Climate Justice and Non-State Actors Corporations, Regions, Cities, and Individuals 2019055890, 2019055891, 9780367368906, 9780367368920, 9780429351877

Citation preview

CLIMATE JUSTICE AND NON-STATE ACTORS

This book investigates the relationship between non-state actors and climate justice from a philosophical perspective. The climate justice literature remains largely focused upon the rights and duties of states. Yet, for decades, states have failed to take adequate steps to address climate change. This has led some to suggest that, if severe climate change and its attendant harms are to be avoided, non-state actors are going to have to step into the breach. This collection represents the first attempt to systematically examine the climate duties of the most significant non-state actors – corporations, subnational political communities, and individuals. Targeted at academic philosophers working on climate justice, this collection will also be of great interest to students and scholars of global justice, applied ethics, political philosophy, and environmental humanities. Jeremy Moss is a Professor of Political Philosophy at the University of New South Wales, Sydney. His main research interests are in political philosophy and applied philosophy. Current projects include: climate justice, the ethics of renewable energy as well as the ethical issues associated with climate transitions. He is Director of the Practical Justice Initiative and leads the Climate Justice Research programme at UNSW. Moss has published several books including: Reassessing Egalitarianism, Climate Change and Social Justice, and Climate Change and Justice (Cambridge University Press). Lachlan Umbers is Lecturer in Philosophy at the University of Western Australia, Perth. He works primarily in moral and political philosophy, with a particular focus upon issues in democratic theory and climate justice. His work has been published in journals such as the British Journal of Political Science, Philosophical Studies, Political Studies, and the European Journal of Political Theory.

ROUTLEDGE ENVIRONMENTAL ETHICS Series Editor: Benjamin Hale, University of Colorado, Boulder

The Routledge Environmental Ethics series aims to gather novel work on questions that fall at the intersection of the normative and the practical, with an eye toward conceptual issues that bear on environmental policy and environmental science. Recognizing the growing need for input from academic philosophers and political theorists in the broader environmental discourse, but also acknowledging that moral responsibilities for environmental alteration cannot be understood without rooting themselves in the practical and descriptive details, this series aims to unify contributions from within the environmental literature. Books in this series can cover topics in a range of environmental contexts, including individual responsibility for climate change, conceptual matters affecting climate policy, the moral underpinnings of endangered species protection, complications facing wildlife management, the nature of extinction, the ethics of reintroduction and assisted migration, reparative responsibilities to restore, among many others. Sustainable Action and Motivation: Pathways for Individuals, Institutions and Humanity Roland Mees Climate Justice and Non-State Actors: Corporations, Regions, Cities, and Individuals Edited by Jeremy Moss and Lachlan Umbers For more information on the series, please visit https://www.routledge.com/RoutledgeEnvironmental-Ethics/book-series/ENVE

CLIMATE JUSTICE AND NON-STATE ACTORS Corporations, Regions, Cities, and Individuals

Edited by Jeremy Moss and Lachlan Umbers

First published 2020 by Routledge 2 Park Square, Milton Park, Abingdon, Oxon OX14 4RN and by Routledge 52 Vanderbilt Avenue, New York, NY 10017 Routledge is an imprint of the Taylor & Francis Group, an informa business © 2020 selection and editorial matter, Jeremy Moss and Lachlan Umbers; individual chapters, the contributors The right of Jeremy Moss and Lachlan Umbers to be identified as the authors of the editorial material, and of the authors for their individual chapters, has been asserted in accordance with sections 77 and 78 of the Copyright, Designs and Patents Act 1988. All rights reserved. No part of this book may be reprinted or reproduced or utilised in any form or by any electronic, mechanical, or other means, now known or hereafter invented, including photocopying and recording, or in any information storage or retrieval system, without permission in writing from the publishers. Trademark notice: Product or corporate names may be trademarks or registered trademarks, and are used only for identification and explanation without intent to infringe. British Library Cataloguing-in-Publication Data A catalogue record for this book is available from the British Library Library of Congress Cataloging-in-Publication Data Names: Moss, Jeremy, editor. | Umbers, Lachlan, editor. Title: Climate justice and non-state actors / edited by Jeremy Moss and Lachlan Umbers. Description: Abingdon, Oxon ; New York, NY : Routledge, 2020. | Series: Routledge environmental ethics | Includes bibliographical references and index. Identifiers: LCCN 2019055890 (print) | LCCN 2019055891 (ebook) | Subjects: LCSH: Environmental responsibility--Social aspects. | Environmental justice. | Business enterprises--Environmental aspects. | Environmental responsibility. | Environmental policy--Citizen participation. Classification: LCC GE195.7 .C55 2020 (print) | LCC GE195.7 (ebook) | DDC 363.7/057--dc23 LC record available at https://lccn.loc.gov/2019055890 LC ebook record available at https://lccn.loc.gov/2019055891 ISBN: 978-0-367-36890-6 (hbk) ISBN: 978-0-367-36892-0 (pbk) ISBN: 978-0-429-35187-7 (ebk) Typeset in Bembo by Taylor & Francis Books

CONTENTS

List of contributors Introduction Lachlan Umbers (UWA) and Jeremy Moss (UNSW)

vii 1

1

Levels of climate action Garrett Cullity (Adelaide)

12

2

Sub-national climate duties: Addressing three challenges Lachlan Umbers (UWA)

29

3

Carbon majors and corporate responsibility for climate change Jeremy Moss (UNSW)

45

4

Sectoral responsibility for climate justice: Is aviation exceptionalism defensible? Elisabeth Ellis (Otago)

5

Corporations’ duties in a changing climate Stephanie Collins (ACU)

6

Individual climate justice duties: The cooperative promotional model & its challenges Elizabeth Cripps (Edinburgh)

61 84

101

vi Contents

7

8

Are we morally required to reduce our carbon footprint independently of what others do? Susanne Burri (LSE)

118

Right-levelling indeterminacy: Environmental problems, non-state actors, and the global economic market Benjamin Hale (CU Boulder)

135

Index

152

CONTRIBUTORS

Susanne Burri is an Assistant Professor in the Department of Philosophy, Logic & Scientific Method at the London School of Economics and Political Sciencee (LSE). Her research interests are primarily within normative ethics and include questions about permissible harming, moral decision-making under risk and uncertainty, and problems about the value of life and the badness of death. Susanne Burri holds a B.A. in Economics from the University of St. Gallen in Switzerland, a MSc in Economics & Philosophy from the LSE, and a PhD in Philosophy from the LSE. Stephanie Collins is an Associate Professor of Philosophy at the Dianoia Institute of Philosophy at the Australian Catholic University. She primarily works on the philosophy of human groups – that is, on how we should conceptualise groups’ ontological status, mental and epistemic capacities, agency, responsibility, and duties. She also has interests in moral demandingness, associative duties, human rights, and the nature of individuals’ obligations in the face of global catastrophes and moral collective action problems. Her first book, The Core of Care Ethics, was published by Palgrave Macmillan in 2015. Her second book, Group Duties: Their Existence and Their Implications for Individuals, was published by Oxford University Press in 2019. Her work on collective responsibility has been published by journals such as Journal of Philosophy, Australasian Journal of Philosophy, Canadian Journal of Philosophy, Political Studies, and Journal of Business Ethics. Elizabeth Cripps is a Senior Lecturer in Political Theory at the University of Edinburgh and the author of Climate Change and the Moral Agent: Individual Duties in an Interdependent World (OUP 2013). She has published widely on topics in climate ethics and justice, including collective responsibility and individual duties, population and justice, justice to non-human animals, and parental responsibilities. Her current research focuses on two areas: the intersection of global justice, population justice and climate justice, and parental duties in the context of global climate change. Elizabeth has a first degree from the

viii Contributors

University of Oxford and a PhD from University College London and is a former British Academy Postdoctoral Fellow. She serves on the Nuffield Council of Bioethics Working Group on Genome Editing (Farmed Animals), is a member of the Demography, Ethics and Public Policy Network, and is a consultant editor of the British Journal of Politics and International Relations. Garrett Cullity is Hughes Professor of Philosophy at the University of Adelaide. He taught previously at the universities of Oxford and St Andrews. A moral philosopher whose work ranges across the theoretical and applied parts of the discipline, his work has addressed topics including the nature of moral judgement, the sources of moral knowledge, the relationship between reasons for action and rationality, the content of the moral virtues, the moral emotions, human rights, fairness and collective action, and the ethical issues surrounding international aid, climate change, and our activities as consumers. He is the author of The Moral Demands of Affluence (OUP, 2004) and Concern, Respect and Cooperation (OUP, 2018), a co-editor of Ethics and Practical Reason (OUP, 1997), and an Associate Editor of the journal Philosophy and Public Affairs. Elisabeth Ellis is a Professor of Philosophy and Politics and Director of the programme in Philosophy, Politics and Economics at the University of Otago. Her work in environmental political theory investigates how we can make policy decisions that serve our collective interests in flourishing now and in the future. This stream of work includes papers in environmental democracy, the collective ethics of flying, the value of biodiversity losses, climate adaptation justice, and species extinction. She has written two books and edited a volume of essays on the political philosophy of Immanuel Kant; other work in the history of political thought includes essays on Hobbes, social contract theory, and modernity. Elisabeth’s work has been supported by the Institute for Advanced Study (Princeton), the National Endowment for the Humanities, the Alexander von Humboldt Foundation, the Andrew W. Mellon Foundation, the Deutscher Akademischer Austauschdienst, and New Zealand’s Deep South National Science Challenge. Benjamin Hale is an Associate Professor in the Environmental Studies Program and the Philosophy Department at the University of Colorado, Boulder. He works primarily in the area of environmental ethics and policy, though his theoretical interests span much larger concerns in applied ethics, normative ethics, and even metaethics. From 2013 to 2019 he was Vice-President and then President of the International Society for Environmental Ethics. Among other credits, he is co-editor of the journal Ethics, Policy & Environment and author of The Wild and the Wicked: On Nature and Human Nature (MIT Press, 2016). Jeremy Moss is a Professor of Political Philosophy at the University of New South Wales, Sydney. His main research interests are in political philosophy and applied philosophy. Current projects include: climate justice, the ethics of renewable energy as well as the ethical issues associated with climate transitions. He is Director of the Practical Justice Initiative and leads the Climate Justice Research programme at UNSW. Moss has

Contributors ix

published several books including: Reassessing Egalitarianism, Climate Change and Social Justice, and Climate Change and Justice (Cambridge University Press). Lachlan Umbers is a Lecturer in Philosophy at the University of Western Australia. He works primarily in moral and political philosophy, with a particular focus upon issues in democratic theory and climate justice. His work has been published in journals such as the British Journal of Political Science, Philosophical Studies, Political Studies, and the European Journal of Political Theory.

INTRODUCTION Lachlan Umbers (UWA) and Jeremy Moss (UNSW)

The 2015–2016 Paris Climate Agreement was hailed worldwide as amongst the most significant developments in the history of the global response to climate change. Article 2a commits each of the 186 states party to the agreement to, inter alia, the twin aims of constraining “…the increase in global average temperature to well below 2°C above pre-industrial levels”, and pursuing “…efforts to limit the temperature increase to 1.5°C above pre-industrial levels…”. Under the agreement, states made a series of pledges to reduce their emissions in line with these objectives, with the intention that those pledges would ratchet up over time, beginning in 2020. Unfortunately, however, several studies have concluded that the emissions-reduction pledges made by the nations party to the Paris agreement are not sufficient to contain global warming to 1.5–2°C (e.g. Rogelj et al., 2016). And, even if they were, no major industrialised nation is presently on-track to meet its pledges (Victor et al., 2017). For example, President Obama pledged to cut the United States’ emissions to 26–28% below 2005 levels by 2025. Yet the Obama administration’s climate policies were only ever likely to cut emissions by 15–19% (Victor et al., 2017, p. 26). President Trump announced his intention to withdraw the United States from the Paris Climate Agreement entirely in June 2017, and his administration is in the process of dismantling Obama-era climate regulations. Recent projections, then, have suggested that temperatures are likely to rise by between 2°C and 4.9°C above the pre-industrial average by 2100 (Raftery et al., 2017), risking many very serious harms (IPCC, 2014). Despite the significance of the Paris Agreement, then, states are simply not doing enough to counter the threat of climate change. In response to this, some have suggested that, if severe global warming and its effects are to be avoided, non-state actors are going to have to step into the breach (e.g. Rogelj et al., 2016, pp. 6–7). Perhaps the most natural candidates are sub-national political communities, many of whom are significant contributors to climate change in their own right. Cities, for example, have been estimated to account for upwards of 70

2 Lachlan Umbers (UWA) and Jeremy Moss (UNSW)

percent of global energy-related CO2 emissions (IEA, 2008). The United States Environmental Protection Agency (2016) reports that Texas emitted 708.81 million metric tons of CO2 in 2014. If Texas were a country, it alone would have ranked as the 8th largest emitter of CO2, worldwide, in that year (Olivier et al., 2016, pp. 42–3). In the same year, the city of Tokyo emitted 70.1 million metric tons of CO2e (CDP, 2018) – roughly equivalent to the emissions produced by the entire nation of Portugal (EEA, 2017, p. 71). More striking, however, is the record of many sub-national political communities in responding to climate change (Bulkeley, 2013). In 1990, for example, Toronto committed to reducing its net emissions to 20% below 1988 levels by 2005 (Harvey, 1993). The world’s first mandatory emissions trading scheme was introduced by the state government of New South Wales in 2003. California’s greenhouse gas emissions have steadily decreased every year since 2007, following the introduction of the California Global Warming Solutions Act in 2006 (Board, 2018). Studies indicate that a carbon tax introduced by British Columbia in 2008 has been responsible for a 5–15% reduction in emissions (Murray and Rivers, 2015, pp. 677–80). Associated with a steep decline in power plant emissions since its introduction, the Regional Greenhouse Gas Initiative (RGGI) is a cap-and-trade system regulating CO2 emissions from power plants in Connecticut, Delaware, Maine, Maryland, Massachusetts, New Hampshire, New York, Rhode Island, and Vermont (Murray and Maniloff, 2015). And, on the very day President Trump announced his intention to withdraw from the Paris agreement, the governors of California, New York, and Washington announced the formation of the United States Climate Alliance, which now comprises 23 states (together accounting for around 25.5% of the United States’ total emissions) committed to curbing their emissions in line with the Paris targets. There are many examples besides these to be found. Corporations, too, are highly significant actors with respect to climate change. The vast majority of global emissions are traceable in some way to economic activity, much of which, in turn, is linked to corporations.1 If global shipping corporations were a country, for example, they (too) would have ranked 8th on the list of the world’s highest emitters of CO2 in 2014 (Olivier et al., 2016, pp. 42–3). Global aviation would have ranked 14th. Heede (2014) has found that around 63% of global emissions over the period 1854–2010 are traceable to the activities of just 90 large petrochemical and concrete corporations. Recent research has, in turn, suggested that these emissions alone are responsible for around 50% of the global warming (around 0.4°C) and around 32% of the global average sea level increases (around 5.7cm) which occurred over the period 1880–2010 (Ekwurzel et al., 2017, pp. 584–6). The corporate response to climate change has been mixed. The need for societies to de-carbonise presents commercial opportunities. Accordingly, certain corporations have invested substantial resources in the development of environmentally friendly technology in anticipation of future demand for electric vehicles, solar technology, and so on (Pulver, 2011, p. 583). Yet regulatory steps to mitigate climate change threaten the interests of other, powerful commercial enterprises. As such, several fossil-fuel corporations (together with others from industries such as steel, forestry, mining, and

Introduction 3

automobile manufacturing) have conducted highly effective campaigns in recent years to undercut attempts to promote climate action by funding conservative think-tanks and contrarian scientists (Dunlap and McCright, 2011, p. 148), and lobbying against environmental regulations (Perrow and Pulver, 2015, pp. 72–4). Finally, consider the role of individuals. The emissions-producing actions of collective entities like states and corporations are, ultimately, the aggregate product of the actions of individual members of those collectives. Moreover, much of the world’s total greenhouse emissions are traceable to individuals’ ordinary conduct. In the United States and Europe, for example, around a third of total carbon emissions are attributable to household energy use (Stern et al., 2016, p. 1). Erhardt-Martinez et al. (2015, pp. 93–9) point out that residential energy consumption in the US nearly quadrupled over the period 1950–2010. Over a similar period, the number of operating vehicles in the US increased 475%, from 43 million to 248 million and per capita vehicle miles driven more than tripled to around 9600 miles per year.2 Other studies indicate that the emissions associated with these activities might be reduced at very little cost. Dietz et al. (2009), for example, estimate that US household CO2 emissions might easily be reduced by 20% with minimal to no impact upon household wellbeing. Many individuals are also taking action in response to climate change. Some have begun to practice ‘green consumption’, altering their purchasing habits in an attempt to reduce their impact upon the environment (Erhardt-Martinez et al., 2015, pp. 113–5). The global climate movement has seen thousands of individuals around the world mobilise to demand action on climate change (Caniglia et al., 2015). On October 24 2009, for example, 5245 separate acts of protest took place in 181 countries in the run-up to the COP15 climate conference in Copenhagen (Jervey, 2009). On March 15 2019, an estimated 1.6 million students from 125 countries went on strike to demand climate action (BBC, 2019). In 2010 there were 467 unique NGOs devoted to pressing for climate action in the United States, alone (Brulle, 2014). Sub-national political communities, corporations, and individuals, then, are all highly significant actors with respect to climate change. Despite this, the majority of the climate justice literature is focused on states – and, more specifically, on the nature and basis of states’ duties to reduce their emissions to mitigate climate change.3 There are good reasons for this. Preceding considerations notwithstanding, states remain the most significant actors in the global response to climate change. Most agree that the ideal solution to the climate crisis would involve a global agreement between states, wherein each agreed to take on a fair share of the associated global burdens. Nevertheless, this focus has come (to at least some extent) at the expense of giving adequate consideration to the many important normative questions concerning the climate duties of non-state actors.4 Such questions are growing ever more significant in light of states’ ongoing intransigence with respect to climate change. This collection represents the first attempt to bring together scholars working in this space, with the goal of giving due consideration to these issues.

4 Lachlan Umbers (UWA) and Jeremy Moss (UNSW)

Perhaps the most fundamental of these questions concerns the fundamental basis of non-state actors’ climate duties. One salient possibility is that such duties might arise (at least partially) in virtue of the relation between such actors and the states to which they belong. Both Cullity and Umbers’ contributions consider this possibility. Garrett Cullity’s chapter seeks to understand the relationship between climate actors at different ‘levels’. What he calls the ‘initial picture’ of the derivation of the duties of national governments, individuals and intermediate actors (e.g. businesses or local government) is flawed. According to Cullity, the initial picture states that the principal bearers of climate duties are national governments who together can reach the requisite international agreement to prevent climate change. The duties of intermediate and individual agents derive from their relationships to national governments on this picture. However, Cullity argues that this picture of the derivation of duties is flawed in several ways. First, individuals might derive duties from their individual contribution to the expected harms of climate change; not just from their governments. If this is the case, then they can have independently derived duties to reduce their emissions and participate in the collective activities of other groups whose goal is to reduce their emissions. Intermediate actors also contribute to harm via their emissions and have the ability to influence others. They too, then, have duties that are independent of national governments. National governments themselves should not see their only duty as being to reach an international agreement. There are other forms of joint activity, such as UN summits, that they might have a duty to join. Cullity argues that there are in fact different kinds of duties that each type of actor has. In respect of solving the global collective action problem that climate change represents, intermediate agents have duties to influence their national government to take effective action. Whereas individuals have duties to participate in collective action designed to prevent climate change. The governments themselves have duties to participate in global efforts to mitigate climate change. Lachlan Umbers’ chapter, similarly, considers the possibility that at least some climate duties of at least some non-state actors might arise in virtue of the relation between these actors, and the states to which they belong. Focusing on the duties of sub-national political communities, Umbers argues that at least some of the duties of such actors arise out of the failure of the states to which they belong to take adequate steps to respond to climate change. These, Umbers claims, are ‘devolved duties’ – duties on the part of the constituent members of some wider collective agent to take steps to discharge a duty which properly applies to the wider collective, where action at level of the wider collective is sufficiently unlikely. States, he argues, have duties to take steps to aggressively reduce their emissions in order to mitigate the threat posed by climate change. At least at present, states are failing to discharge these duties. The sub-national political communities of which they are comprised, then, have devolved duties to reduce their emissions so as to at least partially discharge the climate duties of the states to which they belong.

Introduction 5

He then goes on to consider and respond to three potential lines of objection. The first of these points out that there is substantial uncertainty with respect to the harms with which the emissions of many sub-national political communities’ emissions will be associated. Umbers concedes that there is such uncertainty but denies that this undermines the idea that sub-national political communities have devolved duties to reduce their emissions by appeal to the idea that such duties may be grounded in considerations of expected value and disvalue. The second points out that not all sub-national political communities will take steps to reduce their emissions. Perhaps, then, holding that any such community has duties to take such steps is unfair. Umbers responds that this, at best, constitutes an independent consideration which might weigh against such communities taking such steps, allthings-considered. It does not in any way, however, undermine the existence of such duties, pro tanto. Finally, Umbers considers potential concerns arising out of the possibility of carbon leakage – i.e. the risk that that sub-national political communities taking unilateral steps to reduce their emissions might lead to an increase in emissions produced in other jurisdictions. Umbers points to empirical evidence suggesting that this is unlikely to be a major concern, and might further be obviated by sensible policy design. Even if this was not so, moreover, he argues that the possibility of carbon leakage fails to undermine the considerations upon which such duties are grounded. Carbon leakage may be a significant concern for policy-makers, However, it poses no fundamental difficulties for the idea that sub-national political communities have duties to reduce their emissions. Corporations are amongst the most important actors with respect to climate change, worldwide, given both their enormous contributions to climate change, and the vast resources at their disposal. The papers collected in this volume focus attention upon four central questions concerning the duties of such actors, each of which has been under-explored in the existing literature. First, what are the climate duties of corporations, and upon what kinds of considerations are they grounded? Second, what kinds of actions should they take in response to those duties? Third, ought we adopt a ‘sectorial’ approach to regulating particular industries? And, finally, what is the relation between the climate duties of individuals, and the duties of the corporations with which they are associated? In his chapter, Jeremy Moss argues that major fossil fuel producing corporations or ‘carbon majors’, are liable for climate harms because of their role as producers and suppliers of fossil fuels. While it might seem obvious that carbon majors are morally responsible for their actions and they harm to which they contribute, Moss argues that it is important to be clear why they are responsible. For example, while carbon majors produce huge quantities of fossil fuels, other agents are directly responsible for releasing greenhouse emissions into the atmosphere by burning them. Carbon majors are certainly a major part of the causal chain that leads to the direct production of greenhouse gas emissions, but we need to understand how much responsibility their role in that chain confers. Moss argues that carbon majors contribute to climate change in two key ways: by directly producing and extracting the fossil fuels that

6 Lachlan Umbers (UWA) and Jeremy Moss (UNSW)

create emissions and by influencing the climate related policies and actions that states and other actors adopt. These two types of contribution render them morally complicit in creating climate harms. Yet even if carbon majors do bear moral responsibility for their contributions to climate change, it is not obvious what actions they ought to take in response. In the second part of the chapter, then, he identifies three broad types of actions that corporations ought to take: disgorging their profits, lowering their direct carbon emissions and phasing out fossil fuel related activities. However, these duties are not necessarily co-dischargeable. For example, a duty to disgorge profits to compensate those who have been harmed by climate change may conflict in practice with a duty to cease producing fossil fuels. The paper concludes by offering a framework for deciding which of these duties it is most important to discharge. In her chapter, Lisa Ellis discusses how we ought to account for the greenhouse gas emissions produced by the aviation industry. Ellis first argues that standard approaches to climate justice in philosophy need to be modified to take into account the particular characteristics of sectors such as aviation. For example, the aviation industry’s emissions are, for the most part, not accounted for in the carbon budgets of states. There is thus no authority in global climate politics to properly govern the industry. This presents a challenge for the regulation of the industry. Ellis explores some of the features that such regulation might entail. Emissions from aviation are growing quickly and threaten to absorb an everincreasing percentage of the remaining carbon budget. In the second part of the chapter, then, Ellis considers whether the aviation industry ought to be treated differently to other sectors such as agriculture or whether it ought to have the same obligations to cut its emissions as other sectors. To determine the status of the industry’s claims, she considers whether several general and plausible ethical positions (polluter pays, beneficiary pays, and ability to pay principles) in the climate justice literature are applicable at the sectoral level. She argues that rather than relying on ideal principles, we should begin with principles that are already shared and consult affected communities via a variety of mechanisms, including citizen assemblies, participatory decision-making processes, and organized civil society groups. Stephanie Collins’ chapter, finally, considers the relationship between the collective duties of corporations and the duties of individuals. She begins by laying out a series of challenges to the idea that individuals might have duties to reduce their personal carbon footprint. In particular, she argues that the expected disvalue associated with ordinary individuals’ carbon emissions is unlikely to be significant enough to give rise to a duty to reduce one’s carbon footprint. This, she argues, ought to prompt us to consider the duties of other non-state actors – in particular, for-profit corporations. Corporations, Collins argues, are moral agents, insofar as they have the capacity to recognise and respond to moral reasons. And there are a range of considerations in virtue of which we ought to regard many corporations as having duties to take steps to reduce their emissions. Corporations, in many cases, are significant contributors to climate change, have the ability to make a substantial difference to the

Introduction 7

expected harms of climate change by reducing their emissions, and have benefited substantially from the unjust failure of the states to adopt adequate climate policies. Collins then argues that the duties of corporations, in turn, have significant deontic implications for individuals. Individual members of such corporations, she argues, may not be able to make a significant difference to climate change by reducing their personal emissions. But they may well have the ability to make a significant difference to the likelihood that the corporations to which they belong will discharge their collective duties to reduce their emissions. Many individuals thereby have duties to take steps to as members of such corporations to raise the likelihood of their doing so. Non-members, she notes, may also have similar duties. Discussion of the relation between individual and collective-level duties does not only arise in the context of corporations. Elizabeth Cripps is the most noted defender of the claim that individuals have more general promotional duties with respect to climate change in the climate justice literature.5 These are duties to take steps to raise the probability of collective action in response to climate change being taken (e.g. by engaging in political activism). Cripps’ chapter begins, then, with the claim that individuals have duties to act together with motivated others to promote appropriate global action on climate change, where they can do so without unreasonable cost. As she points out, however, it is far from obvious how individuals ought to go about discharging this duty, in practice. The principal aim of Cripps’ chapter is to address this issue. Cripps sets out the ‘Cooperative Promotional Model’ – a deliberative process she argues individuals ought to go through in determining how they ought to go about discharging these duties. Individuals, she claims, must first establish what an optimal division of labour with respect to the promotion of collective action on climate change might look like, together with the optimal role they might play in such an arrangement in light of their particular capacities. In determining what they ultimately ought to do, however, individuals must then adjust for the likely (possibly sub-optimal) conduct of others, and remain flexible (i.e. be willing to adjust their own plans in light of changing circumstances). Having set out this model, Cripps then considers a range of associated issues. She begins by sketching a sophisticated justification for her approach, drawing (in part) on recent developments in the consequentialism literature. She then goes on to consider whether her approach might justify a duty on the part of individuals to reduce their personal carbon footprints, as well as considering various ways in which the notion of ‘unreasonable costs’ might be spelled out. Finally, she considers three potential challenges to her approach, arguing that while none is ultimately successful, they do point to the fact that climate activists have reason to prioritise achieving effective global action on climate change ahead of other considerations (e.g. fairness, efficiency). While Cripps’ chapter defends the claim that individuals have duties to promote collective action in virtue of the failure of states to take adequate steps in response to climate change, Susanne Burri and Ben Hale’s chapters consider the possibility that individuals might have moral obligations to reduce their personal carbon

8 Lachlan Umbers (UWA) and Jeremy Moss (UNSW)

footprint. Burri’s chapter begins with the common objection that if each token emission of greenhouse gases is unlikely to by itself cause significant harm, we are not under an obligation to reduce our emissions. Her claim is that many philosophers who argue for this conclusion do so at the level of ‘fact-relative’ morality, (whether emissions do in fact cause harm). Whereas, she claims, it would be more appropriate to consider the issue at the level of ‘evidence-relative’ morality (the reasons that apply to us given the evidence available). In the case of an individual’s emissions, she argues that each token emission raises the total expected disvalue of climate related harms. However, this by itself is not sufficient for even a pro tanto duty to reduce our emissions. For instance, even when we shift to an evidence-relative perspective, the harm that a token emission is likely to cause is tiny. In addition, it is not obvious that a person doing something that carries a minor risk of harm is wrongful. For instance, if someone decides to drive their electric car to the cinema they increase the probability that there will be an accident along the way. But so long as they drive carefully, this seems obviously morally permissible. Burri’s argument is that we ought to adopt the “low expected cost principle’ (LEC). The principle states that we are morally required to reduce our emissions if we can do so at little or no expected cost. Burri argues that such a principle can be supported from a variety of normative perspectives (act-consequentialist, rightsbased or virtue-ethical) and is, in fact, less minimal in its implications than it might at first appear. Burri argues that there is good reason to think that its implications are far-reaching, given that many of our emissions-producing activities have loweremitting alternatives, and that our failure to take up these alternatives is often simply a result of a lack of reflection. Finally, in his chapter, Hale aims to challenge the idea that, where states fail to establish appropriate collective solutions to environmental problems such as climate change, individuals ought to respond by taking steps to address the issue independently – e.g. by altering their consumption habits so as to reduce their personal carbon footprint. Drawing on Hardin (2003), Hale argues that markets can produce a wide variety of possible outcomes, depending on the strategic interactions of the various actors in the market. This, he argues, means that the results of individuals’ attempts to intervene in the marketplace (e.g. by reducing their consumption of items produced in emissions-intensive ways) may be simply indeterminate. Hale argues, then, that in order to address environmental problems such as climate change in the absence of state-level solutions, non-state actors (including individuals, but also potentially actors of other kinds – e.g. local governments) ought to seek to form ‘coalitions’ in order that they might co-ordinate their actions with one another. This is the only way, Hale argues, in which the indeterminacy of potential market outcomes might be addressed – and, as such, the only way in which non-state actors can hope to make meaningful progress on environmental problems like climate change. Non-state actors, then, ought to concerntrate their efforts upon establishing the conditions under which discourse and deliberation among the various actors might take place, with a view to seeing such coalitions established.

Introduction 9

Notes 1 Corporations’ activities are often subsidised by governments. Research conducted by the International Monetary Fund, for example, has found that subsidies for the production and consumption of fossil fuels amounted to a staggering $5.3 trillion US globally in 2015 alone, largely due to the failure of governments around the world to tax the associated negative externalities (Coady et al., 2015). 2 Though they also point out that some of these trends have begun to improve – per capita vehicle miles have fallen to 2004 levels, car ownership has fallen, and total miles driven peaked in 2006 (Erhardt-Martinez et al., 2015, p. 98). 3 See, for example, Barry and Kirby (2017), Duus-Otterström (2014), Gosseries (2004), McKinnon (2012), Morrow (2016), Moss and Kath (2019), Page (2006; 2011; 2012), Roser and Seidel (2017), and Shue (2014). 4 We do not mean to suggest that such questions have been entirely neglected. The climate duties of sub-national political communities have been discussed in Umbers and Moss (2018; 2020). The duties of corporations have been discussed in Schwenkenbecher (2018), and Shue (2017). And the duties of individuals to reduce their emissions or engage in political activism have been discussed in, among others, Cripps (2013), Cullity (2015), Sinnott-Armstrong (2005), and Vanderheiden (2016). 5 See esp. Cripps (2013).

Bibliography Agency, U. S. E. P. 2016. State Energy CO2 Emissions [Online]. Available: https://19janua ry2017snapshot.epa.gov/statelocalclimate/state-energy-co2-emissions_.html [Accessed]. Barry, C. &Kirby, R. 2017. Scepticism about Beneficiary Pays: A Critique. Journal of Applied Philosophy, 34, 285–300. BBC. 2019. School strike for climate: Protests staged around the world [Online]. Available: https:// www.bbc.com/news/world-48392551 [Accessed 24/9/2019]. Board, C. A. R. 2018. California Greenhouse Gas Emissions for 2000 to 2016. CA: California Air Resources Board. Brulle, R. J. 2014. Environmentalisms in the United States. In: Doyle, T. & MAcGregor, S. (eds.) Global Perspectives on Environmentalism. New York, USA: Praeger. Bulkeley, H. 2013. Cities and Climate Change, UK, Routledge. Caniglia, B. S., Brulle, R. J. & Szasz, A. 2015. Civil Society, Social Movements, and Climate Change. In: Dunlap, R. E. & Brulle, R. J. (eds.) Climate Change and Society: Sociological Perspectives. Oxford, UK: Oxford University Press. CDP. 2018. 2016 - Citywide GHG Emissions [Online]. Available: https://data.cdp.net/ Cities/2016-Citywide-GHG-Emissions/dfed-thx7/data [Accessed]. Coady, D., Parry, I., Sears, L. &Shang, B. 2015. How Large Are Global Energy Subsidies?: International Monetary Fund. Cripps, E. 2013. Climate Change and the Moral Agent: Individual Duties in an Interdependent World, Oxford, UK: Oxford University Press. Cullity, G. 2015. Acts, omissions, emissions. In: MOSS, J. (ed.) Climate Change and Justice. Cambridge, UK: Cambridge University Press. Dietz, T., Gardner, G. T., Gilligan, J., Stern, P. C. & Vandenbergh, M. P. 2009. Household actions can provide a behavioral wedge to rapidly reduce US carbon emissions. Proceedings of the National Academy of Sciences, 106, 18452–18456. Dunlap, R. E. & McCright, A. M. 2011. Organised Climate Change Denial. In: Dryzek, J. S., Norgaard, R. B. & Schlosberg, D. (eds.) The Oxford Handbook of Climate Change and Society. Oxford, UK: Oxford University Press.

10 Lachlan Umbers (UWA) and Jeremy Moss (UNSW)

Duus-Otterström, G. 2014. The problem of past emissions and intergenerational debts. Critical Review of International Social and Political Philosophy, 17, 448–469. EEA 2017. Annual European Union greenhouse gas invenotry 1990–2015 and inventory report 2017. European Environmental Agency. Ekwurzel, B., Boneham, J., Dalton, M. W., Heede, R., Mera, R. J., Allen, M. R. & Frumhoff, P. C. 2017. The rise in global atmospheric CO2, surface temperature, and sea level from emissions traced to major carbon producers. Climatic Change, 144, 579–590. Erhardt-Martinez, K., Schor, J. B., Abrahamse, W., Alkon, A. H., Axsen, J., Brown, K., Shwom, R. L., Southerton, D. &Wilhite, H. 2015. Consumption and Climate Change. In: Dunlap, R. E. & Brulle, R. J. (eds.) Climate Change and Society: Sociological Perspectives. Oxford, UK: Oxford University Press. Gosseries, A. 2004. Historical Emissions and Free-Riding. Ethical Perspectives, 11, 36–60. Hardin, R. 2003. Indeterminacy and Society, Princeton, NJ: Princeton University Press. Harvey, L. D. D. 1993. Tackling urban CO2 emissions in Toronto. Environment, 35, 16–20, 38–44. Heede, R. 2014. Tracing athropogenic carbon dioxide and methane emissions to fossil fuel and cement producers, 1854–2010. Climatic Change, 122, 229–241. IEA2008. World Energy Outlook 2008. Paris: International Energy Agency. IPCC2014. Climate Change 2014: Impacts, Adaptation, and Vulnerability. Part A: Global and Sectorial Aspects. Contribution of Working Group II to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change, Cambridge, UK: Cambridge University Press. Jervey, B. 2009. The Global Climate Movement Comes of Age [Online]. Good. Available: https:// www.good.is/articles/the-global-climate-movement-comes-of-age [Accessed 24/5/2018]. McKinnon, C. 2012. Climate Change and Future Justice: Precaution, Compensation and Triage, London, UK: Routledge. Morrow, D. R. 2016. Climate Sins of Our Fathers? Historical Accountability in Distributing Emissions Rights. Ethics, Policy & Environment, 19, 335–349. Moss, J. &Kath, R. 2019. Historical Emissions and the Carbon Budget. Journal of Applied Philosophy, 36, 268–289. Murray, B. &Rivers, N. 2015. British Columbia’s revenue-neutral carbon tax: A review of the latest “grand experiment” in environmental policy. Energy Policy, 86, 674–683. Murray, B. C. & Maniloff, P. T. 2015. Why have greenhouse emissions in RGGI states declined? An econometric attribution to economic, energy market, and policy factors. Energy Economics, 51, 581–589. Olivier, J. G. J., Janssens-Maenhout, G., Muntean, M. &Peters, J. A. H. W. 2016. Trends in Global CO2 Emissions: 2016 Report. The Hague PBL Netherlands Environmental Assessment Agency; Ispra: European Commission, Joint Research Centre. Page, E. A. 2006. Climate Change, Justice and Future Generations, UK, Edward Elgar. Page, E. A. 2011. Climatic Justice and the Fair Distribution of Atmospheric Burdens: A Conjunctive Account. The Monist, 94, 412–432. Page, E. A. 2012. Give it up for climate change: a defence of the beneficiary pays principle. International Theory, 4, 300–330. Perrow, C. &Pulver, S. 2015. Organizations and Markets. In: Dunlap, R. E. & Brulle, R. J. (eds.) Change Change and Society: Sociological Persectives. Oxford, UK: Oxford University Press. Pulver, S. 2011. Corporate Responses. In: DRYZEK, J. S., NORGAARD, R. B. & SCHLOSBERG, D. (eds.) The Oxford Handbook of Climate Change and Society. Oxford, UK: Oxford University Press. Raftery, A. E., Zimmer, A., Frierson, D. M. W., Startz, R. &Liu, P. 2017. Less than 2 °C warming by 2100 unlikely. Nature Climate Change, 7.

Introduction 11

Rogelj, J., Elzen, M. D., Höhne, N., Fransen, T., Fekete, H., Winkler, H., Schaeffer, R., Sha, F., Riahi, K. &Meinshausen, M. 2016. Paris Agreement climate proposals need a boost to keep warming well below 2 °C. Nature, 534, 631–639. Roser, D. &Seidel, C. 2017. Climate Justice, New York, NY, Routledge. Schwenkenbecher, A. 2018. Why business firms have moral obligations to mitigate climate change. In: Brueckner, M., Spencer, R. & Paull, M. (eds.) Disciplining the Undisciplined? Perspectives from Business, Society and Politics on Responsible Citizenship, Corporate Social Responsibility and Sustainability. Springer. Shue, H. 2014. Climate Justice: Vulnerability and Protection, Oxford, UK: Oxford University Press. Shue, H. 2017. Responsible for what? Carbon producer CO2 contributions and the energy transition. Climatic Change, 144, 591–596. Sinnott-Armstrong, W. 2005. It’s Not My Fault: Global Warming and Individual Moral Obligations. Advances in the Economics of Environmental Research, 5, 293–315. Stern, P. C., Janda, K. B., Brown, M. A., Steg, L., Vine, E. L. & Lutzenhiser, L. 2016. Opportunities and insights for reducing fossil fuel consumption by households and organisations. Nature Energy, 1, 1–6. Umbers, L. M. & Moss, J. 2018. Going it Alone: Cities and States for Climate Action. Ethics, Policy & Environment, 21, 56–59. Umbers, L. M. & Moss, J. 2020. The Climate Duties of Sub-National Political Communities. Political Studies, 68, 20–36. Vanderheiden, S. 2016. Climate Change and Free Riding. Journal of Moral Philosophy, 13, 1–27. Victor, D. G., Akimoto, K., Kaya, Y., Yamaguchi, M., Cullenward, D. &Hepburn, C. 2017. Prove Paris was more than paper promises. Nature, 548, 25–27.

1 LEVELS OF CLIMATE ACTION Garrett Cullity (Adelaide)

Climate action of different sorts is possible for national governments, for individual persons, and for various other intermediate agents – collective agents operating at a subnational level, such as regional governments, cities, corporations, and nongovernment associations of various kinds.1 Given these three levels of agency – national, intermediate, and individual – what climate duties are located at each level, and what is the relationship between them? An initial answer is this. The principal bearers of climate duties are national governments. Solving the global collective action problem of anthropogenic climate change will require an international climate agreement; national governments are the parties who must reach it. Their duties are duties of difference-making: duties to do what will make a great difference to the welfare of current and future people.2 Intermediate and individual agents then have derivative climate duties – duties deriving from their relationships to the actions of national governments. The climate duties of intermediate agents are duties of influence: duties to perform those actions that, through incentive and example, can influence national agents to solve the global problem. And the climate duties of individuals are duties of participation: duties to join in collective actions initiated at higher levels, complying with the regulation of our consumption activities in the ways needed to mitigate, and ultimately solve, the problem. This initial answer is wrong. The main aim of this paper is to explain the ways in which it is wrong, and to formulate a better view. Nonetheless, it is a useful starting point for thinking about the different climate duties that apply at different levels of agency. The three kinds of duties it emphasizes are importantly distinct, and it is by seeing what needs to be added to and subtracted from the initial answer that we best approach a more adequate one. I hope to vindicate this approach as I go, by showing that it is fruitful. After beginning with a fuller description of the initial picture and the three kinds of duties it ascribes, I work through each of the three levels of climate agency in turn, explaining how the initial answer needs to

Levels of climate action

13

be improved. The last section summarizes the alternative picture that we arrive at if we make these improvements. It is a more complicated picture: while the duties of lower-level agents do depend on their relationships to higher-level ones, the reverse is also true; and while duties of difference-making and influence are important, I shall be arguing that the principal climate duties of agents at all three levels are participatory.

I: The initial picture In between the international and individual levels of agency, I am grouping together a variety of other agents as “intermediate”: regional governments, cities, corporations, and non-government organizations. This may seem misguided. Measured in terms of economic size, the biggest “intermediate” agents dwarf the smallest countries. California’s GDP is $2.75t, just below that of France on the international scale. The smallest nation economically, Tuvalu, has a GDP of $38m, forty times smaller than that of Tokyo and 70,000 times smaller than California’s. Given this, it may seem muddled and unhelpful to specify the “levels” of climate action as I have, placing California at an intermediate level between Tuvalu and individuals. No doubt, some individuals spend more than $38m in a year.3 However, the point of this classification is not to grade agents in terms of their economic power. It is to identify the kinds of climate action that are available to different kinds of agents, the morally relevant reasons there are to perform those actions, and consequently the kinds of duties that they bear. The stringency and scope of an agent’s duties depend on which actions the agent is able to perform. Economic capacity is relevant to this – we will come to that later – but so is normative capacity. Some actions are only available to agents with the right kind of authority to perform them. In the case of climate action, it is only national governments that are qualified to be parties to an international treaty; intermediate agents, whatever their economic size, are not. And intermediate agents, disparate though they are, possess capacities for action that individuals lack (when acting as individuals rather than as institutional role-bearers): capacities that include representation, coordinating the efforts of individuals, and authority over them of various forms. The initial picture of the allocation of climate duties from which we will be working emphasizes this difference between the actions available to agents at the three levels. It relies on a diagnosis of what the problem of anthropogenic climate change fundamentally is: namely, a global collective action problem (Gardiner, 2001). We all inhabit an economic order in which the environmental cost of each agent’s resource use is largely externalized – most of the cost of each agent’s action is borne by other parties (Stern, 2008, p. 1). Some of the bearers of that cost are our contemporaries; many more will bear the much greater costs that are projected into the future.4 Large-scale collective action problems of this form need to be solved by imposing effective regulation: here, since the problem is global, the regulatory solution has to be global too. What we need to achieve is a binding

14 Garrett Cullity (Adelaide)

international agreement that limits future global greenhouse gas emissions to sustainable levels through the application and enforcement of effective economic incentives and supply-side restrictions (Stern, 2007; Green and Denniss, 2018). It is natural to treat the allocation of duties in this case as following the same pattern that we find in other collective action problems. For example, suppose our town’s water supply is limited, and with unregulated water usage we face the prospect of running out. We need the municipal authorities to introduce a scheme of water restrictions. In this sort of situation, the allocation of duties is evidently this: the municipal government has a duty to implement a scheme of water restrictions; other influential agents (such as the local health authorities) have a duty to use their influence to get the government to do this; and individual water consumers have a duty to comply with the restrictions once implemented. The initial picture makes the corresponding allocation of duties in response to our global collective action problem of climate change. The agents with a duty to solve the problem are national governments. They are the agents in whose power it lies to form the international agreement that will constitute the solution we need. Agents at the other two levels, since they are not potential parties to such an agreement, do not bear those duties. However, they are able to support a global climate solution in two other ways. Intermediate agents, given the scale on which they operate, have the capacity to influence national governments to reach the agreement we need; and individuals can conform with it once it is reached. Individuals will then acquire duties to participate in a regulatory solution; but no individual has a unilateral duty to adopt the constraints that would make sense as part of that regulatory solution, prior to its adoption – any more than I have a duty not to water my lawn prior to the implementation of municipal water restrictions.5 We can sharpen this picture by introducing some terminology. “Duties of difference-making”, we can say, are duties to produce positive impacts and avoid negative impacts on others’ welfare. If we are being careful, we should note that causing an impact is not always the same thing as making a difference to whether that impact occurs – but let us use “duties of difference-making” to cover both cases.6 Within this class of duties, we can then make a pair of distinctions. Duties of dependent difference-making are duties to produce positive and avoid negative impacts on others’ welfare, by affecting whether some other agent produces or avoids those impacts. Duties of independent difference-making are the duties to produce positive and avoid negative impacts on others’ welfare directly, without relying on someone else’s agency to do so. A further distinction concerns whether a duty of difference-making is a duty to induce another agent to do what that other agent already has a duty to do. When a duty of difference-making is not of that kind, we can call it a primary duty of difference-making. A secondary duty of difference-making is a duty to induce another agent to fulfil a primary duty of difference-making.7 Notice that these two distinctions do not coincide. We see this in the duties that authoritative agents have to solve collective action problems: these are primary but dependent. The municipal government’s duty to implement water restrictions is primary, since it is a duty to instruct individuals to do what they

Levels of climate action

15

do not already have a duty to do; but it is also dependent, since the difference the government makes is made by affecting the actions of individual consumers. The other class of duties for which we will need a label comprises “duties of participation”. These are the duties that an agent can have to join in the activity of a group. When a group is achieving something important by acting together and is doing so through its members’ willingness to join in, sharing the costs of the collective action, I can thereby acquire a duty to join in on the same terms.8 Duties of participation should be distinguished from duties of difference-making: I can have duties to join in the worthwhile actions of groups that are large enough to absorb the costs of my non-participation, without its having a significant impact on anyone’s welfare. If I ride on public transport without paying, the moral complaint against me is not that I am harming anyone. It is that I am failing to participate in the collective practice on the same terms as everyone else.9 Participatory duties are the duties I have when I lack a good answer to the question “Why aren’t you joining in?” Phrased in these terms, the initial picture of climate duties is as follows. National governments are the bearers of primary duties of difference-making, as the potential parties to the international agreement that is needed as the solution to our global problem: like other agents with the capacity to solve collective action problems, their duties of difference-making are primary but dependent. Intermediate agents have secondary duties of dependent difference-making: duties to influence national governments to do what they already have a duty to do. Some prominent individuals have secondary duties of influence too – opinion-formers and wellconnected people who can influence the decision-making of national governments. However, most individuals are not in that position: their climate duties are duties of participation, not difference-making. In the future, when there is a global climate agreement, individuals will have duties to comply with the regulation it imposes on them. Meanwhile, the climate duties of individuals extend at most to participating in the joint efforts now under way to encourage their governments to do what they ought – supporting climate action campaigns and voting for parties with sound climate policies (Johnson, 2003; Maltais, 2013). To complete the initial picture, we can consider its application to two other important kinds of climate action. So far, we have been focusing on the actions required to solve the global climate problem. But in addition, during the period before we have arrived at that solution, we need to consider the ways in which our action can affect the severity of the problem itself. We face the question what we should be doing to mitigate the climate impacts of our past and current activity, and what we can do by way of adaptation, to protect vulnerable people from being harmed by the changes to the climate that are already underway. The initial picture can be extended to cover these actions too, making the corresponding claims. With respect to mitigatory and adaptive action, national governments are again the primary bearers of duties of dependent difference-making to exercise their authority to change their citizens’ behaviour. They have duties of mitigation, owed to climatevulnerable people around the world, to regulate their own economies in a way that

16 Garrett Cullity (Adelaide)

lessens the harm they are doing; and duties of adaptation, owed to their own citizens, to protect them from the threats that the processes of climate change now under way will expose them to. From these, the mitigatory and adaptive duties of other agents derive in the same ways as before. Intermediate agents have secondary duties of difference-making through influence on national governments; individuals have duties of participation in the actions initiated by their governments. A corollary of this picture is that the climate duties of intermediate agents and individuals may be relatively modest. When one agent has a difference-making duty and a second agent can potentially influence whether it is fulfilled, then that gives the second agent a morally relevant reason to exercise that influence. But whether there is a duty to exercise it will depend on how influential one is likely to be, how burdensome the action will be, and what morally relevant reasons there are to do other things instead. A clear case for the possession of duties of influence will apply only to those intermediate agents for which climate action is not very disruptive. Similarly, getting from the reasons that individuals have to participate in worthwhile joint action to a duty to participate also depends on the cost of participation and the reasons there are to do other things instead. The party with the best climate policy may have other social and economic policies that are bad. And while I have participatory reasons to join in climate action campaigns, I also have participatory reasons to join in campaigns to address poverty, animal welfare, mental health provision, and every other worthwhile movement for social improvement. So the clearest assignment of climate duties, on this picture, is to national governments. Some intermediate agents and individuals may bear the other two kinds of duties: but that depends on what else is at stake for them.

II: Revising the initial picture The initial picture makes some assumptions that others may wish to question, but I shall not. It assumes that institutional agents can have moral duties; that current patterns of greenhouse-gas-emitting activity will harm future generations; and that duties to avoid such harm are not undermined by the fact that the identity of future individuals is contingent on the actions we take now. I also accept its assumption that the scope and stringency of the duties an agent has depend on all of the reasons that bear on that agent’s actions.10 Thus, although the initial picture offers a classification of the kinds of climate duties that are borne by agents at the three levels and the relationships between them, it does not say that all agents at a given level have the same duties. It says that the climate duties that individuals have are participatory, but it does not say that all individuals have participatory climate duties. It allows that some individuals – those with compelling reasons to do other things instead – lack any climate duties at all. My own concerns with the initial picture lie elsewhere. I shall be arguing that it needs to be revised in three main respects. It mischaracterizes the relationship that national governments bear to the global problem of climate change and consequently mischaracterizes the duties they have in relation to its solution. It

Levels of climate action

17

understates the climate duties of lower-level agents. And it oversimplifies the relationships of dependency between the content of climate duties at the three different levels. The first of these errors, I shall argue, comes from a mistake about the kind of collective action problem that anthropogenic climate change presents us with. It is true in many cases that agents with the authority to implement regulation with the potential to solve a collective action problem can have a primary duty of dependent difference-making to do that; and that the agents over whom they have authority can then acquire duties of participation after the regulatory solution has been implemented. However, on closer examination, we will find that this description does not fit the global climate change collective action problem – the problem is too big. It turns out that the duties of national governments are not primary duties of dependent difference-making. The initial picture is not only wrong about the climate duties of national governments; it is also wrong about the climate duties of agents at the other two levels. It sees the duties of lower-level agents as deriving from the relationships they bear to higher-level ones; but that is incorrect too. It underestimates the contribution we are all making to producing the current problem. Once we get that properly into focus, we find that intermediate agents and individuals have climate duties that do not derive from their relationship to national governments. Beyond this, the initial picture needs a third correction. It is not just a mistake to think that all lower-level climate duties depend on relationships to higher-level agents – on the contrary, the content of the climate duties of higher-level agents can depend on their relationships to lower-level ones. And at all three levels, the climate duties we have are primarily participatory. The next three sections explain these points by considering each of the three levels of climate agency in turn – starting with individuals, then considering national governments, and finally turning to the varied group of intermediate agents.

III: Individual action The initial picture sees the climate duties of individuals as deriving from their relationship to higher-level agents. The bearers of primary duties of differencemaking are agents at higher levels: individuals currently have secondary duties of participation in collective efforts to get higher-level agents to discharge their primary duties, and will in future acquire duties of participation in the initiatives that those other agents eventually adopt to solve the problem (and to mitigate it meanwhile). The effects of any individual’s actions on harmful climate change are too small to be themselves the ground of a climate duty.11 However, once we get a better appreciation of the scale of the problem that we are creating, we find a serious case for attributing to individuals duties that do not derive from our relationship to higher-level agents. Epidemiologists already estimate that approximately 300,000 extra deaths are attributable annually to humaninduced climate change (150,000 from the first 0.5°C rise in temperature above

18 Garrett Cullity (Adelaide)

pre-industrial levels, and another 150,000 for the second); alongside these deaths, there are many more individuals who are severely affected, in needing emergency assistance or having their livelihoods significantly compromised (GHF, 2009). If we continue putting greenhouse gases into the atmosphere at the current rate, this figure will rise above 500,000 annual deaths in our lifetimes; and the IPCC tells us that the greenhouse gases we are now putting into the atmosphere can be expected to remain there (unless others act to take them out) for over a millennium (IPCC, 2007, p. 47). Such figures do not allow us to draw conclusions about the amount of harm that is done by any individual emitter: the causal system linking individual emissions to individual harms is too complicated for a connection of this sort ever to be traceable. However, what they do allow us to do is to quantify the expectation of harm that is associated with the lifetime emissions of an average individual living in the developed world. On average, each rich-world individual contributes only about one two billionth of the total emissions being added to the atmosphere globally.12 But the aggregate amount of harm that these emissions are projected to do is very large: it includes more than 500,000 deaths a year for more than 1,000 years. So on these figures, the average contribution is more than a quarter of a death each. The actual causal contribution that any one of us makes might be either higher or lower than this: we will never know. But the expected harm associated with the activity of each rich-world emitter – the probability-weighted sum of the harms it might produce – is the average. To this mortality figure, we then need to add the expected non-lethal harm. Using the same epidemiologists’ data and applying the same methods, the average figure we arrive at for this is more than 290 people needing emergency assistance or having their livelihood significantly compromised. Elsewhere, I discuss this calculation in more detail, and the conclusions we should draw from it.13 It invites questions about the extrapolation that is needed in order to get from the established data to an estimate of future harm and its attribution to one type of action in a very complex causal system. However, it is likelier to be an under- than an overestimate.14 And while the details of the calculation can be questioned, the issue cannot be ignored. If the overall amount of harm we are doing is large enough, then I face moral questions about my own contribution to it. If I am acting in a way that imposes a substantial risk of harm on other people, I need to justify the imposition, and I cannot do so merely by pointing out that many other emitters are doing the same thing. From here, there are two serious arguments for thinking that individuals have a duty to reduce or offset their own personal greenhouse gas emissions. The first, maintained by John Broome and others, is an argument from individual differencemaking (Broome, 2019; Broome, 2012; Hiller, 2011; Morgan-Knapp and Goodman, 2015). I have a duty not to impose on others a substantial risk of harm without good justification. Here, the risk associated with my emissions is substantial – a quarter of a death and 290 other severe harms – and all I have by way of a justification for imposing it is my own convenience; so I have a duty not to impose it.

Levels of climate action

19

A question can be raised about that first argument. There is an expectation of harm associated with many of our actions, but whether we have a duty to avoid them depends not just on the magnitude of the expected harm and the cost to oneself of avoiding it – it depends also on one’s degree of responsibility for any harm that does result from the action. When harm is a result of the causal influences of many different agents, that can lessen the degree to which I am properly held responsible for it. From that general point, we cannot conclude that the first argument fails; but there is a further issue that needs to be investigated before concluding that it succeeds – the issue of responsibility-attribution.15 However, alongside this first argument, we need also to consider a second. This is an argument for an individual duty of participation rather than of differencemaking. Given the risks of harm that are associated with individual emissions, many people around the world are reducing or offsetting their emissions in order to lessen the impact that they have on vulnerable people. So I face the question “Why aren’t you joining in?” When this question is asked, I am not being asked to justify the difference I make if I continue to emit at unconstrained levels. Instead, I am being asked to justify my unwillingness to participate in a joint practice (of constraining emissions) that makes a substantial difference, given the number of other people who participate in it. What they do together is not itself going to solve the global problem of climate change, but it is a joint action of substantial harmreduction in which I could participate on the same terms as the other participants. Why am I not prepared to join this group on the same terms, sharing the costs of what they are doing together?16 Although both of these arguments need to be taken seriously, I find the second more compelling: I cannot see a good answer to the question it poses. Together, they constitute a serious challenge to the initial picture of climate duties. The climate duties of individuals are not limited to those that derive from their relationships to higher-level agents. Individual contributions to the problem are serious enough, and the actions of individuals to lessen that contribution significant enough, to create duties of climate action that are independent of those relationships. In that way, the initial picture’s treatment of individual climate duties is too narrow: it misses the duties that do not derive from our relationships to higherlevel agents. But in another way, it is too broad. There is an important restriction on the participatory duties that we can acquire through standing under the authority of a government. Citizens who stand under the authority of legitimate governments can have a duty of political obedience to comply with their governments’ climate change legislation. On an orthodox understanding of political obligation, this duty is not conditional on the government’s decisions’ being correct or optimal, but only on their being justly arrived at and imposed. But this carries with it an important restriction. Individuals who are subject to the power of illegitimate governments do not have that duty. They can still have climate duties of mitigatory action, of the kind described above – duties to reduce or offset their own emissions. And they could have participatory duties to bring their own conduct into

20 Garrett Cullity (Adelaide)

conformity with a future global climate agreement, if that agreement is worthwhile. But further duties of obedience to the authority of higher-level climate agents depend on the legitimacy of that authority.

IV: International action We can now look more closely at the climate duties of national governments. The initial picture characterizes these as primary duties of dependent difference-making through authority: duties to form an international agreement as the solution to our global collective action problem; after forming it, to implement it through the exercise of regulatory authority over their own citizens; and before forming it, to take effective mitigatory and adaptive action. This picture needs four main amendments. The first involves being more careful about the actions that are available to any one national government, interacting with the others. There is a sense in which the governments of the world, considered collectively, are able to make an international climate agreement; but no individual government can do that. What it can do is to propose such an agreement, to signal its willingness to comply with it, to commit to doing so, and to produce incentives for other countries to do likewise. Given this, the initial picture misdescribes the duties that a national government has in relation to a global climate agreement. No individual government can solve our global problem, so no government has a duty to do that. The principal climate duties of each national government are duties of dependent difference-making through influence on the actions of other governments. And these duties of difference-making through influence are all secondary, since they are duties to induce the other parties to do what they already have a duty to do. The actions each government has a duty to perform in influencing the others – actions of signalling, commitment, and incentivizing – are actions of inducing the other parties to perform actions of the same kinds, which they already have duties to perform. There is no primary duty-bearer. The second respect in which the initial picture needs correction is that it overlooks the importance of the participatory duties that there now are at the national level. Each government can participate in those forms of cooperation that currently exist as worthwhile contributions towards reaching a global agreement: for example, participation in the UN climate summits, compliance with the agreements reached at those summits, and cooperation within the coalitions of states that have been formed to influence climate decision-making, such as the Major Economies Forum on Energy and Climate, the Global Climate Change Alliance, and the Alliance of Small Island States (AOSIS). The importance of these forms of cooperation comes from the way they sustain and strengthen a needed vehicle for solving our collective problem, rather than because they are directly solving it. The magnitude of the problem confers importance on these cooperative efforts, and grounds the characteristic challenge – “What justification is there for not joining in?” – that gives rise to participatory duties. These must be distinguished from the participatory duties that national governments will have to comply with a

Levels of climate action

21

comprehensive global climate agreement once it is reached. In advance of this, their principal climate duties are duties of two kinds to direct themselves towards securing that agreement – secondary duties of dependent difference-making through mutual influence, and the different participatory duties just mentioned. In addition, it is true that they also have duties to take mitigatory and adaptive action in advance of that comprehensive agreement. But here, a third departure from the initial picture should be noticed. National governments have two sets of mitigatory duties. They have duties to mitigate their own climate impacts, through the actions of their employees: these are indeed primary duties of dependent difference-making, as the initial picture says. But they also have duties to regulate the activities of their citizens in a harm-minimizing way – and these mitigatory duties of national governments are secondary, not primary. They are duties to compel us to do what we already have a duty to do. In that respect, the initial picture is mistaken in seeing the mitigatory duties of individuals as deriving from those of national governments. However, this is not to say that there is a derivation running in the other direction. Applying our distinctions, the mitigatory duties of national governments should be classed as duties of dependent difference-making, since they are duties to make a difference by via the actions of other agents over whom they have authority; and they are secondary duties when they are duties to compel us to do what we already have a duty to do. But it is not the case that national governments have mitigatory duties because individuals have such duties. A national government is a locus of decision, and it is the expected consequences of its decisions on the welfare of those who stand to be affected that is the ground of its duty to mitigate that impact. To discharge that duty, a national government must take decisions to regulate the activities of its citizens in ways that lessen its climate impact; and when it does so it will be directing us to do what we already have a duty to do it. But its duty does not derive from ours.17 However – as the fourth and final correction – we should notice that there are indeed some ways in which the content of national governments’ climate duties depends on their relationship to lower-level agents. The duties of governments are constrained by their relationship to the citizens of whom they are the political representatives. The constraints that the rights of individual citizens place on the actions of representative governments add to the moral as well as the political complexity of the climate action challenge. National governments (as well as those of regions and cities) have duties of representative responsibility, owed to those who vest them with authority. They must govern justly, and must protect the welfare of those to whom they are responsible – especially the welfare of those who are least well off. This supports the view that national governments have a duty not to make unilateral commitments that will impose significant sacrifices on their own citizens – especially, their own poor citizens – in advance of effective global action. And more particularly, it strongly supports the view that the governments of developing countries have a duty to their own citizens not to forgo the opportunities for development that developed countries have already taken.18

22 Garrett Cullity (Adelaide)

A further point about political representation is worth making. Some governments have been elected on a policy platform of climate inaction. They ought, morally, not to have adopted this platform, and voters ought not to have voted for it. However, a promise one ought not to have made can be a promise one ought to keep. And when the authority of a government derives from the mandate it receives from the people it represents, its entitlement to act is restricted by what the populace has authorized it to do. This creates the possibility that some legitimate governments have a duty of climate inaction. To say this is not to exculpate them: they act wrongly in advocating the platform they do. But having been instructed to implement it, they may still have a duty to do so. The one part of the initial picture of the climate duties of national governments that emerges unchanged from this examination is its description of their duties of adaptive action. These are indeed duties of primary difference-making through authority. National governments have duties to secure the welfare of their own future citizens; but it is less plausible that when my government exercises its authority over me for the purpose of implementing adaptive measures to lessen the impact of future climate change on future generations, it is thereby compelling me to do something that I already have a duty to do. I have mitigatory duties connected to reducing the expected harm associated with my energy-consuming behaviour; and if there were something I could easily do to assist future individuals to adapt to a warmer world, then that could be the ground of a duty of adaptive action; but it is not clear that there are effective adaptive actions readily available to me. So the adaptive duties of national governments are primary.

V: Intermediate action Having identified the corrections in the initial picture that are needed for a more accurate picture of the climate duties of individuals and national governments, we can turn now to intermediate agents. We can focus first on cities and regional governments, before considering intermediate agents of other kinds. The place to begin is by noticing how the points made in the previous two sections about individual agents and national governments apply to these intermediate agents too. Then we can look at the further important respect in which the initial picture of their duties needs to be corrected. Section III argued that the scale of our climate problem is large enough to give individuals duties of mitigatory action that are independent of their relationship to higher-level agents. If so, the same point applies to cities and regional governments. Since they are responsible for a greater volume of emissions, the expectation of harm associated with their actions, and the strength of the moral reason to reduce it, are correspondingly greater. Since these intermediate agents act by exerting their authority over individuals, these are duties of dependent difference-making; and when the individuals who are subject to that authority already have mitigatory duties of their own, the duties that the intermediate agents have to exercise their authority in this way are secondary, not primary. They do not derive from the mitigatory duties of individuals, nor from those of national governments. At all

Levels of climate action

23

levels, there are agents making decisions with significant enough welfare impacts to give rise to mitigatory duties – duties that do not derive from the corresponding duties possessed by other agents.19 Thus, cities and regional governments acquire mitigatory duties in the same way that individuals do – independently of their relationship to higher-level agents. But then they also acquire duties of adaptive action in the same way that national governments do. As politically representative agents, cities and regional governments have duties to take precautionary action now in order to protect their own residents’ future welfare against the harmful impacts of climate change. These are primary duties of dependent difference-making. That gives us two corrections to the initial picture of the climate duties of intermediate agents. That picture treats all of their duties as deriving from their relationships to national governments; but their mitigatory and adaptive duties do not fit that picture. How about their duties of influence – those which the initial picture characterizes as intermediate agents’ principal climate duties? Here, three corrections are needed: we can examine them in ascending order of importance. First, the initial picture treats this as a point of contrast between national and intermediate agents: intermediate agents have duties to influence national governments, the climate change difference-makers. However, we have seen that the principal climate duties of national governments are themselves duties of influence. So to the extent that intermediate agents have duties of dependent difference-making through influence on governments’ actions, these are duties of the same kind as the principal climate duties of governments themselves. It is true that the kinds of government-influencing action that subnational agents can have duties to perform are restricted: they cannot include signalling a willingness to commit to an international treaty to which they are not potential parties. However, they are duties of the same general type. The other point we made about national governments that transfers to intermediate agents concerns the importance of coalitional action. The most effective way for intermediate agents to exercise their influence is by acting together, in groupings such as the 650-strong Cities for Climate Protection programme (CCP), the C40 Cities Climate Leadership Group of ninety global cities, the “We Are Still In” group of 3,800 American business, political and educational leaders, or the Regional Greenhouse Gas Initiative (RGGI) in which ten US states participate in a cap-and-trade emissions reduction scheme. Once these joint actions are established, the morally relevant reasons for each intermediate agent to join them are participatory, rather than reasons of difference-making. (It is the coalition that is the difference-maker; that is what makes the joint action worthwhile; the reason for each agent to join in that action is participatory.) These reasons give rise to duties in the same way as before. For each coalition that is achieving something significant, each further potential partner to the coalition faces the question “Why aren’t you joining in?” The seriousness of the problem that is being addressed determines the strength of the justification that is needed to address this question satisfactorily. When there is no satisfactory justification, there is a participatory duty.

24 Garrett Cullity (Adelaide)

However, there is a third, more important, respect in which the initial picture needs to be corrected. The importance of coalitional climate action at the intermediate level is not tied to its capacity to influence the actions of national governments (Heijden, 2018). When US city mayors and state governors joined the “We Are Still In” coalition, the point of doing so was not primarily to persuade the current US national government to change course. It was to demonstrate that a commitment to effective climate action was not dependent on national government policy. And the thinking behind C40 is to show that the world’s urban population has the ability and good will to transform the world economy in a way that bypasses inertia at the level of international negotiation.20 While the overall point of such action is to make progress towards an eventual global framework for effectively regulating human impact on the climate, its contribution to doing this is to start to establish the new economic structures that a future global agreement will need to enforce, rather than merely to induce other agents to reach that agreement.21 It can be questioned to what extent considerations of duty actually motivate such initiatives. Long-sighted municipal leaders who want to see their own cities perform strongly in the economy of the future have appreciated the advantages of being early adopters of the new technologies and forms of social organization that will have to be found – or at least not to get left behind, by tapping into the expertise of the other C40 members. Participation in the RGGI has not been a matter of economic self-sacrifice: the economies of the participating states have grown at the same time that their emissions have gone down (Fairfield, 2014). However, our question is whether these agents have climate duties, not how far they are motivated by them. And this point actually makes it easier, not harder, to establish that there are such duties. Every agent has morally relevant reasons to be taking climate action; whether there is a duty to take it depends on whether there is a good answer to the question “Why aren’t you joining in climate action?”; the smaller the sacrifice that such action requires, the harder it is to answer this question. Turning finally from subnational governments to intermediate agents of other kinds, we find that their climate duties overlap with those already mentioned. Corporations are economically powerful enough to have primary duties of differencemaking to take mitigatory action to lessen the impact of their emission-producing activities. If the scale of our climate impact is large enough to generate mitigatory duties for individuals, then commercial enterprises of any size have such duties also, as do a wide range of other associations: unions, clubs, professional bodies and volunteer organizations. In addition, corporations can join coalitions of influence (such as the “We Are Still In” coalition), and this gives them participatory duties of the same sort as intermediate political agents. Where the non-governmental intermediate agents differ is in respect of adaptive action. Since they do not have the same duties to protect the interests of particular future individuals as the political agents who represent those individuals, the case for assigning duties of adaptive action to them is weaker.22 They have the same duties as all agents not to cause harm, but lack the relationship of special responsibility for welfare-protection that marks out the political agents as having a duty to protect their own vulnerable residents from predictable future harm.

Levels of climate action

25

Non-government organizations with a declared platform of climate activism are the intermediate agents whose principal climate duties are secondary duties of influence. The source of these duties comes not just from their capacity to exert an influence, but an implicit promise to their supporters to do so. Through making these commitments, these organizations can assume duties they would not otherwise have. However, notice that here, too, a correction to the initial picture is needed: the duties of activist organizations can extend beyond duties of influence. An organization that promises to speak up against bad policy has a duty to do so, owed to the promisees, independently of whether this is likely to change the policy.

VI: The revised picture According to our initial picture of climate duties, national governments are the difference-making agents with a duty to solve the global problem of climate change; intermediate agents have a duty to use their influence over their own national governments to do that; and individuals have a duty to participate in effective joint activity towards that end. This kind of picture captures the moral structure of many collective action problems; but it is not adequate here. We have found four main deficiencies. It simultaneously overstates the responsibilities of national governments, understates the responsibilities of lower-level agents, oversimplifies the relationships of dependency between the content of climate duties at different levels, and misses the significance of coalitional agents. The last of these points deserves further emphasis. With the formation of coalitions at each level, new climate agents are created.23 The same deliberative mechanisms that bestow on national and subnational governments the capacity to appreciate and respond to reasons, and consequently make their actions morally evaluable, are present among those coalitions of such agents that have adopted agreed procedures for decision-making on behalf of coalition members. If Tuvalu and Tokyo can have climate duties, so can the C40 network and AOSIS. These are climate agents located at the same intermediate level of climate action as their members; once formed, the coalitions are bearers of duties of difference-making, while the duties of their members are participatory. A corollary of this is that participatory actions can be nested within each other; and as a consequence, so can the duties to which they give rise. What can make the action of a collective agent worthwhile is that it is an action of participation in the agreed activity of a larger coalition. If the C40 coalition of cities together are having a valuable impact on global climate politics, the action of the coalition may be worthwhile, generating participatory reasons for my city to join C40; this can in turn make my city’s climate actions as a participant in C40 worthwhile, and that can generate participatory reasons for me to support those climate actions (for example, voting to support them in municipal elections, and joining in local C40 initiatives). This is a further way in which higher-level participatory duties can generate lower-level ones.

26 Garrett Cullity (Adelaide)

We need to replace the initial picture, then. In outline, the improved picture of climate duties we have arrived at is this. We should retain the initial picture’s distinction between the actions that can be taken towards solving to our global collective action problem and the actions of mitigation and adaptation that need to be pursued meanwhile. However, at all levels, the duties to perform actions of the first kind are primarily participatory. The solution being sought is an effective form of cooperation; what is required of agents at every level is cooperative participation in the forms of joint activity that present us with the best ways of reaching that goal. With duties of mitigation and adaptation, by contrast, the duties of higher-level (national and intermediate) agents are duties of difference-making. Economically powerful agents have duties to mitigate the harm done by the emissions activity for which they are responsible; politically powerful agents have duties of adaptation to protect their own populations from climate harms. And although in my view the mitigatory duties of individuals are duties of participation, these go beyond simply complying with the authority of higher-level agents: as individuals, we ought to be participating in joint efforts to reduce the harm that we ourselves do. On this picture, national governments are important climate agents: solving the problem of climate change will indeed require an international agreement. But they are not the source of all climate duties, in the way that the initial picture suggests. All it takes to possess a climate duty is either economic power, or political responsibility, or the capacity for joint action. The content of the climate duties possessed by agents at higher and lower levels is interdependent. The duties that higher-level agents have to exercise economic power are constrained by their political responsibility to those they represent; and lower-level agents’ duties to participate in climate action are shaped by the importance of supporting effective higher-level action, for the sake of our collective future.24

Notes 1 In this essay, I follow the common practice of using “national governments” to refer to sovereign states (whether or not they represent a unitary nation), and “international agreements” to refer to agreements between those states. 2 In addition to the impacts on people, the impacts on other animals and the rest of the environment are also morally significant. For a summary of these, see Intergovernmental Panel on Climate Change (IPCC) (2015), pp. 64–7. 3 Sources: International Monetary Fund (https://www.imf.org), U.S. Department of Commerce Bureau of Economic Analysis (https://www.bea.gov). 4 For an attempt to estimate these costs in welfare terms, see Section III. For estimates of the economic costs, see Mendelsohn (2011). 5 For examples of the initial picture, see Johnson (2003); van de Poel, Fahlquist et al. (2012); Maltais (2013); Sinnott-Armstrong (2005). 6 I can cause an impact without making a difference when I pre-empt another action that would have caused it; I can make a difference to whether an impact occurs without causing it when I refrain from preventing it. For further discussion of this distinction and its application to climate harms, see the appendix to Cullity (2019). 7 Tertiary duties are also possible, where A has a duty to get B to comply with her duty to get C to comply with his duty. I leave these out of the current discussion.

Levels of climate action

27

8 Under exactly what conditions is that so? I offer an answer in Cullity (1995). 9 For fuller discussion, see Cullity (1995), pp. 6, 22. 10 This does not equate an agent’s duties with verdicts about what that agent morally ought to do, all things considered. It allows for cases in which I ought to infringe a duty for the sake of something else that is morally more important (such as another, more stringent, duty). 11 See the authors cited in note 5. 12 Source: Carbon Dioxide Information Analysis Center (http://cdiac.ess-dive.lbl.gov). 13 For the details, see Cullity (2019), Section 1. 14 For an overview of the scale of possible impacts of climate change on human health, see IPCC (2014), Ch. 11. 15 For a fuller investigation of this question, see Cullity (2019), Sections 2–3. 16 For a fuller exposition of this second argument, see Cullity (2019), Section 6. 17 Isn’t this double-counting? Yes, in the sense that it counts the same harms as belonging to the impact of national government decisions, as well as of the decisions of individuals. But attributions of responsibility for harmful outcomes are subject to this kind of double-counting. If A abets B in harming C, A might thereby acquire a measure of responsibility for the harm to C without in any way diminishing B’s responsibility for that same harm. For further discussion, see Petersson (2013). 18 On the importance of energy use for development, see Carbonnier and Grinevald (2011). 19 This now involves not just double- but triple-counting in the attribution of responsibility for harmful impacts. But this remains legitimate in the same way as before: see note 17. 20 For a study of the C40 network and its place within the broader context of transnational municipal cooperation, see Lin (2018), Ch. 5. The corresponding point can be made about the CCP program: see Betsill and Bulkeley (2004), p. 490. 21 On the importance of “knowledge communities” in solving problems in the provision of global public goods, see Brousseau and Dedeurwaerdere (2012). 22 Weaker, but not non-existent. If my past actions have created a threat to you, then that does give me a morally relevant reason to help you to protect yourself from the threat. 23 For a more general analysis of the role of transnational coalitions in global governance, see Forman and Segaar (2006). 24 Earlier versions of this essay were presented to the Climate Justice Beyond the State workshop (Sydney, 2018), World Forum for Climate Justice (Glasgow, 2019) and to a meeting of the Fellows of the Academy of the Social Sciences in Australia (Adelaide, 2019). I am grateful to the organizers and audiences for their help in improving it, and to Jeremy Moss for especially helpful written comments.

References Betsill, M. M. & Bulkeley, H. 2004. Transnational Networks and Global Environmental Governance: The Cities for Climate Protection Program. International Studies Quarterly, 48, 471–493. Broome, J. 2012. The Public and Private Morality of Climate Change. Tanner Lecture. The University of Michigan. 16 March, 2012. Broome, J. 2019. Against Denialism. The Monist, 102, 110–129. Brousseau, E. & Dedeurwaerdere, T. 2012. Global Public Goods: The Participatory Governance Challenges. In: Brousseau, E., Dedeurwaerdere, T., Siebenhüner, B. et al. (eds) Reflexive Governance for Global Public Goods, Cambridge, Mass.: MIT Press. Carbonnier, G. & Grinevald, J. 2011. Energy and Development. In: Carbonnier, G. et al. (eds) International Development Policy: Energy and Development, Basingstoke: Palgrave Macmillan. Cullity, G. 1995. Moral Free Riding. Philosophy and Public Affairs, 24, 3–34.

28 Garrett Cullity (Adelaide)

Cullity, G. 2019. Climate Harms. The Monist, 102, 22–41. Fairfield, H. 2014. Best of Both Worlds? Northeast Cut Emissions and Enjoyed Growth. The New York Times. 6 June, 2014. Forman, S. & Segaar, D. 2006. New Coalitions for Global Governance: The Changing Dynamics of Multilateralism. Global Governance, 12, 205–226. Gardiner, S. M. 2001. The Real Tragedy of the Commons. Philosophy and Public Affairs, 30, 387–416. Global Humanitarian Forum (GHF) 2009. Human Impact Report: Climate Change – The Anatomy of a Silent Crisis. Geneva: Global Humanitarian Forum. Green, F. & Denniss, R. 2018. Cutting with Both Arms of the Scissors: the Economic and Political Case for Restrictive Supply-Side Climate Policies. Climate Change, 150, 73–87. Heijden, J. v. d. 2018. City and Subnational Governance: High Ambitions, Innovative Instruments and Polycentric Collaborations? In: Jordan, A., Huitema, D., van Asselt, H. & Forster, J. (eds) Governing Climate Change: Polycentricity in Action? Cambridge, UK: Cambridge University Press. Hiller, A. 2011. Climate Change and Individual Responsibility. The Monist, 94, 349–368. Intergovernmental Panel On Climate Change (IPCC) 2007. Climate Change 2007: Synthesis Report, New York: Cambridge University Press. Intergovernmental Panel On Climate Change (IPCC) 2014. Climate Change 2014: Impacts, Adaptation and Vulnerability, New York: Cambridge University Press. Intergovernmental Panel On Climate Change (IPCC) 2015. Climate Change 2014: Synthesis Report. Contribution of Working Groups I, II and III to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change. Geneva: IPCC. Johnson, B. 2003. Ethical Obligations in a Tragedy of the Commons. Environmental Values, 12, 271–287. Lin, J. 2018. Governing Climate Change: Global Cities and Transnational Lawmaking. Cambridge, UK: Cambridge University Press. Maltais, A. 2013. Radically Non-Ideal Climate Politics and the Obligation to at Least Vote Green. Environmental Values, 22, 589–608. Mendelsohn, R. 2011. Economic Estimates of the Damages Caused by Climate Change. In: Dryzek, J. S., Norgaard, R. B. & Schlosberg, D. (eds) The Oxford Handbook of Climate Change and Society. Oxford, Oxford University Press. Morgan-Knapp, C. & Goodman, C. 2015. Consequentialism, Climate Harm and Individual Obligations. Ethical Theory and Moral Practice, 18, 177–190. Petersson, B. 2013. Co-Responsibility and Causal Involvement. Philosophia, 41, 847–866. Sinnott-Armstrong, W. 2005. It’s Not My Fault: Global Warming and Individual Moral Obligations. In: Sinnott-Armstrong, W. & Howarth, R. (eds) Perspectives on Climate Change: Science, Economics, Politics, Ethics, Oxford: Elsevier. Stern, N. 2007. The Economics of Climate Change: The Stern Review. Cambridge, UK: Cambridge University Press. Stern, N. 2008. The Economics of Climate Change. The American Economic Review, 98, 1–37. Van De Poel, I., Fahlquist, J., Doorn, N., Zwart, S. & Royakkers, L. 2012. The problem of many hands: Climate change as an example. Science and Engineering Ethics, 18, 49–67.

2 SUB-NATIONAL CLIMATE DUTIES Addressing three challenges1 Lachlan Umbers (UWA)

It is widely accepted that major industrialised nations have generally failed to take adequate steps to respond to climate change.2 Set against this, climate action at the sub-national level has a surprisingly long history (Bulkeley, 2013). In 1990, for example, the city of Toronto committed to reducing its net emissions to 20% below 1988 levels by 2005 (Harvey, 1993). By 2017, the city’s emissions had fallen to 44% below 1990 levels (City of Toronto, 2017, p. 3). In 2003, the state government of New South Wales introduced the world’s first mandatory emissions trading scheme which, by 2010, was estimated to have either reduced or offset the equivalent of over 90 million tonnes of carbon emissions (CO2e, hereafter) (Perdan and Azapagic, 2011, p. 6045).3 Following the introduction of the California Global Warming Solutions Act in 2006, California’s emissions have steadily declined each year since 2007 (Board, 2018). British Columbia introduced a carbon tax in 2008, which recent studies indicate has been responsible for a 5–15% drop in emissions (Murray and Rivers, 2015, pp. 677–680). Associated with a steep decline in power plant emissions since its introduction, the Regional Greenhouse Gas Initiative (RGGI) is a cap-andtrade system regulating CO2 emissions from power plants in Connecticut, Delaware, Maine, Maryland, Massachusetts, New Hampshire, New York, Rhode Island, and Vermont (Murray and Maniloff, 2015). Recently, many American states and cities responded to President Trump declaring his intention to withdraw the United States from the Paris Climate Agreement by committing to taking unilateral action on climate change, even in the absence of an adequate national-level climate policy.4 And there are many examples besides these to be found. Few would dispute that climate action at the sub-national level is valuable, from a moral point of view.5 The purpose of this piece, however, is to defend a stronger and more contentious claim: that many sub-national political communities (i.e. states like California and Queensland, and cities or municipalities like Tokyo and London) have moral duties to take unilateral steps to reduce their emissions, even in the absence of

30 Lachlan Umbers (UWA)

adequate national-level climate policies. Jeremy Moss and I have set out a positive case for such duties elsewhere (see Umbers and Moss, 2018, 2020). This chapter considers and responds to three significant objections that might be raised against this idea. I shall begin by briefly recapitulating the case for such duties. I shall then consider, in turn, what seem to me to be the three most important potential lines of objection – the appeal to uncertainty, the appeal to unfairness, and the appeal to the possibility of carbon leakage. Each raises interesting and important issues. None, however, successfully undermines the claim that many sub-national political communities have duties to take unilateral steps to help mitigate climate change. Four preliminary issues ought to be addressed before we go on. First, it is important to clarify that my aim in this piece is only to defend the claim that many sub-national political communities have pro tanto duties to take steps to reduce their emissions. At least in principle, however, these duties might be outweighed by other considerations all things considered. What each sub-national political communities ought to do at any given time will be a matter of balancing the reasons which apply to it at that time. Since these will vary very widely community-tocommunity, I will have little to say on the matter in what follows. Second, this piece defends the claim that sub-national political communities have duties to reduce their emissions to help combat climate change. That, however, is not intended to detract from the idea that action at the national level – ideally as part of some fair, enforceable global agreement – represents the ideal response to the climate crisis. Indeed, as the following section makes clear, the argument I shall defend here applies only to sub-national political communities belonging to nations that have failed to adopt adequate climate policies. Third, the arguments that follow are intended to apply only to sub-national political communities that belong to wealthy, developed nations. In developing nations, there will often be a range of complicating considerations (e.g. potential trade-offs between emissions-reductions and the alleviation of severe poverty) which will render the case for such duties much more difficult to make out. For simplicity’s sake, then, I set such cases aside. Finally, I will be defending the claim that some sub-national political communities have collective duties to reduce their emissions. Collective duties, in this sense, are duties which apply to collectives qua collectives, rather than to the members of such collectives qua individuals (though, of course, collective duties will often have implications for the duties of individuals).6 As such, the argument to follow presupposes that such communities constitute collective agents. Moral duties, after all, are duties to act, which only agents can do. I cannot defend this claim in detail, here. However, it is widely accepted that states constitute collective agents, given that they seem to possess the features in virtue of which we generally ascribe agency to natural persons – a continuous identity over time, and (through their legislative and executive organs) the capacity to deliberate over, adopt, revise, and pursue particular goals and ends (Goodin, 1995, ch. 2, Erskine, 2001). Sub-national political communities typically possess these same features. I shall therefore assume that such arguments also apply mutatis mutandis to sub-national political communities.

Sub-national climate duties 31

I. Sub-national political communities’ devolved duties to reduce their emissions: a recapitulation In this section, I will briefly and informally recapitulate the positive case – set out in detail in Umbers and Moss (2020) – in favour of the claim that many subnational political communities have duties to take unilateral steps to reduce their emissions, where the states to which they belong fail to adopt adequate nationallevel emissions-reduction policies. It is uncontroversial that many nations have duties to reduce their emissions to help mitigate the threat posed by climate change. Philosophers have appealed to a range of different considerations in defence of such duties.7 Perhaps the two leastcontentious, however, are contribution and capacity.8 Proponents of contribution-based national-level duties to mitigate climate change appeal to the widely accepted claim that agents have general duties not to avoidably cause wrongful harm to others, as well as to the familiar corollary that where agents violate such duties, they have special duties to (among other things) mitigate the harm for which they are responsible. A factory-owner who dumps toxic waste into a river and causes children downstream to develop a severe illness, for instance, would clearly have duties to (again, among other things) clean up the river and cover the children’s medical expenses. Similarly, then, proponents of contribution-based duties to mitigate climate change hold that those wealthy, industrialised nations who have emitted vastly excessive quantities of greenhouse gases and thereby contributed substantially to the harms of climate change have special duties to mitigate those harms by reducing their emissions (e.g. Duus-Otterström, 2014, Moss and Kath, 2019). Proponents of capacity-based duties to mitigate climate change, by contrast, begin by appealing to the idea that agents have positive duties to mitigate serious harms where they have the capacity to do so without unacceptable cost. Singer (1972), for example, asks us to imagine that we are standing next to a pond in which a child is drowning whom we could rescue at the cost of ruining our shoes. There is a powerful intuition, here, that we would have a duty to do so. The child’s most fundamental interests are under threat, while the costs of averting that threat are very minimal. Proponents of national-level capacity-based duties to mitigate climate change (e.g. Page, 2011, pp. 417–420), then, point out that climate change has caused, and will cause, a range of serious harms. Emissions-reductions, particularly by major polluters like the United States, would curb the severity of climate change, mitigating these harms. Achieving these reductions may be costly, but not unacceptably so, given the magnitude of the harms they might help to avoid. Let us assume that some version of each of these accounts is plausible. In that case, each nation will have duties to reduce its emissions in proportion to the degree to which the considerations which give rise to these duties apply to it (other things equal). Moreover, the sub-national political communities of which each is comprised will have duties to take on a share of the costs of implementing those policies in proportion to the degree to which the relevant considerations apply to them.9 Other things equal, sub-national political communities with a larger volume

32 Lachlan Umbers (UWA)

of historical emissions would be required to take on a greater share of the costs of a national-level response to climate change than communities with a smaller volume of historical emissions, for instance. Were nations to adopt adequate emissions-reduction policies, there would be (relatively) few interesting questions to ask concerning sub-national political communities’ climate duties. Presumably, an adequate national-level climate policy would be one which (a) took on an appropriate share of the global burden of responding to climate change, and (b) distributed the costs fairly among the agents belonging to the nation in question, sub-national political communities included. In that case, all sub-national political communities would be required to do with respect to climate change (or, at least, with respect to climate change mitigation) would be to carry out the functions assigned to them under the policy. Yet that, of course, is not the circumstance in which the vast majority of such communities find themselves. Nations, by and large, have failed to adopt adequate climate policies. Given this, I argue, sub-national political communities have what I shall refer to as devolved duties to take steps to unilaterally reduce their emissions to help mitigate the threat posed by climate change. Consider the following case. Ten on the Shore: A company is holding its Christmas party down at the seaside. The company comprises a CEO, a four-person service team, and a five-person sales team. During the party, they see a dinghy capsize, plunging three victims into the water. There are two boats on the shore: a large boat which requires ten people to row and could carry all three victims to safety, and a smaller boat which requires five people to row but could carry only one of the victims to safety. There is only time for one boat trip before the victims drown. The CEO and the service team refuse to act. Nothing can be done to change their minds. The refusal of the CEO and service team to participate in a company-wide rescue attempt makes a profound difference to the duties of the sales team. Were the CEO and service team willing to participate in a rescue attempt, the sales team would obviously be obliged to join in. The company, clearly, has a collective duty to attempt to rescue the victims grounded in the interests of the victims in receiving aid, and their capacity to assist them. The sales team’s duties to participate in such a rescue would be grounded in the same considerations; the interests of the victims in receiving aid, and their capacity to help promote those interests. Clearly, the optimal response to such considerations on the part of the sales team would be to participate in a company-wide rescue attempt. That way, all three victims might be rescued. Yet the failure of the others to co-operate means, effectively, that this course of action is unavailable. As such, the appropriate response to the considerations which apply to the sales team given the failure of the others to co-operate would clearly be to act independently, so as to partially discharge the original collective duty by rowing the smaller boat out to the victims and rescuing one of them.

Sub-national climate duties 33

The structure of Ten on the Shore is instructive. Here, there is a collective duty which applies to the company as a whole in virtue of certain considerations (the interests of the victims, and the company’s capacity to assist). Were the company sufficiently likely to discharge that duty, each of the ‘sub-collectives’ of which it is comprised would have had duties to take play their part in a collective-wide effort to discharge that duty. These duties, in turn, would have been grounded in the same considerations which give rise to the collective duty. The failure of the collective to adequate take steps to discharge that duty, however, gives rise to duties on the part of some sub-collective (the sales team) to take unilateral steps to at least partially discharge the original collective duty, grounded in the very same considerations which gave rise to that collective duty.10 The climate case has a markedly similar structure. Wealthy industrialised nations have duties to reduce their emissions to help mitigate climate change that they are (by and large) presently failing to discharge. The considerations which ground those duties also apply to many of the sub-national political communities of which the nations in question are comprised. Many such communities are significant contributors to climate change in their own right. The United States Environmental Protection Agency (2016) reports that Texas, for example, emitted 708.81 million metric tons of CO2 in 2014. If Texas were a country, it alone would have been the 8th largest emitter of CO2, worldwide, in that year (Olivier et al., 2016, pp. 42–43). Many also plausibly have the capacity to make some difference to the problem of climate change by reducing their emissions – either in virtue of the potential direct effects of such reductions upon the climate, or in virtue of the indirect effects that such reductions might have in spurring other agents to take similar steps (see below). The ideal response to these considerations on the part of sub-national communities would be to play their assigned roles in an adequate national-level response to climate change. A national-level response of this sort would almost certainly be fairer and more effective than unilateral action at the sub-national level (see below). Yet this simply is not the situation in which many sub-national political communities find themselves. And, in virtue of that fact, such communities plausibly have devolved duties to take unilateral steps to partially discharge the climate duties of the nations to which they belong by taking unilateral steps to reduce their emissions.

II. Uncertainty The most obvious response to this line of argument points to a disanalogy between the examples which motivate contribution- and capacity-based duties, and the case of sub-national political communities and climate change. There is no non-trivial uncertainty as to the harms for which the factory-owner who pollutes the river is responsible. Nor is there any non-trivial uncertainty as to whether the agent standing next to the pond can prevent the child’s death. Yet there is substantial uncertainty as to the magnitude of the harms associated with many sub-national political communities’ emissions and (as such) the harm they might potentially

34 Lachlan Umbers (UWA)

avert by reducing them. After all, while some sub-national political communities are major contributors to climate change (e.g. Texas), most are not. In 2013, for example, the city of Auckland produced 10.4 million metric tonnes of CO2e (CDP, 2018) – far less than Texas, or any major industrialised nation. This objection fails. Consider the following cases. Two Shooters: Bonnie and Clyde each fire a single shot at a victim’s legs. The victim is severely injured as a result. Later, they discover that only one of the guns was loaded with a live round. The other was loaded with a blank. There is no way to determine which gun fired which round. As such it cannot be determined which shooter actually caused the injury to the victim. Branch: While out hiking, a large tree branch falls on David, causing him severe pain. The branch has pinned David’s arms to the ground, such that he cannot shift the branch, himself. George comes across David. He is unsure as to whether he has the strength to lift the branch. On the basis of the available evidence, however, he justifiably believes that if he attempts to do so, there is a 0.5 probability that he will succeed, and thereby alleviate David’s suffering. Both Two Shooters and Branch feature uncertainty.11 Bonnie and Clyde are both uncertain as to whether they are (individually) causally responsible for the harm suffered by the victim. George is uncertain as to whether he can alleviate David’s suffering by lifting the branch. Yet, intuitively, both Bonnie and Clyde, and George, have duties to have duties to take steps to attempt to mitigate the harms in question, nevertheless. Notice, moreover, that in Two Shooters, the intuition that Clyde ought to take on a fair proportion of the costs of mitigating the harm to the victim persists even if we specify that, as a matter of fact, Bonnie’s gun was loaded with the live round. After all, for all Clyde knows, there is a 0.5 probability that his action caused a severe injury to the victim. Similarly, even if we specify that, as a matter of fact, George will be unable to lift the branch, the intuition that George ought to attempt to do so persists. After all, for all George knows, there is a 0.5 probability that, by attempting to lift the branch, he will succeed, thereby alleviating David’s suffering. These cases illustrate that contribution and capacity-based duties are (or, at least, may be) grounded in considerations of expected value and disvalue. Both Bonnie and Clyde have contribution-based duties to mitigate the harm done to the victim because both performed actions with a sufficiently high probability of causing serious harm to the victim. George has a capacity-based duty to attempt to lift the branch off David because the evidence available to him suggests that he has a sufficiently good probability of substantially diminishing the suffering David is enduring. The mere fact of uncertainty, then, simply does not show that the considerations which ground national-level duties to mitigate climate change do not apply to the sub-national political communities of which they are comprised. The standard tool used by governments and economists for the purpose of estimating the expected disvalue of emissions is the so-called ‘social cost of carbon’,

Sub-national climate duties 35

estimates of which aim to put an approximate dollar figure on the costs (over a given time period) associated with each tonne of CO2e emitted.12 Nordhaus (2017), whose estimates are among the more conservative, estimates the social cost of CO2e emissions from the present to 2050 at $31 (at the 2010 value of the US dollar) per tonne.13 Accepting this estimate as correct, the expected costs of Texas’ 2014 CO2 emissions are roughly $21.9bn. And Texas is hardly the only plausible case. In 2014, the city of Tokyo emitted 70.1 million metric tons of CO2e (CDP, 2018) at an expected cost of $2.17bn. In 2015, New York City emitted 52 million metric tons of CO2e (LLC et al., 2017, p. 13) at an expected cost of $1.6bn. Even Auckland’s 2013 emissions have a decidedly non-trivial expected cost of $322m.14 As such, there are plausible reasons to think that there will in fact be many subnational political communities to whom the considerations which ground contribution-based duties apply.15 The magnitude of these expected harms, moreover, suggests that even relatively modest emissions-reduction efforts might well have substantial expected value (and thus give rise to capacity-based duties). Accepting Nordhaus’ estimate once again, a mere 5% reduction in Auckland’s 2013 emissions, for instance, would have been expected to save $16.1m in social costs.16 Moreover, it is also important to note that emissions-reduction policies may have other valuable effects. In particular, it may be that unilateral action by sub-national political communities helps contribute to a process of ‘policy diffusion’. Policy diffusion occurs where some government’s adoption of some policy increases the likelihood of the same or similar policies being adopted by other governments.17 Such effects have been observed in many different contexts, including at the subnational level. In a widely cited study of the diffusion of anti-smoking legislation in the US, for example, Shipan and Volden (2006) found that, under certain conditions, the adoption of anti-smoking legislation by municipal governments made the adoption of similar legislation at the state level significantly more likely, and that the adoption of such legislation by some states made neighbouring states significantly more likely to adopt similar policies. We might reasonably hope, then, that unilateral action on climate change by some sub-national political communities might render it more likely that others will adopt similar policies. A number of recent empirical studies offer reason for hope, on this score.18 In separate studies of the adoption of climate and energy policies by US states, for example, Bromley-Trujillo et al. (2016, pp. 551–553, 559) and Chandler (2009, pp. 3278–3279) find the adoption of such policies by ideologically similar states to be among the principal drivers of states’ decisions to adopt similar policies, themselves.19 At the municipal level, Krause (2010) finds that the presence of neighbouring cities who have enacted climate protection policies is among the principal drivers of cities’ decisions to adopt climate protection initiatives, themselves.20 The possibility of such effects clearly adds to the expected value of emissions-reduction efforts at the subnational level.21 The appeal to uncertainty, then, fails to undermine our argument. Nations’ contribution- and capacity-based duties to help mitigate climate change by

36 Lachlan Umbers (UWA)

reducing their emissions arise in virtue of the expected disvalue of their emissions, and the expected value of policies they might adopt to reduce their emissions. There are many sub-national political communities to whom these same considerations apply, and who might plausibly have devolved duties to unilaterally reduce their emissions in consequence.

III. Unfairness Another obvious line of objection would appeal to the fact that – even allowing for policy diffusion effects – not all (perhaps not even most) sub-national political communities will take serious steps to reduce their emissions in the absence of a national-level climate policy compelling them to do so. Perhaps, then, it is unfair to demand that any sub-national political community take such steps unilaterally. After all, by doing so, such communities will effectively deprive themselves of the right to emit some quantity greenhouse gases, thereby imposing costs upon themselves (and their citizens) which others will avoid. Yet issues of distributive fairness arise only with respect to the distribution of goods to which the agents in question have claims (cf. Broome, 1990–1991). Suppose two thieves rob a bank. The next evening one thief steals the others’ share of the money. We should hardly wish to say that the latter would thereby have a fairness-complaint against the former. After all, neither thief has any claim to the money they have stolen. Yet suppose, on the other hand, that two people are marooned on a desert island. If one monopolises all the food and water and refuses to share, the other would have an obvious fairnesscomplaint. Each has a positive, equal claim to the food and water necessary to survive. As such, it is obviously unfair that only one of the survivors should have those claims satisfied. Proponents of this line of objection, then, must necessarily presuppose that the communities to whom our argument is supposed to apply have claims to continue emitting greenhouse gases at roughly the rates at which they do at present. If they do not have such claims, no issue of unfairness can arise in the first place. Yet this straightforwardly begs the question. After all, the argument sketched above aims to establish that, far from having claims to continue emitting at their present levels, many sub-national political communities have duties to aggressively reduce their emissions in order to help mitigate the severity of climate change. A proponent of this line of objection might respond by pointing out that even if sub-national political communities do not have claims to continue emitting at present rates, serious emissions-reduction policies will plausibly involve substantial costs. Adopting such policies, then, may require sub-national political communities to divert resources from programmes for the provision of goods to which the communities in question do plausibly have moral claims (public health, education, etc…). Communities who fail to take such steps will not make any such sacrifices. This, prima facie, is unfair.

Sub-national climate duties 37

In response, it is first important to note that there are many examples of nontrivial emissions-reduction policies adopted by sub-national political communities that do not seem to have entailed particularly heavy costs. For example, studies indicate that British Columbia’s carbon tax has had little effect upon economic growth (Murray and Rivers, 2015, pp. 677–680). The states in which the RGGI is operative continue to enjoy rates of economic growth above the national average (Fairfield, 2014). It is also important to note that such policies will often be associated with important co-benefits which may offset their costs, to at least some degree. For example, steps to reduce carbon emissions may be associated with better public health outcomes from improved air quality and increased uptake of active modes of transport like cycling (Haines et al., 2009–2010). Neither of these points, of course, establishes that there will never be fairnesscomplaints of this sort.22 Yet it is important to notice that this in no way undermines the claim under consideration: that many sub-national political communities have pro tanto duties to take steps to unilaterally reduce their emissions. It is hardly as though complaints of unfairness somehow undermine the degree to which the considerations which give rise to such duties apply to the communities in question. If Diane and Elaine each dump equal quantities of toxic waste into a river, causing children downstream to develop a severe illness, for instance, both will have contribution-based duties to clean up the river and cover the children’s medical expenses. Yet suppose that Diane refuses to discharge these duties. That fact would hardly show that contribution-based considerations no longer apply to Elaine. Elaine may well have a fairness-complaint against Diane. Yet that would surely not show that she has no duty to help mitigate the harms for which she is responsible. Analogous considerations apply with respect to capacity-based duties. At most, then, this line of objection can show only that considerations of unfairness can sometimes weigh against sub-national political communities’ duties to take steps to address climate change. Yet, as Hohl and Roser (2011) have pointed out, given the importance of addressing climate change, it seems likely that fairness-based considerations of this sort will be heavily outweighed all-things-considered in the majority of cases.23 Sub-national political communities who take steps to reduce their emissions, then, may well take on costs that others will (wrongfully) avoid. This may give rise to issues of unfairness – and points to an important respect in which a uniform, national-level climate policy would be superior to unilateral action at the sub-national level. Yet, for the reasons outlined, this simply does not show that sub-national political communities do not have duties of the sort defended in this chapter.

IV. Carbon leakage A final line of potential objection appeals to the possibility of carbon leakage. The phenomenon of carbon leakage occurs where the adoption of emissions-reduction policies in one jurisdiction leads to increases in emissions elsewhere, thereby undermining the efficacy of such policies. Experts, generally, think that carbon

38 Lachlan Umbers (UWA)

leakage may occur for two kinds of reasons (Branger and Quirion, 2014, pp. 54–55). First, emissions-reduction policies may decrease the consumption of fossil fuels in some jurisdiction. This decreases global demand for fossil fuels and may thereby make it cheaper (and thus more attractive) for actors in other jurisdictions to increase their own fossil fuel consumption. Second, emissions-reduction policies such as carbon taxes may impose costs upon carbon-intensive industries. This may cause such industries to relocate from jurisdictions where emissions-reduction policies have been introduced to those in which they have not (or where less demanding policies are in place). Even if such industries do not relocate, the introduction of emissions-reduction policies may render their products less competitive, thereby leading to an increased demand for similar or substitute products produced in lessregulated jurisdictions. The possibility of carbon leakage is among the principal arguments deployed against aggressive unilateral emissions-reduction policies at the national level (see, e.g., Babiker, 2005). Theorists who have discussed unilateral climate action at the sub-national level – including those generally supportive of such action (e.g. Ostrom, 2010, p. 554–555) – often point out that such policies also risk causing carbon leakage. This possibility, in turn, might be thought to undermine the degree to which contribution- or capacity-based considerations can be said to apply to sub-national political communities. The matter is clearest with respect to capacity-based considerations. Capacitybased duties arise where an agent has the capacity to reduce levels of expected harm to some sufficiently substantial degree, without unacceptable cost. Carbon leakage, insofar as it undermines the efficacy of emissions-reduction policies, must clearly diminish the degree to which any such policies can diminish the expected harms of climate change. Suppose Ohio were contemplating introducing a carbon price. If their doing so is likely to simply cause emissions-intensive industries to relocate to some other, unregulated jurisdiction where they can continue emitting as before, the expected value of their doing so would clearly be undermined. Contribution-based duties, on the other hand, only arise where an agent has contributed to some harm, in a manner for which the agent in question is morally responsible. If Gerry dumps a barrel toxic waste into a river because Sandi forced him to do so at gunpoint, for example, Gerry is clearly not morally responsible for the ensuing harm. Nor will he have contribution-based duties to help mitigate the harm associated with his actions. Now, it is often assumed that for an agent to be morally responsible for some harm, they must be causally responsible for that harm. This, in turn, suggests that the possibility of carbon leakage might undermine the degree to which any sub-national political community can reasonably be said to be morally responsible for the expected harm with which their emissions are associated. Suppose, for instance, that had Ohio taken steps to reduce its emissions by 10% a decade ago, this would have precipitated an increase in emissions produced in other American states, such that Ohio’s doing so would have made no net difference to the total emissions produced by the United States. This, it might be suggested, entails that Ohio is not causally responsible for the harms with which

Sub-national climate duties 39

these emissions will be associated. The phenomenon of carbon leakage means that they would have been produced whether or not Ohio had taken steps to reduce its emissions. This implies that Ohio cannot be morally responsible for such emissions, either. And this, finally, entails that Ohio cannot have contribution-based duties to help mitigate the harms with which these emissions will be associated. This line of objection invites two sets of responses. First, the challenge to contribution-based duties I have outlined appeals to the claim that an agent A may be morally responsible for some harm H if and only if A is causally responsible for H. Yet, as Sartorio (2004, pp. 319–325) forcefully argues, views of this sort entail unacceptable results in a range of cases. Suppose, for instance, that Charles is tied to a stake. Alan shoots Charles in the heart, instantly killing him. Yet suppose also that had Alan failed to shoot Charles, Betty would have done so instead, and with the same effect. Simple counterfactual dependence accounts of causation (upon which this line of objection relies) entail that Alan fails to cause Charles’ death.24 Had Alan failed to shoot Charles, Betty would have done so, and Charles would have died regardless. As such, Charles’ death does not counterfactually depend upon Alan’s actions. We should hardly wish to accept that Alan is not thereby morally responsible for Charles’ death. Yet this seems to be precisely what is entailed by the view of the relationship between moral and causal responsibility which underpins this line of objection. Cases of this kind are familiar, and have lead many philosophers to propose more sophisticated accounts of the relationship between causal and moral responsibility. Amongst the more popular of these approaches, for instance, holds that an agent may be morally responsible for some harm where their actions constitute a necessary element of a set of factors sufficient for the harm in question, where the set itself need not be necessary for that harm.25 On this view, Alan will be morally responsible for Charles’ death, insofar as his shooting Charles would certainly have been sufficient to cause Charles’ death, even if (ex hypothesi) his doing so is not necessary for Charles’ death. Similarly, on such approaches, Ohio would certainly count as responsible for the expected harm associated with the emissions that would have ‘leaked’ to other American states. Their permitting such emissions to be produced within their territory is sufficient for the expected harm with which those emissions are associated, even if it is not necessary. Now, whether this approach is ultimately satisfactory is an issue we cannot explore in depth, here.26 Nor need we settle the issue. The point is simply that the view of the relationship between causal and moral responsibility which underlies this objection is implausible. Alan is clearly morally responsible for Charles’ death. We should reject any view which entails otherwise. And we should likewise reject any view which permits sub-national political communities (or any other agent for that matter) to evade moral responsibility for the emissions produced within their borders by appeal to the fact that some other community would have produced those emissions had they not done so.

40 Lachlan Umbers (UWA)

Secondly, and more straightforwardly, we might respond to both sets of challenges by questioning the degree to which carbon leakage is likely to undermine the efficacy of unilateral emissions-reduction policies at the sub-national level. For any emissions-reduction policy, the rate of carbon leakage with which it is associated is given by the increase in CO2e emissions elsewhere brought about by that policy, divided by the decrease in CO2e emissions brought about by that policy in the jurisdiction in which it applies. In a review of studies into the phenomenon, Branger and Quirion (2014, pp. 54–61) find that attempts to model the effects of emissions-reduction policies ex ante typically predict rates of carbon leakage of around 5–20%. Empirical studies of real-world emissions-reduction policies (e.g. the EU’s emissions-trading scheme), however, have tended to find limited evidence of such effects.27 Moreover, even if (as is at least possible) some potential emissions-reduction policies at the sub-national level give rise to a significant risk of carbon leakage, this may simply be an argument for careful policy design. Some industries will be simply unable to shift jurisdictions – companies which provide public transport in particular cities, for example. Others will be very unlikely to do so, given the costs (e.g. transport costs) associated with relocation.28 The risk of carbon leakage, then, might simply argue for carefully targeted emissions-reduction policies, rather than against unilateral action more generally. And, even if emissions-reduction policies at the sub-national level will necessarily be accompanied by substantial carbon leakage, there are steps such communities might take to address the issue. Namely, such communities might take steps to offset those ‘leaked’ emissions by – for instance – helping to funding re-forestation efforts or de-carbonisation projects in developing nations. Carbon leakage points to another important respect – efficacy – in which adequate national-level emissions-reduction policies are likely superior to unilateral action at the sub-national level. It does not in any obvious way, however, undermine the claim that there are pro tanto duties on the part of sub-national political communities to take unilateral steps to help mitigate climate where the nations to which they belong fail to adopt such policies.

V. Conclusion We have now considered what seem to me be the three strongest lines of objection to the idea that sub-national political communities have devolved duties to take steps to unilaterally reduce their emissions in the absence of adequate nationallevel climate policies. Each of these objections raises important issues. Yet none ultimately succeeds in undermining the positive case for such duties. We have good reason to maintain, then, that climate action at the sub-national level is – so long as adequate national-level climate policies are not forthcoming – not merely morally valuable, but morally required.

Sub-national climate duties 41

Notes 1 My thanks to Jeremy Moss for comments on an earlier version of this paper. 2 Victor et al. (2017), for example, present evidence to suggest that no major industrialised nation is presently on-track to meet their pledges under the Paris Agreement. 3 The scheme was closed down in 2012 to make way for the short-lived national carbon tax imposed by the Australian federal government. 4 For a partial overview of these efforts, see Mayors (2018). 5 Indeed, some have suggested that emissions-reductions at the sub-national level will be essential to meeting the Paris targets (e.g. Rogelj et al., 2016). 6 On this point, see Collins (2019; 2020). 7 For an overview, see Roser and Seidel (2017, chs 10–16). 8 These are not mutually exclusive. Many endorse duties of both kinds (e.g. Page, 2011, Roser and Seidel, 2017). 9 This is a natural extension of Collins’ (2016) persuasive account of the basis of individuals’ duties to take on the costs of discharging states’ duties. 10 Though it is important to note that, for the reasons set out in the introduction, duties of this kind will arise only where the sub-collective in question constitutes an agent in its own right. 11 My discussion here has been much influenced by Jackson (1991). 12 For an excellent discussion of these issues, see Broome (2019). 13 Others have produced credible estimates of up to $220 per tonne (Moore and Diaz, 2015). 14 It should be borne in mind, moreover, that these figures do not take into account the emissions associated with products produced elsewhere and consumed within the communities in question, or those associated with products exported from these communities. Plausible arguments have been offered to suggest that states also bear at least partial responsibility for such emissions (Afionis et al., 2017, pp. 3–6, Moss, 2015). Such arguments presumably apply mutatis mutandis to the sub-national case, also. 15 That is not to deny, of course, that there will be many cases in which the considerations in question do not apply to any non-trivial degree. The Danish town of Ærøskøbing emitted just 225 metric tons of CO2e in 2015, for example (CDP, 2018). 16 This estimate is rather imprecise of course, insofar as it takes no account of effects such as policy diffusion or carbon leakage, which can amplify or undermine the impact of emissions-reduction policies. 17 For an overview of the policy diffusion literature, see Shipan and Volden (2012). 18 For a partial overview of this literature, see Lyon (2016, pp. 145–7). 19 Matisoff and Edwards (2014), similarly, show that states with similar political cultures are more likely to emulate one another’s energy policies. 20 Though there is some scepticism as to whether this ‘geographical’ diffusion occurs at the state level. See Matisoff (2008). 21 There are several other possible indirect effects. See Umbers and Moss (2020, pp. 27–28). 22 My argument in this paragraph has been much influenced by Karnein (2014, pp. 599– 600). 23 Indeed, it seems rather more plausible that sub-national political communities might have duties to ‘take up the slack’ – i.e. to take on more costs than they would have been required to do if all sub-national political communities had taken steps to discharge their duties (Umbers and Moss, 2020, pp. 29-31). 24 The canonical statement of the counterfactual dependence account of causation is Lewis (1973). 25 This view is particularly associated with Hart and Honoré (1959), though their principal focus is legal, rather than moral responsibility. It has been taken up in more recent times by several philosophers. It is adopted, with numerous additional conditions, by Braham and Hees (2012), for example.

42 Lachlan Umbers (UWA)

26 For some scepticism, see Barry and Øverland (2015, pp. 174–6). 27 Though, as Branger and Quirion (2014, p. 60) note, this may be because real-world emissions-reduction policies have often been rather modest in their ambitions, and have also often been accompanied by costly efforts to offset the costs of such policies to industry. 28 Barker et al. (2007) hypothesise that this partially accounts for the relatively low observed rates of carbon leakage from unilateral steps taken by European countries.

Bibliography Afionis, S., Sakai, M., Scott, K., Barrett, J. & Gouldson, A. 2017. Consumption-based carbon accounting: does it have a future? Wiley Interdisciplinary Reviews: Climate Change, 8, 1–19. Agency, U. S. E. P. 2016. State Energy CO2 Emissions [Online]. Available: https://19january 2017snapshot.epa.gov/statelocalclimate/state-energy-co2-emissions_.html. Babiker, M. H. 2005. Climate change policy, market structure, and carbon leakage. Journal of International Economics, 65, 421–445. Barker, T., Junankar, S., Pollitt, H. &Summerton, P. 2007. Carbon leakage from unilateral Environmental Tax Reforms in Europe, 1995–2005. Energy Policy, 35, 6281–6292. Barry, C. &Øverland, G. 2015. Individual responsibility for carbon emissions: is there anything wrong with overdetermining harm? In: Moss, J. (ed.) Climate Change and Justice, Cambridge, UK: Cambridge University Press. Board, C. A. R. 2018. California Greenhouse Gas Emissions for 2000 to 2016. California Air Resources Board. Braham, M. &Hees, M. V. 2012. An Anatomy of Moral Responsibility. Mind, 121, 601–634. Branger, F. &Quirion, P. 2014. Climate policy and the ‘carbon haven’ effect. Wiley Interdisciplinary Reviews: Climate Change, 5, 53–71. Bromley-Trujillo, R., Butler, J. S., Poe, J. &Davis, W. 2016. The Spreading of Innovation: State Adoptions of Energy and Climate Policy. Review of Policy Research, 33, 544–565. Broome, J. 1990–1991. Fairness. Proceedings of the Aristotelian Society, 91, 87–101. Broome, J. 2019. Against Denialism. The Monist, 102, 110–129. Bulkeley, H. 2013. Cities and Climate Change, Routledge. CDP. 2018. 2016 – Citywide GHG Emissions [Online]. Available: https://data.cdp.net/ Cities/2016-Citywide-GHG-Emissions/dfed-thx7/data. Chandler, J. 2009. Trendy solutions: Why do states adopt Sustainable Energy Portfolio Standards? Energy Policy, 37, 3274–3281. City of Toronto, E. A. E. D. 2017. TransformTO Implementation Update. Toronto, CA: City of Toronto, Environment and Energy Division. Collins, S. 2016. Distributing States’ Duties. The Journal of Political Philosophy, 24, 344–366. Collins, S. 2019. Group Duties, Oxford, UK: Oxford University Press. Collins, S. 2020. Corporations’ Duties in a Changing Climate. In: Umbers, L. & Moss, J. (eds.) Climate Justice: Corporations, Regions, Cities, Individuals, Routledge. Duus-Otterström, G. 2014. The problem of past emissions and intergenerational debts. Critical Review of International Social and Political Philosophy, 17, 448–469. Erskine, T. 2001. Assigning Responsibilities to Institutional Moral Agents: The Case of States and Quasi-States. Ethics & International Affairs, 15, 67–85. Fairfield, H. 2014. Best of Both Worlds? Northeast Cut Emissions and Enjoyed Growth. The New York Times, 6 June 2014.

Sub-national climate duties 43

Goodin, R. E. 1995. Utilitarianism as a Public Philosophy, Cambridge, UK: Cambridge University Press. Haines, A., McMichael, A. J., Smith, K. R., Roberts, I., Woodcock, J., Markandya, A., Armstrong, B. G., Campbell-Lendrum, D., Dangour, A. D., Davies, M., Bruce, N., Tonne, C., Barrett, M. &Wilkinson, P. 2009–2010. Public health benefits of strategies to reduce greenhouse-gas emissions: overview and implications for policy makers. The Lancet, 374, 2104–2114. Hart, H. L. A. & Honoré, T. 1959. Causation in the Law, Oxford, UK: Oxford University Press. Harvey, L. D. D. 1993. Tackling urban CO2 emissions in Toronto. Environment, 35, 16–20, 38–44. Hohl, S. &Roser, D. 2011. Stepping in for the Polluters? Climate Justice under Partial Compliance. Analyse & Kritik, 33, 477–500. Jackson, F. 1991. Decision-theoretic Consequentialism and the Nearest and Dearest Objection. Ethics, 101, 461–482. Karnein, A. 2014. Putting Fairness in Its Place: Why There Is a Duty to Take Up the Slack. Journal of Philosophy, 111, 593–607. Krause, R. M. 2010. Policy Innovation, Intergovernmental Relations, and the Adoption of Climate Protection Initiatives by U.S. Cities. Journal of Urban Affairs, 33, 45–60. Lewis, D. 1973. Causation. The Journal of Philosophy, 70, 556–567. LLC, C., Pasion, C., Oyenuga, C. &Gouin, K. 2017. Inventory of New York City Greenhouse Gas Emissions in 2015. New York, NY: Mayor’s Office of Sustainability. Lyon, T. P. 2016. Drivers and Impacts of Renewable Portfolio Standards. Annual Review of Resource Economics, 8, 141–155. Matisoff, D. C. 2008. The Adoption of State Climate Change Policies and Renewable Portfolio Standards: Regional Diffusion or Internal Determinants? Review of Policy Research, 25, 527–546. Matisoff, D. C. & Edwards, J. 2014. Kindred spirits or intergovernmental competition? The innovation and diffusion of energy policies in the American states (1990–2008). Environmental Politics, 23, 795–817. Mayors, C. 2018. Climate Mayors: Climate Action Compendium [Online]. Available: http:// climatemayors.org/actions/climate-action-compendium/. Moore, F. C. & Diaz, D. B. 2015. Temperature impacts on economic growth warrant stringent mitigation policy. Nature Climate Change, 5, 127–131. Moss, J. 2015. Exporting harm. In: Moss, J. (ed.) Climate Change and Justice, Cambridge, UK: Cambridge University Press. Moss, J. &Kath, R. 2019. Historical Emissions and the Carbon Budget. Journal of Applied Philosophy, 36, 268–289. Murray, B. &Rivers, N. 2015. British Columbia’s revenue-neutral carbon tax: A review of the latest ‘grand experiment’ in environmental policy. Energy Policy, 86, 674–683. Murray, B. C. & Maniloff, P. T. 2015. Why have greenhouse emissions in RGGI states declined? An econometric attribution to economic, energy market, and policy factors. Energy Economics, 51, 581–589. Nordhaus, W. D. 2017. Revisiting the social cost of carbon. Proceedings of the National Academy of Sciences of the United States of America, 114, 1518–1523. Olivier, J. G. J., Janssens-Maenhout, G., Muntean, M. &Peters, J. A. H. W. 2016. Trends in Global CO2 Emissions:2016Report. The Hague PBL Netherlands Environmental Assessment Agency; Ispra: European Commission, Joint Research Centre. Ostrom, E. 2010. Polycentric systems for coping with collective action and global environmental change. Global Environmental Change, 20, 550–557.

44 Lachlan Umbers (UWA)

Page, E. A. 2011. Climatic Justice and the Fair Distribution of Atmospheric Burdens: A Conjunctive Account. The Monist, 94, 412–432. Perdan, S. &Azapagic, A. 2011. Carbon trading: Current schemes and future developments. Energy Policy, 39, 6040–6054. Rogelj, J., Elzen, M. D., Höhne, N., Fransen, T., Fekete, H., Winkler, H., Schaeffer, R., Sha, F., Riahi, K. &Meinshausen, M. 2016. Paris Agreement climate proposals need a boost to keep warming well below 2 °C. Nature, 534, 631–639. Roser, D. &Seidel, C. 2017. Climate Justice, New York, NY, Routledge. Sartorio, C. 2004. How To Be Responsible For Something Without Causing It. Philosophical Perspectives, 18. Shipan, C. R. & Volden, C. 2006. Bottom-Up Federalism: The Diffusion of Antismoking Policies from U.S. Cities to States. American Journal of Political Science, 50, 825–843. Shipan, C. R. & Volden, C. 2012. Policy Diffusion: Seven Lessons for Scholars and Practitioners. Public Administration Review, 72, 788–796. Singer, P. 1972. Famine, Affluence, and Morality. Philosophy & Public Affairs, 1, 229–243. Umbers, L. M. & Moss, J. 2018. Going it Alone: Cities and States for Climate Action. Ethics, Policy & Environment, 21, 56–59. Umbers, L. M. & Moss, J. 2020. The Climate Duties of Sub-National Political Communities. Political Studies, 68, 20–36. Victor, D. G., Akimoto, K., Kaya, Y., Yamaguchi, M., Cullenward, D. &Hepburn, C. 2017. Prove Paris was more than paper promises. Nature, 548, 25–27.

3 CARBON MAJORS AND CORPORATE RESPONSIBILITY FOR CLIMATE CHANGE Jeremy Moss (UNSW)

1 Introduction Most discussion of the responsibility for addressing climate change centres on the duties of states, and for good reason. States negotiate climate agreements and their policies on climate change are also substantially responsible for the contribution that their (and other) citizens make to the problem of climate change. States also have substantial capacity to compel agents such as corporations or individuals to reduce their level of emissions. But in the context where states are not doing nearly enough to address climate change, attention is increasingly turning to the contribution that corporations make to climate change and what this ought to mean for how we assign the burdens of reducing emissions. This is particularly the case for major fossil fuel producing corporations or ‘carbon majors’. The actions of carbon majors have attracted attention for two main reasons. The first pertains to the huge role they have played in contributing to climate change and the moral responsibility that this implies. Carbon majors have produced an enormous volume of fossil fuels that have led to a high percentage of global emissions (Frumhoff et al., 2015). Heede (2014) claims that 63% of global emissions over the period 1854–2010 can be attributed to ninety large fossil fuel and cement producing corporations. In many cases, their actions have also thwarted robust action on climate change through their watering down or forcing the abandonment of effective climate policies, as well as through lobbying for subsides that make their business more competitive (Rich, 2018). These latter activities are often directed not just at sustaining the use of fossil fuels and making them profitable, but at actively delaying the take up of renewable alternatives. But interest in carbon majors has now increased for a second reason: what ought to be demanded of them in response? Advocates for climate change have

46 Jeremy Moss (UNSW)

demanded that carbon majors cease their exploration for new reserves, stop their extensive lobbying of all levels of government, divest themselves of their fossil fuel assets and, of course, phase out the production of fossil fuels. Understandably, the first of these issues has attracted the most attention. But given the urgent and high-stakes nature of addressing climate change, the issue of what ought to be done by carbon majors is now at least as important, and arguably more important, than the issue of their responsibility. In what follows, I will firstly discuss what I take to be the most plausible framework for understanding why the carbon majors are morally responsible. While it might seem obvious that carbon majors are morally responsible for their actions, it is important to be clear why they are responsible. For example, while they produce huge quantities of fossil fuels, for the most part they do not consume them. Through actions such as driving cars or producing electricity by burning coal, other agents are directly causally responsible for producing emissions. Carbon majors are certainly a major part of the causal chain that leads to the direct production of emissions but we need to understand how much responsibility their role in that chain confers. The discussion presented here is not intended to be exhaustive, but merely outlines a plausible framework for moral responsibility that ought to be set out before addressing any consequences. Having outlined why carbon majors bear some degree of responsibility, I will then outline what actions they ought to take to address their responsibility for their role in contributing to the harms of climate change. Given the limited time in which we have to take action, – a 45% emission reduction over the next 10 years according to a recent IPCC report – this is an urgent question (IPCC, 2018). In what follows I will focus on carbon majors and not corporations more broadly. Carbon majors are the most obvious case of companies who are likely to be liable for climate harms and there is also a great deal of empirical research on the kinds of contribution that they make. I will also assume that corporations are moral agents in the relevant sense and can be held responsible for their actions (Pettit, 2017). While I cannot go into details here, I adopt the argument put forward by Pettit that corporations are agents because they exhibit three features: they have purposes, form reliable representations of their environment and act reliably to satisfy those purposes in light of the representations that they produce. Carbon majors clearly exhibit these features. Carbon majors have a very sophisticated understanding of the markets in which they operate and clearly demonstrate their ability to react and plan for different market conditions, often over very long time frames. Their purposes are clearly set out in shareholder documents and reinforced by their management structure and decisions. The duties that I will be arguing that corporations ought to take will be pro tanto duties and I will mostly leave aside the issue of to whom the duties ought to be discharged. I will also focus on the duties that arise for carbon majors as a result of their causing harm. They might also have duties derived from other sources, such as their ability to address climate change or their having profited from harm, but I leave those arguments aside.

Carbon majors and corporate responsibility for climate change 47

1 The contribution of carbon majors The term ‘carbon majors’ refers to the large fossil fuel and cement producing corporations. Heede (2014) identifies a total of ninety companies that have produced fuels that have led to 63% of the world’s emissions between 1854 and 2010. The corporations are made up of those which are investor-owned corporations (fifty), state-owned (thirty-one) and nation state producers of fossil fuels (nine). The three different groups all share roughly one third of the emissions with the state-owned group accounting for slightly less. Oil and gas companies make up the majority of the carbon majors (fifty-six), coal comes in second with thirty-seven and seven are cement producers.1 While their contribution to the total amount of global emissions is huge, we need to do a bit of ‘moral mathematics’ to establish a framework of attributing responsibility. The most plausible way of understanding how carbon majors are responsible is because their actions harm others via their production and sale of fossil fuels. The duties to others that follow from these particular harmful contributions are grounded in the more general negative duty not to cause wrongful harm. Following Shue (1981), an agent wrongfully harms another agent where (1) their act sets back an agent’s significant interests, (2) they had reasonable options to do otherwise (3) knew or should have known the likely consequences of their actions in relation to the interests of other agents and (4) where they made a contribution.2 Such negative duties not to harm others are powerful and important constraints on our actions. Moreover, it follows that if one agent wrongfully harms another, that the agent has pro tanto duties to cease the action and perhaps assume compensatory duties. The harms that the carbon majors contribute to are of two sorts. There are of course the direct harms that result from climate change; increased severe weather events, the spread of disease and other impacts that have been well documented. But there is also the harm of denying others the opportunity to emit GHGs. This is an important dimension of harm if the well-being of current and future individuals is lessened because they cannot emit GHGs. Individuals who are forced to forego inexpensive power or transport options will be less well off than those who were able to access such high emitting goods in the past. It may well also cost them more to access the same goods. A recent analysis by Carbon Brief claims that if temperature rises are to be kept to 1.5C, then younger generations will only be able to emit around 43 tonnes of lifetime CO2, around eight times less than someone born before 1950 (Hausfather, 2019). If carbon majors have violated a duty not to cause harm, it is not immediately clear precisely how they have done so. It is not always clear whether carbon majors are morally or causally responsible for all or a significant proportion of the emissions that are produced from the fossil fuels (and cement products) that they extract and sell (Frumhoff et al., 2015, Heede, 2014, Heede and Oreskes, 2016). Consider, for example, that CO2 is released into the atmosphere via the burning of fossil fuels – not primarily by their extraction from the earth. In what is perhaps the best known article on carbon majors and responsibility, Frumhoff et al. note that while actors such as governments, emitting industries and individuals bear responsibility as well as

48 Jeremy Moss (UNSW)

carbon majors, they do not articulate the responsibility that attaches specifically to producers of fossil fuels (Frumhoff et al., 2015, p. 167). Henry Shue writes that we should hold carbon majors responsible ‘in proportion to their contribution’ (Shue, 1981, p. 594). More emphatic is the claim by Ekwurzel et al. (2017) that 4 C of global mean surface temperature rise (out of a total rise of 1.1C) can be traced to the emissions of carbon majors between 1880 to 2010. Similarly, they claim that around 32% of sea level rise (5.7cm) can be traced to carbon majors. What is needed is a way of determining the degree of responsibility carbon majors ought to bear given their contribution to producing GHGs. A useful way of understanding how carbon majors are substantially morally responsible is to divide the agents into principal and secondary agents. Where a harmful action occurs, the agent(s) who do the harming are the principal agents. The actions of a principal agent or agents constitute the wrongful harm. For example, in the case where one agent murders another by shooting them, the principal agent is the person who pulled the trigger. Secondary agents on the other hand, contribute to the harm by performing actions that are a part of the causal chain leading to the harmful act but do not in themselves constitute the harmful act (Gardner, 2007, Kadish, 1985, Kutz, 2007, Lepora and Goodin, 2013). In the murder case, a secondary agent might have sold the gun to the principal agent, when they didn’t have to do so and knowing what the gun would be used for. The actions of the gun dealer in this case are ones that make her complicit in the harm committed by the principal agent. In terms of how principal and secondary agents contribute to climate change via emitting GHGs, the simplest way in which an agent can contribute to climate harms is by directly producing emissions through their actions, such as by driving cars, or by an institutional or corporate agent operating a coal power plant. The direct emitters are the principal agents. As we noted, the contributions that carbon majors make to climate change via their production of fossil fuels are indirect. While they can produce significant direct emissions, mostly they contribute by being producers and suppliers of fossil fuels (‘upstream producers’ as they are often called). By providing indirect support a secondary agent becomes an accomplice or accessory and shares some of the liability for the harms done because they assisted or aided the principal agent in some way. The badness of the harm committed by the secondary agent derives from the harm of the principal agent. So, if a carbon major produces and sells fossil fuels that are in turn consumed by a power station, the carbon major is complicit as a secondary agent in the harm caused by the principal agent via, in this case, the direct releasing of GHGs into the atmosphere. Being complicit might involve a combination of elements such as how much an agent contributed, where they stood in the causal chain, the extent of their foreknowledge of the harm and so on. But there are what I take to be the two key general features of any account of complicity: what the agent knew or ought to have known about their involvement, and to what degree an agent contributed to the harm or was at risk of doing so (the ‘knowledge’ and ‘contribution’ conditions).3

Carbon majors and corporate responsibility for climate change 49

On the knowledge condition, there is evidence that many individual carbon majors and the industry as a whole had access to research about climate change well before it was widely available to the public. A report by the Centre for International Environmental Law (2017) outlines how one oil company in the US knew as early as 1957 that CO2 was likely to affect the climate and the US oil industry peak body knew by 1958. By 1968 the oil industry was receiving warnings from its own scientists concerning potential climate risks. The report notes that throughout the 1970s and 80s oil companies were actively managing their own assets to take into account climate risks (e.g., modifying oil rig design in line with potential sea level rise) while at the same time downplaying those same risks in public. Unlike more complex cases of complicity where agents might not realise how they contribute to harm (for instance, by making a film that suggests ways of committing suicide) carbon majors know that the fuels they supply will end up in the atmosphere as GHGs. Even allowing for some level of excusable ignorance, with over 44% of the world’s emissions coming since 1990, there can be little doubt concerning liability based on prior knowledge (Boden et al., 2016, Moss and Kath, 2019). In regard to the contribution condition, attributing degrees of complicity to an agent turns on how a contribution was made. Generally, ways that secondary agents may be complicit include: encouraging, permitting, soliciting or by aiding and abetting (Hart and Honoré, 1959). With respect to carbon majors’ relevant contributory action (producing and selling fossil fuels), the relation between the principal and secondary agent is relatively straightforward: they aid principal agents by supplying fossil fuels. We should note here that not all cases of complicity are so clear cut. One agent encouraging another to, say, cheat on their tax or engage in lewd behaviour might fall into a moral grey area due to uncertainty over how much the secondary agent contributed to the final outcome. But the cases of complicity above are at the other end of the spectrum. Given the explicit knowledge that carbon majors had regarding their actions and the crucial role that their products play in the production of GHGs, it is a very clear case of complicity. I cannot discuss the contributions of each carbon major in detail. But one example is illustrative. The mining giant BHP, which ranks in the top twenty carbon majors (Griffin, 2017), produced fossil fuels that resulted in 596 Mt CO2-e (BHP, 2018) in 2018 alone. To put that in perspective, the latter figure is more than Australia’s entire total projected domestic emissions for 2018 (533.7 Mt CO2-e). If BHP were a country the emissions from its products would be larger than those of 25 million Australians. But there is a further category of action associated with carbon majors that is arguably as significant as supplying fossil fuels, which we can call negative impact. Negative impact occurs where one agent exercises power-over another agent or agents in a way that negatively and indirectly alters their options for action. Tobacco companies have often been accused of this kind of negative impact (Oreskes and Conway, 2010). Through their lobbying, withholding of relevant health information and marketing they have harmed others even though they themselves don’t directly cause the cancers

50 Jeremy Moss (UNSW)

caused by smoking. Carbon majors generate negative impact in three main ways. The first is through having a structural impact on the prevalence of high GHG emitting technologies and practices. For example, through making individuals and businesses dependent on fossil fuel produced stationary energy, or through significant investment in infrastructure and industries that lock-in fossil fuels as the main energy source, thereby excluding renewable alternatives. For example, consider the production and transportation infrastructure required for gas and LNG. Transporting gas requires construction of physical pipelines, while transporting LNG may require building ships or storage to keep it liquefied and facilities to transform it from a gas into a liquid. Producing gas for consumption requires production plants. In addition to these very costly resources, those invested in the gas market will likely promote the development of resources that depend on gas, for example, new power plants that are designed to run on gas, automobile infrastructure developed to support gas vehicles, and so on. Development of gas infrastructure may then also lead to attempts to lengthen the life of the infrastructure for economic reasons, which will then create a competition among resources that may reduce the possibility of developing renewable technologies (Caldecott et al., 2016). To the extent that these resources are invested in, the opportunity to invest in the development of renewable energy sources is lost. In this way, gas infrastructure can create lock-in effects. Lock-in effects occur when longerlived investments in LNG or other fossil fuel distribution systems prevent the deployment of resources for investing in lower-carbon alternatives, and when such existing resources are less well-suited than power plants for carbon capture and storage. Second, carbon majors can have a negative impact on political outcomes that affect the prevalence of fossil fuels. This can take the form of watering down environmental legislation, ensuring fossil fuel favourable domestic legislation is enacted, securing subsidies to make fossil fuels competitive, weakening global climate agreements, or in general via influencing governments to favour fossil fuel production or use (Perrow and Pulver, 2015, Pulver, 2011). This kind of impact allows the fossil fuels to be produced in a profitable and legally advantageous way. Finally, carbon majors can and do have an impact via framing debate about climate change in ways that influence what issues are debated and the terms of those debates. Framing debate can include such activities as effectively casting doubt on climate science or the cost of renewables via funding friendly think tanks, or through attacking the credibility of individual scientists. One example of this kind framing was the involvement of carbon majors such as BP, Exxon and Mobile among others in the creation of the Global Climate Coalition, which lobbied members of the US congress (Oreskes and Conway, 2010, pp. 205–208). Similarly, the American Petroleum Institute prepared a communications plan that had the aim of altering the ‘conventional wisdom’ by getting people to accept ‘uncertainties’ concerning climate science (CIEL, 2017, p. 17). A recent analysis of lobbying data in US congress between 2006–2009 estimates that over US $1 billion was spent lobbying on climate related issues and bills. The biggest spender during that period was Exxon (Delmas et al., 2016). According to a report prepared by Greenpeace (2011), Koch Family foundations have spent over US$127 million

Carbon majors and corporate responsibility for climate change 51

during 1997–2017 supporting groups that have attacked climate science. To provide a sense of how large this contribution from just one company is, the budget of the IPCC for 2018 was around $US 7.7 million (IPCC, 2018). In regard to their negative impact, note that by actively planning and conspiring with primary agents such as power companies, carbon majors are more than just secondary agents, they are co-principals. Co-principals do more than just contribute to a harm, what they do constitutes the harm. Where carbon majors act together to influence government decisions or alter public understandings of climate issues, they are co-operating (planning and executing a wrongdoing together) or colluding (acting in secret to deceive others and obtain a benefit). In these cases, in addition to supplying fossil fuels in a way that is complicit with the actions of a principal agent, they are also co-principals in virtue of their actions in co-operating or colluding to bring about a harmful act. Through influencing government policy and public debate, carbon majors are able to ensure that the production and sale of fossil fuels is more profitable and necessary than it otherwise would be, thereby contributing to its continuation.

2 The duties of carbon majors If carbon majors are complicit in contributing to climate harms or act as principals, they have a pro tanto duty to cease the harmful activities and to accept some of the costs of the harms. However, it is important point to note that in some cases the duty to cease harming and the duty to compensate can conflict, at least in practice. Where liability is incurred through a harmful practice and the only way an agent can pay for that liability is to keep performing (at least for a time) the harmful activity, then the duties may not be able to be discharged simultaneously. For instance, suppose that a factory owner pollutes the land of his neighbour while he produces goods. Suppose further that he agrees that he has harmed his neighbour and that he should stop the harmful activity and pay compensation. Yet, the only way he can do so is by continuing to operate his factory as that is the only means by which he can generate revenue from which the compensation can be paid. The harmed neighbour might think it acceptable to put up with a little more pollution to receive the compensation that he needs. Yet, what is not acceptable is the situation where the factory owner continues to operate and harm third parties to earn the money to pay the compensation. That is because if we have pro tanto duties not to harm that will include the case of harming third parties to compensate past harms to another agent. Doing so would create cases of what we can call Principle of Unjust Restitution (PUR). Thus, It is impermissible for an agent to inflict new harms on third parties to rectify past harms PUR gives carbon majors a pro tanto duty and makes it more likely that cessation is a stronger duty to satisfy than discharging liability in many practical instances. This point will be important when we discuss what actions carbon majors ought to take.

52 Jeremy Moss (UNSW)

Types of response If carbon majors bear significant responsibility in the way that I have described, how ought they to respond? This is an important question for several reasons. As major contributors to climate change, carbon majors ought to bear the liability for what they have done. But we should also be interested in their response because of the difference that their actions can make. Carbon majors control and utilise vast resources that could be used to assist avoiding climate change. We can divide the kinds of actions they can take into three broad types. The first, which I call phase outs, involves stopping (over time or immediately) the production and sale of fossil fuels and those activities that support them, including: lobbying for financial subsidies for production, the exploration for new resources, by seeking support for fossil fuels through the funding of lobbyists, politicians, ‘think tanks’, industry groups and other kinds of activities designed to achieve favourable outcomes for carbon majors. As the name suggests, phase outs can occur over time and need not be immediate (indeed, are highly unlikely to be in many cases). To prevent further climate harms the quicker the phase outs are the better, but many countries cannot stop using fossil fuels overnight. Lest this be taken as a kind of back door policy for business-as-usual, it is important to be clear that phasing out has to be a real and urgent goal. Adopting phase-out targets that stretch far into the future and relegate emissions cuts to the end of a phase-out plan likely reflect dubious commitment (Victor et al., 2017). The second way in which carbon majors might respond is via disgorgement. Disgorgement here means carbon majors giving up the profits, proceeds or assets of the production and sale of fossil fuels, which may then be used to discharge any associated liability.4 This may occur by, for instance, compensating victims or likely victims of climate harms or contributing to mitigation/adaptation efforts. Disgorgement may also encompass simply giving up the profits or proceeds of the activities that might be used for purposes other than compensating victims of climate harms but I set that issue aside here and focus on the narrower use of disgorgement. The third type of action that a carbon major could take involves transforming its activities in one or more ways such that it is no longer a substantial emitter of GHGs (decarbonisation) or converting itself into a company that no longer seeks to have negative influence over climate policies (conversion). For instance, an oil company could de-carbonise its operations by ensuring that the extraction of oil was done in a carbon neutral manner. It could also offset any or all of its scope 1, 2 and 3 emissions via offsets. What the UNFCC calls ‘scope 1 and 2’ emissions are those emissions that are produced within a state’s borders by various types of activity, such as industrial activity, transport, and so on. Scope 3 emissions are the emissions that are produced by the commodities when they are generated outside the state’s territory and these are not part of the state’s emissions budget. To give an example, if a state exports coal, the emissions that are generated by extracting the coal and transporting it to a port are part of that state’s emissions budget

Carbon majors and corporate responsibility for climate change 53

because they occur within its territorial boundary. The emissions that are produced when the coal is burnt are part of the actual budget of the state that burns them (Eggleston et al., 2008). A further type of transformation is the divestment of all GHG emitting parts of its operations and invest in new, less-polluting, enterprises. Instead of drilling for oil it could sell solar panels and wind turbines, thereby continuing to be an energy company, but via renewable technologies. Transforming their activities might also include developing new technologies that assisted with mitigation. It could also keep its fossil fuel assets but retire them. Yet the actions listed above may conflict in practice. For example, transforming a fossil fuel company into a carbon neutral or completely different kind of business might be a hugely costly task and soak up much of the profits that might otherwise be used for compensation. Conversely, selling fossil fuel assets as part of a process of divestment may provide a revenue stream for compensation, but allow the process of emitting GHGs to continue via a different owner. Or, continuing to sell fossil fuels will provide a revenue stream for compensation but expose others to climate harms in the process. The following section will discuss each option in turn and evaluate their respective strengths. There might of course be other measures that the carbon majors ought to take such as admitting their fault or forming with others to promote collection action. However, I will leave these aside and focus on what I take to be the main types of action open to them.

3 Assessing the actions of carbon majors Transformation Starting with the option of transformation (T), while it might seem straightforward that carbon majors ought to transform/decarbonise in some way as a matter of priority, this is not the case. Transformation includes: T1 Decarbonisation: decarbonising fossil fuel operations or offsetting emissions (scope 1+2 or 3). T2 Conversion: cease influencing policies to favour fossil fuels. T3 Divestment: selling the fossil fuel assets of a carbon major.

T1 Decarbonisation On the merits of transformation via decarbonisation, all things being equal, it is good for a company’s operations to be carbon neutral for the obvious reason that it reduces a source of emissions. Future harms are avoided, in one sense, by such a measure and this measure observes the priority of not harming via the production of emissions. Yet when we consider what types of actions best respond to the injunction not to contribute to climate harms we ought also to factor in how likely the actions are

54 Jeremy Moss (UNSW)

to achieve their goal. All things being equal, if an agent is contributing to harm then those actions that most decisively remove the agent from contributing to harm are to be preferred to actions that less certainly reduce harm. If this is the case, then if decarbonisation does not surely and adequately eliminate the risk of contributing to harm then there is a reason not to undertake it as the primary response. These considerations are relevant to assessing the merits of decarbonisation strategies. For example, decarbonisation may still produce lock-in effects for the sector. Lock-in effects occur in this case where the continued use, development and support of fossil fuels leads to their continuing to be the dominant energy source. When states allow their essential energy infrastructure to depend on fossil fuels because, for instance, they have built expensive grid connected coal power plants then switching to another type of energy generation might be costly and disruptive. Lock-in is especially bad where other actors do not offset. Offsetting emissions as a method of transformation may involve similar difficulties. To be effective offsets must be reliable and long-term. Offsetting emissions by creating carbon absorbing forests in high-bushfire-risk areas or in countries subject to uncontrolled land-clearing is clearly undesirable. Such offsetting strategies may also delay the phase out of the industry if they are not seen as a temporary measure. Offsetting measures should also avoid causing harm to others in unacceptable ways, such as by encouraging governments to remove people from their traditional land or by buying all the cheap emission permits when a country is at its least developed and leaving it little room for its own development-associated emissions. Given many offset measures carry some degree of risk, the acceptability of available options depends on their safety and effectiveness. In any case, decarbonisation by offsetting is less effective in reducing risk of harm than stopping the activity entirely. Whether it is an acceptable temporary measure will be returned to below. Yet, even if all these issues could be overcome, offsetting or reducing emissions from the production of fossil fuels does not meet liabilities. While some decarbonisation ought to figure in the carbon majors’ response because there will be fossil fuel production that it is necessary to maintain in the short term, it is minimal and unsatisfactory since by itself this action alone does not adequately address the liability of carbon majors.

T2 Conversion: ceasing negative impact Carbon majors that ‘convert’ would no longer seek to influence energy and related policies in a way that favours fossil fuels. Such a response is, of course, a major and necessary step in mitigating climate change as emissions levels would diminish remarkably if climate policy were free of the undue influence of carbon majors that currently prevails. Yet by itself it is still, at best, only a partial response that is subject to similar objections to those that applied to decarbonisation; while a step in the right direction, not influencing by itself might still allow carbon majors to continue emitting. It also does not address liability issues. It might also be the case

Carbon majors and corporate responsibility for climate change 55

that if markets are structured around the operations of carbon majors, then governments may be hesitant to alter the role of carbon majors in supplying fossil fuels.

T3 Divesting A more radical dimension of transformation is the divestment of those parts of a carbon major that produce fossil fuels. Divestment from carbon majors and related companies has become a hugely popular and relatively successful campaign issue (Bratman et al., 2016); indeed, the divestment movement has succeeded in removing substantial funds from carbon majors and related climate harming companies. What is at issue here is whether carbon majors ought to divest themselves of their fossil fuel assets as part or all of their response to their duties not to harm. We might divide the kinds of divestment into two types: T3.i: Selling the carbonised parts of a business T3.ii: Retiring the carbonised parts of a business and returning the land to the state. Further, selling can also be broken down into: Selling and retaining profits (for shareholders or executives for instance, or to buy new businesses, T3.i.i): and selling and using the proceeds to compensate or mitigate the harms that have been previously contributed to by carbon majors (T3. i.ii). For example, a carbon major might sell its coal production business and use the proceeds to discharge any compensatory or mitigation liabilities. There is obviously some overlap here with phasing out the production and sale of fossil fuels. One dimension that is more clearly a transformation is the selling of fossil fuel assets for the purpose of transforming the activities of a company. How ought we to evaluate the option of divestment? Beginning with selling assets (T3.i), in particular to another company who will continue to exploit those assets, it is doubtful that divesting the fossil fuel parts of a business via a sale satisfies the cessation requirement of contributing to harm for the simple reason that the contribution will simply continue via another company. By selling a fossil fuel production asset as an ongoing concern, a carbon major does not reduce the harms such an asset creates, it merely transfers the operation to some other company where the harms can continue to be generated. Yet, selling (T3.i) does mean a carbon major is not complicit in one way; the carbon major is not involved themselves in the production and sale of fossil fuels if it has sold its assets. However, it does render a carbon major complicit in another way by providing a working fossil fuel business for others to utilise. Hence, neither option is satisfactory. Further, selling and retaining the proceeds for internal company use also seems a worse option than using the funds to compensate those already harmed or at risk of harm. But even the latter option violates PUR. A carbon major that does divest its fossil fuel assets as on-going concerns and uses the proceeds to compensate allows the harms to continue via other owners.

56 Jeremy Moss (UNSW)

Retiring assets (T3.ii) on the other hand, while not compensatory, at least observes the stronger of the two requirements in response to harm and does not violate PUR. For this reason retiring (T3.ii) is preferable to selling (T3.i).5

Disgorgement According to a duty of disgorgement, carbon majors’ accumulated profit (which might otherwise be paid out to shareholders, management or reinvested in the operations of carbon majors), ought to be transferred to either the victims of harm or the mitigation of climate change (possibly adaptation, though this issue is set aside). There are distinct types of disgorgement. For instance, a carbon major could continue to operate its businesses and use the proceeds to compensate or mitigate to the appropriate degree (D1: disgorge and compensate). The benefits of this option are that it at least might be able to compensate those who have been harmed. However, disgorgement for compensation clearly violates PUR as it commits carbon majors to continuing to harm others, albeit for the purpose of generating revenue for compensation. In the case of compensation specifically, there could also be perverse incentives to increase production and profitability to discharge any duties, again this would violate PUR. A variation on this simple duty of disgorgement is disgorging profits from nonfossil fuel parts of a business. Whilst this option is not available for all carbon majors, some carbon majors are energy companies that earn their profits via renewable energy or operations that are completely unrelated to energy. This version of disgorgement (D2: clean disgorgement) does not violate PUR. The disadvantage of this option is that funds may fall short of what is required. However, if funds are sufficient and the fossil fuel component of the business ceases operations, then this option seems to be an acceptable as a way of discharging liabilities. A further variation on the options above is extending the net of disgorgement to the management of carbon majors (D3: wide disgorgement). Bonuses, excessive salaries, etc. of company executives might be returned to the company and disgorged to compensate victims. This option may still not meet all compensatory requirements, but it may well make a non-trivial difference to large groups of people who could share in the often tens of millions of dollars that is paid to the management of carbon majors.

Phase outs As we saw, phase outs involve the shutting down of the fossil fuel operations of carbon majors. Yet if the most important duty associated with having contributed to harm is ceasing to harm, then phase outs are a better response than either disgorgement or transformation. Phase outs are better than transformation because they have less risk of harm. Stopping the production and sale of fossil

Carbon majors and corporate responsibility for climate change 57

fuels means no emissions are generated, whereas offsetting offers more risk that emissions will continue. Disgorging profits from producing fossil fuels violates the principle of unjust restitution even if it does compensate those who have been harmed in the past. Phasing out the operations of carbon majors will mean taking different actions for different companies. While I cannot go into all of the details here, a framework would have to be established that ranked their operations in terms of identifying the worst emitters (oil from tar sands for instance), which of the operations was really necessary for meeting basic energy needs, whether a particular country had more need for allowing a production to continue, whether some carbon majors are extracting unfairly large shares of fossil fuel resources and so on. Some particularly polluting carbon major operations might be phased out immediately, while others might be among the last to be shutdown. Where the activities of carbon majors are unjustified, the actions undertaken in response ought to take should involve immediate phase outs. For operations that are phased out immediately, retiring those assets is preferable to selling them, as we have seen. Where operations are ongoing in line with appropriate climate targets, a mixed or hybrid approach is required. If carbon majors are to avoid harming others, profits should go to ensuring their operations are carbon neutral, perhaps via offsetting. If carbon majors are not to violate PUR, addressing their liabilities to those already harmed ought to be undertaken with the resources that remain after ensuring there no concurrent harms from the operations that it is necessary to continue. This means phase outs ought to be combined with some decarbonisation, followed by disgorgement, whereby the disgorgement does not violate PUR (D2 and D3).6 It might be objected that the kind of actions that carbon majors ought to take are far too demanding. In particular, phasing out their operations is too harsh a consequence, given that in many cases an individual carbon major is ‘merely’ complicit in a harm and therefore shares responsibility for the harm with a primary agent. This objection is an important reminder that the actions of a carbon major ought to be in proportion to their contribution to a harm. In the case of disgorgement for the purposes of compensation, I see no reason why a proportionate response would not be appropriate. For example, in the case where liability was incurred because a carbon major supplied coal to a power station, it is complicit in the harmful actions of the power station but not wholly responsible for compensating victims. But note that while duties to compensate ought to be proportionate, this is not the case with contributing to harm. In standard cases, agents who contribute to harm ought to assume liability and cease contributing to the harm in question. While the duty of compensation is proportional, the duty to cease harming is not. As we saw, agents who contribute to harm, even where they are not the only contributors, have a pro tanto duty to cease their contribution. In the case of carbon majors, this requires them to cease producing and selling fossil fuels as well as ceasing any negative impacts. If this is the case, then it is not too harsh to require them to phase out their operations even if that means that they are no longer a business.

58 Jeremy Moss (UNSW)

Conclusion The past and present contribution of carbon majors to climate harms, is and has been immense. Through suppling fossil fuels on a huge scale and securing of fossil fuel friendly policies, carbon majors have played a major role in ensuring modern societies are dependent on GHG-emitting practices. While there are different possible ways of determining the responsibility that carbon majors have, there are strong grounds for attributing it on the basis of violating the duty not to harm. As recognition of the role that carbon majors have played has increased, so have the calls for them to act to prevent the harms for which they are responsible. I have argued that while each of the three main types of responses has merit, the strongest case can be made for carbon majors to cease the production, sale and support of fossil fuels. While it is clear that carbon majors are not simply going to admit their role and respond in the right way, it is nonetheless fruitful to know what their response ought to be. As with most moral analyses of harmful situations where hope for action is difficult to sustain, it is still useful to know what ought to be done. This is particularly so given the calls for action to stop carbon majors continuing with their impacts on the climate grow louder. If these calls are to be heeded, it is important that carbon majors’ responses be clearly ranked by means of a robust framework which assesses responsibility and clearly ranks responses.

Acknowledgements I would like to thank the audiences at the University of Warwick, University of the Artic, Justice Beyond the State workshop at UNSW, Norway, and Conference on Applied Philosophy, Kyoto. For comments on the chapter I would like to thank Lachlan Umbers and Daniel Burkett. I would also like to thank Persephone Fraser and Kitty Jean Laginha for invaluable assistance in the research for the chapter. The research for the paper was supported by an Australian Research Council Discovery Grant DP180100355.

Notes 1 Heede (2014) notes that not all of the fossil fuel production ends up as emissions. His calculations allow for non-combustible uses such as petrochemicals which effectively store carbon. 2 For related formulations, see Cripps (2011), Feinberg (1984). 3 For discussion of these issues see, see (Gardner, 2007, Hart and Honoré, 1959, Kadish, 1985, Kutz, 2007, Lepora and Goodin, 2013). 4 I leave aside the question of whether disgorgement ought to include profits, assets or revenue. I also set aside the question of whether disgorgement must be used only for compensation of the victims of climate harms and how those victims are identified. For recent treatments see Goodin and Pasternak (2016), as well as the 2014 Special Issue of Journal of Applied Philosophy, 31(4). 5 There might be other duties than the ones that I describe (e.g. duties to shareholders for a return on investment), the workers and communities who depend on the carbon majors for their livelihood or duties to the public to provide reliable energy sources. I will not consider those here. 6 I leave aside the costs to other industries or businesses associated with these actions.

Carbon majors and corporate responsibility for climate change 59

References BHP. 2018. Scope 3 Emissions Calculation Methodology 2018 [Online]. Available: https:// www.bhp.com/-/media/documents/investors/annual-reports/2018/bhpscoper3emissions calculationmethodology2018.pdf?la=en [Accessed 12/11/2019]. Boden, T. A., Marland, G. &Andres, R. J. 2016. Global, Region, and National Fossil-Fuel CO2 Emissions, Oak Ridge, TN: Carbon Dioxide Information Analysis Centre, Oak Ridge National Laboratory, U.S. Department of Energy. Bratman, E., Brunette, K., Shelly, D. C. & Nicholson, S. 2016. Justice is the goal: divestment as climate change resistance. Journal of Environmental Studies and Sciences, 6, 677–690. Caldecott, B., Harnett, E., Cojianu, T., Kok, I. &Pfeiffer, A. 2016. Stranded Assets: A Climate Risk Challenge. Washington: Inter-American Development Bank (IADB). Centre for International Environmental Law. 2017. Smoke and Fumes; The Legal and Evidentiary Basis for Holding Big Oil Accountable for the Climate Crisis. Washington D.C.: Center for International Environmental Law. Cripps, E. 2011. Climate change, collective harm and legitimate coercion. Critical Review of International Social and Political Philosophy, 14, 171–193. Delmas, M., Lim, J. &Nairn-Birch, N. 2016. Corporate Environmental Performance and Lobbying. Academy of Management Discoveries, 2, 1–23. Eggleston, H. S., Miwa, K., Srivastava, N. &Tanabe, K. (eds.) 2008. 2006 IPCC Guidelines for National Greenhouse Gas Inventories – A primer, Geneva: Intergovernmental Panel on Climate Change (IPCC). Ekwurzel, B., Boneham, J., Dalton, M. W., Heede, R., Mera, R. J., Allen, M. R. & Frumhoff, P. C. 2017. The rise in global atmospheric CO2, surface temperature, and sea level from emissions traced to major carbon producers. Climatic Change, 144, 579–590. Feinberg, J. 1984. The Moral Limits of the Criminal Law: Harm to Others, Oxford, UK: Oxford University Press. Frumhoff, P. C., Heede, R. &Oreskes, N. 2015. The climatic responsibilities of industrial carbon producers. Climatic Change, 132, 157–171. Gardner, J. 2007. Complicity and Causality. Criminal Law and Philosophy, 1, 127–141. Goodin, R. E. & Pasternak, A. 2016. Intending to Benefit from Wrongdoing. Politics, Philosophy, and Economics, 15, 280–297. Greenpeace 2011. Koch Industries: Secretly Funding the Climate Denial Machine, 2011Update. Washington D.C.: Greenpeace. Griffin, P. 2017. The Carbon Majors Database: CDP Carbon Majors Report2017. London: Carbon Disclosure Project (CDP). Hart, H. L. A. & Honoré, T. 1959. Causation in the Law, Oxford, UK: Oxford University Press. Hausfather, Z. 2019. Analysis: Why children must emit eight times less CO2 than their grandparents [Online]. Carbon Brief. Available: https://www.carbonbrief.org/analysis-why-children-m ust-emit-eight-times-less-co2-than-their-grandparents [Accessed 12/11/2019]. Heede, R. 2014. Tracing anthropogenic carbon dioxide and methane emissions to fossil fuel and cement producers, 1854–2010. Climatic Change, 122, 229–241. Heede, R. &Oreskes, N. 2016. Potential Emissions of CO2 and Methane from Proved Reserves of Fossil Fuels: An Alternative Analysis. Global Environmental Change, 36. IPCC2018. Global Warming of 1.5°C. An IPCC Special Report on the impacts of global warming of 1.5°C above pre-industrial levels and related global greenhouse gas emission pathways, in the context of strengthening the global response to the threat of climate change, sustainable development, and efforts to eradicate poverty. Geneva: Intergovernmental Panel on Climate Change (IPCC).

60 Jeremy Moss (UNSW)

Kadish, S. H. 1985. Complicity, Cause and Blame: A Study in the Interpretation of Doctrine. California Law Review, 73, 323–410. Kutz, C. 2007. Causeless Complicity. Criminal Law and Philosophy, 1, 289–305. Lepora, C. &Goodin, R. E. 2013. On Complicity and Compromise, Oxford, UK: Oxford University Press. Moss, J. &Kath, R. 2019. Historical Emissions and the Carbon Budget. Journal of Applied Philosophy, 36, 268–289. Oreskes, N. &Conway, E. M. 2010. Merchants of Doubt, London, UK: Bloomsbury. Perrow, C. &Pulver, S. 2015. Organizations and Markets. In: Dunlap, R. E. & Brulle, R. J. (eds.) Change and Society: Sociological Perspectives. Oxford, UK: Oxford University Press. Pettit, P. 2017. The Conversable, Responsible Corporation. In: Orts, E. & Smith, C. (eds.) The Moral Responsibility of Firms, Oxford, UK: Oxford University Press. Pulver, S. 2011. Corporate Responses. In: Dryzek, J. S., Norgaard, R. B. & Schlosberg, D. (eds.) The Oxford Handbook of Climate Change and Society. Oxford, UK: Oxford University Press. Rich, N. 2018. Losing Earth: The decade we almost stopped climate change. The New York Times, August 1. Shue, H. 1981. Exporting Hazards. Ethics, 91, 579–606. Victor, D. G., Akimoto, K., Kaya, Y., Yamaguchi, M., Cullenward, D. &Hepburn, C. 2017. Prove Paris was more than paper promises. Nature, 548, 25–27.

4 SECTORAL RESPONSIBILITY FOR CLIMATE JUSTICE Is aviation exceptionalism defensible? Elisabeth Ellis (Otago)

Introduction1 International aviation is consuming an increasing portion of the global carbon budget; already it consumes a disproportionate share, especially on a per capita basis, since most human beings do not participate in the international air transport regime. As is well understood, the costs of adaptation to climate change fall most heavily on the least well off among us, while the benefits of high-carbon air transport are confined mostly though not exclusively to the relatively well off (IPCC, 2014a; Graver, Zhang, and Rutherford, 2019). While nearly all other sources of greenhouse gas emissions are subject to at least preliminary regulation via their national commitments under the 2015 Paris Agreement, as well as under trans-national sectoral agreements such as the commitment of the International Maritime Organization to net zero emissions by 2050, international aviation’s emissions fall under the International Civil Aviation Organization’s CORSIA, or Carbon Offsetting and Reduction Scheme for International Aviation (ICAO, 2019). CORSIA is a weak, initially voluntary agreement that aims to achieve carbonneutral growth between 2021 and 2035. With some exceptions based on development status and passenger volume, national aviation sectors are to offset emissions that exceed 2020 levels through the ICAO’s global market-based mechanism. Even if the mechanism works perfectly to hold emissions from international aviation at 2020 levels between 2021 and 2035 (an unlikely prospect, as I shall explain shortly), still the aviation sector would be claiming for itself a large and increasing portion of the remaining global carbon budget. The present and nearfuture state of relations among global economic sectors with regard to climate justice thus reflects a fundamental imbalance: most sectors are embarking on the transition to a low-carbon economy, while aviation plans to continue to emit increasing amounts of greenhouse gas.

62 Elisabeth Ellis (Otago)

Does this state of affairs necessarily offend climate justice? Could humanity even hypothetically choose to put all its emissions eggs in the international air transport basket? If so, what principles and institutions would support such a decision? If not, how can we make decisions about the sectoral distribution of the global carbon budget in a principled way? In this chapter I review some theoretical perspectives on climate justice and apply them to these questions. I find that these perspectives concur in their condemnation of the status quo, but that they provide little guidance for moving towards climate justice. I propose an alternative perspective that recognizes the reality that decisions with implications about climate justice are being made at the sectoral level; a sectoral perspective on climate justice suggests a dynamic and adaptive perspective that would allow groups of human beings to make defensible decisions about their emissions behaviour. Sectoral perspectives involve non-state actors such as large airlines and related institutions acting to reduce emissions. Under the Paris Agreement most countries are obliged to reduce their emissions in line with their nationally determined contributions, and to increase their ambitions as needed to avoid catastrophic global warming, and to support international initiatives to promote sustainable development goals like reducing poverty while also transitioning to a low-carbon global economy. Though progress thus far is far from adequate, some countries and many non-state actors, as well as groups like the European Union, subnational groups like the Regional Greenhouse Gas Initiative in the U.S., and multi-player groups like the ‘Powering Past Coal’ initiative are setting net zero targets and changing behaviour to realize them (Climateactiontracker.org, 2019). Individual behaviour change is inadequate to this task, and global and national governance is patchy and slow: non-state action is thus an essential part of any possible transition to sustainability (Higham, Ellis, and Maclaurin, 2018). This chapter approaches the question of sectoral responsibility for climate change by trying to answer the question of whether aviation exceptionalism is defensible.2 In the literature on global climate justice, responsibility is rarely distributed by sector. The climate justice literature considers alternative principles that might govern the distribution of responsibility for greenhouse gas emissions; the entities potentially bearing those responsibilities range from individuals to countries to humanity as a whole. Alternatively, both the literature on climate policy and the reality of climate policymaking in the immediate post-Paris era take economic sectors—agriculture, energy, construction, transportation, services, and so forth—as an important locus of responsibility for achieving emissions reductions and the transition to sustainable economies. Here I adopt the sectoral approach to climate responsibility and apply it to the case of international aviation. I do this for three reasons, having to do with method, the state of play in global climate politics, and the character of the aviation sector itself, respectively. First, in this chapter I am engaging in pragmatic, applied, non-ideal theory; I intend to use philosophical analysis to clarify reasoning about climate policy that is already occurring in the present and soon to occur in the near-future public sphere (Heyward and Roser, 2016; but see Levy, 2016). In my view, engaged philosophy

Sectoral responsibility for climate justice 63

cannot provide critical analysis of the status quo without “meeting people where they are,” and in this case that means recognizing economic sectors as a locus of climate justice responsibility. It is true that policy debates being held around the world are beginning to consider substantial trans-sectoral instruments that would impose a price on carbon sufficient to drive system change in a sustainable direction.3 However, most of the present and near-future work in emissions reduction is being and will be negotiated sector by sector as UN member countries attempt to act on their nationally determined contributions to global emissions reduction efforts. In New Zealand, for example, the sticking point for negotiating cross-party agreement on a Zero Carbon Bill has been the relative size and pace of the contribution expected from the agricultural sector. Moreover, even as countries implement theoretically trans-sectoral instruments like carbon-credit markets and carbon taxes, decisions about the size and pace of expected contributions are again made on a sector-by-sector basis. Worries about carbon leakage, for example, have driven policies that grant gentler transitions to the carbon market for sectors whose production is more mobile or whose demand is more elastic than sectors serving relatively inflexible local markets (Transport & Environment, 2019). In brief, then, the first reason I adopt the sectoral approach here is that this is where the real distribution of climate responsibility is now taking place. To put the contrast a little too bluntly, while climate justice theorists argue about exactly how little the industrial countries owe the rest of the world, policy makers are making decisions about the distribution of responsibility among sectors of their respective economies, decisions that could benefit from a climate justice perspective. The second reason I focus on sectoral responsibility follows from the first one. Ideally the externalities associated with greenhouse gas emissions should be removed by coordinated collective action, as the familiar framework of the tragedy of the commons envisages. Just as we use civil order to solve ordinary commons problems (say, to distribute the exclusive rights to broadcast at particular locations on various radio spectrums), so we should ideally solve the problem of excessive emissions by submitting to a common authority. To make a long story very short, the present state of authority in global climate politics is mixed, with some authoritative signalling from the centre but plural and uneven distribution of most policy-making authority. In such a non-ideal environment, non-state actors and clubs of the willing are where the relevant action is (Nordhaus, 2015). Moreover, as Mark Budolfson has argued, sub-global and sub-national actors and especially groups of them can through their independent actions affect the interest calculations and ultimately the decisions of other actors (Budolfson, forthcoming). For example, the eventual inclusion of aviation in the European Union’s emissions trading system, or ETS, probably contributed to the decision of the ICAO to commit to a global market-based mechanism to offset emissions from growth after 2020 (Ahmad, 2015). The alternative to the ideal of perfectly coordinated global climate action is not anarchy but the non-ideal reality of pluralistic collections of authorities mutually influencing each other (Levy, 2015).

64 Elisabeth Ellis (Otago)

Finally, the third reason I am interested in sectoral responsibility for climate justice is related to our concrete question about whether aviation exceptionalism can be defended. The aviation sector is an outlier among economic sectors along a number of practically and ethically significant dimensions. Even if we do not end up agreeing with aviation industry representatives that voluntary sectoral self-governance is the right model for emissions reduction in international air transport, we can still agree with them that aviation as a sector is different. Most obviously, international air travel is instrumental to some crucial and intrinsically valuable goods, like cross-cultural understanding, prosperity, and world peace. These are the values that defined the Chicago Convention of 1944, which sets aviation apart from other sectors by forbidding taxation of jet fuel, among other policies that cemented global aviation’s exceptional status in the immediate post-war context.4 In addition, demand elasticity for flying varies enormously across space and groups of travellers, such that the burden placed on consumers of international aviation by efforts to reduce externalized costs ranges from negligible to ruinous (Higham, Ellis, and Maclaurin, 2018; McWha and Murray, 2015; Schiff and Becken, 2011). Although there are no precise estimates reported in the public literature, probably no more than 20% of human beings alive today have taken an international flight (Gurdus, 2017; Negroni, 2016). Again, precision is not possible yet, but there is good reason to believe that among the fewer-than-20% of human beings who have flown internationally, a relatively small number of frequent flyers are responsible for the bulk of emissions from international air transport (Kommenda, 2019). I will discuss the implications for climate justice of such disproportionate emissions by a minority of human beings later in the chapter. What matters here is that industry representatives have good reason to expect robust growth over the coming decades, growth so robust that it swamps any gains the industry is able to make in efficiency (Lee et al., 2009; Markham et al., 2018). The aviation industry exhibits a bizarre dynamic in which increased efficiency itself contributes to increasing emissions. As the costs per kilometre are reduced by incremental efficiency gains (lighter seats, better flight management, and so forth), previously uneconomic routes become viable. The gains from increased efficiency in aviation drive costs and emissions per kilometre down, but in the context of extreme unmet demand, this only increases total kilometres flown. For the aviation industry, efficiency gains are swamped by growth (Peeters et al., 2016; Higham, Ellis, and Maclaurin, 2018). These unusual efficiency/emissions relationships in aviation are compounded by the time frames in which transition to new technology can occur. The life span of a typical passenger jet is about 30 years, while any radically new technology to which the industry might transition remains beyond the horizon, and efficiency gains within the present technology have already been realized for the most part (Markham et al., 2018). Present-day batteries are not up to the task of delivering the power needed for most air transport at reasonable weights and costs, and other alternatives such as hydrogen power are decades away.

Sectoral responsibility for climate justice 65

Given the attractiveness of business as usual in air travel from a perspective that discounts its climate impact, it is not surprising that promises of technological innovation that would allow us (that is, those few of us who do fly) to maintain current flying behaviour while reducing emissions continue to beguile. Indeed, the ICAO continues to rely on technologically driven efficiency gains rather than demand reduction for the bulk of its expected “carbon neutral growth” after 2020, all scientific evidence about the demand/efficiency relationship to the contrary. If we are serious about restricting the rise in global mean temperatures to 1.5 or even 2 degrees above the pre-industrial norm, however, GHG emissions must drop sharply in 2020 and to net zero by 2050 (or probably earlier) (IPCC, 2018). Even if the industry’s most optimistic technological scenarios were to eventuate, the time scale for technology-driven emissions reduction without limiting sectoral growth—at least several decades, likely more—does not match the time scale of necessary action. A final reason that the aviation industry is different from other sectors is that governance of international air transport is especially difficult under the distributed plural system in place today.5 As I discuss later in the chapter, the ICAO’s plan for carbon-neutral growth between 2020 and 2035 (Carbon Offsetting and Reduction Scheme for International Aviation, or CORSIA) expects countries to monitor emissions and implement policies for their reduction under the leadership of the international organization. Participation in CORSIA is voluntary for everyone at first (2021–2026), then voluntary for some and mandatory for others (2027–2035) (Pesmajoglou, 2018). While other sectors’ emissions are to be managed under existing national GHG reduction plans, international aviation emissions are managed by agreement among ICAO members and implemented by member states.6 This contrast between most sectors’ climate action plans and aviation’s CORSIA is reminiscent of the difference between the Paris Agreement and the Kyoto Protocol: like the latter, CORSIA’s weak, centrally distributed targets for emission reduction seem designed to protect high-emissions business as usual under the cover of climate friendly language. While industry representatives from all sectors exercise influence throughout the political system, representatives of the aviation industry exercise proportionately more influence over decision making at the ICAO than their counterparts in other sectors exercised at Paris and subsequently.7 The ICAO’s CORSIA regime straddles the line between industry self-regulation and government oversight. Another way to put this is that compared with other sectors aviation is unusually free to determine its own emissions behaviour. It follows, then, that compared with other sectors aviation should be at least as responsible for the consequences of that behaviour. Thus the aviation sector really does represent an unusual case among other sectors of the global economy. Its product plays a special role in facilitating key goods like peace and prosperity. Despite the sector’s enormous size and carbon footprint it has barely scratched the surface of its potential market (at least under business as usual) (Gurdus, 2017). Unusually among sectors aviation cannot expect efficiency gains to reduce emissions (at least not without transformative innovation that is beyond the time horizon relevant to the world’s Paris goals). Finally, due to its

66 Elisabeth Ellis (Otago)

unusual autonomy, the aviation industry is disproportionately responsible not only for the externalized costs of doing business but also for the consequences of decisions made by the ICAO as it effectively governed itself over the past half century.

Global climate justice literatures The literature on climate justice offers a number of contending principles for answering questions about responsibility for climate change, including some that are directly relevant to flying behaviour. This literature provides arguments about responsibility for emissions at the individual level, at the level of the nation-state, and at the global level; these further divide according to whether responsibility is distributed with respect to historical behaviour or only with respect to present-day differences among agents, such as capacity to pay or present-day emissions (BouHabib, 2019; Heyward and Roser, 2016; Gardiner et al., 2010; Moellendorf, 2015). Cosmopolitans such as Peter Singer argue for duties to redress climate injustice regardless of our local memberships, while associativists such as David Miller reply that robust fulfilment of redistributive and other duties happens in relatively homogeneous groups whose members have strong ties to each other (Arneson, 2015; Miller, 2007). Many of these theories of climate justice offer principled answers to questions about who should have to pay the costs of climate mitigation and adaptation, and to whom should these costs be paid, and exactly how much is owed. Under the polluter pays principle, for example, climate justice directs those responsible for harm (polluters) to redress the harms they have caused by paying for them. Myriad difficulties arise in connection with the polluter pays principle, including the fact that many of the parties most responsible are dead, though the underlying rationale about the justice of requiring redress for harms caused remains broadly attractive (see, for example, Moss and Walsh, 2019). Revisions in the spirit of the polluter pays principle have been offered; the beneficiary pays principle, for example, proposes that presently existing individuals or groups who are benefitting from current and historical emissions have a duty to redress climate harm (Shue, 1999; Duus-Otterström, 2017). Other answers to questions about who should pay and how much range from the “ability to pay principle,” which discounts historical responsibility in favour of present capacity, to some “realist” views that also discount historical responsibility but do so in favour of a “Paretian” principle that insists that all participants in present climate action see themselves as better off (Caney, 2010; Posner and Weisbach, 2010). The global climate justice literature provides clarity about alternative ways to assess the morality of how our climate relationships are structured. As I argue in this section and the next one, significant additional work remains to be done before these kinds of high-level principles are able to contribute to answering the question at the heart of this chapter about whether distributing an exceptional proportion of the remaining global carbon budget to the aviation sector (as business as usual foresees) is defensible. The questions motivating the mainstream of existing literature are different than our question, and they are being asked at a much higher

Sectoral responsibility for climate justice 67

level of abstraction. Our question limits us to analysis of the next few decades, from the present moment forward to mid-century or earlier, depending on the rate of depletion of the remaining global carbon budget. Moreover, we are limited to asking how we should distribute a finite resource—emissions within the carbon budget—under conditions of scarcity and uncertainty. What kinds of distribution are defensible, and how should we make decisions about that distribution? To anticipate work done later in the paper, business as usual entails a distribution increasingly skewed towards a single kind of activity—flying—but without anyone accountable for that distribution. Most (not all) existing literature on global climate justice would condemn the status quo in aviation emissions, since it involves a transfer of risk from the least to the most vulnerable.8 However, they offer little guidance for policymakers at the sectoral level, engaged in iterated decisions under conditions of deep uncertainty about the distribution of responsibility among economic sectors, and aiming at near-term change that can contribute to reaching the Paris goals within an urgent time frame. First, the bearers of moral agency in these theories are individuals, national states, and humanity as a whole, while our question involves the potential defensibility of privileging particular non-state actors like the aviation industry over others. Second, even if we were to establish the general moral wrongness of the status quo in the distribution of the global carbon budget across sectors from the point of view of most theories of climate justice, that conclusion would not move us very far towards assisting policymakers in real time. Climate justice directs us not only to reject substantive injustice like transfers of risk to the most vulnerable. It directs us to reject procedural injustice that undermines the agency of those whose prospects are determined by others’ decisions. The question of the defensibility of aviation exceptionalism has to be incorporated into ongoing, iterated, accountable processes, as I discuss in this chapter’s last section. This is not just another case of ideal versus non-ideal theorizing: both the idealists and the non-idealists in the global climate justice literature presume a model in which political philosophy proposes and policy disposes (Edenhofer and Kowarsch, 2015). There are good reasons to worry about this model generally (Levy, 2016; Markowitz, Grasso, and Jamieson, 2015), but in our case of international aviation emissions it is especially problematic. Decisions about climate policy are being made under conditions of deep uncertainty: we are uncertain about the future, of course, but we are also uncertain about how non-linear changes and complicated interactions among many systems will affect how often we have to change course and what kinds of options we will face at each juncture (Marchau et al., 2019). Decisions have to be revisited not just regularly but also as new and unexpected circumstances arise (Stephens, Bell, and Lawrence, 2018). These unpredictably changing circumstances for decision making about aviation emissions policy will have consequences for climate justice every time they arise. Thus analysis of the status quo as violating one or another principle of global climate justice—the polluters are not paying, the beneficiaries of historical injustice are not paying, current policy guarantees neither equal nor sufficient resources nor

68 Elisabeth Ellis (Otago)

capabilities, and so forth—will always be insufficient for our purposes. We need analysis of the status quo, orientation towards alternatives, and a mode of ongoing evaluation under constantly changing circumstances. We need to recognize that questions of justice arise at all spatial and temporal levels, and moreover that framing of climate action decisions in terms of spaces and times already raises questions of climate justice. As Vivek Mathur and colleagues have argued, the global climate justice literature’s focus on North-South issues and on states as actors misses key sites of climate justice, such as multi-level government structures and variation at the local level (Mathur et al., 2014). In the next section, I discuss a promising alternative to the global climate justice literature that has more purchase on our central question. Here I want to highlight just a few contributions from the climate justice literature that do contribute to our analysis. In the first case, most global climate justice perspectives converge on equality and agency as central values, and thus distributive and procedural justice are consistent objects of interest (Ellis, 2018). Second, whether or not theorists are explicitly Rawlsian in their outlook, both high-level and more applied climate justice scholars tend to oppose transfer of climate risk to the most vulnerable (see, for example, the capabilitarian argument for climate justice focusing on vulnerability by Schlosberg, Collins, and Niemeyer, 2017). Finally, more engaged theorists of climate justice tend to follow Sen, Rawls, and others in noting that plural theories of justice are a fact of climate policy (Sen, 2011; Rawls, 2001). From there they look to rules of thumb like the aforementioned rule about risk transfer, as well as to more freestanding political solutions that leave selection of justice criteria to the affected publics (Sen, 2011; Rawls, 2001). The recent resurgence of interest in citizens assemblies and other deliberative mechanisms for climate policy is thus unsurprising in this context (Dryzek et al., 2019).

A sectoral climate justice alternative As we have seen, the global climate justice literature provides some direction about high-level principles that might guide decisions about whether aviation exceptionalism is defensible, but little guidance for how sectoral decision makers can negotiate tensions among its various positions. If we want to meet those decision makers where they are, we should turn to the working principles currently in use at the sectoral level. Interestingly, the main official principles governing emission responsibility under the United Nations Framework Convention on Climate Change (UNFCCC) and the ICAO, respectively, also exhibit substantial tensions. The principle of “common but differentiated responsibility” does substantial work in the climate justice area of global emissions policy making, as it recognizes the common human duty to work towards sustainability while also recognizing variation among human beings according to membership (Romera and van Asselt, 2015). ‘Differentiated responsibility’ cashes out multiple alternative climate justice principles at once: whether you think responsibility should vary according to present ability to pay, for example, or according to historical association with polluting

Sectoral responsibility for climate justice 69

activities, either way you will agree that people living in advanced industrial countries need to take more and faster climate action than those living elsewhere. The ICAO, however, also subscribes to a principle of “non-discrimination,” which in context means that carriers serving the same route are treated the same regardless of where they are registered. Practically the principle means that on routes to countries whose “differentiated responsibility” means that they are exempted from participation in CORSIA, all carriers regardless of origin will be exempted. In a pattern we see repeated when climate policy makes attempts to acknowledge the claims of justice, practical solutions resolve tensions by selecting the accommodation viewed as most immediately feasible (Ellis, 2018). In practice this means the solution that imposes the fewest costs on those most able to resist them. Thus in resolving the tension between the UN principle that would ascribe differentiated responsibility among carriers according to the relative advantage of their state of registration, and the ICAO principle that would promote a free and efficient market in aviation by subjecting competitors on each route to the same rules, CORSIA selects the solution that reduces responsibility for the most advantaged by allowing airlines from high responsibility states to share in the reduced responsibility enjoyed by developing states’ airlines when they compete on the same route. As we have seen, the global climate justice literature offers competing principles for realizing shared values like agency and equality, but tends to converge on general rules of thumb like “do not transfer risk to the most vulnerable,” and the all-affected principle. In a couple of recent papers, Martin Kowarsch and Ottmar Edenhofer argue for a different model of philosophical engagement in climate policy matters. “The discussion of ethical principles has often been isolated from the decisions faced by policy makers, that is from the multitude and diversity of ethically relevant aspects of climate policy” (Kowarsch and Edenhofer, 2016). Philosophical participation is essential to good climate policy making, according to Kowarsch and Edenhofer, but this participation should take the form of membership of integrated interdisciplinary teams (Edenhofer and Kowarsch, 2015; see also Markowitz, Grasso, and Jamieson 2015). Some cutting-edge climate adaptation policy processes integrate ethical assessment of pathways to sustainability, and this is a good first step, though the mode of integration is one-shot and relatively clumsy compared to the subtlety with which people normally reason about climate justice; conventional dynamic adaptive policy pathways planning, or DAPP, for example, distinguishes bluntly among ethically flavoured adaptation paths according to whether the decision makers choose to lean towards equity, environmentalism, or neoliberal values (Haasnoot et al., 2013; Ellis, 2018). In contrast to the conventional DAPP approach, Kowarsch and Edenhofer note that there are significant questions of justice that arise not only at the outset of climate policy decisions but repeatedly. Distributive trade-offs occur at every step, and even setting trigger deadlines and decision time-frames have surprisingly substantial ethical implications. Kowarsch and Edenhofer here confirm the IPCC’s view as reported in the 2014 WGIII report: “ethical judgments of value underlie almost every decision that is connected with climate change” (IPCC, 2014b).

70 Elisabeth Ellis (Otago)

A good example of the kind of work that Edenhofer and Kowarsch call for is the series of papers by an interdisciplinary team of economists, policy scholars, engineers, and philosophers on the topic of carbon prices and inequality (Budolfson et al., 2019; Dennig et al., 2015; Fleurbaey et al., 2019). They demonstrate, among other findings, that modelling to determine the social cost of carbon can take account of ethical values beyond simple utilitarian ones and that more fine-grained, regional attention to inequality data demonstrates that climate change impacts disproportionately harm the poor within states as well as globally (Budolfson et al., 2019; Dennig et al., 2015; Fleurbaey et al., 2019). Despite the inherent difficulty of the project, estimating the costs and the distribution of the costs of different paths of climate action even roughly can provide guidance to policy makers choosing among them. How modellers account for, say, the social cost of carbon depends on weights accorded to variables that are determined at least tacitly by theories of value (Budolfson et al., 2017). While the general rule of not transferring risk to the most vulnerable provides high-level policy guidance, it cannot effectively guide more complex choices about risk transfer inherent in most climate policy decisions. For these questions, ongoing interdisciplinary work including philosophers, political theorists, policy scholars, and others in the human sciences is needed; their efforts should span “high altitude” principles of justice and on-the-ground implementation decisions, meeting people where they are. As I argued earlier, to answer questions about sectoral responsibility for climate change, we need a model of the relation between political philosophy and policy making that improves upon the conventional “conveyor belt” model in which philosophy proposes and policy disposes. A better model recognizes that these are ongoing interactions: as we move through experiences and evolving conditions, we offer arguments connecting principles and policies, recognizing that these cannot be more than provisional given the deeply uncertain circumstances in which we find ourselves (Ellis, 2004). Rather than seeking to discover ideal principles that we then expect policy makers to adopt based on their inherent rationality, we should begin with widely shared principles already applied (in haphazard fashion) by policy makers and encourage repeated consultation and consideration of principled policy stances. Principled analysis meets policy makers where they are by analysing the values inherent in the status quo and suggesting modes of continuing that analysis, pointing out a path forward while acknowledging that today’s conclusions cannot be the last word (Ellis, 2008).

Aviation and the global carbon budget Aviation is an extreme global outlier among sectors with regard to emission reduction targets: even international marine transport, the only other sector not included in the Paris Agreement, has at least set an official goal of reducing emissions by 50% from 2008 levels by 2050 (Graver, Zhang, and Rutherford, 2019). Global aviation produced between 2% and 3% of total carbon emissions in 2018, about as much as Germany. Yet unlike Germany, the aviation sector has no official

Sectoral responsibility for climate justice 71

target for emissions reductions. Some emissions from international aviation are covered by emissions trading systems like the European ETS, but most emissions from flying are omitted from national and international carbon reduction targets.9 More worrying than the raw volume of aviation emissions are the projections of growth over the next half century. The ICAO itself projects that emissions will triple by 2050 under business as usual, while other analysts predict an even higher rate of growth (IPCC, 2014b). More interesting for our question is the projected rate of growth of the proportion of the global carbon budget expected to be consumed by global aviation. Unfortunately this can only be estimated very roughly, though we can say some things about the direction of change under different scenarios and their relative orders of magnitude. In the first place, it is hard to estimate the global carbon budget with any precision (Rogelj and Forster, 2019). The idea of a global carbon budget is that it provides an answer to the question of how much more carbon can we release without experiencing an effect or effects we would like to avoid. As all readers of the IPCC’s series of reports will know, these estimates of relationships between magnitudes of emissions and eventual effects are probabilistic. For example, the IPCC report from late 2018 estimated that for a 50% chance of keeping the global mean temperature below 1.5 degrees above the pre-industrial norm (and thus for a 50% chance of avoiding the worst climate-change-driven disturbances, such as losing all of the world’s coral reefs), we would need to emit at most 580 more gigatons of CO2 from all sources. Total emissions from all sectors in 2018 amounted to 42 Gt CO2. Annual emissions from aviation are approaching the gigaton/year range; emissions in 2018 have been estimated variously at .895 Gt CO2 (ICAO) and .918 Gt CO2 (Graver, Zhang, and Rutherford, 2019), with annual rates of growth under business as usual estimated between 3.5% (IATA, 2018) and 5.7% (Graver, Zhang, and Rutherford, 2019). Given the IMO official target of halved emissions from maritime transport and the commitments under the Paris Agreement to hold the rise in global mean temperature to no more than 2 degrees above pre-industrial levels—that is, given the commitments of every other economic sector to emissions cuts while aviation commits only to “carbon-neutral growth to 2035”—we can expect that between now and 2050, the share of the global carbon budget of every other sector will decline in proportion to the share consumed by air travel. Estimates of how large this shift will be vary, but all involve a massive shift of emissions towards aviation: from the status quo of 2–3% to somewhere between 12% and 25% by 2050 (Pidcock and Yeo, 2016; Transport & Environment, 2019). For our purposes the exact shift is less relevant than the trend: should we spend a large and growing proportion of our global carbon budget on aviation? Can we defend aviation exceptionalism? The answer from the ICAO is a resounding ‘yes’. As we have seen, international aviation emissions are not covered in the Paris Agreement. The ICAO—as the UN body responsible for facilitating agreements about civil aviation among UN member states, including climate policy for international air transport—has negotiated CORSIA, a weak agreement along the lines of now-defunct Kyoto

72 Elisabeth Ellis (Otago)

Protocol, under which the sector commits to “carbon neutral growth” between 2020 and 2035. Even this limited ambition is undermined by the facts that CORSIA’s first few years are mostly voluntary, and that the mechanism for carbon neutrality is not demand reduction but a combination of efficiency gains (which have mostly already been taken, are at any rate swamped by growth, and are necessarily slow given fleet turnover time in the industry) and “market based mechanisms,” or offsets. There is an interesting debate about the value of offsets, but setting that aside we can appreciate what the ICAO is now considering.10 They are considering whether to accept carbon credits from the UN’s Clean Development Mechanism. These credits do not pass a basic “additionality” test (that is, they do not provide carbon reduction that is additional to what would be provided if they were not in force). The EU will not accept them after 2020 (and they have already limited their acceptance of CDM credits from the least effective sources, such as destruction of industrial gases) (European Commission, 2016a). CDM credits mostly fund either projects that do not substantially reduce emissions or they fund emission reducing projects that would have gone ahead anyway because they are inherently profitable. If CORSIA allows credits that do not actually reduce carbon emissions, then the proportion of the global carbon budget effectively consumed by aviation will rise faster than expected. However, even if the ICAO’s new market-based mechanism for offsetting emissions from new growth between 2021 and 2035 ends up using better quality offsets, still, the industry’s proportion of the global carbon budget after 2021 is expected to rise compared with the proportion consumed by every other sector. Barring unforeseen physical or social changes that upend prevailing projections, we must conclude one of two things about the position of the ICAO with regard to future greenhouse gas emissions by the international aviation industry. Under the best available current estimates, achieving a world with a global mean temperature of 1.5 or even 2.0 degrees above the pre-industrial mean requires an immediate sharp decrease in emissions followed by rapid progress to net zero. The ICAO under its most optimistic projections presumes continued increases in aviation emissions, possibly “offset” between 2021 and 2035 but still overall consuming an increasing portion of the global carbon budget. We must conclude that if the aviation sector is committed to the Paris goals of limiting global warming below potentially catastrophic levels, then the aviation sector is presuming that flying is more important than other human activities like driving, sheltering, and eating. More precisely and with maximum charity: the presumption at minimum must be that the value of continued growth in the total amount of flying outweighs the cost of the correspondingly rapid reductions in emissions associated with other human activities (necessary to stay within the projected carbon budget). Alternatively, we could conclude that despite its public communication to the contrary, the aviation sector is not committed to limiting global warming. In that case, the sector is committed to no set of relative valuations of its activities compared with other activities; all of them would be equally unlimited and presumably equally short-lived.

Sectoral responsibility for climate justice 73

In summary, business as usual entails a large and increasing proportion of the global carbon budget be spent on aviation, relative to all other sectors. Though this proportion is expected to rise under business as usual, at present fewer than a fifth of human beings alive today participate in international aviation at all (Gurdus, 2017). Moreover, of those who do fly, an even smaller proportion are the frequent fliers whose many flights emit the bulk of global aviation emissions. Thus under business as usual a relatively small minority of people are set to consume an increasing proportion of the global carbon budget (on environmental justice and preferring minority interests, see Ellis, 2016). It is of course a well-known fact about global greenhouse gas emissions considered across sectors that a relatively small group of very wealthy people consume a disproportionate amount of the global carbon budget, and that under business as usual inequality in both income and emissions are set to rise. The existing literature on climate justice already provides multiple analytical perspectives on this state of affairs. Our sectoral perspective, however, allows us to meet people and policy makers where they are rather than expecting “high altitude” ethical principles about global justice to influence on-the-ground decisions automatically.11 Even when high-level principles about climate justice and responsibility differentiated by history or capacity do influence on-the-ground policy making, they are filtered through a sectoral lens. A good example of this is provided by French president Emmanuel Macron’s successive attempts to use taxes on fuel to drive down national greenhouse gas emissions. In 2018 Macron’s government announced a tax on fuel at the pump, a tax that would disproportionately affect people in France’s car-dependent rural areas as well as long-distance commuters. Popular anger over the fuel tax led to the “gilets jaunes” or yellow vest movement which (roughly) articulated the unwillingness of ordinary people to shoulder the immediate costs of climate transition, especially when these were led by a government perceived as giving a free pass to corporations and the wealthy. Macron was forced to withdraw his fuel tax, and in 2019 his government instead proposed an eco-tax on flights from France. Popular anger was driven at least in part by perceptions of climate injustice associated with the first tax hitting the less well-off and rural people the hardest, but the public discussion and eventual policy response to it both operated through a sectoral rather than general climate justice lens. Questions like: Do we want to make flying more expensive? Do we want to punish people facing near-zero substitutability for personal car-centred transport? Why should aviation be exempted from eco-taxes while personal car-based transport is not? And, How can we reduce national emissions fairly? are relevant to climate policy discussions in France and elsewhere. These questions use a sectoral lens to mediate more general intuitions about climate justice.

Sectoral responsibility for climate justice in practice The global justice literature offers differing perspectives, but provides good reasons to take a critical view of the status quo in international aviation that transfers risk

74 Elisabeth Ellis (Otago)

from the most to the least advantaged. Building on Kowarsch and Edenhofer’s analysis of the shortcomings of conventionally applied climate justice theory, we have a view that recommends ongoing integration of climate ethics into iterative decision making about climate policy. Now we can bring that integrated critical perspective to meet climate publics and policy makers where they are, deciding about emissions management at the sectoral level. The European Union’s emissions trading system has been rightly criticized for making too little difference: it has coddled high-emissions industries with free allocations, and the price of carbon credits is still far too low (as of 2019). Even so, the European ETS shows promise not only because of its record of modestly reducing aviation emissions (17 million tonnes per year according to the European Commission, which is significant in the context of total global aviation emissions of around 895 million tonnes in 2018 – European Commission, 2016b; Air Transport Action Group, 2018), but especially because of its ability to revisit its practices and revise ambitions upward in response to new information about its own performance and conditions in the material and societal spheres.12 Having begun by accepting low-quality credits from the UN’s Clean Development Mechanism, the European ETS no longer accepts the weakest ones and will ban them altogether from 2020, for example (European Commission, 2016a). The EU ETS is an existing venue for ongoing, principled argument about sectoral responsibility for climate justice; moreover, it has some ability to translate these principles into practices over time frames that correspond to the urgent need to reduce global emissions outlined in the IPCC special report of late 2018. While hardly a directly democratic organization, the European ETS is susceptible to influence not only from industry but also from climate justice oriented NGOs and from popular pressure via its member states’ representatives. Thus the European ETS is already a place where the question of aviation exceptionalism can be addressed in both a principled and practical way, and in reasonably good time, even if its results have so far been modest. Recent action at the state level in Europe, such as efforts by Sweden and France to use price signals to encourage substitution of flying with lower-carbon transport, make it more likely that a stronger pan-European system will eventually emerge. The UNFCCC process that led to the Paris Agreement is another surprisingly promising venue for sectoral climate justice policy. Unlike its predecessor framework under the Kyoto Protocol, the Paris Agreement expects actors at the level of member states to negotiate, provisionally realize, and renegotiate series of increasingly ambitious emissions reduction plans. The UN provides coordination, signalling, and leadership, while the member states engage in proposing, monitoring, and taking climate action in their sphere of influence. Under the Paris Agreement most sectors are covered by their nationally determined contributions, though alongside that system there are a growing number of international coalitions focusing on sector-based reductions. For example, the “Powering Past Coal Alliance” includes states, subnational governments, and private groups committed to transitioning from coal energy (PPCA, 2019). The existing commitments of member states and the International Maritime Organization

Sectoral responsibility for climate justice 75

are widely acknowledged to be inadequate to achieve the Paris goal of less than two degrees of temperature rise. Despite that sobering fact, the Paris process combines topdown leadership with bottom-up initiative, includes revisiting of goals and progress to encourage ambition and to align action with that ambition, and through its “common-but-differentiated responsibility” principle instantiates in policy at least some widely held intuitions about climate justice. As long as the ICAO exempts itself from the UN climate justice norms, there is no global venue to ask the question of how much of the remaining carbon budget should we spend on international aviation. The ICAO and the aviation industry, isolated from dialogue with other sectors, never ask this question, for obvious reasons. If, as seems likely, the ICAO and its CORSIA mechanism remain the most important source of international coordination of aviation emissions behaviour, then the question of the aviation sector’s responsibility for climate justice will have to be asked somewhere else. I will suggest another venue in the next part of the chapter, but first I would like to note that as faulty as it is, the ICAO could play a role in coordinating climate justice policy. First and most important for our purposes in this chapter, the ICAO is a well-established institution that exists and that already coordinates international aviation (and coordination of the emissions aspect of aviation is what is needed for sectoral climate justice in this case) (van Asselt and Böβner, 2016). The ICAO solves coordination problems in the industry all the time, in conflicts ranging from electronic passports to airplane safety standards to air traffic management. The ICAO is the existing organization best placed to coordinate sectoral emissions reduction among industry players, if only it had the organizational culture and a substantial mandate to do that work. One thing hampering the ICAO from taking up this climate justice role is its extreme insulation from public scrutiny; another is the excessive influence of industry representatives rather than a broad group of stakeholders including those most vulnerable to costs externalized by the industry at present. A more transparent and accountable ICAO could align itself to the goals of the Paris Agreement rather than excepting itself while free riding on the work of all the other sectors. The work of negotiating the aviation sector’s contribution to climate change mitigation could hardly be more difficult than the work of negotiating nationally determined contributions for states whose emissions are presently similar to those of global aviation, such as Germany. The ICAO already relies on states to enforce its safety and other standards, and it could do the same for emissions reductions to fulfil the aviation sector’s contribution to achieving the world’s Paris goals. In what venue, then, could the question of aviation exceptionalism be meaningfully debated? The UNFCCC and the UN General Assembly are of course two obvious candidates, but there are other alternatives that would be more likely than the previous two to reflect common intuitions about a principled response to this question. I am thinking of citizens assemblies, participatory decision making, and more generally of organized civil society (Dryzek, Bächtiger, and Milewicz, 2011). Before discussing these means, however, I want to say a few things about why they are necessary.

76 Elisabeth Ellis (Otago)

As we have already seen, there are no simple principles of sectoral climate justice that can adjudicate the question of the right allocation of the carbon budget among diverse human activities represented by sectors. Moreover, as we have also seen, international aviation is a good enjoyed disproportionately by a small minority of human beings at the expense of most current and future human beings, while aviation at the same time offers unsubstitutable instrumentally valuable means to some intrinsically valuable goods. We cannot expect any kind of ethical algorithm to produce defensible results from these various considerations. Instead, the answer to how much of the global carbon budget should be spent on international aviation must be answered by people themselves. The sectoral approach to climate justice is a non-ideal, pragmatic method of bringing widely held ethical considerations into the policy process where they have a chance of informing decision procedures and substantive outcomes. Conventional global climate justice theory naturally ignores the problem of distribution among sectors because inter-sectoral distribution is an inherently dynamic process that ought to respond flexibly to changing circumstances. If we had asked in the first half of the nineteenth century (before the internal combustion engine was invented), how much of our global waste allocation should be distributed to animal-powered transport, that answer would have depended on a contingency—dominance of horse-based travel—that was soon to disappear. Sectoral distribution questions are deeply political and always provisional, as the earlier example of New Zealand’s present-day conflict around this same problem of animal waste makes clear. Ideal climate principles can provide some general guidance about substance (don’t transfer risk to the most vulnerable) and process (engage all affected) with regard to sectoral distribution questions, but they cannot pre-empt people’s repeated and changeable decisions about specific allocations. For example, with respect to emissions justice, ideal theory might direct us to think about avoiding harm (for a practical application of this principle to the case of attributing responsibility for climate harm to the major Australian fossil fuel producers, see Moss and Fraser, 2019). We know from the empirics of climate change that if we are committed to keeping the increase in mean temperature to less than two degrees above the pre-industrial norm, then there is a limited amount of carbon and carbon equivalents that we can emit. Ideally, this amount would be calculated and recalculated regularly and permissions to emit distributed according to some consensus norm, so that everyone would know how they can behave without damaging the prospects of future people and the most vulnerable. Of course, nothing like this system is in place, but thinking about it makes the tacit principles governing our present relations with each other clear. When we emit more than our share of greenhouse gases, we are effectively externalizing the cost of our activities onto others, harming them, and thus effecting climate injustice. But how can we identify the share of greenhouse gas emissions that we can emit, promoting our flourishing while avoiding wrongful harm to others? The status quo in distributing emissions allocations in the aviation sector simply ignores the question, but as we have seen, this amounts to tacit approval of an ever-increasing share

Sectoral responsibility for climate justice 77

for aviation and less for every other human pursuit. Instead, we should look for means for people to think through principled allocations of limited emissions among competing sectors, so that people can regularly and repeatedly make principled judgments about how much of the remaining global carbon budget should be allocated to particular sectors. Conventional democratic theory would at this point offer representative government and political parties as the best available means to subject sectoral allocation decisions to the popular judgment about climate justice. As the introduction to this volume reminds us, however, we have not been able to rely on nationstates to constrain greenhouse gas emissions to sustainable levels. Non-state actors have an essential role to play in driving the transition to a low-carbon future, in the aviation sector as elsewhere. As we have seen, business as usual in the international aviation sector means that a large and increasing portion of the global carbon budget will be consumed by the aviation sector. However, we have also seen that the international aviation sector is different from other sectors in ways that might justify some exceptional treatment. Every policy settlement – from business as usual (aviation exceptionalism) through to mandatory net zero emissions without exception – entails trade-offs that raise significant questions of climate justice. The stakeholders include sectoral interests like airlines and airplane manufacturers and their supporting industries, present and potential consumers of international aviation, and, crucially, all those who bear international aviation’s externalized costs. Decisions about how much of the global carbon budget to spend on international aviation should include those stakeholders who are now excluded (the cost bearers). Given the shortcomings of the existing system in this regard, a global citizen assembly along the lines of recent experiments at the national (e.g. Ireland) and regional (e.g. British Columbia) levels could serve to develop, and then to amplify the considered judgements of people about the distribution of emissions permissions among sectors. The global level is only one place where policies of aviation exceptionalism should be evaluated by stakeholders; at the national and regional levels, people should consider the various trade-offs of different policies and select solutions that reflect their values. These participatory decision making processes should be regular and repeated, as people revisit their allocations in ever-changing empirical and societal contexts. The trade-offs for people with access to easy substitutes for international air travel (say, most Europeans) will be different than those for people with no such access (say, most North Americans), and different again for people whose physical circumstances prevent them from developing satisfactory substitutes (say, most people in Oceania). Regular and repeated consideration of the distribution of emissions across sectors at the national and regional levels is therefore necessary. Finally, organized civil society has a critical role to play in assessing aviation exceptionalism, pursuing collective action where possible, and driving the political system to respond to their judgment about the defensibility of the status quo. What had been a trickle of calls for exit from aviation exceptionalism led by people like Eric Holthaus is becoming a flood of coordinated efforts by associations, churches, universities, and even

78 Elisabeth Ellis (Otago)

groups of corporations to reduce the share of the global carbon budget consumed by flying. Together these three means to assessing the defensibility of aviation exceptionalism comprise a deliberative system producing judgements about how much of the global carbon budget we should spend on international aviation.

Conclusion Can aviation exceptionalism be defended? This chapter has argued for a nuanced answer rather than the ICAO’s resounding ‘yes’ or the climate justice literature’s nearunanimous ‘no’. The status quo amounts to a transfer of climate risk to the most vulnerable: international aviation, through its industry-dominated regulatory body, asserts its right to a disproportionate and growing portion of the world’s greenhouse gas emissions, while the rest of the global economy is expected to reduce emissions sharply from 2020 and down to net zero by 2050. Despite the manifest injustice of business as usual in international aviation emissions, we should not conclude simply that there is a global ‘right answer’ to the question of how much of the remaining carbon budget we should spend on flying. Instead, the proportion of emissions allocated to aviation should be an ongoing matter of public judgment at all levels from global through local. While we cannot rely on states to solve this problem for us, the deliberative system can bring public judgment from citizen assemblies, participatory decision making processes, and organized civil society together with constituted authorities to ensure that the decisions about emission distribution among sectors are defensible.

Notes 1 Acknowledgements are due to Emily Mathias, James Maclaurin, James Higham, Emily Nacol, Shmuel Nili, the members of the Otago University Philosophy and Politics departments, and the Global Politics Workshop at Northwestern University. Special thanks to the Otago University “Thirst for Knowledge” group, the Western Political Science Association Climate Friendly Conference Committee, and He Kaupapa Hononga/Otago’s Climate Change Research Network. The infelicities and errors in this chapter are mine alone. 2 Airplanes contribute to climate change by emitting carbon dioxide, nitrous oxides, other gases, particulates, and water vapor, not to mention the GHGs associated with their production, deployment, and maintenance. While it is generally recognized that carbon emissions make up only part of aviation contributions to global warming, carbon emissions are what proposed regulations and most policy discussions focus on. For example, the global market-based mechanism that the International Civil Aviation Organization will employ from 2020 to encourage “carbon-neutral growth” in the industry excludes drivers of radiative forcing other than carbon. 3 None of these supra-sectoral efforts have been sufficiently ambitious as of this writing (2019). A good example of an inadequate step in the direction of trans-sectoral carbon pricing is the recent proposal from the German Klimakabinett to price carbon emissions at 10 EUR/tonne (much too low to drive transition, and much lower than the social cost of carbon; see e.g. Pindyck, 2019). Note that trans-sector carbon pricing mechanisms are almost invariably negotiated sector by sector, with specific provisions (especially exemptions, free and discounted allocations, delayed entry, and so forth) made on a per-sector basis. 4 Of course every sector produces things that are instrumental to intrinsically valuable things: agriculture produces food that supports human life, for example. What sets global aviation

Sectoral responsibility for climate justice 79

5 6

7

8 9

10

11 12

apart here is the idea that regulating it would substantially interfere with the provision of those goods. Some of these arguments have been eroded in the wake of the victory of the European ETS over challengers from the United States who argued that requiring emission credits for flights to Europe amounted to a tax on jet fuel (and that would be forbidden under the Chicago Convention). The European Court of Justice ruled, however, that buying European emissions credits should be considered just a cost of doing business rather than a tax on jet fuel (Court of Justice of the European Union, 2011). Except international marine shipping, which like aviation is governed by a UN sponsored international agency. The IMO has, unlike aviation, committed to an ambitious program of emissions reduction (International Maritime Organization, 2018). The European Union includes flights to and from its cities in its emissions trading scheme, while the ICAO would prefer that all international flights were subject to a single global market-based mechanism, CORSIA (which is expected to be significantly weaker than the already weak European ETS). Industry representatives and the ICAO oppose a “patchwork” system of regulation of international aviation, while climateoriented NGOs and some states in the European Union prefer to use the stronger rules of the European system despite difficulties coordinating with the global one. The ICAO operates under some of the least transparent rules in global governance. NGO representation is strictly limited, while industry representation is generous. In meetings of the ICAO’s committee on aviation environmental protection, NGO participation is limited to a single umbrella representative. Moreover, participants in deliberations of the ICAO’s committee on aviation environmental protection must sign non-disclosure agreements that could make them liable for huge damages if committee documents become public. Together these procedures add up to an agency making decisions about matters of global environmental security with almost no public oversight or accountability (Farand, 2019; Laville, 2019). See, for example, Posner and Weisbach 2010. Some nationally determined contributions to emissions reduction under the Paris Agreement include domestic aviation emissions. At this writing (2019), some countries are adding climate charges to air travel within their jurisdictions, and there is discussion of efforts to use additional charges on domestic aviation to subsidize provision of substitutes for flying such as train travel. While it is too soon to judge whether market-based mechanisms can bring sufficient reductions in carbon emissions to achieve the Paris goals or anything close to that, it is widely held that they should be considered a transitional stopgap between the highcarbon and zero-carbon economies. Meeting policy makers where they are means taking offsets and ETSs seriously. Even so, there are good reasons to worry about the efficacy of offsets, from their historical lack of accountability to their time-shifting of reduction activity to their ability to prevent more substantial transformation in good time. Thanks to my colleague Andrew Moore for this formulation of the contrast between general and engaged climate justice perspectives. The ability of carbon pricing to reduce flying behaviour varies across consumer groups, regions, and institutions. A study of Australia’s experiment with carbon pricing between 2012 and 2014 concluded that “there was no evidence that the carbon price reduced the level of domestic aviation in Australia,” though the price may have simply been too low in that case (Markham et al., 2018). On the other hand, the fact that a recent study found that many leisure travellers rate their flying as unimportant suggests that a substantial enough carbon price could be effective (Gössling et al., 2019).

References Ahmad, T. 2015. Evaluating the Effectiveness of the European Union Emissions Trading System to Reduce Emissions from International Civil Aviation. McGill International Journal of Sustainable Development Law and Policy, 1, 119–182.

80 Elisabeth Ellis (Otago)

Air Transport Action Group. 2018. Facts & Figures. 2018. Available at: https://www.atag. org/facts-figures.html. Arneson, R. J. 2015. Theories, Types and Bounds of Justice. In: Moellendorf, D. (ed.) The Routledge Handbook of Global Ethics, 1st ed. Routledge. Available at: https://doi.org/10. 4324/9781315744520. Asselt, H. van & Böβner, S. 2016. The Shape of Things to Come: Global Climate Governance after Paris. Carbon & Climate Law Review, 10, 1, 46–61. Bou-Habib, P. 2019. Climate Justice and Historical Responsibility. Journal of Politics, 81, 4, 1298–1310. Available at: https://doi.org/10.1086/704325. Budolfson, M. Forthcoming. Political Realism, Feasibility Wedges, and Opportunities for Collective Action on Climate Change. In: Budolfson, M., Plunkett, D., and McPherson, T. (eds), Philosophy and Climate Change, Oxford, UK: Oxford University Press. Budolfson, M., Dennig, F., Fleurbaey, M., Scovronick, N., Siebert, A., Spears, D. & Wagner, F. 2019. Optimal Climate Policy and the Future of World Economic Development. The World Bank Economic Review, 33, 1, 21–40. Available at: https://doi.org/10. 1093/wber/lhx016. Budolfson, M., Dennig, F., Fleurbaey, M., Siebert, A. & Socolow, R. H. 2017. The Comparative Importance for Optimal Climate Policy of Discounting, Inequalities and Catastrophes. Climatic Change, 145, 3–4, 481–494. Available at: https://doi.org/10.1007/s10584-017-2094-x. Caney, S. 2010. Climate Change and the Duties of the Advantaged. Critical Review of International Social and Political Philosophy, 13, 1, 203–228. Available at: https://doi.org/10. 1080/13698230903326331. Climateactiontracker.org. 2019. Countries | Climate Action Tracker. 2019. Available at: https://climateactiontracker.org/countries/. Court of Justice of the European Union. 2011. The Directive Including Aviation Activities in the EU’s Emissions Trading Scheme Is Valid. Available at: europa.eu › rapid › pressrelease_CJE-11–139_en. Dennig, F., Budolfson, M. B., Fleurbaey, M., Siebert, A. & Socolow, R. H. 2015. Inequality, Climate Impacts on the Future Poor, and Carbon Prices. Proceedings of the National Academy of Sciences, 112, 52, 15827–15832. Available at: https://doi.org/10. 1073/pnas.1513967112. Dryzek, J. S., Bächtiger, A., Chambers, S., Cohen, J., Druckman, J. N., Felicetti, A., Fishkin, J. S. et al. 2019. The Crisis of Democracy and the Science of Deliberation. Science, 363, 6432, 1144–1146. Available at: https://doi.org/10.1126/science.aaw2694. Dryzek, J. S., Bächtiger, A. & Milewicz, K. 2011. Toward a Deliberative Global Citizens’ Assembly. Global Policy, 2, 1, 33–42. Available at: https://doi.org/10.1111/j.1758-5899. 2010.00052.x. Duus-Otterström, G. 2017. Benefiting from Injustice and the Common-Source Problem. Ethical Theory and Moral Practice, 20, 5, 1067–1081. Available at: https://doi.org/10.1007/ s10677-017-9845-7. Edenhofer, O. and Kowarsch, M. 2015. Cartography of Pathways: A New Model for Environmental Policy Assessments. Environmental Science & Policy, 51 (August), 56–64. Available at: https://doi.org/10.1016/j.envsci.2015.03.017. Ellis, E. 2004. Provisionalism in the Study of Politics. In: Shapiro, I., Smith, R. M., & Masoud, T. E. (eds) Problems and Methods in the Study of Politics, Cambridge, UK: Cambridge University Press. Ellis, E. 2008. Provisional Politics: Kantian Arguments in Policy Context, Yale University Press. Ellis, E. 2016. Democracy as Constraint and Possibility for Environmental Action. The Oxford Handbook of Environmental Political Theory. Available at: https://doi.org/10.1093/ oxfordhb/9780199685271.013.12.

Sectoral responsibility for climate justice 81

Ellis, E. 2018. How Should the Risks of Sea-Level Rise Be Shared? Wellington, N.Z.: Deep South National Science Challenge. Available at: https://www.deepsouthchallenge.co.nz/ sites/default/files/2018-11/How%20should%20risks%20of%20SLR%20be%20shared%20 Lisa%20Ellis%20Final%20Report.pdf. European Commission. 2016a. Use of International Credits. Climate Action—European Commission. 2016. Available at: https://ec.europa.eu/clima/policies/ets/credits_en. European Commission. 2016b. Reducing Emissions from Aviation. Climate Action—European Commission. November 23, 2016. https://ec.europa.eu/clima/policies/transport/a viation_en. Farand, C. 2019. Aviation’s Black Box: Non-Disclosure Agreements, Closed Doors and Rising CO2. Climate Home News. September 12, 2019. Available at: https://www. climatechangenews.com/2019/09/12/non-disclosure-agreements-closed-doors-risingco2-uns-aviation-body/. Fleurbaey, M., Ferranna, M., Budolfson, M., Dennig, F., Mintz-Woo, K., Socolow, R., Spears, D. & Zuber, S. 2019. The Social Cost of Carbon: Valuing Inequality, Risk, and Population for Climate Policy. The Monist, 102, 1, 84–109. Available at: https://doi.org/ 10.1093/monist/ony023. Gardiner, S. M. et al. 2010. Climate Ethics : Essential Readings, Oxford, UK: Oxford University Press. Gössling, S., Hanna, P., Higham, J., Cohen, S. & Hopkins, D. 2019. Can We Fly Less? Evaluating the ‘Necessity’ of Air Travel. Journal of Air Transport Management, 81 (October), 101722. Available at: https://doi.org/10.1016/j.jairtraman.2019.101722. Graver, B., Zhang, K. and Rutherford, D. 2019. CO2 Emissions from Commercial Aviation, 2018. ICCT working paper 2016–2019. International Council on Clean Transportation. Available at: www.theicct.org. Gurdus, L. 2017. Boeing CEO: 80% of People Never Flown. For Us, That Means Growth. CNBC. 7 December 2017. Available at: https://www.cnbc.com/2017/12/07/boeingceo-80-percent-of-people-never-flown-for-us-that-means-growth.html. Haasnoot, M., Kwakkel, J. H., Walker, W. E. & Ter Maat, J. 2013. Dynamic Adaptive Policy Pathways: A Method for Crafting Robust Decisions for a Deeply Uncertain World. Global Environmental Change, 23, 2, 485–498. Available at: https://doi.org/10. 1016/j.gloenvcha.2012.12.006. Heyward, J. C., & Roser, D. eds. 2016. Climate Justice in a Non-Ideal World. First edition. Oxford, UK: Oxford University Press. Higham, J., Ellis, E. & MacLaurin, J. 2018. Tourist Aviation Emissions: A Problem of Collective Action. Journal of Travel Research, 58, 4, 535–548. Available at: https://doi.org/10. 1177/0047287518769764. ICAO. 2019. Climate Change. 2019. Available at: https://www.icao.int/environmental-protec tion/pages/climate-change.aspx. International Maritime Organization. 2018. UN Body Adopts Climate Change Strategy for Shipping. 2018. Available at: http://www.imo.org/en/MediaCentre/PressBriefings/Pages/ 06GHGinitialstrategy.aspx. IPCC. 2014a. Climate Change 2014: Impacts, Adaptation, and Vulnerability. Contribution of Working Group II to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change. Available at: http://www.ipcc.ch/report/ar5/wg2/. IPCC. 2014b. Climate Change 2014: Mitigation of Climate Change. Contribution of Working Group III to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change. Available at: https://www.ipcc.ch/report/ar5/wg3/. IPCC. 2018. Global Warming of 1.5°C. Available at: http://www.ipcc.ch/report/sr15/.

82 Elisabeth Ellis (Otago)

Kommenda, N. 2019. 1% of English Residents Take One-Fifth of Overseas Flights, Survey Shows. The Guardian, 25 September 2019, sec. Environment. Available at: https://www.theguardian.com/environment/2019/sep/25/1-of-english-residents-takeone-fifth-of-overseas-flights-survey-shows. Kowarsch, M. & Edenhofer, O. 2016. Principles or Pathways? Improving the Contribution of Philosophical Ethics to Climate Policy. In: Heyward, J. C. and Roser, D. (eds) Climate Justice in a Non-Ideal World, Oxford University Press. Available at: https://doi.org/10. 1093/acprof:oso/9780198744047.003.0015. Laville, S. 2019. Critics Attack Secrecy at UN Body Seeking to Cut Global Airline Emissions. The Guardian. 2019. Available at: https://www.theguardian.com/business/2019/ feb/11/critics-attack-secrecy-at-un-body-seeking-to-cut-global-airline-emissions. Lee, D. S., Fahey, D. W., Forster, P. M., Newton, P. J., Wit, R. C. N., Lim, L. L., Owen, B. & Sausen, R. 2009. Aviation and Global Climate Change in the 21st Century. Atmospheric Environment, 43, 22–23, 3520–3537. Available at: https://doi.org/10.1016/j.atm osenv.2009.04.024. Levy, J. T. 2015. Rationalism, Pluralism, and Freedom. First edition. Oxford, UK: Oxford University Press. Levy, J. T. 2016. There is No Such Thing as Ideal Theory. Social Philosophy and Policy, 33, 1–2, 312–333. Available at: https://doi.org/10.1017/S026505251600025X. Marchau, V. A. W. J., Walker, W. E., Bloemen, P. J. T. M. & Popper, S. W. (eds) 2019. Decision Making under Deep Uncertainty, New York, NY: Springer Berlin Heidelberg. Markham, F., Young, M., Reis, A. & Higham, J. 2018. Does Carbon Pricing Reduce Air Travel? Evidence from the Australian ‘Clean Energy Future’ Policy, July 2012 to June 2014. Journal of Transport Geography, 70 (June), 206–214. Available at: https://doi.org/10. 1016/j.jtrangeo.2018.06.008. Markowitz, E. M., Grasso, M. & Jamieson, D. 2015. Climate Ethics at a Multidisciplinary Crossroads: Four Directions for Future Scholarship. Climatic Change, 130, 3, 465–474. Available at: https://doi.org/10.1007/s10584-015-1404-4. Mathur, V. N., Afionis, S., Paavola, J., Dougill, A. J. & Stringer, L. C. 2014. Experiences of Host Communities with Carbon Market Projects: Towards Multi-Level Climate Justice. Climate Policy, 14, 1, 42–62. Available at: https://doi.org/10.1080/14693062.2013.861728. McWha, V. and Murray, K. 2015. Effects of an Increase in Travel Ticket Price on New Zealand Tourism. Sapere Research Group. Ministry of Primary Industries, New Zealand. Miller, D. 2007. National Responsibility and Global Justice. Oxford University Press. Moellendorf, D. & Widdows, H. (eds).2015. The Routledge Handbook of Global Ethics. 1st ed. Routledge. Available at: https://doi.org/10.4324/9781315744520. Moss, J. & Fraser, P. 2019. Australia’s Carbon Majors. Practical Justice Initiative, UNSW. Moss, J. & Walsh, E. 2019. Complicity in Climate Harms: A Case Study of Australia’s Gas Export Industry, Vol. 36, Part V. Environmental and Planning Law Journal, 36, 5, 477–488. Negroni, C. 2016. How Much of the World’s Population Has Flown in an Airplane? Air & Space Magazine. Available at: https://www.airspacemag.com/daily-planet/how-muchworlds-population-has-flown-airplane-180957719/. Nordhaus, W. 2015. Climate Clubs: Overcoming Free-Riding in International Climate Policy, American Economic Review 105, 4, 1339–1370. Available at: https://doi.org/10. 1257/aer.15000001. Peeters, P., Higham, J., Kutzner, D., Cohen, S. & Gössling, S. 2016. Are Technology Myths Stalling Aviation Climate Policy? Transportation Research Part D: Transport and Environment, 44 (May), 30–42. Available at: https://doi.org/10.1016/j.trd.2016.02.004. Pesmajoglou, S. 2018. Carbon Offsetting and Reduction Scheme for International Aviation— CORSIA. Paper presented at the ICAO capacity building seminar on low emissions aviation

Sectoral responsibility for climate justice 83

measures. Available at: https://www.icao.int/environmental-protection/Documents/Pre sentations/1_3_Carbon%20Offsetting%20and%20Reduction%20Scheme%20for%20Interna tional%20Aviation_CORSIA.pdf. Pidcock, R. & Yeo, S. 2016. Analysis: Aviation Could Consume a Quarter of 1.5C Carbon Budget by 2050. Carbon Brief. Available at: https://www.carbonbrief.org/aviation-con sume-quarter-carbon-budget. Pindyck, R. S. 2019. The Social Cost of Carbon Revisited. Journal of Environmental Economics and Management, 94 (March), 140–160. Available at: https://doi.org/10.1016/j.jeem.2019.02.003. Posner, E. & Weisbach, D. 2010. Climate Change Justice, Princeton; Oxford: Princeton University Press. Available at: https://press.princeton.edu/books/paperback/9780691166667/ climate-change-justice. PPCA. 2019. Powering Past Coal Alliance | Working towards the Global Phase-out of Unabated Coal Power. Powering Past Coal Alliance (PPCA). Available at: https://power ingpastcoal.org. Rawls, J. 2001. Justice as Fairness: A Restatement, Harvard University Press. Rogelj, J. & Forster, P. 2019. Guest Post: A New Approach for Understanding the Remaining Carbon Budget. Carbon Brief. 17 July. Available at: https://www.carbonbrief. org/guest-post-a-new-approach-for-understanding-the-remaining-carbon-budget. Romera, B. M. & Van Asselt, H. 2015. The International Regulation of Aviation Emissions: Putting Differential Treatment into Practice. Journal of Environmental Law, 27 (April). Schiff, A. & Becken, S. 2011. Demand Elasticity Estimates for New Zealand Tourism. Tourism Management, 32, 3, 564–575. Available at: https://doi.org/10.1016/j.tourman. 2010.05.004. Schlosberg, D., Collins, L. B. & Niemeyer, S. 2017. Adaptation Policy and Community Discourse: Risk, Vulnerability, and Just Transformation. Environmental Politics, 26, 3, 413–437. Available at: https://doi.org/10.1080/09644016.2017.1287628. Sen, A. 2011. The Idea of Justice. 1. Pbk. ed. Cambridge, Mass: Belknap Press of Harvard University. Shue, H. 1999. Global Environment and International Inequality. International Affairs, 75, 3, 531–545. Stephens, S. A., Bell, R. G. & Lawrence, J. 2018. Developing Signals to Trigger Adaptation to Sea-Level Rise. Environmental Research Letters, 13, 104004. Available at: https://doi. org/10.1088/1748-9326/aadf96. Transport & Environment. 2019. Aviation in the ETS. Available at: https://www.transp ortenvironment.org/what-we-do/aviation/aviation-ets.

5 CORPORATIONS’ DUTIES IN A CHANGING CLIMATE Stephanie Collins (ACU)

Introduction When we consider the harms of human-caused climate change, there are numerous entities to which we might want to assign duties. Perhaps most obviously, we turn to states. Moral and political philosophers have had a lot to say about states’ climate-related duties. But the urgency of the situation calls upon us to investigate other, less obvious, duty-bearers. One natural reaction is to go from the largest agents to the smallest—from states to individuals. Do individuals have duties to fly less, drive less, and eat less meat? In Section I, I argue that this question raises difficult issues about individual difference-making. This prompts us to turn elsewhere. The rest of the chapter focuses on (what I take to be) the third most-salient duty-bearer: large for-profit corporations.1 These entities have largely been overlooked in philosophical discussions of climate-related duties. In Section II, I consider two possible reasons for this neglect, and argue that neither are good reasons. In Section III, I give a positive case for weighty and demanding duties for corporations, to cut back their present and planned emissions and to offset their past emissions. These duties are grounded in (at least) three facts about corporations: corporations are capable of doing something significant about climate change (that is, the difference-making problem does not arise for them), corporations benefit from climate injustice, and corporations are culpable causers of climate change. In Section IV, I bring the discussion full circle: corporations’ duties always imply duties for corporations’ members, that is, for the individuals who constitute the corporation. Drawing on earlier work (Collins, 2019), I give an account of who corporations’ members are and how their duties are structured. In heavily-emitting corporations, members prominently include managers, shareholders, and rank-andfile employees. So a range of individuals are on the hook after all, because the

Corporations’ duties in a changing climate 85

corporations that they constitute bear duties. Although individuals cannot make a significant expected difference to the harms of climate change by reducing their personal emissions, they often can make a significant expected difference to whether corporations discharge their climate-related duties. Additionally, even non-members have duties to act upon corporations, from the outside, with a view to inducing corporate duty-fulfilment. These ‘responsiveness duties’ are held by non-member individuals and non-member collective agents—most prominently, states, who have it within their regulatory and legislative power to reign in corporations’ emissions.

I. Individuals’ duties: the problem of difference-making As is now well-known amongst moral and political philosophers, it’s difficult to argue that individuals make a significant difference to climate harms when they fly, drive, or eat meat. It’s therefore difficult to argue that they have duties to refrain from these activities.2 It’s easiest to argue that individuals make a significant difference if we understand climate change as a threshold problem. A threshold problem is a situation in which harm will result if and only if enough agents do their contributory part in some pattern of behaviours that includes many agents, but where one individual acting alone is not enough. Following an example given by Shelly Kagan (2011), suppose a supermarket will order another crate of one thousand chicken carcasses every time it sells one thousand chicken carcasses. Every time the supermarket orders one crate, one thousand chickens are slaughtered. The relevant pattern of actions, in this situation, is people buying chicken carcasses. An individual who buys one carcass (and who has no information about how many carcasses have been sold since the supermarket last ordered a crate) should have 0.001 credence that her purchase will trigger the slaughter of one thousand chickens. In terms of expected disvalue, a 0.001 chance of causing one thousand chicken deaths is equivalent to a 1.0 chance of causing one chicken death. So the chickenbuyer makes a difference of one chicken death ‘in expectation’—that is, that’s the number of deaths we get when we add the various possible numbers of deaths triggered by her action, where each of those possibilities has been multiplied by the credence she should have that the number of deaths will occur if she performs the action. She has a small chance of making a hugely significant difference to how many chickens die, and she has a high chance of making no difference at all. By averaging across these possibilities, the chicken-buyer makes an expected difference that is somewhat significant: the difference of one chicken’s life. Non-threshold cases are harder. Following Parfit (1984, 76–78), Julia Nefsky (2017, 2743–2744) describes one such case: Imagine that there are ten thousand men in the desert, suffering from intensely painful thirst. We are a group of ten thousand people near the desert, and each of us has a pint of water. We can’t go into the desert ourselves, but what we

86 Stephanie Collins (ACU)

can do is pour our pints into a water cart. The cart will be driven into the desert, and any water in it will be evenly distributed amongst the men. If we pour in our pints, the men’s suffering will be relieved. The problem is, though, that while together these acts would do a lot of good, it does not seem that any individual such act will make a difference. If one pours in one’s pint, this will only enable each man to drink an extra ten thousandth of a pint of water. This is no more than a single drop, and a single drop more or less is too minuscule an amount to make any difference to how they feel.3 In non-threshold cases, each individual’s action has no chance of making a significant difference. There is no configuration of the other individuals’ actions such that any given individual’s action will trigger a threshold that matters. No matter how the pattern of pouring is configured and no matter what others do, each individual makes no more than a tiny, imperceptible difference to the men’s thirst. The person also doesn’t make a perceptible difference in expectation, if we were to average across all the possibilities. Is climate change a threshold case, or a non-threshold case? Holly LawfordSmith (2016) argues it’s a threshold case, with two broad types of threshold events. There are ‘macro’ threshold events, such as the permafrost melting or the Amazon rainforest dying out, and ‘micro’ threshold events, such as a particular weather event happening in a particular place, on a particular day, with a particular level of severity. Macro thresholds are more morally significant than micro thresholds. The less significant the threshold, the higher my chance of triggering it (because fewer contributions are needed to make the difference between, e.g., a storm happening today versus tomorrow); the more significant the threshold, the lower my chance of triggering it (because the more contributions are needed to make the difference between, e.g., the permafrost melting or not). For both kinds of threshold, we can apply the expected-value reasoning from the chickens case. Accounting for all thresholds types and rational credences, Lawford-Smith concludes that individual actions of (refraining from) driving, flying, or eating meat make something of a difference, in expectation. John Broome (2019) reasons similarly. He claims one joyride will definitely make a difference to the most microscopic of micro-thresholds—the precise nature of a weather event, for example—where this difference could be positive or negative. For example, your drive might cause a storm to happen a few minutes earlier than it otherwise would have, flooding a field minutes before a farmer was about to fill it with livestock that would have been killed in the flood if not for your drive. Nonetheless, it’s more likely that your drive will have a negative effect than a positive effect, because the incidence of harms will increase with the global concentration of greenhouse gases. So your drive does expected harm. These approaches rely on thresholds. However, there’s reasonable disagreement amongst climate scientists about how to model the system.4 It might turn out to be false that enough climate-related harms that have sharp enough thresholds to get expected value reasoning off the ground. That is, perhaps many of the harms are of

Corporations’ duties in a changing climate 87

the non-threshold variety. For example, Broome himself mentions the continuous harmful process of falling water tables. Our moral theory should be applicable to a world in which climate change turns out not to have many, or sufficiently significant, thresholds. Second, even if each flight, drive, or steak causes a perceptible expected harm, that harm may not be significant. The problem here is that each and every action we perform has some chance of causing some harm. And many of our possible actions cause net expected harm. If morality demanded that we avoid performing each and every action that causes net expected harm, this would plausibly violate the limits of morality’s demandingness (Lichtenberg, 2010, esp. p. 559). What’s more, the moral imperative to avoid causing harm may not always be more morally important than the moral imperative to produce good. So we have to pick our battles: often, avoiding harm is a matter of choosing the lesser of two evils-inexpectation. We can think of an action as having a ‘significant’ expected harm if it has an expected harm that is worse than the expected harm of any other action that is open to the agent at that moment. In this way, whether a given action makes a ‘significant’ expected difference depends upon what the agent could have done instead, and what expected difference that action would have made. Does each flight, drive, or steak make a significant expected difference? There’s room for debate. If one refuses to fly to a significant family event, is one causing harm to one’s family, or merely failing to provide them with a benefit? If one is causing harm to them, is this harm greater or lesser than the expected harm of the flight? Answering the second of these questions requires comparing values that may be incommensurable. Even before we have considered incommensurability, we need some method of measuring the expected harm of the flight. The British government’s Stern Review placed the expected harm caused by carbon at between $25 and $85 per tonne. The average Australian emits between 3 and 30 tonnes of CO2 per year, which (using the Stern Review’s estimate) is $75 to $2,550 worth of expected harm per year.5 Divided across all of one’s carbonemitting activities in a year, it’s clear that a single flight, drive, or steak does very little expected harm (a few dollars’ worth); plausibly less than is caused by missing a significant family event.6 Can we justify individuals’ climate-related duties via something other than expected harm? Perhaps. Individuals might have duties not to symbolically endorse or condone harm, or they might have duties to express an attitude of concern about such harm. The problem, though, is that in a society in which the vast majority of people have adopted carbon-intensive lifestyles, it’s not clear that simply forgoing one flight or one steak does the work of endorsing, condoning, or expressing these things. Endorsing, condoning, and expressing is communicative. Success relies, at least in part, on the communicator’s audience having the right response to the communicative act. In some contexts, the action of forgoing a flight or a steak is as likely to be perceived as virtue signalling, or as self-righteousness, as it is to be perceived as a genuine act of endorsing, condoning, or expressing the right kinds of values.

88 Stephanie Collins (ACU)

This is not meant as an argument against individuals’ climate-related duties. We might try to deny that difference-making is the best account of causation. Or we might try to argue that the values at stake are such that any perceptible expected difference is significant enough to trigger a duty, no matter the other values at stake. Individuals’ duties might be part of the story. My purpose has been to highlight the challenges faced by defenders of individuals’ climate-related duties, in order to motivate the thought that we would be on firmer ground if we had more tools in our climate-justice shed. The tool I’ll suggest is corporations’ duties.

II. Corporations’ duties: reasons for scepticism? Corporations tend not to feature in the philosophical literature on climate-related duties.7 In this section, I consider two possible rationales for this neglect. Before doing so, I should clarify what I mean by ‘corporation.’ These are organisations with the goal of making profit. An organisation is made of a large number of people, usually alongside other material objects such as buildings, computers, and paper (on the need to include objects, see Epstein, 2015). To make up an organisation, the people and objects must together instantiate an organisational structure. Organisational structures can be modelled by diagrams that show nodes (roles) connected by edges (relations). Organisational structures specify ‘(a) criteria to establish their boundaries and to distinguish their members from non-members, (b) principles of sovereignty concerning who is in charge and (c) chains of command delineating responsibilities within the organization.’ (Hodgson, 2007) For example, a particular organisational structure might include the ‘manager’ node, which is connected to the ‘customer server’ node by a bidirectional edge made up of the ‘gives instructions to’ relation (in the manager-to-server direction) and the ‘is accountable to’ relation (in the server-to-manager direction). Following Katherine Ritchie (2013), when enough people and objects occupy the nodes in the structure, the organisation exists in the world and can be identified with this instantiated structure. This might make a corporation sound inert, frozen in time. In order to act in the world, a corporation must form beliefs (about how the world is) and goals (for how it wants the world to be), where the material constituents (humans and objects) can pursue the goals in light of the beliefs. For this to happen, the corporation needs some procedure for deciding its beliefs and goals. Corporations need procedures to make these decisions, because they are made up of many conflicting agents, each of whom might have different views on what the corporation should believe or prefer. These procedures have been theorised extensively in the literature, most prominently in Peter French’s (1979; 1984) discussion of ‘corporate internal decision (CID) structures’ and, more recently, in Christian List and Philip Pettit’s (2011) discussion of ‘aggregation functions’. As Kendy Hess (2018, 37–38) points out, the decision-making procedures used by actually-existing corporations are messy: they include not just voting, committees, and decrees, but also discretion, debates, water-cooler conversations, bargains, horse-trades, compromises, and so on. The most important parts of the procedure

Corporations’ duties in a changing climate 89

are often not explicit or predictable. This hotchpotch of procedures nonetheless enables a corporation to make decisions, in ways that demonstrate diachronic and synchronic rationality (Rovane, 1998; List and Pettit, 2011). They are therefore agents in the world. In the next section, I will give a positive argument for corporations’ climaterelated obligations. Before that, I want to undercut two potential reasons for moral philosophy’s neglect of corporations’ climate-related duties, as compared with our treatment of states’ and individuals’ duties. The first reason might be that we think corporations are not moral agents, so cannot bear duties.8 For corporations to be moral agents, they must be capable of giving due weight to moral considerations when forming their bundle of beliefs and goals (Hindriks, 2018; Collins, 2019, ch. 6). Perhaps corporations are agents that can process some kinds of considerations, but not moral considerations, when arriving at beliefs and goals. One reason for thinking this is that I characterised corporations as having the goal of making profit. If this is their definitional goal, then perhaps other goals can enter the picture only insofar as they are instrumental to this goal. And if morality—including reigning in and offsetting emissions— conflicts with making profit, then perhaps corporations are constitutionally incapable of acting in accordance with morality. This line of reasoning paints an overly myopic picture of corporations. Corporations definitionally have the goal of making profit, but they don’t definitionally have the goal of maximising profit—nor do they definitionally have no goals other than making profit. If making (some) profit is consistent with abiding by morality, then there’s nothing in the definition of a corporation that rules out its abiding by morality as an additional non-instrumental aim. As long as members are not forbidden from bringing moral considerations to bear on the corporation’s decisionmaking, and as long as the corporation has the basic material and structural resources needed to make decisions on the basis of those moral considerations, the corporation is a moral agent. It can therefore bear moral duties, including climaterelated duties. A second reason for neglecting corporations’ duties might be that we worry about wasting our breath. Corporations tend not to listen to anything except shareholders’ votes and customers’ wallets. The idea that they might change their behaviour because of philosophical argument seems laughable. This, I suggest, fails on two counts. First, the fact that an entity won’t do what it ought, or won’t listen to philosophers, has never stopped philosophers from arguing about what that entity ought to do. Most humans are not convinced by (because they do not engage with) philosophical arguments about their duties. But we make those arguments anyway, for at least two reasons: because the arguments might come in handy later (if humans ever change their minds), and because it’s important to be able to appropriately assess whether people are doing what they ought. These considerations apply to corporations. If there’s one way to ensure corporations don’t heed moral arguments, it’s to fail to provide any.9

90 Stephanie Collins (ACU)

Second, corporations can and do pay lip service to climate-related duties—even if they do so as an instrument to profit-making. For example, the oil company Statoil changed its name to Equinor in 2018, in a bid to market its stated intention to invest 15–20% of capital expenditure into ‘new energy solutions’ by 2030.10 And in 2018, Royal Dutch Shell—one of the largest oil companies in the world— released plans to cut its carbon footprint 50% by 2050. The CEO Ven van Beurden claimed that ‘If you want to be a long-term relevant company that is on the right side of history, you have to be involved in this [climate change] discussion, because it’s the most important discussion of our time.’ (Peters, 2018) Van Beurden’s statement is, without a doubt, a marketing strategy. But it demonstrates that moral considerations do get in, if corporations believe there is something to be gained from letting them in. If we can convince shareholders and consumers that corporations have climate-related obligations, philosophers’ arguments will not have been in vain.

III. Multiple sources of duties On what basis might corporations have duties to cut and offset their emissions? It is now common to distinguish (at least) three sources of climate-related duties: capacity, benefit, and culpable causation (Caney, 2005; Caney, 2010; LawfordSmith, 2014). That is, one can have a duty to cut and offset one’s emissions if one is capable of thereby ameliorating the effects of climate change (the ‘ability to pay’ principle), or if one benefits from climate-related injustice (the ‘beneficiary pays’ principle), or if one has culpably caused climate change (the ‘polluter pays’ principle). I will assume each principle is sufficient to ground a duty. I will also assume that they operate additively: the higher the number of these principles that apply to an entity, and the more fully each principle applies (the greater one’s ability, the larger one’s benefit, and the more culpable one’s pollution), the weightier and more demanding is that entity’s duty. A duty is weighty if it is not easily outweighed by other duties in cases of duty conflict. A duty is demanding if it can ask a high cost from its bearer, should a high cost be necessary to produce the relevant outcome. All three duty-generating principles apply to corporations—so, corporations end up with very weighty and very demanding duties. I’ll work through the three principles in turn.

Capacity-based duties As I’ll now explain, corporations have the capacity to drastically cut and offset their emissions, thereby making a difference to the future effects of climate change. In this way, they do not face the difference-making challenge for individuals that I described in Section I. This difference between corporations and individuals arises for two main reasons. The first has to do with actions versus policies. It’s reasonable for an individual to ask whether a given act will make a difference, in expectation, to the harms of

Corporations’ duties in a changing climate 91

climate change. Taking a drive, booking a flight, or ordering a steak is always a discreet, isolatable choice—one whose effects we can ask about, and which can be found to make an insignificant difference in expectation. Corporations, by contrast, are creatures of planning and policy. They tend not to make such small-scale decisions. Each small-scale action of a corporation—each collection of a household energy payment, each barrel of oil extracted—tends to be performed as a direct and unthinking consequence of some larger plan or policy. So, for corporations, it makes more sense to ask whether a plan or policy makes a difference, rather than whether each particular act makes a difference. Plans and policies make more difference than acts. Now, of course, individuals have plans and policies as well (humans are creatures of habit)—but corporations are more so. We can sensibly give both entity-types recommendations about their actions and their policies, but policies are the more appropriate unit of assessment for corporations. Second, even comparing individuals’ policies with corporations’ policies, corporations are far larger emitters than individuals. So, they can make much more of a difference, by halting and offsetting those emissions, than individuals would, by halting and offsetting theirs. To illustrate, Saudi Aramco (one of the world’s biggest oil corporations) emitted greenhouse gases equivalent to 1,550,000,000 tonnes of CO2 in 2010 (Heede, 2014, 237).11 Recall that the average Australian emits between 3 and 30 metric tons of CO2 per year.12 Even allowing for some vagueness in what counts as ‘significant’, the world’s biggest corporations can make a far more significant difference than individuals.

Benefit-based duties Many corporations also benefit from greenhouse gas emissions: they emit as a means to a profit, after all. But for a benefit to produce a duty, it can’t be just any old benefit. Benefits produce duties when they are received as a result of an injustice. Do corporations’ climate-related benefits result from injustice? In answering this, I’ll place a high bar on what counts as an injustice. That way, the duties that result from benefits from injustice will be relatively difficult to come by, and so we will have a strong case for those duties being weighty and demanding. (This differs from Lawford-Smith’s 2014 argument about benefit-based climate-related duties, on which an injustice occurs if the world falls short of an ideal.) I’ll assume that an ‘injustice’ arises when an agent has failed to perform an enforceable directed duty—where an enforceable duty is a duty that third parties are permitted to coerce the agent into performing and where directed duties correlate with rights. Enforceable directed duties are important, so if you benefit—even nonculpably—from someone’s failure to perform one, then you should disgorge that benefit to whoever had a right to the performance of the enforceable duty.13 Do corporations benefit from climate-related injustices? There’s no way to answer fully without begging the question at issue, namely, the question whether corporations have (enforceable directed) climate-related duties. This is because, to identify all benefits from injustice, we first must identify all injustices, which requires identifying

92 Stephanie Collins (ACU)

all directed enforceable duties. But this is not a major problem. We can assume states have enforceable directed duties to regulate market actors away from greenhouse gas emissions. States owe this to their younger generations of citizens (and perhaps younger generations of non-citizens). This simply follows from the fact that states owe it to their citizens to do what states can to ensure all citizens enjoy a minimally decent quality of life. States’ failures to discharge these duties cause corporations’ benefits. This is true on both a ‘process’ and ‘counterfactual’ theory of causation.14 In ‘process’ language: states’ failures to regulate are the mechanism that legally permit corporations to emit; emissions are the mechanism by which corporations profit in the actual world; so, states’ failures are part of the process by which corporations gain profit. On the ‘counterfactual’ sense: if states had done their regulatory duty, then corporations wouldn’t have been legally permitted to emit; if corporations hadn’t been legally permitted to emit (under technological conditions similar to our own), then corporations would have gained less profit; so the counterfactual possibility in which states did their duty is one in which corporations are less well-off than the actual world (assuming similar technological conditions). Thus corporations benefit from climate injustice. Corporations thereby owe two behaviours to the victims of this injustice (younger citizens): (i) stop receiving the benefits and (ii) disgorge benefits already received.

Causation-based duties Finally, corporations culpably cause climate change.15 As I mentioned above, when we look prospectively into the future, corporations’ expected effects on the climate via their plans and policies are much larger than individuals’ expected effects via their actions. When we look retrospectively into the past, this is even more stark. In a paper from 2014, Richard Heede argued that 63% of global carbon and methane emissions from the last 260 years can be traced to the production activities of ninety collective agents that he refers to as ‘carbon majors’. These are ‘investorowned… state-owned… and nation-state producers of oil, natural gas, coal, and cement.’ Half of these emissions—which is 914 gigatonnes of CO2-equivalent— have been emitted since 1986. This figure is crucial, since 1986 is plausibly around the point at which emissions became culpable: the point at which agents could not plausibly deny that they knew, or could reasonably have been expected to know, about the negative effects of emissions. Of the ninety carbon majors, fifty are corporations (rather than states or stateowned enterprises). These fifty corporations almost certainly each made a significant difference, and each did so culpably. In 2010, the biggest corporate emitters included Saudi Aramco, Gazprom, National Iranian Oil Company, Coal India, Pemex, ExxonMobil, BP, and Shell. You might think these corporations don’t really cause emissions—at least not culpably—because they do so only at the behest of consumers, who buy or use products that depend upon oil, natural gas, coal, and cement. So it’s really consumers (not corporations) that culpably cause emissions, and who force corporations to emit

Corporations’ duties in a changing climate 93

on consumers’ behalf. Yet consumers use products that rely on GHGs only because the socio-economic structures in which they live make it almost impossible to choose otherwise. And what caused those structures to be as they are? We cannot answer this except with reference to the entire causal chain, from extractors to manufacturers to retailers to consumers. Everyone in the chain is somewhat stuck. Placing the blame on consumers attributes too little causal power to socioeconomic structures. Consumers are no more to blame than others in the structure. That said, each agent in the chain can take actions to push the chain in a different direction. The largest corporations—such as those named by Heede—are, and were, particularly capable of doing so. They operate at a huge profit, they often operate as monopolies or near-monopolies within particular countries, and they have the technological capability. This means that they had capacity-based duties in recent decades. Because they did not perform those duties, they now additionally have culpable-causation-based duties. As I discussed in Section I, individual consumers arguably did not have the capacity to make a difference. If so, then they didn’t once have a capacitybased duty, so they aren’t now culpable for failing to perform that duty, so their culpability isn’t a basis for their present-day duties. Corporations are on the hook in ways individuals are not.

IV. Back to individuals: the implications of corporations’ duties In characterising corporations, I said they exist only when their structures (organisational role-charts) are instantiated. The structure becomes instantiated only when there are (enough) individual role-bearers, occupying nodes (roles) in the structure. If there aren’t enough managers, officers, and workers, the corporation simply ceases to exist (that is, ceases to exist agentially, and so for the purpose of moral duties; it may exist according to the law). This role of humans as the material constituents of corporations becomes even more important when we consider how corporations might perform their duties to reduce and offset their emissions. I’ve argued elsewhere that duties of collective agents always imply duties for members (Collins, 2019, ch. 7). We can label the latter ‘membership duties’. Membership duties are duties held by agents in virtue of the fact that they are members of a collective that has a duty, where the performance of the membership duty is a component of the collective’s doing its duty. More specifically, I’ve argued that if a collective agent (whether a corporation or another kind of collective agent) has a duty to see to it that X, then 1. Each member has a duty to use their role, if possible and as appropriate, to put inputs into the collective’s decision-making procedure with a view to the procedure’s distributing roles to members in a way that: if enough members used their roles with a view to seeing to it that X, then that would be sufficient for X in a high proportion of likely futures. These are ‘X-sufficient’ roles. 2. If X-sufficient roles are distributed, then each member has a duty to use their role, if possible and as appropriate, with a view to seeing to it that X.

94 Stephanie Collins (ACU)

I’ll assume the members of a corporation are the agents who meet three conditions: (1) the agents are pro tanto committed (even if only tacitly) to abiding by the results of the corporation’s decision-making procedure; (2) the inputs of the corporation decision-making procedure, and the way it processes those inputs to form decisions, relies on the behaviour of the agents, while being distinct from the set of inputs, and procedures (if any), that any one of the agents uses when deciding for themselves; and (3) the enactment of at least some of the corporation’s decisions requires actions by the agents, where those actions are also properly attributable to the collective (Collins, 2019, ch. 1). That formulation is abstract. How does it work for, say, an energy retailer? It might seem a bit much to impose membership duties on low-level employees of such a corporation, when the employees have little decision-making power. Consider an energy retailer employee, whose job is to visit customers’ houses and read their meters. This person is pro tanto committed to abiding by the corporation’s decisions: she’ll generally go to whichever houses the corporation demands she go to, when it demands she go there. That commitment is overridable, and hence pro tanto, but her presumption in decision-making is that she will stick to it. This person also has some input into the corporation’s decision-making. She can decide the corporation will engage with customers in a friendly way, or not, for example. The inputs into the corporation’s decision of whether or not to be friendly, and the way it makes that decision, depends crucially on that member. In this way, any role that contains discretion is going to enable its bearer to have inputs. Finally, some of the meter-reader’s actions are incorporated by the corporation: when she reads meters, the corporation reads meters. (I’ll say more below about which actions of hers are also the corporation’s actions—to foreshadow, her actions are incorporated when she acts within and because of her role.) Are there members of an energy retailer who are not employees? Yes: crucially, shareholders. It might not look plausible that shareholders meet the first of our three conditions: they might not seem committed to abide, since they don’t tend to receive orders. But notice that shareholders cannot, for example, vote on propositions that are not put to them at the shareholder meeting. The question of which propositions to put to shareholders is answered by the corporation’s decision-making procedure. Insofar as shareholders are committed not to take complete control of the agenda, then, they are committed to abide by the procedure. Shareholders more obviously meet the second condition: the procedure’s inputs and processes depend on their behaviours, namely, contributions to shareholder meetings. Finally, we can attribute actions of shareholders to the corporation: if shareholders vote for an entirely new Board, we might expect the headline ‘Company Overhauls Board Members’. The upshot is that an energy retailer has a lot of members, each of whom has duties based on the corporation’s climate-related duties. What is required of members? This question has two aspects: what is required from members qua individuals, and what is required from members qua members? It’s important that membership duties concern only the latter. Suppose you are the meter reader. We

Corporations’ duties in a changing climate 95

might ask whether you should quit your job. We might ask whether you should exploit or subvert your job, to trick the corporation into doing what it ought. We might ask whether you should blow the whistle on environment-destroying policies and practices. None of these are questions about membership duties. They are questions about the duties of you, as a private individual, who happens to work for the corporation. Perhaps you have such duties. However, as noted in Section I, we encounter difficult difference-making problems in arguing for these duties (at least, some of them: whistle-blowing might be a clear case of difference-making, if it would make a big enough splash). I will set these aside. Instead, I’m interested in duties that you can use your role to perform. By ‘use,’ I mean ‘act within and because of.’ That is, you use your role when you act in ways that are permitted (or required) by your role, because those ways of acting are permitted (or required) by your role. I’m interested in how you can use your role—not in how you can exploit it, or subvert it, or change it, or swindle it—because I assume that only your actions that use your role will count as the corporation’s actions; only those actions are actions of the material constituents of corporations (Collins, 2018). So it’s only your role-using actions that can partly constitute the corporation’s doing its duty. This is important for difference-making, as I’ll now explain. You might think that membership duties encounter exactly the differencemaking problems I discussed in Section I—especially for a large corporation with thousands of employees. But this misunderstands the relationship between roleusing actions and corporate actions. Members’ role-using actions do not merely cause corporations’ actions. Instead, they constitute those actions. This means they automatically make a difference to what the corporation does—at least at a finegrained level of description. When a shareholder votes for a Board member who cares about sustainability, this constitutes the corporation’s action of (partly) voting for that Board member. Additionally, even if we think only about the expected difference members make to the coarse-grained decisions of the corporation, being a member of a 35,000 person organisation is more likely to be significant than is being one of billions of people currently contributing to climate change. Now, the corporation’s partly voting for a sustainability-promoting Board does not ensure the corporation as a whole votes for a sustainability-promoting Board. So you might think the vote of an individual shareholder does not make a significant expected difference to how the corporation will act with regard to the climate. But recall how I characterised ‘significant expected harm’: an action’s expected harm is significant if it’s worse than expected harm of the agent’s alternative actions. In the context of membership duties, those alternative actions are the alternative things one could have done within and because of one’s role. Even the small expected goodness of a singular shareholder’s voting for a sustainability-promoting Board is enough to morally outweigh the expected goodness of that shareholder’s voting purely on the basis of profit, all else being equal. The difference-making story for membership duties, then, is this: the corporation can make a significant causal difference to climate change; individuals can’t make a significant causal difference to that; but members can make a significant

96 Stephanie Collins (ACU)

constitutive difference to what corporations do at a fine-grained level of description; members can also make a significant causal difference to what corporations do at a coarse-grained level of description, because a difference’s significance is determined by how it compares with the other things one could do within one’s role. Importantly, membership duties do not rely on members individually satisfying the capacity, benefit, or causation principles. All that matters is that the corporation does, and that they are members of the corporation. Individuals who are not members may also be on the hook. These individuals’ actions never constitute corporations’ actions, so their duties aren’t membership duties. But sometimes, non-member individuals can make a significant expected difference to what corporations do. Again, this is because whether an action’s difference is significant depends on what the agent’s alternatives are. Those living in affluent countries must buy their energy from somewhere. (They can go ‘off-grid’— but a duty to do so plausibly breaches the limits of moral overdemandingness.) More generally, non-members must engage with some corporations. In doing so, it makes sense to choose the action from amongst one’s alternatives that has the best expectation of inducing a corporation to do its duty. In doing this, one is ‘responsive’ to that corporation. As I have defined it elsewhere (Collins, 2019, p. 98), an agent, A, is responsive to another, B, just in case A acts upon B with a view to B responding to the reasons or duties that (A believes) B holds. It’s easy to see that individuals have duties to be responsive to one another. If my friend would be able to reduce her carbon emissions if only I gave her some spare solar panels that I have lying around, and if I could give her the solar panels at reasonable cost to myself (here satisfying the capacity principle), and if this the best expected difference I could make with the panels (so the difference is ‘significant’), then I should do so. By giving her the solar panels, I nudge her towards reducing her carbon emissions, since it is now much cheaper for her to discharge that duty. The same goes for individuals’ duties to act upon corporations from the outside, nudging the corporation towards doing what it ought. The membership duties and responsiveness duties that I’ve advocated have partly the same content as what individuals’ personal consumption-reduction duties would have had, if we’d been able to solve the difference-making problem. Membership duties are duties to act within and because of your role—so they will be rolefocused. These tend not to be the duties we normally think of when we think of individuals’ climate-related duties. But it depends what one’s role is: if one’s role usually involves flying, or attending catered corporate events, then one’s membership duty will call for (trying to) refuse to fly, or not eat animal products at corporate events. If one’s role is reading meters, then one’s membership duty may simply involve using one’s role (insofar as one can) to communicate one’s views to coworkers and line managers. Individuals are heavily influenced by others regarding workplace norms (Herzog, 2018), so this could make a large difference as it trickles through the corporation. But such workplace actions are not the first thing we think of under the banner of individuals’ climate-related duties. So by viewing individuals’ duties through the lens of membership duties, we get a change of focus.

Corporations’ duties in a changing climate 97

When it comes to individuals that act upon the corporation from the outside, thereby performing responsiveness duties, the required actions will look more like what we usually think of as individuals’ climate-related duties. For example, actions might include changing to a more climate-aware energy retailer. Perhaps most centrally, responsiveness duties require non-member individuals to vote for political parties that will better-regulate duty-bearing corporations. Assuming one will vote for some party or other, voting for climate-focused parties does make a significant difference (i.e., a difference that’s more significant than voting for other parties). With this proposal, we have come full circle: states are back in the picture, although indirectly, as a means to enabling corporations to discharge their duties.

Conclusion I started this chapter by setting states aside, and by expressing scepticism about individuals’ duties due to problems of difference-making. I argued that corporations have weighty and demanding duties to dramatically curtail their emissions, and to offset the effects of their past emissions. When we open the black box of the corporation, however, we find that we are once again face-to-face with individuals. But these are individuals under a different guise: individuals qua members, not individuals qua individuals. As members, individuals incur membership duties whenever their corporation has a duty. And even as non-members, individuals will sometimes have responsiveness duties: duties to act upon corporations from the outside, with a view to the corporations doing their duty.

Acknowledgements I thank Lachlan Umbers for helpful feedback on a draft of this chapter. I thank Holly Lawford-Smith for helpful feedback on a draft, for numerous discussions about these issues, and for sharing her draft paper ‘“I Just Work Here”: The Case for Employees’ Obligations to Mitigate Climate Change,’ presented at Climate Justice Beyond the State, 29–30 November 2018, University of New South Wales. Although we came at these ideas independently and have different understandings of some key issues, I owe much to Holly’s treatment of the topic. I acknowledge financial support under the Australian Research Council’s DECRA scheme (project number DE200101413).

Notes 1 One might think the third most-salient duty-bearer is an unorganised group, such as ‘carbon emitters,’ ‘humanity,’ or ‘the rich.’ In Collins (2019, chs 2–3), I argue these kinds of groups cannot bear duties. 2 On difference-making generally, see Nefsky 2019; Lawford-Smith and Tuckwell forthcoming; on climate change specifically, Sinnott-Armstrong, 2005; Cripps, 2013, 119– 124; Kingston and Sinnott-Armstrong, 2018; cf. Lawford Smith, 2016; Broome, 2019.

98 Stephanie Collins (ACU)

3 Similar cases include Quinn’s (1990) ‘harmless torturers’ and Glover and Scott-Taggart’s (1975) ‘bean thieves’. Kagan (2011) argues there are no non-threshold cases. Nefsky (2012) provides a compelling reply. 4 The IPCC concludes that warming and total emissions are ‘approximately linearly related,’ which suggests a non-threshold model. But the relationship between warming and harms must contain at least some thresholds, for the simple reason that a death is a threshold. 5 https://www.epa.vic.gov.au/agc/r_emissions.html#/! 6 Broome’s (2012) discussion implies that you may attend the family event but must offset the flight’s harm by donating to a climate-related charity. Here the question of significant expected differences simply re-emerges: does your donation prevent enough expected harm to justify donating to offsetting rather than to, say, poverty relief? The answer isn’t obviously ‘no’, but it isn’t obviously ‘yes’. 7 Three exceptions are Shue (2017), Schwenkenbecher (2018), and Moss (this volume). 8 Schwenkenbecher (2018) also replies to this objection, with a different theory of corporate moral agency. 9 That is, by neglecting to argue for corporations’ duties, we license the words of Darren Woods, CEO of ExxonMobil: ‘fundamentally, if you look at what society is asking for, is not for companies like ourselves to go into those [renewable energy] sectors. Instead what they’re looking for is solutions to the risk of climate change.’ (Bloomberg, 2019) 10 https://www.equinor.com/en/news/15mar2018-statoil.html 11 I thank Holly Lawford-Smith for drawing my attention to Heede’s paper. 12 https://www.epa.vic.gov.au/agc/r_emissions.html#/ 13 For climate-related duties, you might think there’s a non-identity problem: how can corporations disgorge benefits to victims of climate injustice, when the victims don’t exist? After all, the people who exist given that the duties of justice have not been performed are different from the people who would have existed if those duties of justice had been performed. But climate change is also a problem between contemporaries: the duties of justice are owed from older currently-existing agents to younger currentlyexisting agents. 14 On these two theories see Schaffer, 2016. 15 Shue (2017) also argues for this.

References Bloomberg. 2019. Answer to Climate Change Will Be Found in Labs, ExxonMobil CEO Says. Japan Times. 9 May. Available at https://www.japantimes.co.jp/news/2019/05/09/ business/answer-climate-change-will-found-labs-exxon-mobil-ceo-says/#.XZGL10YzaUk Broome, J. 2012. Climate Matters, New York: WW Norton. Broome, J. 2019. Against Denialism. The Monist, 102, 110–129. Caney, S. 2005. Cosmopolitan Justice, Responsibility, and Global Climate Change. Leiden Journal of International Law, 18, 4, 747–775. Caney, S. 2010. Climate Change and the Duties of the Advantaged. Critical Review of International Social and Political Philosophy, 13, 1, 203–228. Clean Energy Regulator. 2017. 2016–17 Annual Report: Accelerating Carbon Abatement for Australia. Available at http://www.cleanenergyregulator.gov.au/DocumentAssets/Docum ents/Clean%20Energy%20Regulator%20Annual%20report%202016-17.pdf Collins, S. 2018. Who Does Wrong When An Organisation Does Wrong? In: Hess, K., Igneski, V. and Isaacs, T. (eds), Collectivity. Rowman and Littlefield. Collins, S. 2019. Group Duties: Their Existence and Their Implications for Individuals, Oxford: Oxford University Press.

Corporations’ duties in a changing climate 99

Cripps, E. 2013. Climate Change and the Moral Agent: Individual Duties in an Interdependent World, Oxford: Oxford University Press. Cullity, G. Climate Harms. The Monist, 102(1), 22–41. Epstein, B. 2015. The Ant Trap: Rebuilding the Foundations of the Social Sciences, Oxford: Oxford University Press. French, P. 1979. The Corporation As A Moral Person. American Philosophical Quarterly, 16, 3, 207–215. French, P. 1984. Collective and Corporation Responsibility, New York: Columbia University Press. Glover, J. & Scott-Taggart, M. 1975. It Makes No Difference Whether or Not I Do It. Aristotelian Society Supplementary Volume, XLIX, 171–190. Heede, R. 2014. Tracing Anthropogenic Carbon Dioxide and Methane Emissions to Fossil Fuel and Cement Producers, 1854–2010. Climatic Change, 122, 229–241. Herzog, L. 2018. Reclaiming the System: Moral Responsibility, Divided Labour, and the Role of Organizations in Society, Oxford: Oxford University Press. Hess, K. 2018. The Peculiar Unity of Corporate Agents. In: Hess, K., Igneski, V. and Isaacs, T. (eds), Collectivity. Rowman and Littlefield. Hindriks, F. 2018. Collective Agency: Moral and Amoral. Dialectica, 72, 1, 3–23. Hodgson, G. M. 2007. Institutions and Individuals: Interaction and Evolution. Organization Studies, 28, 97–116. Kagan, S. 2011. Do I Make a Difference? Philosophy and Public Affairs, 39, 2, 105–141. Kingston, E. & Sinnott-Armstrong, W. 2018. What’s Wrong With Joyguzzling? Ethical Theory and Moral Practice, 21, 169–186. Lawford-Smith, H. 2014. Benefiting from Failures to Address Climate Change. Journal of Applied Philosophy, 31, 4, 392–404. Lawford-Smith, H. 2016. Difference-making and Individuals’ Climate-related Duties. In: Heyward, C. and Roser, D. (eds), Climate Justice in a Non-Ideal World. Oxford: Oxford University Press. Lawford-Smith, H. & TuckwelL, W. Forthcoming. Act Consequentialism and the NoDifference Challenge. In: Portmore, D. (ed.) Oxford Handbook of Consequentialism. Oxford University Press. Lichtenberg, J. 2010. Negative Duties, Positive Duties, and the ‘New Harms’. Ethics, 120, 3, 557–578. List, C. & Pettit, P. 2011. Group Agency: The Possibility, Status, and Design of Corporate Agents, Oxford, UK: Oxford University Press. Moss, J. Forthcoming. Carbon majors and corporate responsibility for climate change. In: Moss, J. and Umbers, L. (eds) Climate Justice and Non-State Actors: Corporations, Regions, Cities and Individuals, Routledge. Nefsky, J. 2012. Consequentialism and the Problem of Collective Harm: A Reply to Kagan. Philosophy and Public Affairs, 39, 364–395. Nefsky, J. 2017. How You can Help, Without Making a Difference. Philosophical Studies, 174, 2743–2767. Nefsky, J. 2019. Collective Harm and the Inefficacy Problem. Philosophy Compass, 14, 4, 1–17. Parfit, D. 1984. Reasons and Persons, Oxford, UK: Clarendon Press. Peters, A. 2018. Is It Possible for an Oil Company to Help Fight Climate Change? Fast Company. 1 November. Available at https://www.fastcompany.com/90249937/is-it-possible-for-an-oilcompany-to-help-fight-climate-change Quinn, W. 1990. The Puzzle of the Self-Torturer. Philosophical Studies, 59, 79–90. Ritchie, K. 2013. ‘What Are Groups?’ Philosophical Studies, 166, 2, 257–272.

100 Stephanie Collins (ACU)

Rovane, C. 1998. Bounds of Agency: An Essay in Revisionary Metaphysics, Princeton: Princeton University Press. Schaffer, J. 2016. The Metaphysics of Causation. In:Zalta, E. N. (ed.), The Stanford Encyclopedia of Philosophy, Fall 2016. Available at https://plato.stanford.edu/archives/fall2016/ entries/causation-metaphysics/. Schwenkenbecher, A. 2018. Why Business Firms Have Moral Obligations to Mitigate Climate Change. In: Brueckner, M., Spencer, R. and Paull, M. (eds), Disciplining the Undisciplined? Perspectives from Business, Society and Politics on Responsible Citizenship, Corporate Social Responsibility and Sustainability, Springer. Shue, H. 2017. Responsible for What? Carbon Producer CO2 Contributions and the Energy Transition. Climatic Change, 144, 4, 591–696. Sinnott-Armstrong, W. 2005. It’s Not My Fault: Global Warming and Individual Moral Obligations. In:Sinnott-Armstrong, W. & Howarth, R. B. (eds.) Perspectives on climate change: science, economics, politics, ethics. Advances in the Economics of Environmental Resources; vol. 5. Amsterdam; London: Elsevier JAI.

6 INDIVIDUAL CLIMATE JUSTICE DUTIES The cooperative promotional model & its challenges1 Elizabeth Cripps (Edinburgh)

What should individual moral agents do in the face of global failure to act adequately on climate change? Duties to promote effective collective action have been widely defended but this conclusion leaves individuals not much the wiser when it comes to choosing between the very many promotional actions open to them. This dilemma is only exacerbated by recent events in global environmental politics. The current chapter draws on a nuanced, collectively sensitive consequentialism to present one appropriate and apparently promising way forward. Let us start with the assumption that the global elite should act together to secure climate change mitigation and adaptation, at least to the extent of protecting fundamental human interests. By the ‘global elite’ I mean most people in more developed countries and the rich minority in less developed ones. This claim has been defended philosophically as a shared or ‘weakly collective’ duty or as the putative duty of a putative group: a duty to organize as necessary to achieve some collective result (Cripps, 2013, pp. 48–82, Isaacs, 2011, pp. 144–153). This chapter asks what it means for individuals.2 It is frequently claimed that individuals should promote collective climate action (Johnson, 2003, Sinnott-Armstrong, 2005). I have elsewhere defended this as the individual’s primary climate justice duty, on the basis of fairness, efficiency and effectiveness (Cripps, 2013, pp. 140–155).3 However, this leaves the individual still unsure of what she should actually do. Her options range from what we might call ‘pure’ promotional actions, designed to bring about institutional change – voting, campaigning, marching, starting and signing petitions, writing to politicians, writing for newspapers, and so on – to lifestyle changes as part of wider movements such as veganism, or investing in renewable technology. One individual cannot do everything. So how does she choose? This chapter both defends and challenges my response to this: the Cooperative Promotional Model (CPM).4 It understands promotional duties in a broad sense

102 Elizabeth Cripps (Edinburgh)

(beyond merely promoting governmental action) and accounts not only for a duty to perform ‘pure’ promotional actions (such as campaigning or voting) but also to cut one’s own carbon footprint. Having outlined the model and set it in philosophical context, the chapter will highlight three key challenges and indicate briefly why a modified CPM is worth pursuing despite these.

Introducing the Cooperative Promotional Model The CPM requires individuals to fulfil the following duty. I present it here in a preliminary format; by the end of the chapter, a slight adjustment will have become necessary.

Cooperative Promotional Duty (CPD) Act together with motivated others, so far as possible at reasonable cost to oneself, to promote fair, efficient, effective global-level progress on climate change mitigation and adaptation.

I will return later to the ‘reasonable cost’ condition. ‘Acting together’ is intended to include a range of possibilities: strongly collective action via some institutional structure, less formal joint or collective action (with each individual thinking of herself as contributing to a collective result), or reliable coordination. A small-scale rescue case can provide illustration. A number of experienced climbers approaching several tourists in need of rescue across a hillside might be the local mountain rescue group following an established collective decision-making process or they might be a collection of individuals not previously constituting a group. In the latter case, they might act intentionally together by deliberating and determining between them who will rescue each, or they might coordinate in some mutually obvious way without the need for prior discussion (for example, each rescuing the person closest to her). In the climate change case, the situation is complicated by the global-level nature of the challenge. Two distinctions must be drawn. The first is between the global elite taken as a whole, who incur the shared or weakly collective duty, and the subset actually motivated to fulfil it, whom we can call The Motivated. Note that this model focuses on the duty to cooperate with other motivated agents but does not thereby assume that the long-term collective scheme will involve only The Motivated. Quite the contrary, since this would be unfair. I will return to this point. It is not currently feasible for The Motivated to act as a collective. The group has global span but overlaps with states and institutions, rather than neatly encompassing them, and has no decision-making structure of its own. Nor is it feasible for an individual to facilitate such collective action alone.5 Thus, applying the CPM requires drawing a second distinction: between The Motivated as a whole and the many smaller motivated groups or potential groups ((P)SMGs)

Individual climate justice duties 103

within it who are or could easily become capable of collective action. These range from small, often informal collections of individuals through to bigger scale movements, such as the vegan movement, or more or less formally structured collectives, such as NGOs. Achieving progress on fair, efficient, effective mitigation and adaptation will involve more or less strongly collective action by these (P) SMGs and more or less reliable coordination across them. Against this background, there are four steps to fulfilling the CPD.

Step 1 The individual must establish what effective coordination and distribution of individual contributions across the different (P)SMGs would look like. This depends on the chances of success of each effort (that is, whether each group achieves its own end), and how they fit with one another in terms of the overall aim (which, recall, is taken to be fair, efficient, effective global level progress on mitigation and adaptation). With regard to the latter, the actions of different (P)SMGs can be complementary or the reverse. They can be complementary in the straightforward sense of focusing on different aspects of the challenge, for example developing renewable or adaptation technology, or pursuing either institutional or widespread lifestyle change. However, they can also be complementary in a different sense: keeping different options open while it remains unclear which has the potential to succeed. For example, different technologies might be simultaneously researched or different political options explored, evaluated and promoted. However, there might be insufficient resources to pursue or promote all these possibilities adequately. Then, the activities of different (P)SMGs could conflict and, in combination, set back the overall end of fair, effective, efficient global progress. Suppose, for example, there are insufficient resources adequately to promote two major adaptation projects. Conflict can also arise more directly between the goals of different (P)SMGs, even if all are ultimately concerned with climate justice. Compare ‘sustainable intensification’ (Budolfson, 2018) with organic, wildlife-friendly agriculture combined with much lower meat and dairy consumption.6 There is a fine line between different groups pursuing different options complementarily (as it is not yet clear which is most feasible) and actively campaigning against one another in a way which risks setting back the overall end. Moreover, previously complementary subgroups might come into conflict as further evidence about feasibility emerges, or the demand for resources increases. All this is important to calculating how different (P)SMG efforts can optimally combine, but also generates potential challenges for the model, to which I will return.

Step 2 The individual must then ascertain where to devote her own effort, depending on the following factors:

104 Elizabeth Cripps (Edinburgh)

 





Her own skillset and that of other motivated individuals. The costs to herself and to motivated others of acting within different (P)SMGs. This is relevant because of the reasonable cost threshold: the costs of different potential cooperative actions will determine how much an individual can contribute before reaching that threshold, and so where she can be most effective. Her own position and influence, and that of motivated others.7 This is not just about occupying certain political or institutional roles, although that is important: the actions of the US president, the Pope or the CEO of a FTSE 100 company are more likely to bring about political or widespread behaviour change than those of most individuals. It is about influence in more general terms. Consider for example the potential for celebrities to promote social or even political change.8 Nor is the point that more effort is necessarily required of these individuals (though it might be of political or corporate leaders, because of the special liability of certain institutions for harm). Rather, more might be achieved in certain areas by certain individuals at an equivalent level of personal cost. The needs and salience of needs of different (P)SMGs. This follows from Step 1.

Step 3 Having worked out the optimum allocation of effort across the different (P)SMGs, the next step for the individual is to estimate and adjust for likely deviation from that optimum allocation by other motivated individuals. This might result from bias towards certain (perhaps local) ends or inadequate ability to communicate, acquire knowledge, or coordinate optimally.

Step 4 Finally, the individual is required to be flexible. This is not some one-off decision to which she afterwards unwaveringly adheres. Rather, she must adjust her behaviour and her cooperative actions as either the facts or her awareness of them changes. For example, it might become apparent that one option is technically infeasible or another has the momentum to succeed. To illustrate the process, consider the following analogy:

Islanders Fifty people are stranded on a small island where a dangerous storm is reliably forecast for the next day. They have no adequate boats. Another twenty people are on the mainland in a position to attempt a rescue. All are motivated to help. This might be approached in two ways: R1. Using one big boat. This is most likely to be effective but will not be possible in certain (unlikely but possible) weather conditions. R2. Using several smaller boats. This is riskier and less likely to succeed but is the only option in the above weather conditions.

Individual climate justice duties 105

Step 1 is for the individual to work out how the different subsets within Mainlanders could most fairly and efficiently coordinate for the rescue. Some might focus on different aspects of preparing for the rescue: equipping the boat(s), preparing to act as crew, and so on. If there are sufficient Mainlanders and resources to prepare for R1 and R2, a subgroup should focus on each, reliably coordinating to keep both options open. If this is not the case, two rival groups, each pursuing one of the two options would be collectively irrational. In Step 2, the individual estimates where she and others best fit in this effort, according to the needs of the different (P)SMGs, as well as her skills and those of others (are they experienced sailors or trained engineers?), her/their position or influence (is she particularly popular, or valued as a leader in some other sphere?), and the costs of taking on different roles. At the extreme, it might be unreasonably demanding for a breastfeeding mother or someone suffering from acute sea-sickness to take active part in the rescue. However, communication difficulties or widespread bias among others might mean that an individual should deviate from the allocation implied by (1) and (2) in order to contribute most effectively to the overall end (Step 3). Suppose there is no time for a formal decision making process to assign some Mainlanders to R1 and others to R2, but it would be optimal to prepare for both. An individual whose own skills would best fit her for mending and preparing the big boat might see that most others are gravitating that way, and instead prepare to crew a small boat. The last step is for the individual to adjust to changes. For example, if the weather becomes clearly settled for long enough to use the big boat, the individual will shift her efforts from preparing for R2, to helping to bring about R1 as quickly as possible.

Philosophical context Having outlined the CPM, I have now to provide a normative justification for it. As noted above, this chapter assumes a shared or weakly collective duty or, as has alternatively been defended, a putative duty of a putative group (Cripps, 2013, pp. 48–82, Isaacs, 2011, pp. 144–153). This is a duty to act together as necessary to prevent climate change from undermining central human interests. It has been defended as a positive duty, grounded in a collectivised principle of beneficence, a duty to organize to secure basic rights, or a duty to protect the vulnerable (Cripps, 2013, Held, 1970, Shue, 1980). It has also been defended as a negative duty to act together to prevent the harm resulting predictably from the combination of individual actions or from participation in harmful patterns of action (Ashford, 2006, Cripps, 2013, Gardiner, 2011a, Kutz, 2000). In each case, there is moral reason for some group or potential group to achieve something together and this gives rise to individual reasons to act, including to promote collective level progress (Isaacs, 2011, pp. 140–154). The duty to render any collective action fair and efficient is grounded separately in obligations to one’s fellow duty-bearers (Cripps, 2013, pp. 143–150).

106 Elizabeth Cripps (Edinburgh)

Given the focus on promoting a particular state of affairs, the natural justification for an individual’s course of action would be a consequentialist one (Pettit, 1991, p. 340).9 However, the CPM by no means corresponds to a straightforward maximising consequentialism, and certainly does not collapse into utilitarianism. The moral salience of the outcome being pursued – fair, efficient, effective global-level climate change mitigation and adaptation – is ultimately grounded in the normative priority of protecting basic rights or interests. Moreover, the CPM incorporates a demandingness limit in terms of individual costs and relationships, and is compatible with some deontological constraints. A later section will elaborate on this. With these caveats in mind, let me turn to the version of consequentialist-style reasoning underlying the CPM. Simple rule consequentialism will not justify the CPD, because of the partial compliance objection. In this context, the individual rule-consequentialist would calculate what all duty-bearers should be doing and play what would be her own part in that. However, individuals know that many unmotivated persons and groups in the global elite will not do their part (call these The Unmotivated). Accordingly, it would better promote global climate action if The Motivated attempted to motivate others, or ‘took up the slack’ by doing more themselves, than if they attempted to contribute to a combination of actions which will not in fact be forthcoming. The CPM bypasses this problem by focusing on what individuals can achieve by acting together with motivated others. (To reiterate, in drawing this distinction between The Motivated and all duty bearers, I am not assuming that in any long-term global level collective action The Motivated, taken as a whole, must pick up all the slack. I will come back to this.) An alternative consequentialist approach, avoiding the partial compliance objection, would be a modified rule consequentialist reasoning on which the individual calculates how everyone in The Motivated should act to secure the overall result and does her ‘bit’ of that. However, this also fails to justify the CPD, for two reasons. Firstly, unlike the CPM, this approach neglects the possibility that those who want to contribute optimally to combined and collective action might fail to do so effectively. This could result from communication issues, lack of mutual knowledge, or bias by some motivated individuals.10 For example, individuals might tend towards projects local to themselves, even if these are less central to the overall goal. Given all this, it could be more effective for the individual to cooperate with some others to compensate for likely deviations. Secondly, any viable model of individual climate duties must make space for strongly collective action or fail to capture all viable and necessary means of achieving progress. Consider the difference between combined emissions cuts resulting in mitigation gains and the development of (say) a global adaptation fund, which requires strongly collective action. (Even in the former instance, strongly collective action will often be fairer and more efficient (Cripps, 2013, pp. 143– 150).) The CPM allows explicitly for cases in which the best course of action requires acting intentionally as a collective. However, in such cases it becomes impossible or pointless simply to ‘do one’s part’ acting as an isolated individual.

Individual climate justice duties 107

Clearly, then, the CPM can only be grounded in a model on which what it is right for an individual to do depends partly on how others act in cooperative situations, which makes space for more or less strongly collective action rather than simply coordination or combination of individual actions, and which adjusts for failure by even motivated others to act according to the optimal scheme. The obvious question is then: is the CPM justified by act consequentialist reasoning? The answer is less obvious: it can be, but needn’t. In fact, progress can be made on justifying the CPM by considering alternative ways of responding to an important challenge to the idea of individual promotional duties (or individual consequentialism more generally). This is the no difference challenge. The objection goes as follows: while it is clear what we should do as a collective and clear that if there were a collective level response to be part of, the collective duty could indeed mediate individual duties, the situation is relevantly different when there is no duty-fulfilling collective-level scheme.11 Whether via ‘pure’ promotional activities such as voting or marching or cutting individual emissions (to which I will return below), an individual’s actions will make no difference to whether global level progress is made on climate change mitigation or adaptation.12 One response to this appeals to expected consequence reasoning: to the chance of one’s individual contribution triggering a significant collective level change (Broome, 2019, Hiller, 2011, Kagan, 2011). In terms of individual emissions cuts, the idea is that there are thresholds of emissions at which harms get worse and the individual, in emitting, runs the risk of causing such a threshold to be crossed. More generally, the point is that an individual’s actions have a non-trivial probability of causing the global elite more closely to approximate fulfilment of its collective duties. On this interpretation, act consequentialist reasoning justifies the CPD. Calculating what action would have the best expected consequences, in terms of bringing about fair, effective, efficient action on climate change, would require completing steps (1) to (4) above.13 However, it remains contested whether there are collective action cases where threshold reasoning does not apply, and whether the climate case is one of these (Kingston and Sinnott-Armstrong, 2018, pp. 178–181, Nefsky, 2011). If so, an alternative justification for the CPD can be found in a recent account which detaches causal significance from difference-making (Nefsky, 2017). On this, an individual has moral reason to contribute to some combined or collective effort so long as the following hold: it remains uncertain whether the worthwhile outcome will be achieved, part of what determines whether or not it will be achieved is whether enough individuals act in the way the individual is considering acting, and it is still uncertain whether enough individuals will so act. The individual would thus be a non-superfluous part of the cause of a worthwhile event. To put it another way, she would help to bring it about. If this argument convinces, individuals can have moral reason to act in nonthreshold collective action cases. But that does not make it a moral duty to do so (Nefsky, 2017, pp. 2744–2745). There are many such potential endeavours from which an individual could choose, even in the context of promoting fair, efficient,

108 Elizabeth Cripps (Edinburgh)

efficient collective action on climate change. The CPM provides a way of choosing between them, according to the moral salience of the different aims, the relative salience of their need for the individual’s type of actions and their likelihood of being under or oversubscribed.14 The point is not to reduce this to expected consequence calculation: rather the individual is motivated by being part of the ‘herself-and-others’ who could, for example, rectify the balance given general deviation from optimal coordination. I doubt whether this can be called ‘act consequentialism’. Indeed, an advantage of this model – one particularly suited to the CPM – is that it is designed for determining long-term individual patterns of action and does not require the individual to assess each act on a case by case basis (Nefsky, 2017, p. 2765). However, what ultimately matters is not whether the CPM can be fitted into a particular terminology but whether it can be justified philosophically, within the broader moral framework with which this chapter began. This I hope to have demonstrated.

Individual emissions cuts With this in mind, let us briefly consider how fulfilling the CPD might involve cutting one’s own carbon footprint. This both acts as an example of the CPD in action, and demonstrates how the model can incorporate individual emissions cuts, given that these are sometimes consistent with, but sometimes rival, so-called pure promotional actions. Firstly, emissions cuts can play a promotional role, especially if they are part of a collective effort made with motivated others. They can enhance the progress of social movements, for example the vegan movement, by incentivizing others to reduce their own carbon footprint. They can also promote institutional change, either via social impetus or by acting as a ‘signal’ of willingness to accept regulatory change (Lawford-Smith, 2016). Secondly, coordinated or collective actions to reduce emissions through combined lifestyle choices can, at a big enough scale, contribute to mitigation. Individuals can be part of this. There is an apparent tension between signalling for institutional change and for social movement change, which makes it more likely that the latter will be required by the CPD. It has been argued that an individual ‘signal’ must come at significant cost if it is to be taken as a meaningful guarantee of willingness to accept institutional reform, such as higher prices for fossil fuels (Lawford-Smith, 2016, pp. 324–325). Such costly changes may be unreasonably demanding for some. However, lifestyle changes such as becoming vegan, cycling, or holidaying by train are more likely to be imitated by other individuals if they are presented as low cost, even beneficial or enjoyable.15 Since some ‘green’ lifestyle choices are arguably beneficial to the individuals themselves, they could play this role in promoting social movements, all the more so if they take place within an SMG of signallers. Moreover, even such low-cost changes could promote, if not top-down political change, at least some institutional change at an arguably equally influential level: corporate and economic change. This in turn can promote further lifestyle changes,

Individual climate justice duties 109

as the associated individual costs fall. Consider how plant-based products have become increasingly available as the number of vegans and flexitarians increase, in turn facilitating a further lifestyle shift (Hancox, 2018). Finally, thinking purely in terms of contribution to the harm done by climate change – rather than participating in combined or collective action to prevent such harm – one’s individual emissions arguably have some moral significance. However, the reasons for such cuts (whether rationalized in terms of expected consequences or helping to reduce harm) must be balanced against other possible consequences available through cooperative action within the limits of demandingness (Cripps, Forthcoming).

Demandingness As I have stressed, the CPM is not consequentialist ‘all the way down’. The duties it entails are pro tanto. The CPD also incorporates a demandingness limit. This section will briefly elaborate on both of these considerations. Firstly, fulfilment could permissibly be constrained by some other moral duties, for example the individual no-harm principle or certain special duties, such as the duty to care for one’s severely disabled partner.16 Fulfilling the CPD might also come into conflict with other duties of global justice, such as aiding the current poor. I acknowledge this whilst leaving open either of two possible responses: allowing for such duties to take priority sometimes, for some individuals, or widening the scope of the CPM to apply to global justice more generally. The individual would use the four-step process detailed above to determine where to allocate her own efforts within that broader context. However, given the urgency and pervasiveness of the climate emergency, mitigation and adaptation would still be salient collective aims.17 Secondly, the CPD’s ‘reasonable cost’ condition limits the sacrifices an individual can be expected to make, in terms of her own interests, projects and relationships. Broadly interpreted, this also would rule out some negative impacts on those closest to the individual, such as her children. I have not specified what the demandingness threshold is and cannot do more here than gesture in that direction. However, one reasonable line of thought starts from two widely shared moral principles governing individual action: the no-harm principle and the principle of beneficence (Mill, 1859, p.14, Singer, 1972, p. 231). Given the anthropogenic nature of climate change, perhaps individual climate justice duties should be considered in line with the former, thereby putting the reasonable cost threshold very high (Broome, 2012, p. 57, Cripps, 2013, pp. 11–12). However, this is disputed (Cripps, 2013, p. 157, Fragniere, 2018). Even if the stringent demandingness threshold applied in instances of strongly collective harm, there is a morally relevant difference in control between cases where the actions of individuals aggregate predictably to cause serious harm, and those where a group is acting intentionally as a group (that is, as a moral agent in its own right), with harm a foreseeable consequence.

110 Elizabeth Cripps (Edinburgh)

One might turn instead to the CPD as positively derived and so to the notion of ‘significant cost’, borrowed from the moderate principle of beneficence (Singer, 1972, p. 231). I have elsewhere interpreted this as serious, even if temporary, interference with a central human interest (Cripps, 2013, pp. 13–14). This would, for example, rule out an individual having to take on a debilitating but temporary illness, break a leg, or be separated even for a few weeks from their young children. I do not think this can be rejected as too stringent. As in the negative case, appeal might be made to the difference between individual beneficence cases (for example, ‘Anne can pull Bob out of the water at relatively low cost to herself’) and cases such as this one, where the individual does not have sufficient control over the situation to be sure of bringing about the desired result. However, in the context of positive duties it would be a mistake to assume that less sacrifice is required of individuals simply because of the capacity for action (or inaction) by others. There is not space to go into the interesting and important questions around partial compliance and slack-taking here, so I will simply refer to one compelling recent argument (Karnein, 2014), according to which the behaviour of other duty-bearers can change the content of an individual’s duty, but not the demandingness limit. Indeed, up to that limit, an individual might be required to do more than she would otherwise have done because of the failure of others to act. So far, so reasonable. However, this remains an under-theorized area in climate ethics (one, moreover, in need of interdisciplinary input). For example, one might question whether extrapolating from small scale principles is the right way to assess demandingness at all, in the face of the climate emergency. Given extreme global and intergenerational urgency, perhaps climate justice duties are radically more demanding than I have suggested: more akin to what individuals have previously been expected to sacrifice in a national emergency such as war. It is sufficient to leave the issue open for the purposes of this chapter, with the placeholder above provisionally in place. Even if individuals only had a duty to promote climate justice up to the point of fairly modest sacrifice, they would need a way of deciding what to do up to that limit. The CPM provides that. The same would apply if the duties were much more demanding than I have suggested. Moreover, even a low threshold for demandingness would require a great deal more of affluent individuals than most of us currently achieve.

Three challenges for the Cooperative Promotional Model The first challenge facing the CPM (Epistemic Challenge) is as follows. Recall how much information the individual has to acquire and process before taking action. She must assess: the chances of success of different (P)SMGs, their fit with one another, her own and others’ skillsets, costs, position or influence and relevance to the aims of different (P)SMGs. She has also to adjust for the degree to which she can expect others to distribute their efforts optimally. Her ability to rely on her own estimates will depend on the extent and reliability of her knowledge regarding the likely actions of others, and the degree of communication possible with others,

Individual climate justice duties 111

including across (P)SMGs. The CPM does not require individuals to undertake all this epistemic labour no matter what the cost to herself: this would conflict with what was said above. But there is a concern that the motivated individual may not be able to do anything within the limits of reasonable demandingness, because even gathering the information to decide how to act would take her beyond the limit. A further challenge (Conflicting Views Challenge) results from differences of opinion within The Motivated. The discussion above not only implicitly treated the size of The Motivated as fixed (which both this and the following challenge will question): it also implicitly assumed a single scale of progress towards fair, efficient, effective action on climate change, on which all motivated individuals agree and which provides a universal measure for progress. In practice, this isn’t the case. By definition, The Motivated agree on some key moral starting points: the intergenerational and global injustice of climate change threatening central human interests, and the need to respond to that (broadly speaking) fairly, efficiently and effectively. However, motivated individuals differ on precisely what counts as a ‘fair’ collective effort. Differences also result from adherence to moral values beyond the shared points above, with further implications for interpretation of fairness, as well as the degree and understanding of ‘efficiency’ sought by individuals. For example, some will attach moral significance to the flourishing of nonhuman individuals and species;18 others attach religious significance to certain ways or life or parts of the world. These increase the likelihood of conflict between SMGs. For example, the UK’s Royal Society for the Protection of Birds (RSPB) supports climate action in general but opposes most tidal barrages and lagoons, as well as some windfarms, because of damage to wildlife (RSPB, 2016). Some otherwise motivated individuals may in practice be inflexible to the point of either not being motivated to pursue collective progress on climate change if this comes at the cost of some other moral value important to them (thereby reducing the size of The Motivated), or playing ‘chicken’ with other SMGs by continuing to pursue one otherwise unpopular option even at the risk of undermining overall progress. This also exacerbates the Epistemic Challenge, since the motivated individual will have to factor all this into her reasoning even if she is not so conflicted herself. Finally, The Motivated is only a subset of the global elite (Challenge of the Unmotivated). As already stressed, fulfilling the CPD does not mean bringing about long term global level climate action which imposes all the burdens on The Motivated. While there is a moral case for ‘taking up the slack’ within the limits of demandingness (Cripps, 2013, pp. 157–160, Karnein, 2014, Roser and Hohl, 2011), there is an important difference between working with motivated others to get progress going and assuming that long term The Unmotivated will remain off the hook. Recall that the morally required long term end is fair, effective, efficient globallevel climate change mitigation and adaptation. The global elite owe it to the victims of climate change to secure an effective end to climate injustice and to one another to do it fairly and efficiently (Cripps, 2013, pp. 143–150, Karnein, 2014). Given the need for global level mitigation, it would be inefficient and very possibly

112 Elizabeth Cripps (Edinburgh)

ineffective to attempt to protect basic interests from climate change, long term, without the participation of many currently in The Unmotivated.19 Moreover, in that case agents best placed to take action and most responsible for it would carry a lower burden than almost any reasonable conception of fairness would require (Caney, 2010, Page, 2012, Shue, 1999). Given this, the situation facing the motivated individual is complicated in a way which exacerbates both the previous challenges. The Motivated need to get The Unmotivated on board, whether by persuasion or institutional reform to require participation. Both would effectively increase the size of The Motivated. Failing that, they must find alternative ways of imposing a fair share of burdens on The Unmotivated (Caney, 2016).20 However, it will most likely be impossible to achieve full participation, or even fully fair burden sharing. Thus, some long term slack taking will be required (within collective demandingness limits). The three ends of fairness, efficiency and effectiveness will come apart, and The Motivated must choose between them. Now recall the Conflicting Views Challenge. Persuading members of The Unmotivated to join collective action, or responding adequately to the non-ideal situation wherein they remain unmotivated, may well require sacrificing other moral views, for example by damaging certain species, landscapes or ecosystems.21 (Recall that by ‘other’ I mean moral views beyond the shared adherence to basic justice and some level of fairness.) The individual must also include a whole additional set of factors into her decision making process, such as potential or skills to influence The Unmotivated. This, again, exacerbates the Epistemic Challenge.

Conclusions, resolutions – and a final puzzle This chapter has outlined the CPM and explained it philosophically. It has also raised three significant challenges. Despite these, I consider the model worth preserving. This final section will briefly explain why, and acknowledge a last, related puzzle. The Epistemic Challenge is not unique to the CPM. It applies to any model of moral duties requiring the agent to compare the consequences of different patterns of action in a complex scenario extending across time and space. In practice, approximation is necessary, combined with some division of labour. Individuals might begin by promoting the establishment of groups specifically tasked with comparing the effectiveness of different approaches.22 Alternatively, they might rely on existing SMGs to do the due diligence on how they can best coordinate. Each individual could then estimate which (P)SMGs her skills best fit her for, with approximate adjustments insofar as she should reasonably be aware of wide-scale bias.23 Sufficient agreement is likely among those actually committed to fairness (as opposed to using the term as a cover for, say, grandfathering ambitions) for the first component of the Conflicting Views Challenge – differing views over fairness – not to be a limiting factor for individuals determining how to act.24 However, the CPM needs amending in light of the second aspect of the Conflicting Views Challenge and the Challenge of the Unmotivated. As the previous section acknowledged, long-term

Individual climate justice duties 113

full fairness is implausible given the attitude of at least some members of The Unmotivated. Given the different views and values upheld within The Motivated (a tension exacerbated by the need to respond to the (in)action of The Unmotivated), it is also unlikely to be optimally efficient in terms of securing basic justice. I therefore suggest the following:

Amended Cooperative Promotional Duty Act with motivated others, in so far as possible at reasonable cost to oneself, so as best to promote effective progress on global level climate change mitigation and adaptation as fairly and efficiency as feasible.

This significantly modifies the original CPD by acknowledging that, in practice, The Motivated may find full fairness, efficiency and effectiveness to be incompatible. In prioritizing the former – actually avoiding severe climate harms – over fairness and efficiency, the amendment is in line with arguments made extensively elsewhere. Protecting fundamental human interests or basic rights ultimately takes priority over other morally important) considerations such as fairness across duty bearers (Cripps, 2013, pp. 157–160, Roser and Hohl, 2011).25 This amendment allows the CPM provisionally to move beyond the challenges. However, it also prompts a final philosophical dilemma which I acknowledge here but cannot fully resolve. The previous section raised the danger of conflicting moral values as though they were features of other motivated agents, of which the individual carrying out her calculations needed to be aware. But, in practice, almost all motivated persons have moral values beyond the shared commitment to fairly protecting fundamental interests threatened by climate change. We must therefore ask how far an individual can permissibly be guided by hers in determining how she promotes mitigation and adaptation. Suppose, to use the earlier example, she believes strongly in the moral importance of preserving species and populations of birds. Can such considerations legitimately allow her to deviate from maximum efficiency, if that would dictate promoting tidal lagoons for renewable energy? Consider three alternatives. (1) An individual’s (reasonable but comprehensive) moral values should not make any difference to fulfilment of the CPD. She should act only according to justice and efficiency: appealing to her skills and the needs and salience of different (P)SMGs: in determining where to allocate her efforts. (2) The individual must not support efforts liable actively to undermine collective progress (for example, actively campaigning against an otherwise feasible approach). However, she can lend her support to a less directly related area of climate action rather than actively promote a course she deems morally wrong. For example, the bird lover might focus on adaptation aid rather than campaign for or against tidal lagoons. (3) An individual can campaign for the route to collective progress on climate change that she deems morally preferable (so long as it is compatible with otherwise fair collective action) right up to the point where a rival scheme is adopted by collective consent.

114 Elizabeth Cripps (Edinburgh)

My own inclination is for option (2). Option (3) gives insufficient priority to protecting basic justice by permitting individuals to participate in action likely to slow progress towards it. Option (1) has the appeal of similarity to the Rawlsian requirement that the individual consider only the narrow political liberal conception of the good when acting as a citizen (Rawls, 1993, pp. 212–294). However, there is a distinction between refraining from vetoing collective legislation which permits activities incompatible with one’s own comprehensive values, and being required actively to promote such legislation. This would arguably be too demanding, so long as there are other options collectively available which would protect the fundamental interests at stake. Suppose for the sake of argument that ‘sustainable intensification’, including meat farming were the lowest emitting feasible food production system, given general apathy or refusal to shift away from a meat or dairy diet. Must the principled vegan actively promote this? In practice, individuals will generally struggle to promote courses of action incompatible with their own deeply held moral values. With some exceptions, they are also likely to develop skillsets in line with their own interests. Thus, simply fulfilling the CPD as laid out would generally keep the individual’s actual decisions in line with Option (1). However, this chapter does not purport to resolve this final puzzle. I highlight it as yet another outstanding question in the literature around individual climate justice duties. It is to this literature that I hope more broadly to have contributed, by focusing on what the individual can achieve not in isolation but as one of a motivated set.

Notes 1 Earlier versions of this argument were discussed at the Moral Philosophy Seminar, University of Oxford, the Climate Justice Beyond the State Workshop, University of New South Wales, the Centre for Ethics, Philosophy and Public Affairs, University of St Andrew, the Political Theory Colloquium, Goethe University, Frankfurt and the Edinburgh University Political Theory Group. I gratefully acknowledge the constructive feedback on all occasions, as well as helpful comments from Kieran Oberman and the editors of this volume. 2 Were a fair, efficient, effective collective scheme in place to fulfil the shared or weakly collective duty, each individual should (obviously) play her part. But this is not the case. 3 See Cripps (2013, pp. 115–139, Forthcoming) for my rejection of a number of arguments for a priority-taking duty to minimise one’s own carbon footprint (individual harm, fairness, Kantian or virtue ethic-based). I previously distinguished between mimicking, direct and promotional duties (Cripps, 2013, p. 116). Here, the three are subsumed within the overall promotional end. 4 For an introduction to this model for a broader audience, see Cripps (Forthcoming). 5 Barring, possibly, some exceptionally situated or talented individuals, whom I will exclude for the purposes of this chapter. 6 It might be contended that ‘sustainable intensification’ cannot be a genuine part of pursuit of climate justice because it is an oxymoron, at least if it involves farming animals. However, I set this aside for the purposes of the example. 7 For an earlier discussion see Cripps (2013, pp. 162–164). 8 Actor-activists Emma Thompson and Leonardo di Caprio are obvious examples. UK Premier League footballer Hector Bellerin also used a radio interview to highlight the importance of action on climate change (BBC, 2019).

Individual climate justice duties 115

9 To put it another way, I am effectively recommending that those motivated by collective harm or beneficence considerations should ‘think like consequentialists’. This is in contrast to Dale Jamieson (2010) who argues that rule-consequentialists should make themselves think like virtue theorists (to motivate themselves to ‘go green’). For my critique of his argument, see Cripps (2013, pp. 124–127). 10 There is evidence of bias in terms of environmental goals more generally, for example the tendency to care disproportionately about the fate of ‘cute’ species in protecting biodiversity (Small, 2011, 2012). 11 It is important to distinguish between the collective level, where the moral argument can be backwards or forwards looking, and the individual level. Mediated by the collective obligation, the individual has a forward looking or indirect reason to pursue collective progress on climate change, via subset endeavours. Thus, the puzzle is the familiar one of making sense of why an individual should add her effort to a worthwhile group effort where it will make no (perceptible) difference (as with, for example, Derek Parfit’s Principle of Group Beneficence (1984, pp. 76–83)). 12 That is not to say that the objection is equally strong with respect to ‘pure’ promotional actions (Cripps, 2013, p. 147). 13 On this reading, (2) and (3) might be combined into a single step: effectively, the individual works out how she could best approximate the combination of actions identified in step (1), and so optimally bring about progress on climate change given the likely errors of motivated others. 14 An alternative moral reason for participating in collective action is fairness based: the individual should avoid seeking to make an exception of herself, equivalent to free riding in collective self-interest cases (Cullity, 2019). However, the individual still needs to determine which collective efforts are most morally salient and which she has a duty to be part of. 15 A study of another environmentally beneficial diet change (eating insects) found that neither the environmental benefits nor the health benefits were as persuasive as taste or ‘trendiness’ (Berger et al., 2018). 16 I have elsewhere rejected the view that individual harm reasoning gives a direct, priority taking individual duty to eliminate one’s own carbon footprint (Cripps, 2016, Forthcoming). 17 See also Cripps (Forthcoming). 18 It might seem problematically anthropocentric to debate the place of non-humans or ecosystems in these determinations purely on the basis of individual humans happening to value them. I accept that. This paper prioritises protection of basic human interests, as an uncontroversial moral imperative. However, it is arguable that protecting non-humans from serious harm should take moral priority over some other ends acknowledged here (even, arguably, fairness). This would significantly complicate the model. 19 Stephen Gardiner (2011b, pp. 95–98) makes this point in relation to major polluter states. 20 Simon Caney (2016) identifies six possible responses to non-compliance with climate justice duties: increasing compliance, imposing burdens on non-compliers, sacrificing other moral ideas, reassigning responsibility, reassigning burdens to third parties, or adjusting the target (e.g. to allow a temperature rise of more than 1.5°C). 21 The third best response to non-compliance, according to Caney (2016, p. 38). 22 Compare the various Effective Altruist calculators available for those looking to save most lives with their charitable donations. 23 This might seem similar to satisficing consequentialism (Hurka, 1990, Rogers, 2010, Slote and Pettit, 1984). The CPM already advocates one aspect of satisficing insofar as it incorporates a demandingness limit, but I am not advocating satisficing in the sense of knowingly accepting a suboptimum outcome when a better one is (reasonably) within one’s reach. Rather, individuals must approximate to make the fulfilment of the duty at all achievable within the limits of demandingness.

116 Elizabeth Cripps (Edinburgh)

24 Despite ongoing debate over details, there is general philosophical consensus that the global elite should take on the greatest burdens of mitigation and adaptation (Caney, 2010, Page, 2012, Shue, 1980). 25 As the previous section acknowledged, Anja Karnein (2014) rightly distinguishes the question of what duty-bearers owe to the victims of climate change (and the demandingness of those duties) from that of what duty bearers owe to one another. However, where it is not a matter of one person ‘picking up the slack’ by rescuing an additional victim, but of many cooperating to forge long-term collective schemes which satisfy (ideally) both duties, it is necessary to take an explicit stand on whether it is legitimate to prioritise duties to victims.

References Ashford, E. 2006. The inadequacy of our traditional conception of the duties imposed by human rights. Canadian Journal of Law and Jurisprudence, 19, 217–235. BBC 2019. All about Arsenal with Aubameyang, Mkhitaryan, Bellerin and Pepe Football Daily. 7 August. https://www.bbc.co.uk/programmes/p07k1h7q. Berger, S., Bärtsch, C., Schmidt, C., Christandl, F. & Wyss, A. M. 2018. When Utilitarian Claims Backfire: Advertising Content and the Uptake of Insects as Food. Frontiers in Nutrition, 5, 1–7. Broome, J. 2012. Climate Matters: Ethics in a Warming World, W.W. Norton & Company, New York, NY. Broome, J. 2019. Against Denialism. The Monist, 102, 110–129. Budolfson, M. 2018. Food, the Environment, and Global Justice. In: Barnill, A., Budolfson, M. & Doggett, T. (eds.) The Oxford Handbook of Food Ethics, Oxford, UK: Oxford University Press. Caney, S. 2010. Climate change and the duties of the advantaged. Critical Review of International Social and Political Philosophy, 13, 203–228. Caney, S. 2016. Six Ways of Responding to Non-Compliance. In: Heyward, C. & Roser, D. (eds.) Climate Change and Non-Ideal Theory, Oxford, UK: Oxford University Press. Cripps, E. 2013. Climate Change and the Moral Agent: Individual Duties in an Interdependent World, Oxford, UK: Oxford University Press. Cripps, E. 2016. On Climate Matters: Offsetting, Population, & Justice. Midwest Studies in Philosophy, 40, 114–128. Cripps, E. Forthcoming. Intergenerational Ethics and Individual Duties: A Cooperative Promotional Approach. In: Gardiner, S. (ed.) The Oxford Handbook of Intergenerational Ethics, Oxford, UK: Oxford University Press. Cullity, G. 2019. Climate Harms. The Monist, 102, 22–41. Fragniere, A. 2018. How Demanding is Our Climate Duty? An Application of the No-Harm Principle to Individual Emissions. Environmental Values, 27, 645–663. Gardiner, S. M. 2011a. Is No-One Responsible for Global Environmental Tragedy? Climate Change as a Challenge to Our Ethical Concepts. In: arnold, D. G. (ed.) The Ethics of Global Climate Change, Cambridge, UK: Cambridge University Press. Gardiner, S. M. 2011b. A Perfect Moral Storm: The Ethical Tragedy of Climate Change. Environmental Ethics and Science Policy. Oxford University Press. Hancox, D. 2018. The unstoppable rise of veganism: how a fringe movement went mainstream. The Guardian, 1 April. Held, V. 1970. Can a random collection of individuals be morally responsible? Journal of Philosophy, 67, 471–481. Hiller, A. 2011. Climate Change and Individual Responsibility. The Monist, 94, 349–368.

Individual climate justice duties 117

Hurka, T. 1990. Two kinds of satisficing. Philosophical Studies, 59, 107–111. Isaacs, T. 2011. Moral Responsibility in Collective Contexts, New York, NY: Oxford University Press. Jamieson, D. 2010. When Utilitarians Should be Virtue Theorists. In: Gardiner, S., Caney, S., Jamieson, D. and Shue, H. (eds.) Climate Ethics: Essential Readings, Oxford, UK: Oxford University Press. Johnson, B. L. 2003. Ethical Obligations in a Tragedy of the Commons. Environmental Values, 12, 271–287. Kagan, S. 2011. Do I Make a Difference? Philosophy & Public Affairs, 39, 105–141. Karnein, A. 2014. Putting Fairness in Its Place: Why There Is a Duty to Take Up the Slack. Journal of Philosophy, 111, 593–607. Kingston, E. &Sinnott-Armstrong, W. 2018. What’s Wrong with Joyguzzling? Ethical Theory and Moral Practice, 21, 169–186. Kutz, C. 2000. Complicity: Ethics and Law for a Collective Age, Cambridge, UK: Cambridge University Press. Lawford-Smith, H. 2016. Difference-Making and Individuals’ Climate-Related Obligations. In: Hayward, C. & Roser, D. (eds.) Climate Justice in a Non-Ideal World, Oxford, UK: Oxford University Press. Mill, J. S. 1859. On Liberty. In: Gray, J. (ed.) On Liberty and Other Essays, Oxford, UK: Oxford University Press. Nefsky, J. 2011. Consequentialism and the Problem of Collective Harm: A Reply to Kagan. Philosophy & Public Affairs, 30, 364–395. Nefsky, J. 2017. How you can help, without making a difference. Philosophical Studies, 1747, 2743–2767. Page, E. A. 2012. Give it up for climate change: a defence of the beneficiary pays principle. International Theory, 4, 300–330. Parfit, D. 1984. Reasons and Persons, Oxford, UK: Oxford University Press. Pettit, P. 1991. Consequentialism. In: Singer, P. (ed.) A Companion to Ethics. Oxford, UK: Blackwell. Rawls, J. 1993. Political Liberalism, New York, NY, Columbia University Press. Rogers, J. 2010. In Defense of a Version of Satisficing Consequentialism. Utilitas, 22, 198–221. Roser, D. &Hohl, S. 2011. Stepping in for the Polluters? Climate Justice under Partial Compliance. Analyse and Kritik, 33, 477–500. RSPB 2016. The RSPB’s 2050 energy vision: Meeting the UK’s climate targets in harmony with nature. Royal Society for the Protection of Birds. Shue, H. 1980. Basic Rights: Subsistence, Affluence and U.S. Foreign Policy, Princeton, NJ: Princeton University Press. Shue, H. 1999. Global Environment and International Inequality. International Affairs, 75, 531–545. Singer, P. 1972. Famine, Affluence, and Morality. Philosophy & Public Affairs, 1, 229–243. Sinnott-Armstrong, W. 2005. It’s not My Fault: Global Warming and Individual Moral Obligations. In:Sinnott-Armstrong, W. & Howarth, R. B. (eds.) Perspectives on climate change: Science, economics, politics, ethics. Advances in the Economics of Environmental Resources; vol. 5. Amsterdam; London: Elsevier JAI. Slote, M. & Pettit, P. 1984. Satisficing Consequentialism. Proceedings of the Aristotelian Society, 58, 139–176. Small, E. 2011. The new Noah’s Ark: beautiful and useful species only. Part 1. Biodiversity conservation issues and priorities. Biodiversity, 12. Small, E. 2012. The new Noah’s Ark: beautiful and useful species only. Part 2. The chosen species. Biodiversity, 13.

7 ARE WE MORALLY REQUIRED TO REDUCE OUR CARBON FOOTPRINT INDEPENDENTLY OF WHAT OTHERS DO? Susanne Burri (LSE)

1 Introduction We have good reason to believe that climate change is occurring, and that it is fuelled by greenhouse gases in the atmosphere whose emission is a side-effect of various human activities. The scientific evidence suggests that governments around the globe are urgently required to come together to work out and implement policies that will effectively reduce the amount of greenhouse gases emitted. This way, it might still be possible to keep global warming in check, thus limiting the extent to which it leads to rising sea levels, ocean acidification, and an increase in extreme weather events, all of which are highly likely to detract from human and animal welfare. Yet even if we grant that governmental action is urgently required, this does not settle what our moral responsibilities are as individuals. By ‘our’ moral responsibilities, I mean the responsibilities of relatively affluent individuals who live high-emission lifestyles, e.g. because we travel by plane or regularly use a car, heat or cool down our generously sized homes, or consume other goods and services whose energy-intensive production may reasonably be said to add to our carbon footprint. When talking about our individual moral responsibilities to fight climate change, it is useful to distinguish between our responsibilities as citizens and our responsibilities as consumers.1 As citizens, we ought to do our share to work towards a political solution to the problem of climate change. Each of us might, for instance, have a duty to vote for Green parties, to join demonstrations, or to put pressure on their government in other ways to help ensure that just institutions are put in place that lead to a society-wide and, ultimately, globally coordinated reduction in greenhouse gas emissions. Our civic duty to work towards a political solution to the problem of global warming might also provide us with moral reasons in favour

Are we morally required to reduce our carbon footprint? 119

of reducing our carbon footprint in conspicuous ways. Doing so might allow us to signal to our governments our firm and considered support for policies aimed at curbing greenhouse gas emissions, and it might allow us to avoid charges of hypocrisy from political adversaries who are lobbying against the introduction of such policies (see e.g. MacLean, 2019, p. 7). Finally, to the extent that reasonably just institutions aimed at fighting climate change are in place and generally respected, we plausibly are morally required to abide by their constraints, both because we are under a duty to support just institutions (see e.g. Renzo, 2018) and for reasons of fairness (see e.g. Cullity, 2015). It is controversial whether we have any additional moral responsibilities to reduce our greenhouse gas emissions simply in our role as consumers. Suppose that you are doing your share—whatever this amounts to, precisely—to work towards and support just institutions aimed at providing a political solution to the problem of climate change. On a sunny Sunday morning, it occurs to you that you would love to take your family’s cherished old-timer for a leisurely drive. You know, however, that such an excursion would involve the emission of greenhouse gases that you could rather easily avoid by going on a bicycle trip instead. Are you morally required to leave your old-timer in the garage and hop on your bicycle instead? In a thought-provoking article, Walter Sinnott-Armstrong (2005) argues that you do no wrong if you take your old-timer for a drive. Sinnott-Armstrong’s (2005, pp. 291–294) key contention is that your drive-related emissions do no harm, as all climate change-related harms that will eventually occur are going to occur whether or not you take your car for a drive. He also argues that besides avoiding harm, there is no other morally relevant reason in favour of reducing your individual emissions (Sinnott-Armstrong, 2005, pp. 294–302). In a more recent article, Ewan Kingston and Sinnott-Armstrong (2018) present a number of further arguments, and reaffirm Sinnott-Armstrong’s earlier conclusions. While similar verdicts are reached also by other philosophers (see e.g. Johnson, 2003, Tan, 2015 or, more cautiously, Cullity, 2015), not everyone agrees. Some philosophers argue, contra Kingston and Sinnott-Armstrong (2018), that individual emissions make a difference to the climate-change related harms that will eventually occur, and that this fact provides each person with a reason in favour of minimising their carbon footprint (Broome, 2012). Others contend that even if individual emissions make no difference to the harm that will eventually occur, individual choices to emit can nevertheless contribute to climate change-related harms in morally relevant ways (see e.g. Kutz, 2000; Cripps, 2011). A further group of philosophers draw our attention to the fact that even if token emissions do not always make a difference to the climate change-related harms that will eventually occur, each token emission raises the total expected disvalue of such harms (Almassi, 2012; Broome, 2019). This suffices, they imply, to provide us with a pro tanto reason in favour of avoiding token emissions. In this paper, I argue that the philosophers who focus on the expected disvalue of individual emissions are on the right track. Given our epistemic limitations, we

120 Susanne Burri (LSE)

cannot know, for any particular token emission, what impact—if any—it has on the eventual occurrence of climate harms. Whether we have harm-related reasons to reduce our emissions is thus a question that needs to be asked at the level of evidence-relative morality, i.e. in light of the moral reasons that apply to us given the incomplete information that we have access to at the time of acting. At the level of evidence-relative morality, it is most appropriate to conceptualise the relationship between token emissions and harmful outcomes as one where each token emission raises the total expected disvalue of climate change-related harms. But while this relationship is morally salient, I claim that it is not, by itself, sufficient to provide us with a pro tanto reason in favour of avoiding token emissions. Only in combination with further considerations might it ground a requirement to reduce our carbon footprint. My key contention in this paper is that we are morally required to curb our emission of greenhouse gases whenever we can do so at little or no cost to ourselves. As each token emission adds to the expected disvalue of climate change related harms, we must at the very least not add to the level of greenhouse gases in the atmosphere in a needless manner. This seemingly rather minimal requirement in favour of reducing our emissions receives support not only from a maximising act consequentialist perspective, but also from a rights-based and a virtue ethical point of view. In addition to being broadly supported, the requirement is moreover more demanding than it might at first appear. This becomes clear once we keep in mind that most of us frequently have only a rough idea of the extent to which different alternatives for action would add to our expected welfare. In such cases, we might often be able to switch from a high-emission to a lower-emission alternative without incurring a tangible loss in our expected welfare. Many of the token emissions that we release into the atmosphere might in this sense be emitted needlessly, and ought for this reason to be avoided. The structure of the paper is as follows. In section 2, I critically discuss representative examples of arguments according to which each token emission of greenhouse gases either makes a difference to the climate harms that will eventually occur, or else contributes in some other morally relevant manner to these harms. In section 3, I explain how the arguments presented in section 2 proceed at the level of fact-relative morality, and I argue that when it comes to the relationship between token emissions and climate harms, a move away from fact-relative morality and towards evidence-relative morality is called for. From an evidence-relative perspective, each token emission raises the total expected disvalue of climate change-related harms. I contend that while this relationship between token emissions and possible harmful outcomes is morally salient, it is not sufficient to provide us with a pro tanto reason in favour of avoiding token emissions. In section 4, I put forward and defend a principle according to which we ought to reduce our emissions whenever we can do so at little or no expected costs. I argue that this seemingly rather weak principle has more significant implications than one might initially think. Section 5 concludes.

Are we morally required to reduce our carbon footprint? 121

2 How token emissions contribute to climate harms: fact-relative arguments 2.1 Broome’s difference-making argument In his influential book Climate Matters, John Broome (2012, pp. 49–96) argues that whenever we produce token emissions that we do not fully offset, we thereby make worse the climate change-related harms that will eventually occur, thus violating other people’s rights not to be harmed. His argument runs as follows. The average rich person emits something like 800 tonnes of CO2-equivalent gases into the atmosphere over their lifetime (Broome, 2012, p. 74). Based on current projections about overall emissions, the global warming this will lead to, and the subsequent harms that this will cause, the average rich person’s share of the total harm that climate change will cause can be estimated to deprive other people of a total of six months of good life, or 0.5 QALYs (ibid.). John Nolt (2011) performs a similar calculation for an average contemporary U.S. resident, and estimates that through his or her emissions, the average contemporary U.S. resident makes it the case that one or two people will die prematurely who would not otherwise have died prematurely. While obviously very rough, both of these estimates suggest that each of us makes a considerable negative difference to the lives of others through their emissions. According to Broome (2012, p. 54–55), we have directed duties towards other people not to cause them harm. Whenever we emit greenhouse gases, we fail in our directed duties and violate others’ rights, as the negative difference we make to others’ lives is a predictable and unjust result of our deliberate actions. Admittedly, we cannot easily avoid all of our emissions, and this would seem to render at least the harm that we cause through our subsistence emissions—i.e. the emissions that we depend on to live a minimally decent life (see Shue, 1993)—a mere infringement of the rights of others. But against this, Broome (2012, p. 80) argues that we have the possibility to offset our emissions, and that it is not costly to do this. If we offset our emissions, we have a zero net impact on the overall amount of greenhouse gases in the atmosphere, thus making no one’s life go worse than it would otherwise have gone. Broome (2012, p. 96) concludes that as long as we offset our emissions, we do not add to climate change-related harms and thus do not violate any rights, but that we do add to climate change-related harms in a way that violates others’ rights whenever we emit greenhouse gases without offsetting them.2

2.2 Why Broome’s (2012) argument fails To argue that our individual emissions make a difference to the climate harms that will eventually occur, Broome (2012) invalidly employs a simple ‘share of the total’ calculation (Kagan, 2011, p. 114). The basic assumption behind such a calculation is that there is a neat correspondence between the amount of greenhouse gases released into the atmosphere and the total climate harms that will eventually occur.

122 Susanne Burri (LSE)

More precisely, the thought is that climate change-related harms increase linearly as the concentration of greenhouse gases in the atmosphere increases, such that each token emission makes an—albeit incredibly tiny—difference to the overall amount of climate change-related harms. The key problem with this way of looking at things is that it does not, in general, match the reality of how an increase in atmospheric greenhouse gases leads to the occurrence of harm. The relationship between greenhouse gas emissions and climate change-related harms is complicated by a number of factors. For one thing, many predicted climate harms are related to an increase in extreme weather events, such as storms or floods, that global warming is expected to bring about. While we are able to predict that an increase in global temperatures will lead to an increase in extreme weather events, it is not possible to predict much in advance where and when such events will occur, as this depends on changes in the atmosphere that spread and develop in a chaotic manner. In a chaotic system such as our atmosphere, tiny disturbances in one place can at some later point in time have huge ramifications in other places. It is thus entirely possible that the presence or absence of some token emission can make all the difference to, say, whether some storm comes into existence, or to the geographic location over which it unfolds (Broome, 2019, pp. 112–114). It follows that particular token emissions might have a hugely positive or a hugely negative impact on climate change-related harms, just as they might have no significant impact at all. Admittedly, not all climate harms are weather-related. Take, for example, the expected rise in sea levels, the increase in ocean acidification, or rising average annual temperatures. All of these phenomena are associated with global warming, but they are climate-related as opposed to weather-related, and they are expected to detract from human and animal welfare even in the absence of an increase in extreme weather events. For these climate-related phenomena, it is moreover much more reasonable to assume that they will grow increasingly more pronounced as the level of greenhouse gases in the atmosphere rises. It does not follow, however, that a ‘share of the total’ calculation is more appropriate with respect to the harms that these phenomena will bring about. In a nutshell, many of the harms that these phenomena will bring about are ‘threshold harms’ that occur once some threshold is reached, but that grow no worse once we move beyond the relevant threshold. For such threshold harms, it will frequently be the case that they are overdetermined (see e.g. Kingston and Sinnott-Armstrong, 2018, p. 173; Cullity, 2015, p. 149; Barry and Øverland, 2015, p. 165). Consider the following example. Suppose that farmers in some location are relying on a crop that can no longer be grown economically once average annual temperatures increase beyond some threshold. As this threshold is approached and eventually surpassed, the farmers incur substantial adaptation costs, as they are forced to switch to growing crops that they are not yet familiar with. In this example, it is misleading to say that all token emissions increase by some tiny amount the harm that the farmers suffer. Instead, it is more accurate to maintain that once the concentration of greenhouse gases in the atmosphere reaches a sufficiently high level to raise average

Are we morally required to reduce our carbon footprint? 123

temperatures by some critical amount, further emissions make no difference to the costs incurred by the farmers. Moreover, when harm is overdetermined in this way, there is a sense in which no token emission makes a difference to the harm that occurs. More precisely, if the relevant temperature threshold is surpassed, then this threshold would have been reached, and the farmers would have been burdened with the relevant adaptation costs, quite independently of what any individual greenhouse gas emitter chose to do. Finally, it is possible that at least some harms grow more or less continuously worse as the level of greenhouse gases in the atmosphere increases. Suppose, for example, that sea levels rise continuously as the concentration of greenhouse gases in the atmosphere increases, and that this leads to a continuous loss in arable land in some parts of the world. If there are any such continuous harms, then a ‘share of the total’ calculation is appropriate at least with respect to these harms. But even if there are such harms, it is not clear, pace Broome (2012), that we violate anyone’s rights if we predictably inflict these harms through leading high-emission lifestyles. As Broome (2012, p. 75) admits, ‘[t]he harm [that we] do to each particular person is minuscule.’ But, and this is the crux, it is far from obvious that people have a right against having minuscule harms inflicted on them. If I walk through Central London in a hurry, I can predict that I will give a number of tourists a fright, and it is a well-known fact that being exposed to stress detracts from people’s life expectancy. Yet it seems that I am nevertheless allowed to hurry across London, as we do not usually assume that we are under a duty not to inflict minuscule harms on other people. In other words, while it is uncontroversial that people have a right against having substantial harms inflicted on them, there are no individuals on whom we inflict such harms by contributing to any continuously increasing climate harms such as the gradual loss of arable land. In recent years, many philosophers have recognised that our token emissions might not make a morally relevant difference to any harmful outcomes. Some of these philosophers argue that when we choose to live high-emission lifestyles, we nevertheless contribute to harmful outcomes in a way that renders us morally responsible for these outcomes (see e.g. Kutz, 2000; Cripps, 2011; MacLean, 2019). In the next subsection, I turn to such group-based accounts of our moral responsibility for climate harms.

2.3 Are we morally responsible for climate harms even if our contributions make no difference? Elizabeth Cripps (2011) assumes that climate harms are mostly threshold harms that tend to be overdetermined, such that our individual choices about whether or not to live high-emission lifestyles tend to make no difference to the harmful outcomes that will eventually occur. She nevertheless puts forward a sophisticated argument in favour of the conclusion that each individual who lives a high-emission lifestyle thereby contributes to climate change-related harms in a way that renders the individual morally responsible for these harms. Her key claim is that as that as

124 Susanne Burri (LSE)

members of what Thomas Pogge calls the ‘global elite’—Westerners and the rich minority in poor countries—we are collectively responsible for the harm that is a predictable result of the aggregation of our individual emissions (Cripps, 2011, p. 172). In a nutshell, Cripps argues that we are collectively responsible for the harm that our aggregated emissions bring about because we ought to have acted in a coordinated manner to prevent the harm from occurring. In light of the fact that Cripps assumes that climate harms are overdetermined, her view faces an important challenge. If climate harms are overdetermined, then each member of Pogge’s ‘global elite’ can claim that they do not form part of the group that collectively brings about harm. After all, climate change-related harms occur even if any one particular individual does not emit anything, and all such harms could be avoided if only enough others coordinated their actions to make sure that no harm occurs (Cripps, 2011, pp. 178–180). Against this, Cripps (2011, p. 180) argues that each member of the set of Pogge’s ‘global elite’ is a member of many possible subsets of which it is true that if all the members of this subset were to significantly reduce their emissions, then climate change-related harms could thereby be avoided. According to Cripps, this fact suffices to render each member of the global elite a member of the group that is morally responsible for whatever climate harms that will eventually occur. Cripps’ argument is ingenious, but not entirely convincing. According to Cripps, a failure to coordinate our actions with others to prevent some harm renders us morally responsible for the harm if it occurs. But, for one thing, there are many harms that could be averted or mitigated if sufficiently many individuals came together and made a concerted effort. This would doubtlessly frequently be a good thing, but more is needed to establish that we are morally obliged to help avert or mitigate some specific harm, not least because our resources are scarce, such that we cannot attend to the prevention or mitigation of all harms. Granted, if we share in the responsibility for bringing about some harm, then this can ground an obligation to help mitigate the harm. But it does not seem to be Cripps’ aim to provide an argument along these lines. Moreover, in the case of climate harms as Cripps conceptualises them, even if it can be shown that we are morally obliged to help avert them, this fails to establish that we are morally obliged to reduce our individual emissions. After all, Cripps assumes that reducing one’s individual emissions would be wholly inefficacious. Our duties to help avert such harms might thus be limited to civic duties—we might for example be morally required to lobby for an effective political solution to the problem of global warming, thus coordinating our efforts with others. In sum, if we are under a duty to reduce our individual emissions, Cripps’ account does not do enough to illuminate what this duty might be grounded in. Cripps is not the only philosopher who has put forward an argument according to which we can assume moral responsibility for an outcome even when our individual contributions make no difference to the outcome. Christopher Kutz (2000, p. 105) argues that if you participate in a group action that will predictably cause harm, you become an ‘inclusive author’ of that harm—the harm is at least to

Are we morally required to reduce our carbon footprint? 125

some extent attributable to you. According to Kutz (2000), this holds true even if you do not intend to bring about any harm, and even if your action makes no difference to the harm that will eventually be brought about (see also MacLean, 2019). Garrett Cullity (2000; 2015) argues that in cases of overdetermined harm, if at least some individuals refrain from engaging in the action that in aggregation leads to harm, then these individuals are involved in a worthwhile project, and you have a fairness-based reason to refrain from the relevant action as well, thus doing your share to support their worthwhile project. The basic argumentative strategy shared by all of these accounts is as follows. First, they assume that if you perform an act that leads to a morally salient outcome if it is performed by a sufficiently large number of individuals, then this fact suffices to make you a member of the group that collectively brings about the outcome in cases where the outcome is, in fact, brought about. Second, they contend that as a group member, you are morally implicated in what the group collectively does. This, it is then argued, can provide you with a reason to ensure that you are a member of the relevant group (if the morally salient outcome is desirable) or with a reason to ensure that you form no part of the group (if the salient outcome is objectionable). The problem with this argumentative strategy is that it attributes moral responsibility to individuals based on a very thin notion of what it means to act as a group. Consider harmful outcomes first. As Kutz (2000, p. 89) readily recognises, when we participate in a collective action in a morally relevant manner, this usually means that we coordinate our activities with other individuals with the intention to ‘contribute to a collective end.’ It is quite plausible that when these elements are present, our participation in a collective action can render us morally responsible for its outcome. But these elements are all absent when it comes to the sort of ‘unstructured collective harms’3 that we threaten to bring about if e.g. many of us choose to lead a highemission lifestyle. In addition, it is assumed that with respect to such unstructured collective harms, an individual’s actions are wholly inefficacious—they make no difference to the eventual harmful outcome. This makes it even harder to see in what sense an individual who decides to perform some relevant action thereby becomes a member of some morally salient group (see also MacLean, 2019, esp. p. 10). When it comes to morally worthwhile outcomes such as the prevention of harms, the situation is slightly different. Arguably, individuals who decide to do their fair share to help bring about a worthwhile outcome can be said to coordinate their activities with other individuals with the intention to contribute to a collective end. A more robust notion of collective action thus seems to be at work in such cases. The main problem, however, lies with the claims that individuals ‘contribute to a collective end’ or ‘do their fair share’ by reducing their emissions, i.e. by performing an action that is assumed to be wholly inefficacious. To insist that individuals can contribute to a worthwhile outcome, and that they have moral reason to do so, by doing something that makes no difference to the outcome, involves quite a stretch (see also Nefsky, 2015, pp. 257–259). At the very least, it seems that if the same individuals could make an actual difference elsewhere or in other ways, then they ought to shift their attention to where their efforts are efficacious.4

126 Susanne Burri (LSE)

While the group-based accounts discussed in this section suffer from important weaknesses, they are not necessarily doomed to fail.5 But, at a minimum, more work is needed to render them compelling. In the next section, however, I will argue that with respect to global warming, such accounts are misapplied anyway. When it comes to the relationship between climate harms and individual choices to live more or less carbon-intensely, the relevant consideration is not that our individual choices make no difference to the harms that will eventually occur. Rather, the relevant consideration is that for any individual choice, we do not know what difference—if any—it will make.

3 Token emissions and the total expected disvalue of climate change-related harms In his book Climate Matters, Broome (2012) purports that each token emission increases by a tiny amount the overall disvalue of the climate harms that will eventually occur. By contrast, philosophers such as Cripps (2011) assume that our individual lifestyle choices make no difference to the climate harms that will eventually occur. In this way, both sides purport to have access to information that is not, in fact, within our reach. The simple truth is that we do not know, neither for particular emissions nor for any individual’s lifetime emissions, how large a difference these will make to the climate harms that will eventually occur. Global warming and its associated extreme weather events and natural disasters are all complex processes whose functioning we have only a general understanding of (see section 2.2 above). In addition, what outcomes particular emissions will eventually contribute to depends on how we deal with climate change-related risks from a prevention, mitigation, and adaptation perspective, both from an individual as well as from an institutional point of view. These considerations make it impossible, in practice, to establish what will be the effects of any particular emission. It follows that whether we are morally required to reduce our emissions due to the contribution they make to harmful outcomes is a question that we appropriately have to ask not at the level of fact-relative morality, but at the level of evidence-relative morality. If an action is morally required in the fact-relative or objective sense, we ought to perform it in light of all the pertinent facts, irrespective of whether these are epistemically accessible to us at the time of acting. If an action is morally required in the evidence-relative or subjective sense, we ought to perform it in light of the information that is available to us at the time of acting (Parfit, 2011, pp. 150–151). Once we are clear about the fact that we are considering individual obligations to reduce greenhouse gas emissions from an evidence-relative perspective, the relationship between token emissions and the occurrence of harm is rather straightforward. Greenhouse gases in the atmosphere are understood to be an important driver of global warming, and global warming is understood to be associated with changes in the environment and an increase in extreme weather events and natural disasters that are expected to have a negative effect on human and

Are we morally required to reduce our carbon footprint? 127

animal welfare. In light of this, our best conceptualisation of the relationship between token emissions and the occurrence of harm is that each token emission— as well as each set of token emissions, such as an individual’s lifetime emissions— raises the expected disvalue of climate change-related harms, albeit only by a small amount. We can express a very similar idea by saying that from an evidence-relative perspective, the probability that there will be some particular increase in global temperatures is higher if we produce some token emission instead of avoiding it, i.e. for example P (2◦C threshold reached|Ei = 1) > P (2◦C threshold reached|Ei = 0), where Ei is a random variable that takes the value 1 if some token emission is produced, and the value 0 if the token emission is not produced. If it is more likely that some particular increase in global temperatures will be reached, this in turn raises our best estimate of the extent of climate change-related harms. Importantly, the thought that many climate change-related harms might be overdetermined does not affect this conceptualisation of the relationship between token emissions and harmful outcomes. Even if many harms are in fact overdetermined, it will still be true that these harms are caused by some subset of our token emissions. As we are unable to identify whether any particular token emission will in fact form part of a causal chain that leads to the occurrence of harm, we do best from a subjective perspective if we assume that each token emission raises the expected disvalue of climate harms (Broome, 2019, pp. 118–121; Almassi, 2012, pp. 11–19; Kagan, 2011, pp. 119–128). Consider the following simplified example. Suppose that according to your own best estimate, some harm H will occur if at least 800 people contribute to it. You further estimate that 1,000 other people will in fact contribute to it. Should you conclude that you do not raise the probability of H occurring if you decide to contribute as well, i.e. that P (H|Ci = 1) = P (H|Ci = 0), where Ci is a random variable that takes the value 1 if you decide to contribute, and 0 otherwise? The answer is that you should not, unless you are 100% certain that there are at least 800 other people contributing. If you are less than 100% certain about this, you should assume that your contribution raises the probability that the threshold will be reached, i.e. that P (Σ C ≥ 800|Ci = 1) > P (Σ C ≥ 800|Ci = 0), which implies that you should assume that your contribution raises the probability that H will occur. Moreover, even if you are certain that at least 800 others will contribute, as long as you suspect that the harm in question might be exacerbated if more than 800 people contribute to it, you should leave room for the possibility that your contribution might make things worse, or that the expected disvalue of the harm is larger, or more negative, if you decide to contribute, i.e. EV (H|Ci = 1) < EV (H|Ci = 0). In sum, unless you are 100% sure that a particular harm to which you might be contributing is overdetermined, you are not entitled to conclude that your contribution makes no difference.6 Similar considerations apply to weather-related climate harms that are due to chaotic processes. With chaotic processes, it is possible—though by no means certain—that small disturbances in the system can lead to large differences later on (see section 2.2 above). This means that token emissions might make a significant positive or negative difference to the disvalue of climate harms that are brought

128 Susanne Burri (LSE)

about through extreme weather events and natural disasters (Broome, 2019, p. 114, Fig. 3). Yet despite these complexities, the fact remains that from an evidencerelative perspective, each token emission is most sensibly conceptualised as raising, by some small amount, the expected disvalue of climate harms. As global warming is associated with an increase in extreme weather events, and as extreme weather events tend to detract from human and animal welfare, this allows us to conclude that token emissions raise the expected disvalue of weather-related climate harms (Broome, ibid.). Our very limited understanding of weather-related phenomena means that we are not entitled to conclude much beyond this. The thought that we have to conceptualise the relationship between greenhouse gas emissions and climate harms from an evidence-relative perspective is not new. Shelly Kagan (2011) argues that shifting from a fact-relative to an evidence-relative perspective can help us make sense of cases of overdetermined harm. Ben Almassi (2012) focuses on climate harms in particular, and contends that each token emission raises the expected disvalue of such harms. Importantly for our purposes, while Broome (2012) for the most part talks of actual harms and the difference we make to these harms through our carbon-intensive lifestyles, he sometimes shifts his language and seems to endorse a framework similar to Almassi’s (see Broome, 2012, esp. pp. 115–132). In a recent contribution, Broome (2019) makes this shift explicit, and discusses in detail the relationship between token emissions and climate change-related harms from an evidence-relative perspective. While the relationship between token emissions and harmful outcomes is straightforward from an evidence-relative point of view, this relationship is not, in itself, morally highly significant. Among the philosophers who advocate a shift from the fact-relative to the evidence-relative perspective, this is only insufficiently recognised. Instead, these philosophers generally assume that if an action threatens to make a harmful outcome worse or more likely, then this suffices to provide the agent with a pro tanto reason against performing the action (see e.g. Kagan, 2011, p. 129; Broome, 2019, p. 126). But this is incorrect. For one thing, the expected disvalue associated with particular token emissions is tiny. More importantly, there are many actions that bear a similar relationship to harmful outcomes from an evidence-relative point of view, but that are nevertheless clearly subjectively permissible (see also MacLean, 2019, p. 2; Budolfson, 2019). Consider the example of getting into your electric car and driving to the cinema. Doing this increases the subjective probability that there will be an accident in your town. But if you drive carefully, it is nevertheless morally permissible for you to drive to the cinema. The fact that from an evidence-relative perspective, performing an action raises the probability that harm will occur thus needs to be conjoined with further considerations if it is to render performing the action subjectively impermissible. For the example of driving an electric car, this is rendered impermissible if you drive negligently or recklessly. Crucially, driving negligently or recklessly is subjectively impermissible whether or not it results in an accident; what matters from an evidence-relative perspective is not the eventual outcome of an action, but how the agent should have proceeded in light of the information available to them at the time of acting.

Are we morally required to reduce our carbon footprint? 129

When it comes to the question whether we are morally obliged to reduce our individual emissions, much ink has been spilled on the relationship between token emissions and harmful outcomes, but not enough has been said about what further considerations might render it impermissible to produce a particular token emission. In the next section, I propose that producing a token emission is subjectively impermissible in cases where the expected costs of not producing it are low.

4 A moral requirement to reduce our emissions There are many things that we can do to reduce our carbon footprint. We can eat less meat, live in smaller homes, do less to cool down our homes in the summer or heat them up in the winter, or switch to less carbon-intensive modes of transportation. In this section, I propose that we ought to take such measures whenever doing so is not expected to be costly. More precisely: The Low Expected Cost Principle (LEC): As agents, we are morally required to reduce our carbon footprint if we can do so at little or no expected cost. LEC states a sufficient condition for when we are required to reduce our emissions. As clarified in the introduction, we likely have other reasons to reduce our emissions that might sometimes be sufficiently strong to yield an obligation to do so. We might, for example, be required to abide by the constraints of reasonably just institutions that have been put in place in an attempt to provide a political solution to the problem of climate change. On the face of it, LEC appears rather minimal. My aim in this section is to argue that LEC is well supported from a variety of normative perspectives, and that its reach is less minimal than might at first appear.

4.1 Act consequentialist support for the principle From a maximising act consequentialist perspective, it is standardly assumed that we are subjectively required to maximise the expected moral value of our actions (see e.g. Kagan, 2011).7 The expected moral value of an action is the sum of the moral values of the possible outcomes that the action might bring about, weighted by the probability with which the outcomes are expected to occur if the action is performed. LEC is consistent with a maximising act consequentialist framework. As argued in section 3 above, each token emission raises the expected disvalue of climate harms, albeit only by a tiny amount. If we can reduce our emissions at little or no cost to ourselves, the expected moral value of doing so is higher than the expected moral value of continuing to emit. It follows that from an act consequentialist perspective, undemanding reductions of our emissions are morally desirable.

130 Susanne Burri (LSE)

While LEC is consistent with a maximising act consequentialist outlook, it also appears rather minimal. Surely whenever we are wondering whether we are morally required to reduce our emissions, we expect to benefit from emitting, and we thus believe that choosing not to emit would constitute a sacrifice, in which case LEC fails to apply? If so, LEC lacks bite. But while this worry appears forceful initially, it overestimates the extent to which our decisions are well-informed and our actions uniquely optimal from a prudential point of view. To see what I mean, consider that many of us might sometimes turn up the heating without carefully considering the alternative of putting on a sweater. If we took the time to reflect on the issue, putting on a sweater might appear to us equally attractive from a prudential point of view. To give a different example, some individuals might seek out Bali as a holiday destination because they have never been to Indonesia, when Greece—much closer to home, yet a novel destination as well—would actually hold just as much appeal on reflection.8 More generally, for many activities and routines that form part of our daily lives, we might simply have no very nuanced grasp on how the different alternatives that are available to us compare to each other in terms of their overall prudential value. Suppose that if I commute to work by bicycle, then I estimate that this is more dangerous than driving, and I may sometimes be tempted to give in to laziness and hop in my car instead. At the same time, my health and my sense of wellbeing are improved whenever I cycle, and cycling to work fills me with a sense of accomplishment. In light of these conflicting considerations, I may be unable to reach a definite verdict about the sort of commute that I prefer all things considered. Read charitably, LEC proceeds from the assumption that our knowledge about the prudential value of our different alternatives for action is frequently imperfect in this way, leaving the different alternatives prudentially on a par (see Chang, 2002). LEC then asks us to carefully consider what changes we could make to our lifestyle without having a clear sense that we would thereby be missing out. On reflection, many such changes might be available to us, and LEC insists that we ought to implement them.

4.2 Rights-based support for the principle To understand the appeal of LEC from a rights-based perspective, consider why emitting greenhouse gases is associated with harmful outcomes. Whenever we emit greenhouse gases, we rely on the earth’s carbon sinks to absorb our emissions in such a way that the earth’s climate remains more or less unchanged. Due to our carbonintensive lifestyles, the earth’s carbon sinks are currently overused: they are unable to absorb a sufficient amount of greenhouse gases to keep the earth’s climate stable. The earth’s carbon sinks are an unowned or commonly owned resource. If we used the earth’s carbon sinks sustainably, they would have the potential to contribute to human and animal welfare for an indefinite time to come. But as we are not currently using them sustainably, each time we use them by producing a token emission, we thereby raises the expected disvalue of climate change-related harms.

Are we morally required to reduce our carbon footprint? 131

From a rights-based perspective, the salient question is what rights we have to the use of an unowned or commonly owned resource. Arguably, we may use whatever share of the resource is allocated to us through an institutional mechanism that was implemented to ensure that the resource is used fairly and sustainably (see e.g. Johnson, 2003; Tan, 2015). If there is no such mechanism in place, or if a mechanism is in place but it is ineffective, it is unclear to what extent we are entitled to use the resource. Having said that, the following seems to me highly plausible: if we reasonably perceive that some resource is at risk of being overused, then we have no claim to using it when we could easily refrain from doing so. Whenever we could easily refrain from using the resource, we lack a substantive justification in favour of using it that could ground a relevant use right. If this is correct, it validates LEC.

4.3 Virtue ethical support for the principle From a virtue ethical perspective, LEC is validated if a mindful attitude towards the use of scarce natural resources is a character trait worth cultivating (for arguments along these lines, see e.g. Jamieson, 2007; Sandler, 2010). With a continuously increasing and increasingly wealthy world population, the risk that we overuse natural resources has continuously intensified. In light of the fact that such overuse tends to be associated with harmful consequences, it is thus very plausible that some restraint in the use of these resources is virtuous. Granted, one might argue that self-restraint is superfluous in the presence of effective institutions that regulate how natural resources may be accessed and used. But as such institutions can be difficult to establish and maintain, individuals’ thoughtful restraint can provide at least some protection against the overuse of natural resources when effective institutions are lacking. Importantly, if a morally virtuous person uses natural resources mindfully, then their mindful attitude will not be limited to their use of the earth’s carbon sinks. Instead, it will extend to their use of other energy sources as well, and will permeate their behaviour quite generally. This seems valuable. For even if we manage to move away from a carbon economy, it seems likely that a continuously increasing and increasingly wealthy world population will continue to put significant pressure on the environment.

5 Conclusion In this paper, I have argued that when we think about the duties that we might have to reduce our carbon footprint, we must not get hung up on the question whether our lifestyle choices make a difference to the climate change-related harms that will eventually occur. Given our epistemic limitations, this is a question that we cannot answer. Whether we are morally required to reduce our footprint is thus a question that we should ask not at the level of fact-relative morality, but at the level of evidence-relative morality. In light of what we know, it is most

132 Susanne Burri (LSE)

appropriate to conceptualise the relationship between token emissions and harmful outcomes as one where each token emission raises the expected disvalue of climate change-related harms. From an evidence-relative perspective, this type of relationship between an action and possible harmful outcomes is morally salient, but it is not in itself sufficient to render an action impermissible. I have argued that in light of this morally salient relationship, we are morally required to reduce our emissions whenever we could do so at little expected cost. This requirement is well-supported not only from a maximising act consequentialist point of view, but also from a rights-based and a virtue ethical perspective. As different alternatives frequently appear to us on a par from a prudential point of view if we take the time to reflect on it, the requirement is moreover more demanding than it might at first appear. To take the requirement seriously, we have to carefully consider what changes we could make to our lifestyle without having a clear sense that we would thereby be missing out. On reflection, many such changes might be available, and I have argued that we are morally required to implement them.

Acknowledgments: For insightful written comments on earlier versions of this chapter, I thank Dmitry Ananyev, Liam Kofi Bright, Guy Lipman, Jeremy Moss, and Lewis Williams. For helpful discussions of my ideas, I thank Kartik Upadhyaya, as well as participants at the 2018 UNSW Sydney workshop on Climate Justice Beyond the State and participants at the 2018/19 LSE Philosophy WIP seminar.

Notes 1 This distinction is likely not exhaustive. We might, for example, have responsibilities also in our roles as employees, entrepreneurs, educators, or parents. 2 In this paper, I investigate whether we have duties beyond our civic duties to reduce our emissions of greenhouse gases. I set aside the question whether we have a duty to offset our remaining emissions, just as I set aside the issue whether offsetting one’s emissions might be a substitute for reducing them. For a careful discussion of issues related to offsetting, see Cullity, 2019. 3 This is Kutz’ terminology; see Kutz, 2000, p. 166. 4 One might argue that our fairness-based reasons to make some sacrifice apply to us in all cases where similarly situated others are making the sacrifice. If this is correct, then our fairness-based reasons might speak in favour of making some sacrifice even if doing so would be wholly inefficacious. This thought strikes me as plausible, but it would seem to render fairness-based reasons rather weak. Similarly to what I argue in the main text, if an individual could either act in accordance with their fairness-based reasons but make no difference, or could alternatively make a positive difference elsewhere, it seems that they should disregard their fairness-based reasons and do what makes a difference. I thank Jeremy Moss for discussion on this issue. 5 Though see Nefsky (2015) for a rather thoroughgoing critique. 6 Kingston and Sinnott-Armstrong (2018, pp. 178–180) argue against the idea that individual emissions should be conceptualised to raise the expected disvalue of climate harms. Their main objection is that even if it is true that an aggregate of emissions causes harm, we are not entitled to infer from this that each ‘tiny increment’ in aggregate emissions

Are we morally required to reduce our carbon footprint? 133

‘must therefore have some additive property (such as risk raising)’ (ibid. p. 180). This objection strikes me as confused. To claim that individual emissions raise the expected disvalue of climate change-related harms is not to claim that each individual emission has some ‘additive property (such as risk raising)’. Instead, it is simply to claim that given our epistemic limitations, it makes most sense to conceptualise token emissions as raising the expected disvalue of harm, even if we have strong reason to assume that not all token emissions do in fact play a role in bringing about harm. 7 For simplicity, I assume here that risk neutrality is a morally appropriate attitude. If risk aversion is morally appropriate as well, or possibly even preferable to risk neutrality, this only strengthens the argument that I put forward. 8 It is sometimes argued that getting on a plane does not raise the expected disvalue of climate change-related harms, as the plane would have taken off with or without a particular passenger. For effective replies to this argument, see Almassi (2012) and Kagan (2011).

References Almassi, B. 2012. Climate Change and the Ethics of Individual Emissions: A Response to Sinnott-Armstrong. Perspectives: International Postgraduate Journal of Philosophy, 4, 4–20. Barry, C. and Øverland, G. 2015. Individual Responsibility for Carbon Emissions: Is There Anything Wrong with Overdetermining Harm? In:MOSS, J. (ed.) Climate Change and Justice, Cambridge, UK: Cambridge University Press. Broome, J. 2012. Climate Matters. Ethics in a Warming World, New York; London: W. W. Norton & Company. Broome, J. 2019. Against Denialism. The Monist, 102, 110–123. Budolfson, M. 2019. The Inefficacy Objection to Consequentialism and the Problem with the Expected Consequences Response. Philosophical Studies, 176, 1711–1724. Chang, R. 2002. The Possibility of Parity. Ethics, 112, 659–688. Cripps, E. 2011. Climate change, collective harm and legitimate coercion. Critical Review of International Social and Political Philosophy, 14(2), 171–193. Cullity, G. 2000. Pooled Beneficence. In: Almeida, M. J. (ed.) Imperceptible Harms and Benefits, Dordrecht: Kluwer Academic Publishers. Cullity, G. 2015. Acts, Omissions, Emissions. In: Moss, J. (ed.) Climate Change and Justice, Cambridge, UK: Cambridge University Press. Cullity, G. 2019. Climate Harms. The Monist, 102, 22–41. Jamieson, D. 2007. When Utilitarians Should be Virtue Theorists. Utilitas, 19, 160–183. Johnson, B. L. 2003. Ethical Obligations in a Tragedy of the Commons. Environmental Values, 12, 271–287. Kagan, S. 2011. Do I Make a Difference? Philosophy and Public Affairs, 39, 105–141. Kingston, E. and Sinnott-Armstrong, W. 2018. What’s Wrong with Joyguzzling? Ethical Theory and Moral Practice, 21, 169–186. Kutz, C. 2000. Complicity, Cambridge, UK: Cambridge University Press. Maclean, D. 2019. Climate Complicity and Individual Accountability. The Monist, 102, 1–21. Nefsky, J. 2015. Fairness, Participation, and the Real Problem of Collective Harm. Oxford Studies in Normative Ethics, 5, 245–271. Nolt, J. 2011. How Harmful Are the Average American’s Greenhouse Gas Emissions? Ethics, Policy & Environment, 14(1), 3–10. Parfit, D. 2011. On What Matters. Volume I. Oxford, UK: Oxford University Press. Renzo, M. 2019. Political Authority and Unjust Wars. Philosophy and Phenomenological Research, 99(2), 336–357.

134 Susanne Burri (LSE)

Sandler, R. 2010. Ethical Theory and the Problem of Inconsequentialism: Why Environmental Ethicists Should be Virtue-Oriented Ethicists. Journal of Agricultural and Environmental Ethics, 23, 167–183. Shue, H. 1993. Subsistence Emissions and Luxury Emissions. Law & Policy, 15(1), 39–59. Sinnott-Armstrong, W. 2005. It’s not My Fault: Global Warming and Individual Moral Obligations. In: Sinnott-Armstrong, W. & Howarth, R. B. (eds.) Perspectives on climate change: science, economics, politics, ethics. Advances in the Economics of Environmental Resources; vol. 5. Amsterdam; London: Elsevier JAI. Tan, K.-C. 2015. Individual Duties of Climate Justice Under Non-Ideal Conditions. In: Moss, J. (ed.) Climate Change and Justice, Cambridge, UK: Cambridge University Press.

8 RIGHT-LEVELLING INDETERMINACY Environmental problems, non-state actors, and the global economic market Benjamin Hale (CU Boulder)

The past thirty years of climate negotiations – from Kyoto through Copenhagen to Paris and beyond – have produced only muted success at achieving a global international agreement that will be necessary to tackle climate change. Frustrated by volatile political strife and seemingly intractable bureaucratic hurdles, more than a few in the environmental community have given up on state responses to global problems, reasoning that states just aren’t equipped to make meaningful progress on climate when they are so caught up in the economic sector, so under the spell of national interest and GDP. Many propose, often without much argument, that in the face of the intractability of states, individual actors should go it alone, exercising their personal liberties and responsibilities to take action (Jacobsen and Dulsrud, 2007, Hiller, 2011).1 Frequently this is defended as a form of conscious consumerism that, in principle, should empower individual consumers to address climate change or other environmental problems by changing their consumption habits. In this paper I aim to suggest that this response, while compelling, is unsatisfactory. The uncoordinated actions of individuals – even large aggregate collections of individuals – cannot, without strategic agreement and near-complete compliance, address widespread environmental issues. We must instead pursue a third pathway for resolving large-scale collective action problems: to encourage and form substate or non-state coalitions that function across markets to regulate the behaviour of those markets (Ostrom, 2010). We might hope, for instance, to encourage local municipalities to band together to set policies that change the way resources are extracted and spent; or create conditions that prompt oil producers to forge an agreement to reduce or minimize their oil production (Duggan, 2019). We should not and cannot, however, rely on the activities of any single oil producer, or any ragtag collection of individual oil consumers, to make this happen.

136 Benjamin Hale (CU Boulder)

My objective in this piece is to argue that at least one main reason to seek solutions to these problems at a non-state level is related to causal indeterminacy engendered by markets. Most arguments turn on attempts to avoid negative outcomes, and in doing so bend in the direction of either of the two options I mention above – state-level coordination or individual-level conscious consumerism. The first, most natural response to indeterminacy in collective action is to look to Hobbesian or statist solutions. When state intervention appears to be less probable, the next natural place to look for avoiding negative outcomes is at the individual level, essentially empowering all actors to identify for themselves or agree upon the efficient outcome and pursue it individually. This emphasis on efficient outcomes, on avoiding tragedies of the commons (when conceived as a loss in environmental or social value), pushes experts to think in terms of these polarities. I argue here instead that the most reasonable way to deal with the indeterminacy engendered by markets is to coordinate at the right action level, by which I mean targeting actors who are in the “best place” to overcome problems of indeterminacy through discursive coordination. In this context, what it means to be in the “best place” is to be positioned to overcome problems of indeterminacy, which will depend, in large part, on the market in question. The distinction is critical because, at present, a heavy proportion of climate change solutions turn on binding individuals to imaginary agreements in pursuit of presumed shared objectives, encouraging a suite of practices that, absent an actual agreement of relevant strategic actors, are positioned to be fantastically self-undermining. The problem here is that strategy and complexity – indeterminacy, effectively – interfere with coordination; and the problem in particular with individual-level coordination is not, as is typically assumed, that there are no enforcement or monitoring mechanisms. It is, rather, that intervening agents manipulate the market to adapt to changing individual strategies.

Preliminary comments About ten years ago I authored and published a paper that took aim at issues of causal impotence in fossil fuel consumption. That paper, Non-renewable Resources and the Inevitability of Outcomes (2011), utilizes basic economic theory to suggest that the outcome of individual changes in fossil fuel consumption will result in consequent changes to fossil fuel production, thus offsetting, and even nullifying, the desired positive outcomes. My objective was to advance a causal impotence objection to consequentialist arguments for addressing climate change in hopes of clarifying what I take to more compelling deontological reasons to take action. This paper builds on a related theme. The consequentialist logic that motivates the concern with which I shall be occupied here proposes the following: that if individual actors refrain from participating in some market, they can thereby prevent (or at least, play a very minor but nevertheless morally meaningful role in preventing) some bad. Such a claim is supposed to be generative of duties to take preventive mitigation measures,

Right-levelling Indeterminacy 137

particularly in the form of conscious consumption in large global markets. The line of reasoning I have in mind is particularly prominent in defending supposedly effective strategies for addressing climate change, though it applies to a range of other collective action problems as well, including the deleterious effects of factory farms, the application of pesticides, or the use of conflict minerals. Such arguments have a secondary, albeit more complicated, relationship to deontological reasoning, splitting unevenly between rights and respect theories, so other caveats are in order. I don’t have space in this chapter to establish the more pressing point regarding the mechanisms underlying causal indeterminacy, which I instead develop elsewhere (Hale, 2019). As a somewhat newer causal impotence objection in the ethics literature, however, indeterminacy poses serious worries for theories of moral responsibility that advance the position on consequentialist grounds that individual actors can responsibly resolve market-driven problems by becoming conscious consumers. My general suspicion, which I also do not attempt to defend here, is that consequentialists are smuggling in moral suppositions that otherwise go uninterrogated, when in fact they affect the way we approach a slew of problems. As I say, these are not my primary concerns in this piece, but rather may offer insight into my motivations. Lest my argument in this chapter fall victim to the mistaken apprehension that this is a form of denialism – as philosopher John Broome has recently argued of those who advance causal inefficacy objections (2019) – it is important to emphasize that the position I hold is not an argument against all individual duties to do good, nor ought it to imply scepticism about state-based collective efforts. It primarily serves as an attempt to shift our gaze away from individual solutions to collective problems that are ensconced in complicated economic markets. That is, the scope of this paper is narrow enough not to take on the whole subfield of normative theory that understands ethical problems through the lens of individual responsibility. It is still important to be a good person and to address smaller environmental problems associated with one’s actions. One should not pollute or waste food or throw trash in the streets, just as one should not punch people in the face, torture puppies for gustatory pleasure (Norcross, 2004), or pour toxins in the water supply (Hale, 2013). The causal indeterminacy concerns I introduce here and elsewhere are primarily directed at individualist solutions to collective action problems, and specifically target the consequentialist rationale that motivates such thinking. More substantively, by emphasizing the indeterminacy of the market, I seek to undercut the reasoning that individual actions have determinate effects on collective outcomes. I’ll proceed next by offering a short introduction to “market indeterminacy,” borrowing heavily from work by Russell Hardin. I’ll then discuss two features of indeterminacy – strategy and complexity – that contribute to the instability of outcomes. Following that discussion, I’ll utilize several cases of increasing strategy and complexity to suggest that one pathway to resolve indeterminacy is through discursive coordination at the right action level. In keeping with work that I’ve done in other areas – work that suggests that we should promote conduits and

138 Benjamin Hale (CU Boulder)

channels for discursive engagement of moral and practical issues – I will aim to suggest that these conduits and channels are equally important in the private nonstate sector, though they are often not taken to be.

Market indeterminacy Consider first the Hotelling Rule of Exhaustible Resource Economics (not to be confused with “Hotelling’s Law”) (Hotelling, 1931). The rule itself is complicated and difficult to parse, but in its less formal articulation suggests, among other things, that producers of non-renewable resources will seek to maximize economic rents when extracting those resources. If there is a limited stock of some resource, in other words, producers will extract that resource to optimize profits. The practical implication of these “Hotelling Rents” is that as conditions in the market change, producers will strategize to extract as much value out of their limited resource as is possible. In the case of fossil fuels, oil producers will do whatever they can, whatever is in their power, to extract as much of their resource out of the ground, as many dollars out of their resource, so long as it remains profitable for them to do so. Things obviously get much more complicated when turning attention to renewable resources, but shifting focus can illuminate the indeterminacy of markets more clearly. There are many kinds of renewable resources in the environmental field, and those that interest me for this analysis are “soft commodities” – wheat, coffee, sugar, corn, soybean, livestock, and so on. They also include some renewable energy technologies, like solar, wind, and arguably hydropower. These latter commodities are very much also related to climate change insofar as they serve as replacement goods for fossil fuels. In principle for every kilowatt of energy produced through solar, this is one kilowatt’s worth of fossil fuel that will not be burned. There are differences, to be sure, between markets in renewable and nonrenewable resources. What renewable and non-renewable markets share, however, are a few basic market dynamics. A bunch of people are producing goods of value (producers) for a bunch of people consuming goods to enjoy that value (consumers). They all aim to do so as efficiently and rationally as possible – which is to say, as inexpensively as possible. And they all aim to do so in part by paying close attention to what everyone else in this madcap configuration does. It is tempting when modelling markets to think about the relationship between an individual consumer and the entire supply of goods in terms of simple quantities and uncertainties. At any given decision-juncture for a consumer, there exists a discrete if incomprehensibly large quantity of some good. If the consumer refrains from consuming this good, then there will be more of this good left in the overall stock of goods, notwithstanding any uncertainties that might be associated with the production and supply of that good and the various mechanisms through which decisions of abstention affect such mechanisms. These are fairly standard lessons from microeconomics.

Right-levelling Indeterminacy 139

Political theorist Russell Hardin argues, however, that many of these systems do not function according to such determinate logic (Hardin, 2003). He suggests instead that many social systems, but economic systems in particular, suffer from an indeterminacy that is frequently mischaracterized as uncertainty about determinate outcomes. In a more recent book, economist Yanis Varoufakis offers a similar such analysis specifically with regard to markets (Varoufakis, 2013). Their critical observations are that markets are not the determinate systems that so many models and heuristics depict, but rather indeterminate systems composed of millions, sometimes billions, of strategic actors acting in an environment of extraordinary complexity. This idea of indeterminacy has been floating around the economics and game theory literature for some time, and in fact is the very basis of game theory (Bicchieri, 2009). Hardin suggests, however, that in many of these discussions indeterminacy “is constantly swept under the rug because it is often disruptive to pristine social theory.” He continues with emphasis: “The theory is fake: the indeterminacy is real” (Hardin, 2003). Even given Hardin’s commitment in the later years of his career to exploring indeterminacy, the phenomenon of indeterminacy continues to be widely disregarded, not least by ethicists, whether in the climate literature, the animal ethics literature, or the policy literature more broadly. A few notable commentators have dabbled with questions at the intersection of uncertainty and indeterminacy, though often fail to spell out indeterminacy as the complicating mechanism (Lazarus, 2009, Gardiner, 2006, Wynne, 1992, Galvin and Harris, 2014).

Strategy and complexity Consider, then, a simple determinate system, like that of a light switch. Under most circumstances, if the light switch is connected to a working circuit, it is simple enough to say, with a reasonable degree of certainty, that when the switch is flipped, the light will come on. There may, of course, be physical or structural factors that make this outcome less than guaranteed – say, for instance, that there is a short in the wire, or that the bulb has burned out – but in systems such as these there are clear causal pathways between the flipping of the switch and the lighting of the light. We are surrounded by determinate systems and make hefty use of them in many science and engineering contexts, from aerospace engineering to chemical modelling to computer programming. Such systems provide the practical basis of our technical infrastructure. Add an intervening agent to this configuration, however, and the predictive complications grow enormously. The presence of an intervening agent is the hallmark of indeterminacy. Think again of a light switch. If you are standing near a light switch and I want to turn on the light, I cannot simply touch your nose and expect the light to come on. I have to do quite a bit more to make this happen. I have several possible courses of action at my disposal. I might, among other things, bark orders at you to turn on the light. This is a tenuous pathway to get the

140 Benjamin Hale (CU Boulder)

light on, and my success will be dependent on how compliant or tolerant you are. You might, in response to my antics, respond in a range of ways. Naturally, you might just turn on the light, thinking that I am a loudmouthed fool. But such an outcome is far from guaranteed. You might also be startled by my hostile and unclear directives. You might be offended or affronted by my tone. You might fail to hear, see, or notice me, despite my bellowing commands. You might aim to spite me, given some fact about our history together. What is important to see here is that the outcome – the flipping of the switch, the lighting of the light – is far from guaranteed in this case, simply because there is at least one more agent added to the configuration. The outcome is very much affected not only by my actions, but by your response to my actions. My success will depend a great deal on how I try to get you to turn on the light. This worry, at first glance, may appear to be only epistemic in nature: that regardless of what any interfering agent chooses to do, the world will turn out some particular or definite way. The only problem, so it seems, is that the outcome is unknown to the actor. But this is actually not Hardin’s concern. Hardin’s concern is metaphysical as well: because the initiating action changes the nature of the configuration, in strategic circumstances the responding actor (or actors) has an opportunity to shift the eventuating outcome so that it is always out of reach. Consider the card game War, which is played by pitting two players against one another to flip cards and determine the card of higher value. In that game, when a player flips a card, an opposing player flips over a card simultaneously. There are probabilities and uncertainties about what the outcome of the match will be, but the game is fundamentally determinate. It is a closed system that a supercomputer or a Laplacean demon with perfect information might well anticipate. Even if the deck is shuffled before each flip, only uncertainty stands in the way of the next draw. If, however, the opposing player can reorganize the deck in response to each flip, or perhaps even take other actions to upset the game or the rules of the game, the outcome ceases to be a simple matter of uncertainty. The outcome of any given flip, or of the game itself, is indeterminate – determined only by the respective strategies of the players and their responses to one another. This is similar to the circumstances that govern the light switch + agent case. In the light switch + agent case, I can, of course, do many other things to get you to turn on the light. A better, more productive strategy than barking at you might be to ask you politely to turn on the light. In constrained circumstances – if you and I share a wider set of objectives, say – this may be a far more successful way to gain your compliance. In fact, it may be the best way to get you to turn on the light, and I can bank on your compliance under most, but certainly not all, circumstances. You still might be startled by my politeness, or you might suspect me of something underhanded, or you might grow fearful and run out of the room. I just don’t know what you’re going to do, so I use language, which clarifies and illuminates my intentions, to try to coordinate with you. It is, of course, possible that I might ask you to turn on the light in still yet other ways. I might ask in another language, like Russian. Unless you are a decent

Right-levelling Indeterminacy 141

Russian speaker, this approach is likely to be unsuccessful, even if we share wider objectives. Alternatively still, I might simply gesture in your direction, indicating to you that I would like you to turn on the light. My success again will be dependent on the nature of my gesture, how well it conveys to you my desire to have the light on, and how willing you are to satisfy that desire. The point here isn’t so much that there are many possible ways for me to get you to do my bidding, but rather that once there is an intervening agent introduced into the system, there is more than mere uncertainty about how the system will respond to various inputs. There’s not only the response of the intervening agent, but also the manner in which the agent is prodded or motivated to act. More importantly: the best mechanism we have not just to coordinate, but also to minimize indeterminacy in the system, is to communicate (Habermas, 1987b; a). Now then, it is important to acknowledge that many social systems can be modelled as if they were determinate systems, and many disciplines do this for reasons of simplification. Economics, political science, sociology, demography, medicine, and so on, all depend on simplifying assumptions to make sense of how the social systems that they study work. Where such systems accurately capture the dynamics of human populations, these explanations serve an important descriptive purpose, but they often fail to be predictive. We see time and again attempts to anticipate how various diverse populations will respond or react to stimuli, how individual actors within those populations will react to stimuli, and yet repeatedly these predictions fail to come true, or only come true with very wide error bars (Kaplan, 1940, Nussbaum, 1997, Sarewitz and Pielke Jr, 1999, Hofman et al., 2017). Economies, private businesses, industries, stock markets, public health policies, all can be simplified and modelled fairly successfully for retrospective applications, but when attempting to manipulate or nudge those systems prospectively, these same simplifying assumptions, more often than not, fail to produce the anticipated results. For the purposes of this chapter, there is no need to develop this much stronger position relating to the variety of social systems. All I want to suggest in this chapter is that it makes more sense to understand global markets less as determinate, but more aptly as indeterminate. To understand the fully confounding nature of indeterminacy in global markets it may take a bit more reflection on the complexity of agency. Says Hardin: “The issue is that the basic rationality that makes sense of or fits individual choice in the simplest contexts of choosing against nature does not readily generalize to contexts in which individuals are interacting with other individuals” (Hardin, 2003). His analysis continues to assess a variety of social theories – contractarianism, contractualism, utilitarianism, Kantianism, Rawlsianism, marginalism, among others – and how these social theories have coped with indeterminacy. For our purposes, we can set that analysis aside. However, it will be helpful to borrow from Hardin the insight that essentially two factors make the determinate approach to social and market systems overly narrow: Strategy and Complexity. First and foremost, market interactions are curious beasts. They’re not simple relations in which all actors cooperate to achieve mutually beneficial ends.2 When

142 Benjamin Hale (CU Boulder)

they are, it’s because actors bring to the relationship a shared set of assumptions and goals. In many cases, however, actors respond to one another strategically. When we decide what to do, we can’t just “set it and forget it.” We have to choose strategies for achieving that set of objectives, anticipating that other interactants will respond predictably. If they don’t respond cooperatively, if they respond more selfishly to our actions, we are forced to change course. This grows all the more difficult as we introduce multiple actors. The more actors we introduce into a system, the more complicated the situation gets. Much of the time in our everyday interpersonal interactions, we have fairly reliable systems in place to help us achieve our objectives, particularly in a market system. If I need to have a tree removed, I call a tree removal service. For the right price, they’re usually happy to comply. These systems work because our instrumental reasons are aligned fairly tightly, and in part they are aligned tightly because the task is simple enough that we are either able to make the exchange by choosing a product and paying a price, or able to articulate clearly what we want to have done. If a task is more complicated, we are often but not always able to write contracts and wills that lay out what we want to have done; but even then, we often find ourselves unsettled on the details and, as a consequence, have developed massive legal institutions to address disputes. Such systems work particularly well when we have direct contact with those who might help us achieve our objectives. Like a child’s game of telephone, however, as contact and communication grows more diffuse, and as more intervening actors enter the picture, the complexity of the system grows substantially. So this is really the second bit of Hardin’s observation: that markets are extremely complex constellations of many millions, and sometimes billions, of interactions. It is tempting to think of social complexity in terms of randomness or stochasticity, as, for instance, in a pachinko machine or in the human metabolism. If the laws of physics and metabolism were understood perfectly, we could presumably plug all relevant data into a quantum computer to calculate outcomes. Such a computer might be able to figure out how the balls are going to drop or how a given chocolate bar will affect our nutrient uptake. If these systems are understood as closed and essentially determined, absent randomness (in which case any uncertainty would be understood strictly epistemically), then the problem is just one of information. Or even if these systems involve some level of stochasticity (in which case such a quantum computer would have to model uncertainty probabilistically), the problem becomes not only a problem of information but also the laws that govern the randomness. The integration of a strategic respondent – or additionally chains upon chains of strategic respondents – into this complexity, however, makes social systems radically different than mere stochastic systems like pachinko machines. In these cases, even a very powerful quantum computer would have to anticipate what responses one, three, ten or quite a few more respondents will offer, figuring that as one response is activated, strategies for each actor will change in response.

Right-levelling Indeterminacy 143

And this is no small assumption. Many in the climate community talk about climate systems as though they are complex or stochastic in precisely this way. Teams of climate scientists apply principles from geology, biology, climatology, fluid mechanics, geomorphology, and so on to try to model the trajectory of the planet as accurately as possible. The prevailing assumption is ceteris paribus: Keeping everything equal, what will the planet look like if nobody changes their behaviour? But social systems like those that affect climate change are multi-layered and complex. We often can’t easily gain clarity on what’s happening in a given system, sometimes simply because we’re looking at the wrong scale or perhaps at the wrong conceptual level. Hardin argues that these two components – complexity and strategy – lead invariably to indeterminacy.

Cases and solutions Now then, I don’t want to get caught up in a much more involved discussion of indeterminacy. What I want to do, instead, is turn our attention to solutions for cases of manifest indeterminacy. These are cases that will be characterized, following Russell Hardin, by increasing the extent to which strategic actors are involved, on one hand, and the complexity of the problems they face on the other. Consider just a few. Take a simple but classic determinate case, familiar to all philosophers: Drowning Boy. You are walking along a pathway to University and notice that a young boy has fallen into a shallow pond. Wading into the pond to save the boy will involve little effort from or cost to you, but it may mean that you get wet and muddy (Singer, 1972). Says Singer: If it is in your power to prevent something bad from happening, without thereby sacrificing anything of comparable moral importance, then you ought, morally, to do it (Singer, 1972). Most others agree: you should do what you can to save the drowning boy. Though you might mess up your suit, if you don’t jump into the water to save the drowning boy, he will surely drown. So far so good. Now take a simple strategic system: Drowning Man. Suppose now that there is a man who is drowning a boy in the same shallow pond you passed before. This isn’t just a man who is at the present moment drowning a boy. This is a man who derives great pleasure from drowning boys in shallow ponds. That’s what he does: he’s a “drowning man.” As a man who drowns boys, if you try to stop him, he will just move along and find another way to drown the boy, or he will find another boy to drown. Do you have an obligation to save the boy? On one hand, yes, it would seem that you do have an obligation to save the boy. But it is not clear from the circumstance how best to achieve that outcome. One of the ways to save the boy is to

144 Benjamin Hale (CU Boulder)

remove the threat: to stop the man from drowning the boy. If you do succeed in temporarily stopping him – say, leaving him to wander away from the boy he has just tried to drown – then he will certainly come back and make another attempt, or find another way, to drown the boy. His objective, again, is to get as much pleasure as he can from drowning boys. Your objective is not to save the boy, but to stop the drowning of boys. Neither one of these two cases is particularly complex, but Drowning Boy differs from Drowning Man in that the latter requires you to engage with an actor who will respond strategically to you. This is not so different than: Hunt Saboteur. Unhappy with the animal rights violations of hunters, you hear word that tomorrow there shall be a fox hunt. To save a fox from being hunted, you decide to thwart the hunters by heading out into the woods with horns and drums, banging and tooting madly, alerting foxes to their pursuers. Spoiling hunts by beating drums or blowing horns may seem rude and classless, but “spoil sports” engage in such practices with clear objectives: not just to save foxes, but to ruin hunting, which they find cruel and inhumane. Sometimes these spoil sports are successful at thwarting particular hunts, sometimes they are less successful. So then, that’s the first thought here. Many indeterminate systems are relatively manipulable if the right tactics and strategies are employed. Without too much difficulty, we can anticipate how a drowning man or a fox hunter will respond to interventions. Strategic problems require strategic solutions, and hunts are no different. Since reasoning with hunters will presumably not work, the committed dogooder will need to find another way to prevent the harmer from harming. But what is the practical effect of hunt sabotage? Even in a relatively simple case like Hunt Saboteur, it is difficult to say with confidence. A hunt saboteur might prevent a single hunt from succeeding, but taken alone, probably will not prevent a hunter from hunting again. In some cases, a saboteur may even encourage a hunter to hunt harder, or to hunt differently, or to double down in his efforts to hunt. The ultimate question about a hunt saboteur’s effectiveness is a question of strategy and tactics: what will work best to reduce the hunting? More importantly though, the Hunt Saboteur and the Drowning Man cases are straightforward strategic relations in which the two parties do not share objectives. A better way to stop such practices is not necessarily to interfere directly, possibly pushing the offending party to double down on efforts or to react strongly to the saviour, but rather to change the conditions that enable the harm: either to impose an enforceable rule restricting such harms, or to try to gain the attention of the hunter and the drowning man to reason with them about the cruelty of their practices so that they change their practices on their own. This may not go over particularly well in many instances, but a savvy hunt saboteur would be advised to avoid direct confrontation so as to narrow the possibility that he will make things worse. So far so good. Now take a more complex, but still fairly determinate, system:

Right-levelling Indeterminacy 145

Museum. A museum is rigged with a bomb set to blow in 15 minutes. There are 100 priceless artworks in the museum and 100 bystanders outside of the building, of which you are one. Though a task demanding some level of coordination, you and these ninety-nine bystanders could rush into the museum, run through the halls, and rescue all of the artworks. You can be assured that you will be safe if you get out before then. Should you do it? Such a rescue mission will be complicated and risky, but if everyone understands their job and does not make a mistake, it can be done. What’s the plan of attack? First in, first out? First in, last out? All rescuing parties hug the left side of the hallway? All rescuing parties approach through the main hallways first, peeling off to branch hallways as they are encountered? Again, this looks like a case in which all that is required is a simple strategy, but the ambiguity associated with finding and agreeing on a strategy presents a conundrum for practical reason. That’s part of what’s at issue here. Any one of these strategies may work, but none of the strategies will work if all of the strategies are in play. Above all, this particular situation requires a strategy because it is a matter of great complexity. All motivations of the actors are aligned, but the strategies that they might pursue are not. Their challenge is to agree on which strategy to use. My intuition is that, absent a clear plan, you probably should not run into the building to save the artworks, though it could be argued that you should. Given that there is a discrete bomb with a known detonation time, it’s better to save some than none. Someone should take action. But look again. Even to come to this conclusion, we are reducing away the complexity of the real-world so that the circumstances are constrained to a very small number of conditions. In what real-world scenario are there actual bomb timers? In what real-world museum bombing scenario are there direct causal pathways between aspirational art heroes and outcomes? Now take an even more complex case with a touch of stochasticity built in: Hospital. There are 100 people trapped in hospital beds in a burning hospital and 100 people outside the burning building. You are one of those people outside the building. If each of you runs in to help an ailing patient, and everything goes smoothly, then all patients can be saved. Remember, however, the building is burning, the patients are scared, and almost nobody knows how fire will progress through the building. If you shut a door behind you, this could stop the fire, or it might exacerbate the fire, sending it somewhere else. Should you run in and save them? One thought here is that you run the risk of making the outcome much worse if you run into the building. Typically speaking, the advice is: do not run into the building. This is for a reason: fire rescues are complicated. There’s a lot that you don’t know about how fire will respond, a lot that you don’t know about how scared patients might respond. Fire rescues take a lot of coordination and usually a

146 Benjamin Hale (CU Boulder)

fair bit of training as well. Given the additional indeterminacy of the scared patients, adding one more human to the mix may fail to make the outcome better, and may easily make it worse. Take an even more complicated case, unfortunately all too common in the United States: Active Shooter. There is a school of 100 children. Inside the school is one active shooter, you know not what he looks like. Presumably he carries a gun. You, a lone rescuer, stand outside the building with a gun. Theoretically, you could save the children by running into the school and killing the shooter. This would stop the shooter and prevent the threat. Should you run into the school with a gun? Running into the school may, in fact, get the shooter to back down, but it will also make you a target, possibly of the shooter, but possibly of whomever else shows up with a gun. Additionally, by running into the school with a gun, you risk scaring children out of hiding, possibly into the shooter’s path. Children respond to you. They see you and do things that they wouldn’t otherwise do. The generally accepted advice in this situation is that you should not run into the school with a gun (Frazzano and Snyder, 2014). Let’s ramp up the complexity and strategy even further: Battlefield. You are a lone civilian who wakes up one morning to discover that a battalion of soldiers has parachuted into your small town. They are organized. They intend to take the town. It would behove you to prevent this from happening. Scenarios such as these are widely dismissed as intuition pumps by ethicists on grounds that their complexity makes it too difficult to understand any principle by which an individual actor might abide. The chaos confuses our intuitions, goes the reasoning. The strategic responses of other actors aren’t our responsibility, the reasoning continues. What often also happens, however, is that the principle, say, that one should maximize utility, is applied to such cases after having been derived from much simpler intuition pumps. Any self-respecting consequentialist will recognize that one should in most cases not run onto battlefields without having a plan. Such actions won’t bring about the right outcome. And it may well be so that such cases are too confusing to be of use when isolating philosophical principles like whether one should use others as a means, or whether one should kill one to save five, but it is precisely because they are so complex and strategic as to obscure the ethical principle they are intended to test (in this case that individual actors have a consequentially derived obligation to assist) that we should pay closer attention to them. This is a feature, not a bug, of such cases. Cases in which indeterminacy is in play – and indeed, these are not all cases, but indeterminacy does affect quite a few circumstances in which markets are

Right-levelling Indeterminacy 147

involved – makes drawing a neat and tidy connection between an actor and her actions virtually impossible. Mistakes in moral guidance come in seeking to apply ethical principles drawn from oversimplified, and frequently disanalogous, cases more generally and prospectively. There is likely no simple consequentialist answer to the question of whether one should intervene in an active shooting case or in a battlefield case (Jackson, 1991). Such answers can only be arrived at by working through the various scenarios. Even then, attaining the so-called “right” answer will not only be difficult, but will be made difficult, because all subsequent events will be affected by intervening actors who act strategically. No matter what course of action an actor chooses, there will be no determinate outcome. In all cases there will be an outcome, but the outcome in indeterminate systems can only be understood as a determinate outcome when evaluated retrospectively. So what is one to do? The proper response to complicated, strategic cases like these is not to throw up our hands and declare defeat, but instead to turn our gaze away from individual responsibilities and develop mechanisms for coordinating coalitions of actors – and particularly to find coalitions of actors that are in the best position to coordinate. Such coalitions could be formed by creating facilitated environments in which well-positioned market actors come together to forge mutually agreeable solutions; it could be done by encouraging and negotiating legally-binding contracts between well-positioned subnational governments; or it could be done by approaching the carbon majors and encouraging them to coordinate. The models for encouraging such coordination vary (see, for instance the UNFCCC’s NAZCA database)3, but energy must be focused on these sorts of right-level solutions. Such coordination can enable collectives of actors to anticipate the strategic responses of intervening actors: to reduce complexity by seeking actors at the appropriate action level and to create channels for them to meet minds and cooperate – in other words, to reduce to as low a level as possible both complexity and strategy. Unlike many arguments that push for coordinated solutions on grounds that they will pave the way for a preferable outcome, this argument suggests that the reason to coordinate is not to achieve a preferable outcome, but because the indeterminate nature of markets admits of no other solution. Observations of shifting strategy and complexity, of course, have been made by many others, and as I mentioned at the outset, are the very basis for game theory. Little is novel in this regard. We must find a way to coordinate action to assure that the outcome is best. Both lines of argument I discuss at the beginning of the chapter – the statist and the individualist responses – function along these lines. But such arguments defer to a fundamentally different mechanism than the one I present here. Those arguments suggest that we should seek to find the efficiency point, the good outcome. I am arguing instead that we should avoid, where possible, that temptation. Indeterminate systems are characterized by two features (at least according to Hardin): complexity and strategy. If we elevate either of these two features, we enter into circumstances of greater not lesser uncertainty. This is precisely because the nature of indeterminacy makes it such that strategic solutions beget strategic responses. The error

148 Benjamin Hale (CU Boulder)

of many theorists in the social sciences has been to acknowledge strategy, but to view it primarily as an obstacle preventing achievement of the best outcome. It is important instead to find a strategy upon which the greatest number of parties, with the greatest influence over the arrangement, can agree. By far the best way for us to make indeterminate systems do what is best is to coordinate together; to try, where we can, to cooperate, ideally by coming to an understanding with the various participants. Actors need to coordinate, to meet minds,4 in order to overcome indeterminacy and the concomitant features of complexity and strategy. No doubt, this is difficult. But indeterminacy cannot be overcome using the two standard approaches that we have used in the past. Our markets are too strategic. Our world too complex. There is, as it happens, empirical evidence to suggest that smaller non-governmental entities in fact naturally do this. Elinor Ostrom and her brood of academic offspring have demonstrated time and again that governance structures of Garrett Hardin’s effectively Hobbesian solution aren’t necessary to resolve tragedies of the commons, that smaller communities can manage common pool resource problems quite well when left to their own devices (Hobbes, 2009, Hardin, 1968, Ostrom, 1990). It goes without saying in this analysis that we all have obligations to save drowning boys if we can do so without sacrificing anything else of moral significance. It’s probably not unreasonable to suggest that we have obligations to save them even if we have to sacrifice some things of moral significance. But the challenge that the world poses to us isn’t whether we can save drowning boys without sacrificing anything of moral significance. It is whether we, as individuals, acting in a complex and indeterminate world, can actually save those drowning boys when we, as coordinated collectives, acting at a level appropriate to tame that complexity and indeterminacy, are not just better positioned, but are rightly positioned, to be the ones to do it.

Conclusion At the beginning of the chapter I mentioned that the two most oft-encountered solutions to environmental problems typically operate along either statist Hobbesian lines – that we must forge a global agreement between states to align the economic incentives of sub-state actors and bring their strategies into alignment – or along individualist conscious consumerist lines – that we must act as individuals by endorsing non-binding, self-imposed commitments to be more conscientious in the market. But the analysis above points to at least one problem with this latter approach: it sidelines complexity and strategy by assuming model determinacy and focusing on a desired outcome. A third position emerges when we embrace model indeterminacy. This third position suggests that a more viable approach is to try, where possible, to reduce complexity and to tame strategy. Non-state and sub-state agreements between industry actors can do precisely this. The challenge is to design systems that can bring all the relevant actors together so that they might abandon, or at least reduce, their commitment to strategic action.

Right-levelling Indeterminacy 149

In the case of global carbon markets, we are often talking about the production and consumption of oil from the major emitters, like the US, India, China, and Europe. If even a somewhat large collection of individuals cut back on their oil consumption, the effect is not necessarily to reduce overall the consumption of oil. Rather, if strategy and complexity are not also taken into account, such reductions will affect oil production in unpredictable ways. In many cases the net effect is to shift the production of oil to other markets and, through the miracle of market pricing, to shift the price of oil in local markets (Piggot et al., 2017). Indeed, oil production has climbed consistently almost every year since oil has been extracted, apart from a temporary dip in the 1970s (Dudley, 2017).5 This is where actor coalitions at the appropriate action level make much more sense. To overcome the problem of indeterminacy we cannot depend on individuals to take action against the tidal pressures of global markets. There can be no assurance that there will not be defectors or cheaters. The more possible it is for the strategies of intervening agents to disrupt the strategic objectives of individual actors, whether maliciously or inadvertently, the greater likelihood of failure. But if all (or many) of the major actors in a game agree to behaviours and gain some assurances that actors in that market will be acting in accordance of that agreement – in other words, if complexity can be reduced and strategy coordinated – then a solution to the problem may emerge without the overbearing concern of defection. For this reason, not for the reasons that are standardly given, industries and policymakers must seek to create the conditions for discourse and deliberation amongst themselves, perhaps with the aid of others (Sagoff, 2004, Bäckstrand and Kuyper, 2017, Dryzek and Pickering, 2017, Dryzek, 2017). They have a moral/ethical obligation to do this, not only to prevent negative outcomes, but also to ensure that any actions they take with the objective of producing good outcomes – that is, any actions that they take for ethical reasons – may themselves not be undercut by the strategies of complex networks of intervening agents (Habermas, 1991, Forst, 2012). Individual actors may have such an obligation as well, but given the arrangement of deckchairs, the primary obligation must be to facilitate discourse and justification so that agreements between right-level actors can develop. The current state of discourse is anathema to this discussion: with lifestyle environmentalism and individual self-promotion, with moralizing and identity curation, with social media, the discursive dimensions of the civil sphere are regularly and continually undercut. Our moral responsibilities are to one another, to try to find space with one another to justify our actions and make sure that our actions are justified.

Notes 1 Notably, the references cited here do provide arguments, and in some cases quite sophisticated arguments. That they provide arguments primarily for individual action, however, seems to be a consequence in part of a wider trend in the discourse to turn to individual action when efforts to coordinate state activities fail.

150 Benjamin Hale (CU Boulder)

2 One referee objects that many market interactions are in fact simple relations. Joe might swap his apples for Suzie’s oranges if and only if he’d get more utility out of the oranges than he would get out of his apples. But this objection misses the point about markets and market interactions. Markets are complicated. Joe’s relationship to his applies and Suzie’s to her oranges rests on a long, interweaving chain of relationships that make simple characterizations of their interaction as a binary decision misleading. Joe’s valuing of the apples depends in part on their price, which is a proxy value that he uses to measure his applies against other objects, like Suzie’s oranges, that he may then choose to buy or not buy. The simplification of these exchanges as primarily between two actors with full command of their preferences hearkens to a barter economy (Hart, 2008). It is best limited to introductory economics discussions aimed to illuminate principles of comparative utility. It is far from characteristic of exchanges in a market economy. 3 https://climateaction.unfccc.int/ 4 I don’t have space in this chapter to discuss the nature of this meeting of minds. There are several ways in which this could be accomplished, but that is well outside the scope of this discussion. I hope here only to flag the meeting of minds as the critical, and I would argue, the best, mechanism for overcoming circumstances of indeterminacy, particularly if this meeting of minds is done in a non-strategic and more communicative fashion. 5 There is an additional, non-insignificant effect of cutting back on oil consumption. Namely, those who do so deprive themselves of relatively inexpensive energy, conceivably putting themselves at an economic disadvantage against these others. This is, in fact, part of the rationale that motivates states to resist climate legislation.

References Bäckstrand, K. &Kuyper, J. W. 2017. The democratic legitimacy of orchestration: the UNFCCC, non-state actors, and transnational climate governance. Environmental Politics, 26, 764–788. Bicchieri, C. 2009. Rationality and indeterminacy. In: Ross, D. & Kincaid, H. (eds.) The Oxford Handbook of Philosophy of Economics. Oxford, UK: Oxford University Press. Broome, J. 2019. Against Denialism. The Monist, 102, 110–129. Dryzek, J. S. 2017. The meanings of life for non-state actors in climate politics. Environmental Politics, 26, 789–799. Dryzek, J. S. & Pickering, J. 2017. Deliberation as a catalyst for reflexive environmental governance. Ecological Economics, 131, 353–360. Dudley, B. 2017. BP statistical review of world energy. London: WorldPetroleum Congress. Duggan, J. 2019. The Role of Sub-state and Non-state Actors in International Climate Processes: Subnational Governments. Chatham House Research Paper, London: Royal Institute of International Affairs, Jan 23, 2019. Available at: https://www.chathamhouse. org/sites/default/files/publications/2019-01-23-Duggan.pdf Forst, R. 2012. The Right to Justification: Elements of a Constructivist Theory of Justice. Translated by Jeffrey Flynn. New York: Columbia University Press. Frazzano, T. L. & Snyder, G. M. 2014. Hybrid targeted violence: Challenging conventional ‘Active Shooter’ response strategies. Homeland Security Affairs, 10. Galvin, R. &Harris, J. R. 2014. Individual Moral Responsibility and the Problem of Climate Change. Analyse & Kritik, 26, 383–396. Gardiner, S. M. 2006. A perfect moral storm: climate change, intergenerational ethics and the problem of moral corruption. Environmental Values, 15, 397–413. Habermas, J. 1987a. The theory of communicative action, Volume 1: Reason and the rationalization of society, Boston, MA: Beacon Press. Habermas, J. 1987b. The theory of communicative action, Volume 2: Lifeworld and system, Boston, MA: Beacon Press.

Right-levelling Indeterminacy 151

Habermas, J. 1991. Discourse ethics. In: Lenhardt, C. & Nicholson, S. W. (eds.) Moral Consciousness and Communicative Action. Cambridge, MA: MIT Press. Hale, B. 2011. Non-renewable Resources and the Inevitability of Outcomes. The Monist, 94. Hale, B. 2013. Polluting and Unpolluting. In: BOYLAN, M. (ed.) Environmental Ethics. 2nd edition. Hoboken, New Jersey: Wiley-Blackwell. Hale, B. 2019. Unconscious Consumers: Individual Actions and the Indeterminacy of Outcomes (in preparation). Paper presented at the University of Reading, the University of New South Wales, the University of Colorado, Boulder. Hardin, G. 1968. The Tragedy of the Commons. Science, 162, 1243–1248. Hardin, R. 2003. Indeterminacy and society, Princeton University Press. Hart, K. 2008. Barter. The New Palgrave Dictionary of Economics: Volume 1–8, 384–386. Hiller, A. 2011. Climate Change and Individual Responsibility. The Monist, 94, 349–368. Hobbes, T. (ed.) 2009. Leviathan, Oxford, UK: Oxford University Press. Hofman, J. M., Sharma, A. &Watts, D. J. 2017. Prediction and explanation in social systems. Science, 355, 486–488. Hotelling, H. 1931. The Economics of Exhaustible Resources. The Journal of Political Economy, 39, 137–175. Jackson, F. 1991. Decision-theoretic consequentialism and the nearest and dearest objection. Ethics, 101, 461–482. Jacobsen, E. &Dulsrud, A. 2007. Will consumers save the world? The framing of political consumerism. Journal of Agricultural and Environmental Ethics, 20, 469–482. Kaplan, O. 1940. Prediction in the Social Sciences. Philosophy of Science, 7, 492–498. Lazarus, R. J. 2009. Super Wicked Problems and Climate Change: Restraining the Planet to Liberate the Future. Cornell Law Review, 94. Norcross, A. 2004. Puppies, pigs, and people: Eating meat and marginal cases. Philosophical Perspectives, 18, 229–245. Nussbaum, M. C. 1997. Flawed Foundations: The Philosophical Critique of (A Particular Type of) Economics. The University of Chicago Law Review, 64, 1197–1214. Ostrom, E. 1990. Governing the commons: The evolution of institutions for collective action, Cambridge, UK: Cambridge University Press. Ostrom, E. 2010. Beyond markets and states: polycentric governance of complex economic systems. American Economic Review, 100, 641–672. Piggot, G., Erickson, P., Lazarus, M. &Van Asselt, H. 2017. Addressing fossil fuel production under the UNFCCC: Paris and beyond, Stockholm Environment Institute. Sagoff, M. 2004. Price, Principle, and the Environment, New York: Cambridge University Press. Sarewitz, D. &Pielke JR, R. 1999. Prediction in science and policy. Technology in Society, 21, 121–133. Singer, P. 1972. Famine, Affluence, and Morality. Philosophy & Public Affairs, 1, 229–243. Varoufakis, Y. 2013. Economic Indeterminacy: A personal encounter with the economists’ peculiar nemesis, Routledge. Wynne, B. 1992. Uncertainty and environmental learning: reconceiving science and policy in the preventive paradigm. Global Environmental Change, 2, 111–127.

INDEX

ability to pay principle 66, 76, 90 act consequentialist perspective 120, 129–130, 132 adaptive action 15, 20–24 adaptive duties 16, 22–23 Almassi, Ben 128 appeal to the possibility of carbon leakage, the 30 appeal to uncertainty, the 30, 35 appeal to unfairness, the 30 aviation 2, 6, 61–78 benefit-based duties 91 beneficiary pays principle 6, 66, 90 Broome, John 18, 86–87, 121–123, 126, 128, 137 carbon budget 6, 61–62, 66–67, 70–78 carbon footprint 6–8, 65, 90, 102, 108, 118–131 carbon majors: 2, 6, definition, responsibility of, emissions 45–58 carbon sinks 130–131 capacity-based duties 31, 34–35, 90, 93 causal indeterminacy 136–137 causal inefficacy objections 137 causation-based duties 92–93 climate action 3, 12–26, 29, 38, 40, 63–70, 74, 100, 106, 111, 113 climate duties 3–5,12–26, 29–40 climate harms 5–6, 26, 46–58, 85, 113, 120–129

climate policy 16, 29, 32, 36–37, 54, 62, 67–74 collective action problem 4, 12–17, 20, 26, 135–137 collective agents 12, 30, 85, 92–93 collective duties 7, 30, 107 co-principals 51 Cooperative Promotional Model (CPM) 7, 101–114 compensation, compensatory duties 47, 51–57 complicity 48–49 consumption-reduction duties 96 contribution-based duties 31–39 corporate responsibility 45–58 Cripps, Elizabeth 7, 123 Cullity, Garrett 125 deontological: constraints 106, reasoning 137 devolved duties 4, 31–40 diachronic rationality 89 difference-making problem 84–88, 95–96, 121 difference making, duties of 13–25, 84–88 differentiated responsibility 68–69, 73–75 discursive coordination 136–137 disgorgement 52–57 distributive fairness 36 distributive justice 68 divestment 53–55

Index 153

Edenhofer 67, 69, 70, 74 emissions trading scheme/s 2, 29 enforceable directed duties 91–92 Erhardt-Martinez et al. 3 evidence-relative morality 8, 120, 126–128, 131–132 exceptionalism (sectoral) 61–78 expected value 34–38, 86, disvalue 6–8, 25, 34, 36, 85, 119, 120, 126–132 fact-relative morality 120, 126, 131 green consumption 4, 14 Hardin, Russell 137–148 Heede, Richard 43, 92–93 Hobbesian solutions 136, 148 Hotelling Rule of Exhaustible Resource Economics 138 indeterminacy 8, 135–149 individual responsibility (for climate harms) 137, 147 intermediate: actors 4, agents 4, 11–17, 22–25 Kagan, Shelly 85, 128 Kingston, Ewan 119 Kowarsch, Martin 69, 70, 74 Kutz, Christopher 48, 124–125 liability 48–57, 104 lock-in effects 50, 54 low expected cost principle (LEC), The 8, 129 market-based mechanism/s 72 market indeterminacy 137–139 mitigatory duties 21–26 nations, national governments (duties) 4, 12–26 negative impact 14, 49–58, 109, 122 no–harm principle 109 not to cause harm, duty 24, 47

offsets 52, 54, 72 partial compliance objection 106, 110 participatory duties, duties of participation 4, 14–15,19–26 phase outs (of fossil fuels) 52, 56–57 Pogge, Thomas 124 policy diffusion 35–36 polluter pays principle 6, 66, 90 principle of beneficence 105, 109, 110 Principle of Unjust Restitution (PUR) 51, 56, 57 procedural justice 68 protest (see also strike) 3 putative duty/group 101, 105 reasonable cost (condition/threshold) 7, 96, 102, 104, 109, 113 rights-based perspective 120, 130–132 risk transfer 68, 70 rule-consequentialism/ist 106 sectoral (approach, climate justice, perspective, responsibility, exceptionalism) 6, 61–78 Singer, Peter 31, 66, 143 shipping 6 statist solutions 136 strike 3 sub-collectives 33 sub-national actors 63 Sinnott-Armstrong, Walter 119 social cost of carbon, the 34, 70 sustainable economies 62 synchronic rationality 89 token emissions 119–132 transformation (of activities) 53–56 unilateral action 29, 33, 35, 37, 40 unstructured collective harms 125 virtue ethical perspective 120, 131–132 weakly collective duty 101, 105