCC climate projections rather under-than overestimate anthropogenic climate change (WBGU 2006; WBGU 2009; Allison et al. 2009). The ethical issues resulting from these long-term impacts of human action have long since been acknowledged (cf. Gardiner 2004, 2010a) and have triggered philosophical controversies concerning our responsibility towards future generations (Birnbacher 1995), the acceptability of discounting future damage and benefits (Birnbacher 2001; Ott 2004), the appropriate evaluation of the impacts of climate change (Broome 2004, 2006), the handling of risks and uncertainties (Gottschalk-Mazouz et al. 2003), and the fair distribution of emission rights (Shue 2008; Ott et al. 2008). These ethical investigations are mainly concerned with the design of and our moral obligation to implement mitigation and adaptation policies. Recent developments of the climate policy debate are, however, posing new challenges to climate ethics.

2In his 2006 article Albedo Enhancement by Stratospheric Sulfur Injection: A Contribution to Resolve a Policy Dilemma?, atmospheric chemist and Nobel laureate Paul Crutzen suggests compensating anthropogenic climate change by targeted large-scale interventions in the climate system (Crutzen 2006). Indeed, such socalled geoengineering measures had occasionally been discussed before Crutzen’s article was published (Schneider 1996; Keith 2000, 2001a); and, not least against the background of military considerations, technical interventions for deliberate weather modification had been thought about time and again in the course of the 20th century (Fleming 2010). Crutzen’s deliberate plea for research into measures such as these, however, has made the controversy a permanent issue in scientific journals and the public media. So as to differentiate between the concepts underlying the latest proposals that are aimed at offsetting anthropogenic global warming and previous proposals for large-scale restructuring of the environment, we will refer below to “climate engineering” instead of using the term “geoengineering”. Basically, there are two kinds of climate engineering measures (CE measures): Those intervening in the global carbon cycle to reduce the concentration of atmospheric carbon dioxide (Carbon Dioxide Removal, CDR) and those intervening directly in the radiation balance of the Earth, for example to increase the share of the reflected incident solar radiation (Solar Radiation Management, SRM). In September 2009, the British Royal Society published a report that introduces and examines the different proposals (Royal Society 2009). A second report by the Royal Society, published in 2011, focuses on governance schemes for CE research (Royal Society 2011). Another survey, although limited to policies influencing the radiation balance and to drafting a tentative research agenda, is given by Blackstock et al. (2009).

3Two central questions are raised by the proposal to offset anthropogenic climate change by CE measures:

[Deployment] Should CE measures be deployed? (Under which conditions would deployment be appropriate, and how is one supposed to differentiate between the various CE technologies?)
[R&D] Should CE measures be researched into? (What would be the time frame for such research? What should be the dedicated purpose of research into the relevant technologies? How do we set research priorities?)

4These two issues represent the central focus and starting point of the reconstruction of the CE controversy put forward in this study.

1.2. An Introductory Note on Ethics

This study investigates the ethical aspects of deploying and researching into climate engineering. An ethical analysis assesses the moral reasons in favour of or against taking certain action or policies. Moral reasoning appraises actions or policies from an unbiased point of view which takes the interests of all persons involved equally into account. Moral arguments differ fundamentally from economic or legal ones.

5This study investigates the ethical aspects of deploying and researching into climate engineering. What, though, is understood by “ethical aspects”? And which form can ethical expertise assume at all?

6“Ethics” is understood as the systematic theory of moral evaluation and reasoning. It is thus a scientific discipline focusing on morals as the subject of investigation. But then, what does “moral”, or “morals”, mean? The adjective “moral” describes a certain manner of evaluating, arguing, and acting. While ethical considerations are mainly academic, all of us – more or less - reason and act morally. To be more precise, an action is evaluated as morally right or wrong if evaluation is made from an unbiased point of view, i.e. from a viewpoint considering, in particular, all relevant interests of all persons that may be involved. Moral evaluations are characterised by their detachedness. Obviously, this is a rather general (or philosophically spoken “purely formal”) explication of the concept of morality: Such an abstract explication is, for instance, not capable of deciding matters of moral disagreement and does in no way provide definite moral evaluations of a specific situation or action. Moral disagreement is resolved by giving and taking concrete arguments that are related to the corresponding particular case – and not by recourse to abstract definitions of what is morally right or wrong.

7Moral reasoning and evaluation clearly differ from other manners of analysing and evaluating political measures. Legal evaluation, for example, assesses whether, in accordance with applicable laws, bans may have to be imposed on a certain technology. Apparently, however, conformity with applicable law does not imply that the relevant action is also permissible from an unbiased point of view that takes into account the interests of all persons involved. Also, geopolitical and economic analyses must be distinguished from moral evaluations. If, as a political-science analysis may reveal, some policy is – allegedly – in the national interest of a country, it is thereby shown to be beneficial from a narrow and biased perspective (which acknowledges the country’s interests) only, but not yet from an unbiased moral point of view. In the same way, economic analyses, assessing the effects of political measures on economic factors, do not reveal whether some specific measure considers all relevant interests adequately. It is necessary to distinguish between moral, on the one hand, and legal or social-science analyses and reasoning on the other hand. Nevertheless, as will be shown below, the results of the two latter may pass on, as assumptions, into moral arguments.

8This said, we may now turn to the function and nature of ethical expertise. First of all, ethical expertise is a type of scientific policy advice. Consequently, it is based on and must fulfil the same standards as natural science, legal or economic policy advice. In democratic societies, these standards include, in particular, the ideal of value-freedom. Accordingly, it is not the scientific expert but the democratically legitimised decision-maker who must determine which targets are to be pursued and realised through policy measures. Scientific policy advice must hence be free from (non-epistemic) normative assumptions. Now, ethical expertise deals inevitably with normative (namely moral) considerations, which, however, it must not assume to be correct according to the ideal of value-freedom. Therefore, ethical expertise can only set forth conditional normative recommendations (Grunwald 2008:317 et seq.). In other words, ethical advice makes explicit the basic moral assumptions without asserting them, e. g.: If the basic normative assumptions A are shared and the forecast F is accepted, then the objection O is defeated and the policy measure P should be taken; if, in contrast, the basic normative assumptions A’ are shared and the prediction F’ is accepted, then the objection O’ is defeated and the policy measure P’ should be taken. Ethical expertise strives to make as transparent as possible the manifold relations between our normative beliefs against the background of a given, more or less certain factual knowledge. It thus seeks to enable decision-makers and the public alike to grasp the relevant moral aspects of an upcoming decision and to found their choices on a coherent basis of normative beliefs.

1.3. Methods Applied

In this study, the moral reasons in favour of and against R&D into and deployment of CE methods are analysed by means of argument maps. These argument maps give a transparent overview of the CE controversy. Besides structuring the extremely complex debate, they help, moreover, to determine and evaluate the positions held by proponents.

9So as to clarify the manifold moral aspects and assumptions pertaining to CE, along with their interrelations, this study has applied the method of argument mapping (Betz 2010). Notably, the results presented below have been obtained from the following steps:

  1. Compilation of a commented bibliography of texts addressing aspects of CE.
  2. Mapping of the logico-dialectical landscape of the CE controversy in the form of argument maps, based on the respective text sources.
  3. Presentation of intermediate results on the occasion of a symposium on ethical aspects of CE (Ethische Aspekte des CE) held in Greifswald, Germany (September 2010) and during a separate workshop (November 2010) within a scoping project of the German Federal Ministry of Education and Research; revision of argument maps in the light of the feedback received.

10Without doubt, step 2 is the most comprehensive one, and the corresponding approach requires detailed explanation.

11Firstly, complex argumentation can be analysed and mapped at different levels of detail. In a relatively coarse-grained analysis, the respective arguments and theses are identified, the basic ideas of the arguments are outlined in a few sentences, and the assumed relations between the theses and arguments (support or attack) are sketched. The analysis becomes, in contrast, far more exact if the individual arguments are reconstructed as premiss-conclusion structures. Such, and only such, reconstruction can make the hidden assumptions explicit and reveal how the arguments are actually related to each other. This study sketches the entire moral controversy about CE, which is documented in the commented bibliography. By contrast, only a few selected arguments are reconstructed in detail. The argument maps presented hereunder must hence be regarded as preliminary results that can be revised more or less extensively through further detailed analyses.

12Secondly, the reconstruction of the CE controversy makes use of placeholders. As a matter of fact, there are more than one dozen different CE technologies that vary substantially regarding, for example, their respective operating cost or the extent of potential side-effects. Hence, the arguments in favour of and against one CE measure may differ radically from the arguments that pertain to another CE measure. This would suggest to reconstruct separate argument maps for each individual CE technology. Since such argument maps would share some albeit not all arguments, the already complex reconstruction below would probably have become impenetrable. Now, the placeholder method offers a solution. Instead of referring to a specific CE technology, e. g. ocean fertilization, the reconstructed arguments remain abstract and speak of CE technology T. As a consequence, full-fledged arguments that, for example, can be checked for soundness (i. e. whose premisses and conclusions may be true or false) are not obtained unless all occurrences of placeholder “T” are substituted by a concrete technology such as ocean fertilization or air capture. Arguments that may become very convincing through one such substitution may well become implausible through another. And yet, without any substitution whatsoever, the argument maps can reveal how the different types of considerations are related to each other argumentatively. Now, the placeholder method has not only been used for taking the different CE technologies (placeholder “T”) into account but also for dealing with the variety of R&D strategies and targets. In fact, there are quite different ways of conducting research into a technology: Research may, for example, prepare future deployment of a technology – in which case we would also speak of “technology development”. In contrast, technologies can also be researched into by just taking a closer look at risks and side-effects. The R&D mode is abbreviated in the arguments and theses through placeholder “R”. This second placeholder, too, requires substitution (e. g. “technology development”, “risk assessment”, etc.) before the arguments can be evaluated for soundness and theses can be checked for plausibility and truth.

13Thirdly, argument reconstructions are always interpretations. This being definitely true for coarse-grained analysis, it is also fact that, when compiling a detailed argument reconstruction, there is always room for interpretation that one can fill in one way or another. Hence, there is no such thing as the one and only correct reconstruction of a controversy. This, in turn, does not imply that argument reconstructions are completely arbitrary. In particular, the analysis carried out within this study is based on the principle of charity, i. e. it attempts to render the arguments as strong and convincing as possible. This is to ensure that reasons are not being prematurely rejected merely due to a biased interpretation. In addition, any whatsoever room for interpretation presents a much smaller impediment to controversy assessment (as compared to reconstruction) than may first be assumed. For sure, it is a matter of interpretation how accurately a proponent’s consideration is represented. Whether, however, given the reconstructed debate, a position is still coherent at all remains largely unaffected by issues of interpretation and can be determined objectively relative to the arguments identified.

14Fourthly, and finally, the reconstruction itself does not judge or value and can never be concluding or complete: It does neither decide on who is right nor on who has the final say in a debate.

15On the one hand, the reconstruction is not judging: Firstly and directly, the reconstruction itself entails only if-then claims: If certain statements are true, then certain other statements that occur in the debate must also be true. The argument map does not reveal which statements are true; it is, thus, neutral and open to different evaluations (depending on which statements one considers to be true, false or uncertain). In other words, the argument map discloses the questions to be answered in order to adopt a position in the debate and merely points out the consequences of different answers to these questions.

16Because of this, a thesis that is supported by many arguments is not necessarily true. And, by the same token, a thesis that is attacked by many arguments is by no means bound to be false. This applies equally to arguments: An attack on an argument does not imply that the very argument is definitely refuted. It may be, for example, that the attacking argument itself implies – from an evaluative perspective – absurd premisses that can easily be criticised by adding further arguments.

17Argument maps can be used to determine positions proponents may adopt and to check these positions for coherence (cf. Chapter 5). Moreover, argument maps can be important tools for coping with conflicting positions. As a matter of fact, dissent can have two causes: (i) The proponents have overlooked arguments put forward by their respective opponent. (ii) Some arguments and theses are evaluated differently. Ad (i): Should it be found that dissent arises, among other things, from not yet having considered certain arguments, the argument maps should be completed by the corresponding considerations and the positions held by the opponents should be re-evaluated thereupon. At best, dissent is dissolved right after that. Ad (ii): If there is dissent in spite of agreement on the set of relevant arguments, one may proceed as follows. Using the maps, one firstly identifies the theses and arguments mutually agreed on by the opponents. Based on this common ground, one then tries to determine or develop consensual policies. In other words: The argument maps can be used for developing robust proposals for action, i. e. proposals that are compatible with many different positions and sets of basic moral assumptions.

1.4. How to Read Argument Maps

Argument maps consist of arguments (filled-in boxes) and theses (framed boxes) which may support and attack each other (green and red arrows, respectively).

18Arguments are (i) roughly sketched in a few sentences (coarsegrained analysis) or are (ii) reconstructed as deductively valid premiss-conclusion structures (detailed analysis). The dialectical relations (attack, support) between arguments are either sketched (coarse-grained analysis) or are determined as follows (detailed analysis): An argument supports another one if the supporting argument’s conclusion figures as premiss in the supported one. And an argument attacks another one if the attacking argument’s conclusion negates a premiss of the attacked one.

19An argument map visualises these relations. It contains two types of elements: Simple sentences (e. g. central theses) are visualised as framed white boxes while arguments (containing sentences as premisses and conclusions) are shown as filled-in boxes. To ensure clear arrangement, the maps do not show all the sentences occurring in the arguments as sentences but only visualise the sentences that are of special relevance in the debate (e. g. are referred to in multiple arguments).

20If a sentence that has been visualised in the argument map at the same time occurs in an argument as premiss, this is indicated by a green arrow pointing from the sentence towards the argument. If the negation of such a sentence occurs as premiss in an argument, this is visualised by a red arrow. In the same way, the relations between arguments are represented: If an argument supports another argument, this is shown by a green arrow; a red arrow indicates that one argument attacks another.

21The outlined and reconstructed arguments and theses are consecutively numbered. In the running text, “T1”, “T2”, etc. refer to the respective theses while “A1”, “A2”, etc. point to the respective arguments. Brief designations (e.g. “T1 READINESS FOR DEPLOYMENT DESIRABLE”) of the arguments or theses may be given in addition to facilitate readability. The appendix lists all argument abbreviations and titles and compiles all argument sketches and reconstructions. Individual premisses and conclusions, numbered also, are referred to as “(n)”, where n is the corresponding sentence number if the sentence is found in the argument listed last. “(m. n)” with n being the sentence number and m the number of the argument is used otherwise; “(23.4)” hence refers to the sentence numbered 4 of argument A23.


Le texte seul est utilisable sous licence CC BY-NC-ND 4.0. Les autres éléments (illustrations, fichiers annexes importés) sont « Tous droits réservés », sauf mention contraire.


Open access


Rechercher dans OpenEdition Search

Vous allez être redirigé vers OpenEdition Search