Moral Enhancement and the Public Good 9781032018850, 9781032020860, 9781003181798

Citation preview

Moral Enhancement and the Public Good

Currently, humans lack the cognitive and moral capacities to prevent the widespread suffering associated with collective risks, like pandemics, climate change, or even asteroids. In Moral Enhancement and the Public Good, Parker Crutchfield argues for the controversial and initially counterintuitive claim that everyone should be administered a substance that makes us better people. Furthermore, he argues that it should be administered without our knowledge. That is, moral bioenhancement should be both compulsory and covert. Crutchfield demonstrates how our duty to future generations and our epistemic inability to promote the public good highlight the need for compulsory, covert moral bioenhancement. This not only gives us the best chance of preventing widespread suffering, compared to other interventions (or doing nothing), it also best promotes liberty, autonomy, and equality. In a final chapter, Crutchfield addresses the most salient objections to his argument. Parker Crutchfield is Associate Professor of Medical Ethics, Humanities, and Law at Western Michigan University Homer Stryker M.D. School of Medicine. He writes in bioethics and epistemology, teaches medical ethics to medical students and resident physicians, and provides clinical ethics consultations.

Moral Enhancement and the Public Good

Parker Crutchfield

First published 2021 by Routledge 605 Third Avenue, New York, NY 10158 and by Routledge 2 Park Square, Milton Park, Abingdon, Oxon, OX14 4RN Routledge is an imprint of the Taylor & Francis Group, an informa business © 2021 Parker Crutchfield The right of Parker Crutchfield to be identified as author of this work has been asserted by him 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. Library of Congress Cataloging-in-Publication Data A catalog record for this book has been requested ISBN: 978-1-032-01885-0 (hbk) ISBN: 978-1-032-02086-0 (pbk) ISBN: 978-1-003-18179-8 (ebk) Typeset in Bembo by Apex CoVantage, LLC

Contents

Prefaceviii Acknowledgmentsix Introduction

1

  1 Disease and Treatment Treatment and Enhancement  12 Health and Disease  15 Comparative Health  16 The Traditionalist Response  18 Contextualism About Health and Disease  20 Contextualism and the Treatment/Enhancement Distinction  26

10

  2 Preventing Harm Ultimate Harm  29 Preventing Extinction  29 Duties to Future Persons  30 Duty to Protect  35 Duty to Protect in Nonidentity Cases  38

28

  3 An Epistemic Argument for Compulsory Moral Bioenhancement The Problem  44 Epistemic Burdens  50 What’s on the Menu  51 The Epistemic Burdens of Available Options  52 From Epistemic Burdens to Enhancement  54 Bioliberal Moral Psychology  58 Compulsory Moral Bioenhancement  61 Conclusion 63

42

vi Contents

  4 A Moral Argument for Compulsory Moral Bioenhancement Agar on Moral Post-persons  64 Moral Exemplars  66 We Should Engender Moral Post-persons  68 From Moral Post-persons to Compulsory Moral Enhancement 70

64

  5 The Proposal

74

  6 The Epistemology of Moral Bioenhancement The Reflection Problem  80 The Moral Hallucination Problem  82 The Trustworthiness Problem  83 The Rational Irrationality Problem  84 The Free Rider Problem  86 Recommendation 87

80

  7 Covert Moral Bioenhancement Compulsory Moral Bioenhancement Is a Matter of Public Health 90 Public Health Ethics Frameworks  92 Covert Moral Bioenhancement  93 Values in Public Health Ethics  93 Conditions in Public Health Ethics  97 Implementation 98 Objections 101 It’s Unrealistic  101 It’s Dishonest  102 It’s Unlike Other Public Health Programs  102 It’s Paternalistic  104 It Diminishes Autonomy  104 It’s Inegalitarian  107 Incompetent Administrators  108 Conclusion 109

90

 8 Transparency What Justifies Transparency?  112 Respect for Persons  112 Trust and Transparency  116 Accountability 119 Public Justification  120 Stealth Democracy  122

111

Contents vii

Covert Compulsory Moral Bioenhancement and Stealth Democracy 126 Paternalism 127 Private Institutions in Public Health  128   9 Libertarian Covert Compulsory Moral Bioenhancement Liberty and Ultimate Harm  132 Clean Hands  134 Clean Hands and Covert Compulsory Moral Bioenhancement 136 Minor Intrusions  139 Compatibility Between Covert Compulsory Moral Bioenhancement and Libertarianism  142 Covert Compulsory Moral Bioenhancement Is a Minor Intrusion 144 What Difference Does Covert Make?  147 Right to Privacy  149 Egalitarianism 151 Conclusion 152

131

10 Conclusion Objections 154 The Covert Enhancement Hypothesis  162 The City Walls  163

154

References164 Index172

Preface

I argue for one central claim: someone should administer a biological intervention upon our moral capacities and should do so covertly, without the people who are receiving the intervention believing that they are receiving it. This is a radical and, initially, deeply counterintuitive claim. I am often asked if I really believe it. The answer is: I think the arguments in favor of it are better than the arguments against it. The arguments that follow are as much about the limits of the burdens we are willing to accept and the sacrifices we are willing to make to ensure survival as they are about moral enhancement. What I hope emerges is the idea that our moral capacities are extremely limited, relative to the task of ensuring survival. I attempt to offer a method of repairing this limitation. Maybe there are other ways. Do I walk around believing that we really should covertly manipulate everyone’s moral capacities? No, but the alternative needs defense, especially in the face of what I think are good arguments against it. In part, this project is an invitation to others to offer such a defense. The project began merely as a way of testing the limits of bioethics. The central claim challenges the foundation of the field, something which those in bioethics do too rarely. An underlying premise for my central claim is that, taken as a collective, we are all quite bad at figuring out what to do. We are bad at thinking and bad at acting in ways that achieve the things we think we want to do. In short, we are incompetent. I am no exception. I have been persuaded by the arguments and I hope that they persuade those who read this book. However, my central aim of the book is not to convince others of the central claim. It’s simply to make the claim seem less counterintuitive, less reactionary. Perhaps others will pick up where this project fails and develop something that may end up saving us. As the argument developed, so did my confidence that the conclusion is right. Unsurprisingly, my confidence that we will all be fine went down. I am a realist and this is a realistic book. But all professed realists are just pessimists in bad disguise. There is no reason for optimism. Realistically, we, or the next generation, or the generation after that, are all doomed. Evidence for hope or optimism is easily defeated by the mountains of evidence that we are lost.

Acknowledgments

I am grateful to many colleagues and friends for their role in helping to shape the ideas in this book. I am especially grateful to Tyler Gibb, Mike Redinger, and Scott Scheall for offering support, feedback, and discussion over the several years that the ideas in this book developed. I  also thank the anonymous reviewers for Routledge, Jonathan Anomaly, Jason Eberl, Andy Khoury, and Mark Navin for offering feedback on some of the chapters. Fabrice Jotterand and Devan Stahl brought to my attention some of the objections in the final chapter, for which I am grateful. Thank you to Andrew Beck at Routledge, whose support made this book possible. Chapter  7 was originally published in Bioethics, 2019, 33(1), 112–121, and is reprinted here with permission from John Wiley and Sons. Other chapters were adapted from a series of papers. Chapter 3 is adapted in part from Crutchfield (2016). Chapter 4 is adapted from Crutchfield (2020b). Throughout the book, I have incorporated considerations partly drawn from Crutchfield (2020a). Some of the objections in Chapter 10 are motivated by and adapted from considerations originally explored in Crutchfield (2018).

Introduction

Human society will collapse someday. At the latest, it will collapse when the Sun dies. Probably it will happen sooner, much sooner. This is an event we should try to avoid. Human society enables valuable experiences, relationships, pleasures, and pains. It is good to be alive and do the things that humans do. But these experiences, relationships, pleasures, and pains will certainly all come to an end. Not as certain, but still probable: they will come to an end soon, if we don’t figure out how to work together to avert collective risk. Societal collapse has happened many times before. The conditions that cause societal collapse are complex and interdependent and include factors related to the physical environment, culture and customs, and external forces. It is far more complex than simply noting that the Rapa Nui society collapsed on Easter Island because they used too many trees to transport their famous statues. But societal collapse has occurred and it will occur again. In our case, it may already be occurring, or may be likely to occur soon, if we don’t intervene on collective risks such as pandemics or climate change. But there are other possibilities. An asteroid may hit us; a terrorist may trigger nuclear war; a super volcano may erupt. To avert these events, collective action is required. If we want to prevent extinction from an asteroid, then we will need to devote significant resources to figuring out how to do that. If we want to prevent another global pandemic, then we will need to sacrifice the benefits that result from a globalized economy. If we want to avoid the harm and suffering that is likely to result from climate change, everyone needs to make individual sacrifices. If we take no collective action, it seems likely that this will all come to an end quite soon. Climate change will destabilize the food supply and other sources of livelihood. This will destabilize states. There will be mass migration, pressuring other states. These states will collapse, and so on. This is a risk everyone bears. It takes collective action to avert it. To continue to achieve moral value, and to avoid the significant moral disvalue that comes with famine, disease, war, and insecurity, we must act collectively. There is no reason to suppose that we are currently able to do this.

2 Introduction

It seems the only way to avert this collective risk is to change our ability to behave collectively. My suggestion is that we change our ability to behave collectively by biologically enhancing our moral capacities. Right now, our moral capacities are not up to the task of preventing catastrophe. To bring them up to the task, we need to engage in widespread moral bioenhancement. Much has been written on moral bioenhancement. There is no universally accepted definition of moral bioenhancement. Raus et al. (2014) provide a thorough and helpful taxonomy of the definitions. The definition of moral bioenhancement that I use is this: Moral bioenhancement: The improvement of a person’s moral attitudes, motivations, or behavior through biological means. This definition may not appear helpful, as it includes “improvement.” And there are ambiguities about what it means to improve an attitude, typically a moral belief, or moral motivation or moral behavior (Agar, 2014). It is similar to the one that defines it as the improvement of a person’s moral capacities (DeGrazia, 2014). A  person’s moral capacities are presumably exhausted by his or her moral attitudes, motivations, and behaviors. One might also wish to amend the above definition to include dispositions and emotions (Jebari, 2014). Whatever one’s potential protests, the above definition is general enough to capture the notion of moral bioenhancement and specific enough to be useful. Given this definition, the range of potential methods of moral bioenhancement is wide. The proposal of engaging in moral bioenhancement has generated significant controversy, resulting in a substantial literature debating the ethics of moral bioenhancement. One of the points of disagreement in this literature is whether we should even engage in it. Here are some of the reasons people have offered to think that we should not engage in moral bioenhancement: it undermines our freedom; it undermines our autonomy; it undermines equality; it is not technologically possible; it is not conceptually possible; it is an enhancement; it undermines our moral status; it undermines our identity. I address all of these reasons to not engage in moral bioenhancement in what follows. The science of moral enhancement is not currently developed enough that it is possible to biologically improve all persons’ moral capacities. Some people think it may never be so developed. Others have articulated the reasons why it’s not currently possible and may never be so (Crockett, 2014; Wiseman, 2016). And other than simply stating that I think it will be possible at some point to biologically intervene on a person’s moral capacities, this discussion is beside the point. I agree that it is not currently possible—there is no known substance that can improve all persons’ moral capacities such that they behave in ways to avert collective risk. Some substances have been suggested, though. The

Introduction 3

one that gets the most attention is oxytocin or the “love drug.” The science of oxytocin does not support its use as a widespread moral enhancement. The going hypothesis of the effect of oxytocin on social behavior is that it makes social cues more salient. But social cues trigger both pro-social and antisocial behavior. If oxytocin makes these cues more salient, then it will promote both pro-social and antisocial behavior. Indeed, this is exactly what is observed in studies of oxytocin and its effect on social behavior. It promotes pro-social behavior, such as altruism, among “in-group” interactions (De Dreu & Kret, 2016). This is why it has been described as the “love drug”—the studies were of subjects in the same social group. But it promotes antisocial behavior among “out-group” interactions (De Dreu et al., 2011). For this reason, it can’t be an effective widespread moral enhancer, because collective action to avert collective risk requires pro-social behavior toward members of other social groups. However, it has been shown to improve the pro-social behavior of those with antisocial personality disorder (Alcorn et al., 2015; Timmermann et al., 2017) and may therefore be useful as a moral bioenhancement for individuals who have this disorder. Others have suggested other substances. Psychedelics, in particular psilocybin, may be one potential moral enhancer, as it has been shown to improve social cognition and empathy (Pokorny et al., 2017). More generally, we think we can improve upon a person’s moral capacities. If we didn’t, then we would not engage in widespread moral enhancement through education, both formal and informal. So it’s not as though we think we have these static moral capacities and that they can’t be intervened upon and improved. Furthermore, there may be a variety of methods by which to biologically improve upon a person’s moral capacities. One way would be to administer a substance that has a direct effect upon them. But it may also be possible to intervene indirectly, such as by substances that cause, for example, vomiting, when a person behaves antisocially. Another indirect route to moral enhancement may be by way of cognitive enhancement. The idea is that improving a person’s cognitive capacities will inevitably also improve their moral capacities. The person will be better at reasoning, including moral reasoning, which will improve their ability to behave morally. I take no stand on what the method of enhancement will be, however, only that it will be possible at some point to biologically improve a person’s moral capacities. That it will be possible to at some point biologically improve a person’s moral capacities is one assumption upon which my argument rests. There are others. One is that the justification for the main claim depends in part on the idea that such an intervention is a matter of public health. There are many ways one might segment public health ethics off from other areas of applied ethics (Faden et al., 2020). By “public health intervention” I simply mean any intervention intended to improve or protect the overall health of a population. This is compatible with narrow accounts of public health,

4 Introduction

such as Anomaly’s (2011) account that says public health interventions are those interventions that distribute public goods to the population; and it is compatible with broader accounts that might consider interventions aimed at remediating systemic racism a matter of public health. The more relevant point is that the present concern is the prevention of suffering from collective risks such as those associated with climate change, pandemic, a meteor, or an exploding Sun. The concern is over interventions intended to prevent hunger or other types of biological insecurity or being burned alive or whatever else might be the cause of suffering from the occurrence of these collective risks. Preventing being burned alive is pretty clearly a matter of public health, as is preventing hunger or infectious disease or early death. Preventing the suffering associated with the collective risks that this book is about is straightforwardly a matter of public health. I return to public health in greater detail in Chapters 7 and 8. I make another assumption, but don’t return to it in subsequent chapters: whatever that public health intervention ends up being will be safe and effective. This is to say that it won’t have frequent and deleterious side effects and that it will reliably improve a person’s moral capacities across a wide population. It will intervene at a point upstream of a person’s moral capacities such that it compensates for any cross-cultural differences. Some might worry that a moral enhancement might make us all worse off than we were before the enhancement—that it will take our morality in the opposite direction (Fabiano, 2020). Built into the notion of safety and efficacy is that science can rule this possibility out. Thus, to assert that it’s possible the moral enhancement makes us all worse off reduces to the assertion that science cannot rule out the possibility that the moral enhancer makes us all worse off. If granted these assumptions, I believe the case that we should engage in covert compulsory moral bioenhancement—moral bioenhancement administered to everyone without them knowing it—is strong. The aim of this book is to make this case. It may be reasonable to assume that we can discover a safe and effective means of moral bioenhancement. But this is not to say that whatever method of enhancement we land on won’t have drawbacks or negative side effects. Likely any candidate method will have some undesirable effect. Using oxytocin as a form of boosting empathy is now widely viewed as a poor candidate for moral bioenhancement, because of the likelihood that it will increase conflict between groups, in spite of the fact that it may decrease it within groups. SSRIs and lithium, other proposed substances, have welldocumented side effects. Other psychoactive compounds such as lysergic acid diethylamide or psilocybin may be the best candidates as of now, but if consumed in larger doses, such as those common to recreational use, can have side effects that are highly disturbing to the user. Compounds that enhance morality by enhancing intelligence or cognition are not necessarily any more likely to avoid negative side effects. The

Introduction 5

anti-narcoleptic, Modafinil, is now used in some social circles as a cognitive enhancer, but it can cause sleeplessness and headaches, and nausea. In the movie, Limitless, a novelist suffers from writer’s block and is offered a cognitive enhancer. The pill makes him extremely intelligent and productive and upon taking it his social location and welfare rapidly improve. But after some time, he begins to develop debilitating headaches and discovers he’s probably going to die from the pill. Such a pill could not be considered safe and effective, but the story illustrates the general problem that whatever ends up being safe and effective is still likely to have some negative side effect and what this effect is will have to be balanced against the weight of the benefit that it brings about. Some of these side effects may not be experienced by individuals, but by the group. Enhancing moral capacities by boosting empathy is an example of this effect. Cooperation between groups may be undermined by increases in empathy, given its ties to parochialism. In Chapter 3, I argue that considering the epistemology of collective action warrants compulsory moral bioenhancement. One significant factor in whether groups can sustain cooperation is whether enough members feel and are motivated by shame (Bowles & Gintis, 2011). Thus, on the one hand, we might view one candidate method of enhancement as a substance that increases a person’s sense of shame. But on the other hand, shame is deeply unpleasant, which is bad in and of itself, and we don’t know what happens to cooperation when there is too much shame. It could undermine cooperation even further. Or suppose that instead of administering a substance to people to enhance their cognitive or moral capacities, their genes are edited, either before or after birth. Such interventions are the subject of numerous texts, fiction and nonfiction, and movies, many of which use awful social effects of gene editing as a plot device. One scientific worry is that by intervening upon the genetics of the human germline, we may end up introducing an unknown heritable trait to the species and that this trait may compromise our survival and proliferation. Anomaly and Jones (2020) argue that genetic enhancement can positively affect cooperation and that this effect amplifies as more and more people receive the enhancement. But it is possible that genetic enhancements have some negative side effects, and that these are also amplified as more and more people receive the enhancement. When the enhancement occurs by way of germline editing, which comes with the potential for unknown off-target effects, these negative “network” effects seem more likely. It is a truism that what we ought to do depends on descriptive properties. Thus, any prescription for moral enhancement will ultimately depend on what the facts pertaining to the intervention are. This is to say that the ethics of enhancement will depend on the science of the intervention. The extent to which the science of a candidate intervention—the risks, burdens, and benefits to the individual and the group—differs from other candidates

6 Introduction

is the extent to which the permissions and obligations of administering that intervention will differ. But there are further constraints on what types of enhancement might be compatible with the prescription of this book. I argue that our moral capacities ought to be enhanced and that this ought to be done in secret. Thus, for this to happen the intervention must feasibly be administered in secret. This prescription eliminates any candidate that could not be administered secretly. So, for example, any intervention that can be administered only in the form of a pill is a poor candidate, because it seems infeasible to secretly administer pills to a large portion of the population. Similarly, most genetic enhancements are likely poor candidates, as these interventions often involve prolonged and repeated commitments on the part of the person receiving the intervention. For covert compulsory moral bioenhancement to be feasible, the intervention must be deliverable during the normal course of the person’s life. When institutional review boards (IRBs) assess risks to research subjects, they do so relative to the subject’s normal life activities. If a study requires a spinal tap and the subject wouldn’t be getting a spinal tap outside of the research, the spinal tap represents a significant risk. But if the research requires a spinal tap, and the subject was already going to be getting a spinal tap, say, for a medical diagnosis or treatment, then the spinal tap for research is considered minimal risk. Similar assessments in the administration of a moral bioenhancement are appropriate. The intervention must be deliverable in the normal course of the person’s life for the purposes of secrecy, but what counts as “normal course of life” might differ significantly from person to person or group to group. It’s hard to see how genetic enhancements could be delivered in the person’s normal course of life unless they can be administered through an injection. This is not to say that genetic enhancements lack moral justification, just that they are unlikely to be the best candidate for a covert compulsory moral bioenhancement. Given that behavior is driven by a combination of genetics, biological changes that occur after birth, and a person’s environment, if genetic enhancements are unlikely to work, which leave only pharmaceutical interventions on the person’s biology and interventions upon the person’s environment. It’s possible that interventions upon a person’s environment may produce the sort of effects that we need to stave off societal collapse, and I discuss Powell and Buchanan’s (2016) account of how to do this in Chapter  3. But interventions on the person’s environment are the traditional means of enhancement. As such, they are not likely to be effective at doing what we need an intervention to do. That leaves primarily pharmaceutical interventions upon the person’s biology as candidates for covert compulsory moral bioenhancement. I otherwise have no specific recommendation for what should be used to morally enhance people. Buchanan (2011, p. 7) draws a distinction between those enhancements that come through the “front door” and those that

Introduction 7

come through the “back door.” Enhancements that come through the front door are those that are developed and advertised and administered for the purposes of enhancement. Enhancements that come through the back door are those that enhance but only as side effects of interventions that aim to treat. Most enhancements that are available now come through the back door. But there’s a third category that has so far been unrecognized: those enhancements that come not through either door, but through the pipes, undetectable to the inhabitants. I am arguing that we should pursue this type of enhancement, figuratively and, potentially, literally. Maybe introducing a drug that induces greater sensitivity to shame is worth the cost. How does the unpleasantness of feeling shame compare to the unpleasantness of struggling to survive? Or maybe psilocybin is the answer. Or maybe cognitive enhancement will enable greater cooperation, resolving the problem of preventing and mitigating the suffering associated with collective risks. Though I  make no attempt at identifying a potential moral bioenhancement, I am optimistic that the science of morality, as broad as it is, can discover a safe and effective intervention, given enough resources. I discuss in more detail the conditions of such an intervention in Chapter 5, which lays out the specific proposal—that a substance be administered to them without their knowledge that functions to improve their moral capacities. This chapter puts the proposal in the context of other common ways in which society permits and relies on information withholding and deception. It also describes potentially feasible methods of administration, such as by packaging the substance with compulsory vaccines or by putting it in the water supply like fluoride. This chapter marks the turning point in the book. In the preceding chapters, I argue that we ought to administer moral bioenhancement and that the administration ought to be compulsory. In the succeeding chapters, I argue that moral bioenhancement ought to be covert. One of the objections to moral bioenhancement is that it is an enhancement rather than a treatment and that we should only administer treatments. This claim rests on a distinction between those interventions that count as treatments and those that count as enhancements. This distinction rests further on what counts as health and what counts as disease. In Chapter 1, I argue for a new theory of health and disease according to which whether a person counts as healthy is a matter of the conversational context of the health attributor. This contextualism about “healthy” has significant advantages over traditional theories of health and disease. One consequence of this contextualism is that it becomes less clear that the distinction between treatment and disease is defensible, and even less so that the distinction is clear enough to be morally relevant. In Chapter 2, I argue that we have a moral obligation to prevent suffering and that this is impossible without widespread moral bioenhancement.

8 Introduction

I argue further that current generations have a duty to protect future generations. This duty to protect helps to resolve some of the difficulties associated with moral obligations toward people who don’t yet exist, such as those obligations that the nonidentity problem potentially undermines. Thus, the chapter not only presents an argument for widespread moral bioenhancement but also offers a novel treatment of the nonidentity problem and what it implies about moral obligations to future generations. In Chapter 3, I argue on epistemic grounds that moral bioenhancement is necessary to avert collective risk and that it should be compulsory. This claim rests on a notion that my colleague, Scott Scheall, and I developed over a series of papers. The notion is that of an epistemic burden, which is simply that which one is ignorant of but which is required to deliberately pursue some goal. It’s that which we don’t know but need to. My claim in this chapter is that the epistemic burdens associated with averting the suffering that is likely to result from climate change are so overwhelming that they are impossible to overcome in the absence of enhancement. I then argue that this supports the idea that enhancement should be compulsory. Person and Savulescu (2014b) argue that moral enhancement should be compulsory. The argument in Chapter 3 is a novel argument for this conclusion. I offer a second novel argument for this conclusion in Chapter 4. The argument in this chapter is that we are best off—our interests are best served—by engendering moral post-persons, or persons whose improved moral capacities confer upon them a higher moral status. In doing so, we make things better for ourselves, and that we are obligated to do this. Since satisfying this obligation requires moral enhancement, moral enhancement should be compulsory. Thus, both Chapters 3 and 4 argue that moral enhancement ought to be compulsory, but one argument is on epistemic grounds and the other on moral grounds. In Chapters 6 and 7, I make the case that moral bioenhancement should be covert. In Chapter 6, I argue that the epistemology and psychology of moral capacities are such that the only way for moral bioenhancement to be effective is by administering it covertly. Overt administration simply will not do what it is intended to do, if the intention is to avert catastrophe. The crux of this argument is that informing subjects that they are receiving moral bioenhancement triggers some epistemological problems, including the problem that such information is likely to cause the subject to mistrust the intervention, which undermines its effectiveness. Chapter 7 further establishes the preferability of a covert administration by arguing that if moral bioenhancement is to be compulsory, it is morally preferable for it to be administered covertly rather than overtly. Given that moral bioenhancement, as an intervention to avert collective risk, is a matter of public health, a covert administration better conforms with the norms of public health ethics than does an overt administration. That is, if moral

Introduction 9

bioenhancement is going to be compulsory, it’s morally better for it to be administered covertly. In Chapter 8, I argue that covert compulsory moral bioenhancement is not as politically problematic as it may initially seem. I claim that there is little moral foundation for an obligation to transparency in public health interventions. I argue further that we have good reasons to think that people don’t want transparency in their political decision makers, if those decisionmakers are empathetic and non-self-interested, which, if my argument in Chapter 4 is correct, they would be if they were morally enhanced themselves. In Chapter  9, I  continue with the politics of covert compulsory moral bioenhancement and argue that such a program is compatible with a version of libertarianism that is highly protective of individual liberties. This argument rests on the idea that covert compulsory moral bioenhancement protects the use of others’ individual liberties. If covert compulsory moral bioenhancement is compatible with even the most protective form of libertarianism, then it is likely compatible with other political philosophies that place less weight on individual liberties. From the very beginning of the book, much of the work in establishing the main claim is by responding to objections, which I do often along the way. In Chapter  10, I  address in detail one objection: that moral bioenhancement may change who we are. I spend the chapter arguing that this is not something we should be concerned about. I also address a range of other objections in this chapter and finish with some concluding remarks.

Chapter 1

D isease and Treatment

The task of this book is to justify the administration of a covert compulsory moral bioenhancement program. For the project to get off the ground, it cannot be true that all enhancement of human capacities is wrong. This chapter undercuts the assertion that enhancing human capacities is wrong, clearing the way for the rest of the argument. What counts as an enhancement rather than a treatment depends on what counts as health and disease. In this chapter, I develop an account of health and disease that undermines the foundation of any objection to moral enhancement on the grounds that only treatments are permissible. There are multiple ways one might distinguish health from disease. There are also ways one might distinguish between different accounts of health and disease. One way to distinguish accounts of health and disease is by whether they hold that disease and health are primarily biological (Simon, 2007; Broadbent, 2019; Glackin, 2019). Objectivism is roughly the view that a person is sick, ill, or unhealthy in virtue of the fact that, first, their biology is in some sense malfunctioning and, second, that this malfunction makes their life go worse than it otherwise would. According to objectivism, disease and illness are concepts analyzed primarily biologically. Of course, not all biological malfunction is a disease; the malfunction must also make the person’s life go worse for it to be considered a disease. Whether a person’s life goes worse can be, for objectivists, a biological matter—intractable intense pain intuitively diminishes a person’s well-being. But whether a person’s life is going worse than it otherwise would can also be a matter of how society views that condition and its consequences. The other view, constructivism, analyzes disease and illness differently. Instead of analyzing disease and illness primarily in biological terms, constructivists hold that being sick is a matter first and foremost of social evaluation of a condition. Once society has determined that a certain set of behaviors or conditions diminishes a person’s well-being, we then go on a search for some biological cause of that condition. Objectivists and constructivists can agree that being sick is a matter of both descriptive properties of the person, such as properties of their functioning,

Disease and Treatment 11

as well as evaluative properties, such as properties pertaining to how society views certain conditions or behaviors. What they can’t agree on is which is conceptually prior. For objectivists, it’s biological functioning; for constructivists, it’s society’s view of the condition or behavior. A second way to distinguish types of accounts of health and disease is by whether they include some notion of value in the analysis. Boorse’s biostatistical theory (1976, 1977) is objectivist, but purports to be value-free. However, other objectivist accounts, such as Wakefield’s proper functionalist account (1992), include value-laden conditions. Constructivist accounts are even more value-laden (Reznek, 1987; Cooper, 2002). A third way to distinguish types of accounts is by whether they are fundamentally comparative. Traditional theories view health as the fundamental concept to be analyzed. Some analyze it by incorporating notions like statistical normalcy, proper function, or social construction. But once health is analyzed, other related concepts, such as “healthier than,” can be analyzed in terms of “health.” But comparative accounts of health reverse this order of explanation. According to comparative accounts, “healthier than” is conceptually prior to “health.” That is, rather than explaining “healthier than” in terms of “health,” comparative accounts explain “health” in terms of “healthier than.” Currently, there is only one comparative account of health and disease, and that comes from S. Andrew Schroeder (2013), who claims that “health” is gradable and comparative like adjectives such as “tall,” “flat,” “old,” and “bald.” In this chapter, I  argue for a contextualist account of health, which builds upon Schroeder’s foundation. Fundamentally, he is right that health is comparative. But his novel account is incomplete, which exposes it to a range of problems and objections. The contextualist account I develop in this chapter not only lacks this exposure, it holds several advantages over other accounts of health and disease, regardless of how one sorts those accounts. Specifically, I argue that the meanings of terms such as “sick,” “healthy,” “ill,” or “diseased” are context-sensitive. The truth of statements embedding these terms is determined by the context in which those statements are uttered. These terms are like “tall,” “bald,” “old,” “flat,” “empty,” and many other similar terms. They are gradable adjectives whose meaning is determined in part by the class of things they are being compared to. At a college party, a 35-year-old professor is old; at a conference with her peers, she is not old (and might even win an award for being a young scholar). Both attributions of oldness are true. Whether she is old or not old depends in part on the comparison class. For people at a college party, 35 is old. For people at an academic conference, 35 is not old. I think “sick,” “health,” and the like are just like this. For people in the intensive care unit (ICU), those who have a cold are healthy. But that same person in a preschool is not healthy. If this is right, then it makes as much sense to provide a conceptual analysis of

12  Disease and Treatment

sickness as it does to provide a conceptual analysis of oldness, which implies that the traditional theories are looking in the wrong direction. The first section introduces the purported distinction between treatment and enhancement. In the section that follows I outline Schroeder’s comparative account of health and disease. I then discuss its exposure and weaknesses and the weaknesses of other traditional accounts of health and disease. In the remainder of the sections, I support contextualism about health and disease, show its advantages, and argue that it undermines any objection to moral enhancement on the grounds that moral enhancement runs afoul of the treatment/enhancement distinction.

Treatment and Enhancement Intuitively, there is some important difference between interventions that treat disease and those that only promote health. There seems to be a difference between repairing something so that it is how it ought to be and making something that is already how it ought to be better. This intuitive difference also seems to have moral significance: it seems better to bring a person back to normal than it does to take a normal person and make them better than they were. This is the moral significance of the treatment/ enhancement distinction. The distinction obviously relies on what counts as normal, or on how a person’s health ought to be. The theories of health and disease provide the conditions for when a person’s health is as it should be— when a condition counts as normal and when it doesn’t. In this way, the treatment/enhancement distinction and its moral significance, whatever that significance amounts to, depends on accounts of health and disease. That people can be sick and healthy and that how we intervene on the sick seems different than how we intervene on the healthy are the main reasons to think that there is a distinction between treatment and enhancement. Whether a given intervention is a treatment or enhancement may vary according to what counts as healthy and what counts as sick. Suppose a condition that results from a malfunctioning organ (perhaps that has developed from genetic mutation) is highly prevalent among a population. For a theory that grounds disease in function, that condition will count as a disease. Interventions upon that condition may then count as treatments. But for a theory that grounds disease in statistical normalcy, that same condition will count as not a disease. Interventions upon that condition may then count as an enhancement. If there is a morally relevant difference between treatment and enhancement, then different accounts of health and disease will imply morally relevant differences in which interventions are permissible or obligatory. For a theory that constructs disease out of values, which interventions count as treatments and which count as enhancements will depend on how those values inform disease. For example, Rachel Cooper argues for an account according to which a disease is a “condition that it

Disease and Treatment 13

is a bad thing to have, that is such that we consider the afflicted person to have been unlucky, and that can potentially be medically treated” (2002, p. 272). On her account, as on mine below, there may still be a morally relevant difference between treatment and enhancement. But where the line is drawn between the two may change according to what is considered “bad” or “unlucky,” whereas for objectivist accounts that line generally won’t move. There are multiple reasons one might object to enhancement but not to treatment. One reason rests on the claim that the goal of medicine is merely to return a person to how they biologically ought to function. That is, the goal of medicine just is to treat, not to enhance. This goal demarcates the interventions that are medically permissible and those that are not. Interventions outside this boundary are not permissible. Although the primary goal of medicine may be to restore a person to normal, having this goal cannot be a reason to think that enhancements are impermissible. Otherwise, cosmetic surgery would be impermissible. So would vaccines. Vaccines don’t treat anything; they prevent disease. Presuming that cosmetic surgery and vaccines are compatible with goals of medicine, the goals cannot simply be to restore to normal. They must at least also include prevention. But once prevention is compatible with the goals of medicine, it’s less clear that using the goals of medicine to draw the line between permissible treatments and impermissible enhancements is possible. Some interventions aim at preventing suffering rather than alleviating it. Some of those interventions, such as vaccines, may do so by causing a person’s normal function to be greater than it otherwise would be. More generally, once vaccines and cosmetic surgery are included among the interventions compatible with the goals of medicine, it’s not clear that one can draw a morally relevant distinction between treatments and enhancements. If vaccines are permissible because they enhance the functioning of a person’s immune system, why would it be impermissible to enhance the functioning of a person’s cognitive functioning by giving them a drug? One answer is that vaccines don’t exacerbate unjust social inequalities, but some are concerned that interventions like cognitive enhancements may be more affordable to rich people, which may exacerbate inequality that is already unjust. One might say the same thing for cosmetic surgery. Ugly rich people are better able to use medicine to intervene on their ugliness, which will improve their well-being and further exacerbate the social advantages they already have. I address concerns related to enhancement and egalitarianism in subsequent chapters. But here I note that vaccines would exacerbate unjust social inequalities in very significant ways if they weren’t given to everyone at little to no cost to the person and that they were compulsory. That is, vaccines fail to exacerbate unjust social inequalities because they are provided to people. If they had to be purchased and were as expensive as, for example, cosmetic surgery, rich people would very quickly

14  Disease and Treatment

exacerbate any unjust social inequalities they already enjoy, presumably by living longer, healthier lives. There may be other ways to draw the line such that vaccines are on the permissible side and interventions on the mind are on the other side. For example, one might think that the relevant difference is that one intervention is on the immune system and the other on the mind, and that the mind maintains morally relevant significance. It’s somehow more closely related to who we are, whereas the immune system is mostly unrelated to our identity or personality. This alleged difference between the immune system and the mind, however, is not a plausible way of distinguishing those interventions that are permissible from those that aren’t. First, drawing the distinction like this would preclude any intervention on the mind, even those that restore function. Second, it can’t support any morally relevant difference between treatment and enhancement. To support such a distinction, one must rely on some account of health and disease. There is (at least) one more general problem with the claim that we should allow treatments but not enhancements. Bostrom and Ord (2006) introduce the Reversal Test, which is a thought experiment meant to test whether intuitions are grounded in a cognitive bias toward the status quo. They introduced it as a challenge to those objecting to cognitive enhancement, but it works just as well to identify status quo bias in the objection to any enhancement (or even most intuitions). But since this is a book about moral enhancement, I use that as the relevant proposal. The test is a challenge to those who think that we shouldn’t morally enhance. Suppose instead of morally enhancing, we reverse the direction of the intervention or policy such that the proposal is that we downgrade our moral capacities. Then ask, is this moral downgrade intuitively acceptable? Presumably, it is not acceptable to intervene on a person’s moral capacities such that they behave more frequently and more severely immoral. Thus, the opponent of moral enhancement doesn’t want to make our moral capacities better, and they don’t want to make them worse. It thus looks like they have a bias toward the status quo. The challenge of the Reversal Test is that it requires the opponent of moral enhancement to justify why they think the status quo is a better state of affairs than either widespread moral enhancement or widespread moral downgrade. Why is the status quo morally optimal? The test partly functions to shift the burden of evidence upon the opponent of enhancement. Instead of simply arguing that moral enhancement is wrong, they have to support the notion that the status quo is optimal, which is a difficult task, given all of the bad things that people do and all of the suffering that could be prevented by deviating from the status quo. The task that the opponent of moral enhancement is similar to the task of the person who wants to argue that God exists even though there is a lot of human and animal suffering in the world. A theist has to explain why in spite of all of this suffering this is

Disease and Treatment 15

still the best possible world. The opponent of moral enhancement, by way of the Reversal Test, has to explain why in spite of all the preventable suffering the status quo is morally optimal. Maybe the strategies will be the same. If they are, there is not much hope for the opponent of moral enhancement, given the recurring failure of theodicies. But in the absence of such support, it seems that opposition to moral enhancement is driven by a bias toward the status quo.

Health and Disease Schroeder’s account of health and disease is comparative. ‘Health’ is defined in terms of “healthier than.” Non-comparative accounts define health and disease first. The two accounts that are most influential are Boorse’s biostatistical theory, which grounds health and disease in statistical normalcy, and Wakefield’s functional account, which grounds health and disease in proper function. Boorse’s theory is the following: (1) The reference class is a natural class of organisms of uniform functional design; specifically, an age group of a sex of a species. (2) A normal function of a part or process within members of the reference class is a statistically typical contribution by it to their individual survival and reproduction. (3) A disease is a type of internal state which is either an impairment of normal functional ability, i.e., a reduction of one or more functional abilities below typical efficiency, or a limitation on functional ability caused by environmental agents. (4) Health is the absence of disease (Boorse, 1997, pp. 7–8). Boorse first defines disease as a departure from normal function, relative to a reference class. The relevant references class is members of one’s species that are the same age and sex. Whether a condition is a disease depends on whether the condition is statistically typical among that reference class. If the condition is statistically typical, then the condition is not a disease. A fouryear-old boy’s condition is a disease if the condition is statistically atypical among four-year-old boys and the condition impairs the boy’s functioning. It doesn’t matter if it is common in adults or men or infants. The relevant reference class is boys of a similar age and how prevalent that condition is among them. Note that Boorse’s account purports to be value-neutral. None of the conditions embed terms that refer to values. This means that on Boorse’s account, a condition is a disease even if someone wants it, or enjoys, or derives significant pleasure from it, which may be counterintuitive. A condition could both be a disease and improve a person’s well-being on his account. And there’s no way to block this problem, because to do so one of

16  Disease and Treatment

the conditions would need to embed terms relating to desires or well-being, which of course would make it value-laden. The rejection of any value as a condition of health and disease is what most separates Boorse’s account from Wakefield’s, which is: A condition is a disorder if and only if (i) the condition causes some harm or deprivation of benefit to the person as judged by the standards of the person’s culture (the value criterion), and (ii) the condition results from the inability of some internal mechanism to perform its natural function, wherein a natural function is an effect that is part of the evolutionary explanation of the existence and structure of the mechanism (the explanatory criterion). (Wakefield, 1992, p. 384) Wakefield’s account explicitly includes a value-laden condition (the value criterion). The second difference between Wakefield’s account and Boorse’s is that for Wakefield normal function is not a matter of statistical frequency among a reference class but a matter of whether something is doing what it is supposed to do, where “supposed to do” is determined by the evolutionary history of the thing. There are many notions of proper function, and any of them could be plugged into this account. The important difference is that function is a matter of evolution for Wakefield and a matter of statistical frequency for Boorse. The important similarity is that they both ground the analysis of health in function. This fact implies that for any two individuals, to say that one is healthier than the other would first require that their individual health first be assessed according to how well they function. With that, it would then be possible to say that one is healthier than the other. These are two of the most influential accounts of health and disease. Schroeder introduces a problem for both of them, the solution to which suggests that it is better to develop a comparative account.

Comparative Health The problem Schroeder identifies is that non-comparative accounts seem committed to implausible claims about assessments of intergenerational health. He uses the following example (p. 138): Alys was a very successful early medieval farmer of child-bearing age. In addition to the vegetables and grains which composed the bulk of her peers’ diet, she also had regular access to meats and dairy, and consequently to a nutritionally superior diet—including, among other things, much higher quantities of iron. As a result, she had higher energy levels, greater physical endurance, a longer attention span, was less susceptible

Disease and Treatment 17

to certain pathogens, and was at lower risk for complications during pregnancy. Allie is a 21st-century factory worker. Thanks to fortified foods, her diet contains more iron than Alys’s, although it is still deficient by modern standards. As a result, compared to her fellow workers she suffers from reduced energy levels, reduced physical endurance, a shorter attention span, and is more susceptible to certain pathogens and complications during pregnancy. It seems clear that relative to her contemporaries, Alys was healthy. It also seems clear that relative to her contemporaries, Allie was unhealthy. However, Allie was healthier than Alys. For example, Allie still has a longer life expectancy, more energy (from the fortification of the diet), is less likely to suffer complications during pregnancy (relative to Alys), and many other health advantages. The example supports three claims, each of which is independently plausible. However, they can’t all be true. For every two of the positions, the third must be false. The claims are (p. 139): (1) Alys was healthy. (2) Allie was unhealthy. (3) Allie was healthier than Alys. For non-comparative accounts, (1) and (2) imply that (3) is false; (2) and (3) imply that (1) is false; (1) and (3) imply that (2) is false. Schroeder’s strategy is to show that for non-comparative accounts, denying any of the positions is implausible. Denying (1) is implausible because if Alys was unhealthy, she must be so relative to modern standards. But it is very likely that future developments in health promotion will be such that people currently living won’t be able to meet them. Thus, if Alys is unhealthy by modern standards, then most everyone currently living is unhealthy by future standards. But it’s implausible that most everyone currently living is unhealthy. So, it doesn’t seem plausible to deny (1). It doesn’t seem plausible to deny (2), because if Allie is healthy, then far too many people count as healthy, including many who we would typically describe as being unhealthy. And it doesn’t seem plausible to deny (3), unless one denies the possibility of making intergenerational assessments. But then such a denial makes it impossible to makes sense of claims regarding improving health over time, or of claims that people living now are healthier than people living 50 or a hundred years ago. The alternative, Schroeder claims, is to hold that “healthy” is fundamentally comparative, just like gradable adjectives like “tall,” “old,” “bald,” and “flat” are. The meaning of “healthy” is determined by a comparison class in the way that the meaning “tall” is determined by its comparison class. If viewed this way, then it is possible to resolve the paradox above by claiming

18  Disease and Treatment

that what “healthy” means in (1) is different from what it means in (2). There is no inconsistency because the meaning of “healthy” shifts as the comparison class shifts from a class of individuals who lived generations ago to a class of contemporary individuals. A comparative account has several advantages over non-comparative accounts. First, it can easily resolve the paradox. Non-comparative accounts are seemingly committed to untenable claims about health and disease. Second, it can easily make sense of species-typical disease, or conditions that are highly prevalent among the relevant population, but that we may want to nevertheless count as disease. For example, it’s plausible that environmental factors such as pollution may cause the same harmful condition in the entire population, such as a condition that undermines respiratory function. In that case, for accounts like Boorse’s, the condition wouldn’t count as a disease (Boorse, 2014). Third, a comparative account can also easily account for borderline cases. On a non-comparative account, it’s not clear whether to count, for example, 20/16 vision as a disease. Such impairment is highly prevalent among the population. Maybe it’s hard to say whether someone with 20/16 vision is healthy. But it’s not hard to say that someone with 20/16 vision is, other things being equal, healthier than someone with 20/10 vision. These considerations lead Schroeder to conclude that a comparative account of health and disease is well motivated and may enjoy several advantages over non-comparative accounts. He further specifies his account, starting with more fundamental analysis (p. 144): X is healthier than Y in respect R, iff X’s R-functioning is superior to Y’s R-functioning (given the goals of survival and reproduction). He then accounts for “health” in terms of “healthier than” (p. 145): X is healthy in respect R iff X’s R-functioning is superior to the R-functioning of a sufficient number of the members of some relevant comparison class.

The Traditionalist Response Schroeder’s comparative account has several advantages over noncomparative accounts: it can resolve the paradox of intergenerational health assessments; it can account for species-typical disease; and it can account for borderline cases. There are, however, some problems with his account. Boorse (2014) identifies three of them. One problem that Boorse identifies is that on Schroeder’s comparable account, there doesn’t appear to be any conceptual space for someone to be in perfect health. If health and disease are relative to others and how their processes function, including how one’s

Disease and Treatment 19

own health improves or degrades over time, then it would be impossible to ever be perfectly healthy. One could always improve one’s function. So, one could always be healthier, which means no one can ever be in perfect health. But it seems like it is appropriate in some cases to say of a person that they are perfectly healthy. A second problem relates to people who are high-functioning. Consider two Olympic athletes running the 400-m dash. One person will win, barely beating the second-place finisher. They will win because they function better. On Schroeder’s comparative account that means that the winner is healthier than the second-place finisher. And it doesn’t seem right to say that the second-place finisher is not only less healthy than the winner, but also that they lost because they were less healthy. Schroeder identifies this objection himself. His response is to bite the bullet, but note that doing so isn’t especially problematic, because all accounts have to say that health and disease are a matter of degree. A third problem Boorse identifies is that Schroeder’s account seems to allow for some odd comparisons, given his definitions of “healthier than” and “healthy.” For example, on Schroeder’s account, it would be right to say that a woman is healthier than a man with respect to gestating a fetus. It would be right to say on Schroeder’s account that a philosopher is healthier than an infant with respect to cognition. For non-comparative accounts, no such claims would be true. For Boorse’s account, the relevant reference class is those who are of the same age and sex. For Wakefield, it is proper function that grounds health, and both infants and philosophers are functioning as they should. The problems Boorse identifies are reasons to think that discourse about health and disease is not comparative. We just don’t say of elite athletes that the gold medalist is healthier than the silver medalist. And we don’t say of a woman that she is healthier than a man because she can gestate a fetus. Indeed, of both athletes, we are likely to think that they are perfectly healthy. None of these problems, however, show that a comparative account can’t be compatible with how we think and talk about health and disease. The three problems that Boorse identifies all arise from a fourth problem: Schroeder’s account is merely a foundation. He has not provided a full account of health and disease. For example, he has not offered what, on a comparative account, picks out the relevant comparison class and, once that comparison class is picked out, which properties of its members fix the meaning of “healthier than.” In the case of other gradable adjectives, doing so is straightforward. When attributing tallness to a person, the relevant comparison class is picked out by the conversational context, specifically, the conversational context of the tallness attributor. And once that class of individuals is picked out, it is clear what standard is used to measure tallness—their height. The same is true of the other gradable adjectives. Conversational context picks out

20  Disease and Treatment

the comparison class, and once that’s determined there are straightforward standards by which to attribute baldness (number of hairs on the head), oldness (time alive since birth), and flatness (change in elevation relative to change in distance). But Schroeder’s account neither says how we pick out the relevant comparison class nor does it say how to attribute health to members of that class. He does say that it is a matter of functioning processes. But this is far too vague, especially when compared with the specific ways in which other gradable adjectives are attributed to people and things. In the remainder of this chapter, I offer a contextualist account that can do all of these things. The account can not only resolve the problems associated with Schroeder’s foundational, yet incomplete, account but also maintain all of its advantages. Once I establish it, I use it to show that even if there is a treatment/enhancement distinction, it is not a morally relevant distinction.

Contextualism About Health and Disease Consider the following set of claim, which cannot all be true: (1) S knows that she has hands. (2) If S knows that she has hands, then she knows she is not a brain in a vat. (3) S does not know that she is not a brain in a vat. It’s impossible for all three statements to be true. (1) and (2) imply that the (3) is false; (2) and (3) imply that (1) is false; (1) and (3) imply that (2) is false. This, of course, is the skeptical paradox. There are multiple ways one might resolve it. One is to deny (1). But this comes at a cost, namely that it denies that S and everybody else know things that we ordinarily would want to say they know, such as that they have hands, that it is Thursday, and that it’s cold outside. It gives in to the skeptical worry. Denying (2) requires denying a highly intuitive logical principle, namely that knowing that p, and knowing that p entails, implies knowing that q. Denying (3) doesn’t give appropriate credit to the skeptic. There was something important about Descartes’ consideration that he could be deceived by an evil demon. We can’t prove that we aren’t in the Matrix. All three statements are intuitively and independently plausible, but they can’t all be true.1 One prominent solution to the paradox is to claim that “knows” works like gradable adjectives. Thus, “knows” in (1) means something different than it does in (3). In ordinary, getting-around-in-the-world contexts, S does know that she has hands. By the standards in that context, it is appropriate to attribute to her knowledge of that proposition. But as one moves from (1) to (3), the context shifts, and with it the standards for knowledge attributions. Once one is considering the possibility of being in the Matrix or being deceived by an evil demon, the standards for knowledge have gone

Disease and Treatment 21

up. In that context, it is true that S does not know that she is not a brain in a vat. The meaning of “knows” shifts with the context in the same way that the meaning of “tall” shifts when first talking about preschoolers and then talking about basketball players. Contextualism about “knows” allows one to hold all three statements, resolving the skeptical paradox. Obviously, there are similarities between Schroeder’s account of health and disease and the contextualist solution to the skeptical paradox. But Schroeder stops well short of developing a complete account of health and disease. But just as there are robust contextualist accounts of “knows,” a parallel account of “healthy” is available. This shouldn’t come as a surprise. Contextualism is partly a response to the total failure of trying to provide an analysis of knowledge. The concept of health is similar enough to the concept of knowledge that it is not unreasonable to expect the conceptual analysis of health and disease to also be a total failure. Recall that one of the weaknesses of Schroeder’s account is that it both fails to say how the relevant comparison class is picked out and, once it is, what the standards are for health attribution. A  contextualist account of health and disease can resolve both of these problems. Like it is with other gradable adjectives, the comparison class is picked out by the attributor’s conversational context. For example, if the attribution is in the context of adult women living hundreds of years ago, then plausibly the comparison class is adult women living hundreds of years ago. If the attribution is in the context of contemporary adult women, then plausibly the comparison class is adult women. Schroeder has it right about these attributions: “health” means different things in these contexts. But the context doesn’t shift, and with it, the meaning of “healthy” merely from time period to time period. It can also shift between other conversational contexts. The context can shift with shifts in standards related to function, statistical or proper, sex, age, and even acuity of disease and the degree to which the subject of the attribution might benefit or not. Suppose Rob’s dad is in a critical care unit being treated for complications related to end-stage heart failure, a result of decades of smoking cigarettes. Rob’s big toe is broken, and he’s in a walking boot. But he’s well-fed, fit, moving around easily, and looks well-rested. When the critical care physician comes in to check on Rob’s dad’s ventilator settings, he looks at Rob and says, “Don’t smoke, you’ll stay healthy.” The physician has attributed health to Rob, and this attribution seems true. But suppose that, conveniently, Rob has an appointment in the same hospital that same day with his orthopedic surgeon. As Rob enters the waiting room, he bangs his boot against the door frame hard. The physician sees him do this and says, “Be careful, Rob, or you’ll never be healthy.” This attribution of health also seems true. What changed was the conversational context of the attributor. In critical care, the context of one in which the patients are old and often near death; in an orthopedic surgery, they typically are not.

22  Disease and Treatment

One might complain that in this case, it’s right to say that when attributed to the whole person, the meaning of “healthy” changes according to context, but that it doesn’t change according to a specific function and therefore doesn’t show the context sensitivity of “healthy.” But this is simply an admission that “healthy” is context-sensitive. One feature of the conversational context that helps determine the meaning of health attributions is the relevant function or process. Orthopedic surgeons talking about health and diseases are usually talking about the function of bones, other hard tissues, and how they fit together. Critical care doctors are typically talking about integrated and dynamic organ systems. These are just features of the conversational context that help determine the meaning of “healthy.” There’s no need to define “health,” as Schroeder does, by reference to specific functions or processes. Consider again young, bright Rob. He has a runny nose from the common cold. One of his bright, young friends says, “Rob, stay away from me, you’re sick!” This attribution seems true. Now suppose Rob does what his friend says and walks home. On the way home, he walks by a young couple that is suffering from biological and social consequences of drug abuse. One of them asks for spare change for a bus ticket. Rob wipes his nose with his sleeve, covers his mouth, and says, “Sorry, I’m sick.” The drug addict looks at him and says, “You’re not sick.” It seems right that the drug addict’s attribution is true. But it also seems like Rob’s self-attribution is true. The only way both can be true is if the meaning of health attributions is context-sensitive. For the addict, the standards for being healthy are much lower than they are for Rob. To avoid this conclusion, one must deny one of the above health attributions. The most likely candidate for denial seems to be the drug addict’s attribution that Rob is not sick. Rob has a mild symptom of the common cold. The drug addict is wrong—Rob is sick. But consider what this implies about the drug addict’s other health attributions. Suppose the addict observes a friend undergo addiction treatment and over the course of the treatment observes the friend’s condition improve. As he improves, the addict continues to make health attributions until one day the friend is young and bright and no longer suffering the social consequences of drug addiction and is only suffering one biological consequence: a runny nose from the common cold. When the addict says of his friend, “He’s not sick anymore,” that attribution seems true. The person who claims that the addict is wrong about Rob cannot say that the addict is right about his friend. But it seems obvious that the addict is right about his friend. Or suppose instead that Rob was born with a genetic mutation that caused severe deformity of his left arm. When born, it seems wrong to say that he was born a healthy baby boy. But there is a point in his life at which point it would be right to say that Rob is healthy, even perfectly so. He doesn’t even need treatment on his arm for that attribution to be true. The deformity just needs to be figured into the ordinary conversational context

Disease and Treatment 23

in which he is a subject. Indeed, if he never receives treatment, the only way he could ever appropriately said to be healthy is if the conversational context shifts so that such attributions are true. The meaning of “healthy” shifts according to the context of the attributor. The relevant features of the context include the standards for function. Just as the meaning of “knows” shifts when one enters the philosophy classroom and its higher standards or as the meaning of “flat” shifts as one drives through the plains of Nebraska to the Colorado Rockies, the meaning of “healthy” shifts with standards related to function. The functional standards for a true health attribution are lower among drug addicts, but also among the elderly, such as those in assisted living or a nursing home. But the context can include more, such as features related to the disvalue of the condition. In this way, contextualism about “health” can accommodate both the idea that health and disease are partly a matter of function and partly a matter of value. These standards may vary from context to context, but how they vary will help fix the conversational context and with it the meaning of “healthy.” Contextualism about “health” maintains several advantages over Schroeder’s comparative account. First, as above, it stakes a claim on what makes “health” change meaning. Second, it can help to resolve the other problems with Schroeder’s account, those that Boorse identifies. Boorse claims, and Schroeder admits, that his view requires saying that among elite athletes, the race winner is healthier than the second-place finisher, and that, indeed, that’s why they won the race. But it seems wrong to say that one is healthier than the other and that this difference in health accounts for the difference in athletic performance. Contextualism is committed to no such implication. According to contextualism, the standards for what counts as “healthy” are established by the conversational context of the attributor. While it is open to the contextualist to say that the first-place finisher is healthier than the second-place finisher, she is by no means committed to that. Simply, both athletes are healthy, and there’s nothing more that needs to be said. For Schroeder’s comparative account, more needs to be said, because “healthy” is analyzed in terms of “healthier than.” But contextualism makes no such commitment. Relatedly, contextualism can allow that some people are perfectly healthy, whereas Schroeder’s cannot. This was another of Boorse’s complaints. Just as it’s possible for something to be perfectly flat, a person can be perfectly healthy. This is just to say that there are contexts in which the statement “S is perfectly healthy” is true. What counts as perfect health will vary according to the attributor’s context, however. So although it’s true that function could always improve, this doesn’t preclude attributions of perfect health as it does according to Schroeder’s account. Boorse’s final complaint about Schroeder’s account was that it licenses one to say that women are healthier than men relative to gestating fetuses,

24  Disease and Treatment

for example, and that this seems wrong. This is no problem for contextualism. Suppose an obstetrician says to one of her patients that the pregnant woman and her fetus are perfectly healthy. Then the woman asks, “But am I healthier than my husband?” Schroeder’s account suggests that the physician ought to answer that, yes, she is healthier than her husband, at least in one way. This seems wrong. The physician would likely say instead that she doesn’t know, or that he should see his doctor. But on a contextualist account of health and disease, this reply is perfectly fine: the woman’s question about her husband shifts the conversational context from mothers and fetuses to something that includes adult men such as her husband. The same is true in any case in which Schroeder’s account implies strange health comparisons. According to contextualism, there is no need to claim that women are healthier than men relative to giving birth or that adults are healthier than babies relative to walking. The conversational context of the health attribution typically wouldn’t require such attributions. Contextualism has advantages over Schroeder’s account while also being able to account for intergenerational assessments of health, resolving the paradox he identifies. It has this advantage over non-comparative accounts. It also accounts for judgments about cases, such as those above, which are problematic for non-comparative accounts. But there is another type of case that is a problem for the traditional, non-comparative case that contextualism can easily account for: food poisoning. Suppose Rob eats some tainted meat that wasn’t cooked to a hot enough temperature. A few hours later, he gets violently ill with vomiting and diarrhea. His food poisoning lasts for almost an entire day, but then he returns to normal. Was Rob sick? It seems obvious that he was. But this is not something that objectivist accounts can allow. For Boorse, a condition is healthy if, and only if: (1) The reference class is a natural class of organisms of uniform functional design; specifically, an age group of a sex of a species. (2) A normal function of a part or process within members of the reference class is a statistically typical contribution by it to their individual survival and reproduction. (3) A disease is a type of internal state which is either an impairment of normal functional ability, i.e., a reduction of one or more functional abilities below typical efficiency, or a limitation on functional ability caused by environmental agents. (4) Health is the absence of disease. Food poisoning meets all of these conditions. Rob’s vomiting and diarrhea are a statistically typical contribution to his survival and reproduction. Expelling potential toxins from the body is the function that contributes to an individual’s survival and reproduction, and it is typical for human bodies

Disease and Treatment 25

to respond this way. So, food poisoning meets Boorse’s first two conditions. It also meets the third. The tainted meat doesn’t cause any limitation of functional ability. Everything functioned exactly as it should have. Rob’s body’s resources were marshaled to expel the potential toxin, then they did so and Rob got back to whatever he wanted to do. Things went perfectly, and this is how they would go in most people of his age group and sex. Of course, Rob couldn’t walk or do much of anything else while he was vomiting and having diarrhea. But that doesn’t imply that his functional ability was limited. If this was sufficient to make him sick, then by Boorse’s conditions one is sick while they sleep. Rob’s body was self-regulating in the way that it ought to, and this required conditions preventing him from getting up and walking around. It’s no different from when a person is so tired that they have to sleep. But people who are sleeping aren’t sick. It would be no different if his neighbors’ loud music kept him up all night so that the next night he fell into a deep sleep early in the evening. His body’s statistically normal and properly functioning self-regulation required that he sleep, which prevented him from getting up and walking around. But we wouldn’t want to say that Rob was sick while asleep.2 So, on Boorse’s account, Rob is not sick. Wakefield’s account has the same problem. Wakefield’s account has both a harm condition and a condition on proper function. It’s true that Rob is being harmed, but everything is functioning perfectly. His body is doing what it has evolved to do, and it’s this function that has contributed to his ancestors’ survival and proliferation. Thus, on Wakefield’s account, Rob isn’t sick. Contextualism can easily explain the attribution of sickness to Rob in the case of food poisoning. In the attributor’s conversational context, Rob is sick. Perhaps in this context, the harm dimension is weighted more heavily and the function dimension is less heavily weighted. Although everything is functioning as it should for Rob, he is undoubtedly experiencing harm. Coupled with the fact that the harm is so obviously a matter of biological function, this may swamp any notion of proper function to make felicitous the attribution of sickness. Attributions of health and disease, if the view sketched here is correct, are elliptical, in that there are multiple respects in which one might be healthy or sick. One might be healthy with respect to objective features of the person and their function. Or one might be healthy with respect to normative features. But attributions of tallness, baldness, and flatness are not elliptical. When a person attributes baldness to another, there is only one respect in which one could be bald—the amount of hair on the head. It is my burden to account for the elliptical nature of health attributions and the fact that we typically get by just fine in using “health” in conversation. There are multiple ways to account for these facts, and they correspond to the varieties of epistemic contextualism that have been offered over the

26  Disease and Treatment

past four decades. But here is my preferred method, though it needs to be developed and defended in more detail elsewhere. The respect in which one is healthy (e.g., objectively healthy versus normatively healthy) is picked out by the practical interests of the speaker (Stanley, 2005). The practical interests of the orthopedic surgeon pick out a more objective respect in which Rob is healthy, whereas the practical interests of the substance abuser pick out a more normative respect in which he’s healthy. Usually, there is no confusion among conversational participants about what a speaker’s practical interests are, and there’s little reason to think that making health attributions would introduce such confusion. In the contexts in which a speaker attributes health, these practical interests pick out the respect in which one is healthy.

Contextualism and the Treatment/Enhancement Distinction Contextualism about “healthy” and other health attributions has lots of advantages, not only over non-comparative accounts of health and disease but also over Schroeder’s foundational but incomplete comparative account. Whether a person is healthy or sick can vary according to the context of the health attributor. The features of this context may include considerations of function (statistical or proper), harm, source of harm, age, sex, or even genotype or phenotype. A  contextualist account of health and disease has implications for the purported distinction between treatments and enhancements. First, if whether an intervention is a treatment or enhancement depends on whether a person is healthy, and whether a person is healthy depends on the conversational context, then whether an intervention is a treatment or enhancement shifts with the meaning of “healthy.” Unless one defines the morally objectionable contexts, if shifting contexts changes whether an intervention is a treatment or enhancement, the purported treatment/ enhancement distinction cannot ground a moral objection to enhancements. This fact neutralizes one significant objection to moral “bioenhancement.” Second, and although this is obviously true, that the treatment/enhancement distinction is sensitive to the conversational context suggests that medicine doesn’t only deal in treatments. Medicine treats and enhances. Here is another set of claims, which can’t all be true: the proper goal of medicine doesn’t change; the proper goal of medicine excludes enhancements; whether an intervention is a treatment or enhancement changes according to context. At least one of these claims is false. I’ve argued that the third is not. Third, given that health and disease attributions are context-sensitive, it’s plausible that capacities that moral bioenhancement intervenes upon are diseases. That is, it is plausible that in the context of collective action to

Disease and Treatment 27

mitigate collective risk, our cognitive and moral capacities malfunction and that we experience harm from this dysfunction. The moral and cognitive capacities associated with collective action to mitigate collective risk are not healthy. If this claim is true, then interventions upon them aren’t enhancements at all. They’re treatments, which further undermines the idea that we shouldn’t biologically intervene upon moral capacities because we shouldn’t allow enhancements. In what follows, I refer to interventions upon moral capacities as moral “enhancements,” but this is only out of convenience. They could equally be called moral “treatments,” and readers should not take my use of “enhancement” as endorsement of the treatment/enhancement distinction. The above considerations are some of the ways to neutralize the potential objection to moral enhancement on the presumption that such a distinction exists. But if health and disease are context-sensitive, it’s better to simply claim that there is no distinction at all. Suppose someone tried drawing a morally relevant distinction based on baldness or tallness or oldness. That’s what drawing a treatment/enhancement distinction is like, if contextualism about health and disease is true. Unless the moral relevance of the treatment/enhancement distinction shifts with context, which would suggest an unacceptable sort of moral relativism, it is better to say that there just is no such distinction between treatment and enhancement. Making such a distinction in a way that is morally relevant requires a sharp and clear separation between those interventions that are treatments and those that are enhancements. But given the context-sensitivity of health and disease, there is no such sharp and clear separation. Thus, no distinction is possible, and all of the hand-wringing about whether an intervention counts as a treatment or an enhancement is only apt to chap some hands rather than guide the administration of interventions aimed at improving others’ well-being.

Notes 1. The literature on contextualism and the skeptical paradox is vast. For entry into the topic, see Rysiew (2016). 2. Boorse considers similar cases of poisoning, but in those cases the organ system in question is dysfunctional. In this case it is not.

Chapter 2

P reventing Harm

The argument in the previous chapter blocks the objection that, overall, we shouldn’t endorse enhancements. The argument in this chapter blocks another potential objection. One might agree that enhancements are generally permissible, but disagree that we should use them, or anything, really, to avert the suffering associated with collective risks. One might agree that these risks are indeed pressing and that only collective action can help us avoid them, but still deny that we should do so. For example, climate change is bad, but that doesn’t mean we should do anything about it. This chapter is supposed to block this claim. These risks are significant, and we should do something to try to prevent the suffering that is associated with them. On our current course, we can expect very bad things to happen to lots of people. If we can prevent these bad things from happening in a way that doesn’t require current generations to bear a burden that is comparable to these bad things, then we should prevent them. There are several considerations that motivate the need for moral bioenhancement. What they have in common is that they all pertain to the bad things that will happen to people if there is no widespread moral bioenhancement. The first of the three considerations below is that, in the absence of moral bioenhancement, we can expect “ultimate harm,” which is a condition in which all or most humans have died or are living lives that are not worth living. This is the consideration that motivates Persson and Savulescu’s claim that moral bioenhancement should be compulsory. The second consideration is that ultimate harm may not be as bad as it gets: humans could go extinct, which would be very bad. If we can prevent this from happening, we should. The third consideration is that we have a duty to protect future generations. Many believe that we have a duty to not harm future generations. But I argue that the grounds for preventing the bad things from happening to future generations is not that we have a duty to non-maleficence, but rather that we have a duty to protect. Even if ultimate harm and extinction are not likely, they are still live possibilities. We must shield future generations from these possibilities since it is our actions that are bringing

Preventing Harm 29

them about. Or, it is our moral duty to protect future generations from preventable bad things.

Ultimate Harm Persson and Savulescu offer details of ultimate harm elsewhere (2014b). Simply, ultimate harm is the condition in which all or most of humanity is dead or living a life not worth living. Their contention is that with advances in science, technology, and globalization, we are closer than ever to causing it. A single bad actor can now kill millions or billions of people with a nuclear weapon or other weapons of mass destruction. As scientists synthesize highly infectious diseases immune to vaccine, one could get out and cause a global epidemic. It used to be that these methods of killing high numbers of people were not so easily accessible. But as science advances and some states collapse, these become more accessible to terrorists. All it takes is one person to get their hands on a weapon of mass destruction and the means to use it to cause a potentially global epidemic or worldwide nuclear war. Climate change is another avenue of ultimate harm. Inaction will likely result in living conditions much worse than the current living conditions, even for those who are most well off. Seas will rise; refugees will die; states will fail; mosquitos will infect; ecosystems will collapse. Rather than dying by climate change or terrorism or pandemic, perhaps artificial intelligence will doom us all. Or maybe a particle accelerator will cause a black hole into which we are all pulled, the gravity separating our bodies as it does so. In short, there are lots of ways we might all die, and as science and technology and globalization develop these ways become more likely to occur. We should do the things that are necessary to avoid these occurrences.

Preventing Extinction Ultimate harm is very bad. When ultimate harm occurs, there will be catastrophic suffering and loss. It is worth preventing at almost any cost. Persson and Savulescu don’t need anything worse in order to make their point that moral bioenhancement ought to be compulsory. But there is something that is worse than ultimate harm: extinction. Compare three scenarios. Scenario A  is no catastrophic suffering and loss, just the ordinary amount of suffering and loss. Scenario B is ultimate harm, a scenario in which climate change results in the death of the vast majority of humans, and those that have not undergone considerable suffering. Scenario C is like scenario B, except no one is left—everyone dies. Obviously, Scenario C is the worst of the three, and Scenario B is very bad, much worse than Scenario A. But between which events is the difference greater? There is some difference in badness between Scenario A and Scenario B. There is also a difference in badness between Scenario B and Scenario C. Which of the two differences

30  Preventing Harm

is greater? Parfit (1984) claims that the second difference, the difference between ultimate harm and extinction, is greater. Others disagree and think instead that going from 0% loss of life to 80% loss of life and the added suffering is much worse than going from 80% to 100%. The point of the thought experiment is to draw out the notion that there is some loss of value from extinction that is over and above the simple loss of life of individuals. Johan Frick (2017) claims that humanity has intrinsic value, that it is valuable for its own sake. He also claims that the appropriate response to things of intrinsic value is to ensure their survival, noting that we go to great lengths to preserve the existence of endangered plants and animals, and culturally and historically important sites because it is good for these things to exist apart from the value individual members achieve from them. It’s thus plausible that we may have good reasons to ensure humanity’s survival, apart from the reasons that support the preservation of individual humans. The intrinsic, final value of humanity grounds each generation’s pro tanto moral duty to ensure its survival. That humanity’s existence has intrinsic value seems obvious. And it seems equally obvious that we have a duty to ensure that that intrinsic value is not lost. In the thought experiment above, does it compensate for the extra 60% difference if loss of life? Supposing at the time of catastrophe there are 10 billion people, the difference between Scenario A and B is a difference of 8 billion people. The difference between Scenario B and C is then only 2  billion people. Is the intrinsic value associated with humanity’s survival greater than the value of 6 billion people? Maybe it is, especially if we add to the intrinsic value of humanity’s survival the value associated with humans who, in the event of extinction, wouldn’t exist. In any case, if there is an intrinsic value to humanity’s existence, then we plausibly have a duty to ensure mere survival, in the same way that we have a moral duty to protect other endangered species or heritage sites.

Duties to Future Persons Ultimate harm is bad; extinction is worse. On the assumption that we should prevent bad things from happening, so long as doing so doesn’t require the sacrifice of something of equally morally valuable, we should do so (Singer, 1972). Persson and Savulescu, in their discussion of ultimate harm and how it motivates compulsory moral bioenhancement, presume that the harm to be prevented is experienced by currently living individuals. Because the scope of harms is restricted to those which existing persons may experience, they are able to escape the tricky problems associated with satisfying moral duties to future people, especially those future people whose very existence is causally dependent upon the things we do or fail to do in response to existential threats.

Preventing Harm 31

The problem is the nonidentity problem, typically attributed to Derek Parfit (1984). The problem centers around the idea that the existence of future people is in some cases dependent on procreative decisions, decisions which bring about some suffering in the person created. Suppose that Heather is planning on having a child. She knows her family has a history of a rare genetic variant that causes the onset of severe and debilitating physical symptoms in middle age and a comparatively early death. Because of this history, her doctor recommends genetic testing prior to conceiving. Heather undergoes the genetic testing and learns that she has the variant and that in the absence of any intervention, she will pass it on to her offspring, who, like Heather, will experience the pain and suffering and early death. In light of this, her doctor recommends gene therapy followed by selective in vitro fertilization to guarantee that her offspring do not suffer from their inheritance. Because of her family’s unique genetic history, the university is covering the cost of these treatments. Heather does not undergo gene therapy and instead conceives through intercourse with her partner. Later, her child, Lisa, is born with the variant. Many people confidently judge that what Heather did was wrong. She should have undergone the gene therapy and in vitro fertilization, and she was wrong to not do so and instead conceive Lisa through intercourse. But here’s the problem: if Heather had done that, Lisa would not have been conceived and born. Some other child would have been conceived and born instead. Since Lisa would not have been born, and living with the variant and the symptoms it causes is better than not living at all, it’s not clear who, if anyone, Heather wronged. But if there was no one who Heather wronged, it’s difficult to see how what she did was wrong. The problem is reconciling the strong moral intuition that what Heather did was wrong with the seemingly sound argument that she did not. Parfit offers other examples that involve policymaking and the environment. Suppose that a community elects to deplete resources rather than conserve them, where conserving them wouldn’t represent a great burden to the community. As a result of the decision to deplete, those who live during the time the resources are being depleted do not suffer, but that those who live after the resources are depleted suffer greatly. While it seems that the community ought to have conserved rather than depleted the resources, it also doesn’t seem that they harmed anyone. Had they chosen instead to conserve the resources, the identities of those who live hundreds of years later would have been different. Opting to deplete influences procreative decisions such that the people who end up suffering would not have existed if the community instead opted to conserve. And since it seems like not existing at all would be even worse, it’s not clear who the community harmed. And if they didn’t harm anyone, it’s not clear how they wrong future generations.

32  Preventing Harm

There is an enormous body of literature devoted to responding to the nonidentity problem in a way that preserves both the idea that what the community did was wrong and the idea that those who suffer as a result are harmed. Much of this literature is devoted to developing accounts of harm whereby it makes sense to say that the community harmed its ancestors. For example, Harman (2009) develops an account of harm according to which one person harms another if they cause the harming event—it has nothing to do with comparing how their lives otherwise might have gone. Others, for example, Meyer (2003), argue that doing things that cause a person to drop below a certain threshold of well-being is to harm that person, again regardless of how their lives might have otherwise gone. However, much of this literature presumes that the reason that the community’s choice to deplete rather than conserve is harm-based. That is, the wrongness of depleting is in some a matter of the community harming its ancestors. The trick is, then, just a matter of coming up with an account of harm whereby a person or community can harm a future generation. But the wrongness may instead be due to a failure to satisfy a duty to protect future generations. An important feature of the nonidentity problem and the cases that drive it is that we, as observers, can be certain about the outcome of the subject’s choices. In the case given earlier, we are certain that if Heather doesn’t undergo gene therapy and in vitro fertilization, her offspring will have the variant and the subsequent suffering. Consider other cases that drive the nonidentity problem (Roberts, 2019). One of Parfit’s original cases is of a 14-year-old girl who decides to conceive a child rather than wait some number of years later to do so. We judge her choice to conceive as wrong because we know (or, in the 1980s, we knew) that when 14-year-olds have children they don’t typically have the maturity or personal resources to give the child a “good start.” But had she waited, she would have had a different child, so it’s not clear how she has wronged the child she has as a 14-yearold mother. Kavka’s slave child case is similar. In this case, a couple enters into a binding, enforceable contract with a person who will pay them some large sum of money to bear a child who, once born, will go to the person and live the life of a slave (Kavka, 1982). It seems wrong for the couple to enter such a contract. But if they had not entered the contract, the child would not have been born. Plausibly, a life of slavery is still better than no life at all. The terms of the contract provide us with certainty that the child will end up a slave. Or consider two cases from Rivera-López (2009), which he calls “Parfitian.” In both cases, we can be certain of the outcome of the subjects’ choices. Anna’s Choice: Anna is planning to have a child. The doctor tells her that any child she conceives without medication will suffer from condition X. X is a serious physical disability, which

Preventing Harm 33

is nonetheless compatible with a worthwhile life. But Anna can take some medicine that prevents X. Bertha’s Choice: Bertha is planning to have a child. The doctor tells her that the child will suffer from condition X if she conceives now, but she can prevent X by waiting one month to conceive. The same is true of the above depletion case. As observers, we know that the community’s ancestors will suffer, and we know that the community itself will not. Parfit offers a similar example that has a structure similar to that of the depletion case. In that case, a community develops a policy that will result in the emission of nuclear waste hundreds of years after the decision, at which point it is known that the people living near it will suffer, while it is also known that the people living prior to that will not. This is all to say that a feature of the cases, both those that pertain to individuals and those that pertain to larger populations, is that they build in outcome certainty. Once the certainty of outcome is changed, however, it’s much less clear that what grounds the wrongness of nonidentity cases is that the decision of the subjects harms others, or that the reason it is wrong is that it violates a duty to non-maleficence. Consider instead a case in which the outcome is uncertain. Suppose a couple is planning on having a child. Before doing so, they take genetic tests to screen for any heritable diseases. The genetic counselor tells them that both of them have genes that are known to cause severe muscular atrophy in middle ages. However, there is only a 25% chance their child will inherit the gene. They are offered the option of screening the egg and sperm and in vitro fertilization to guarantee that their child won’t inherit it. However, they decline that option and instead elect to conceive without screening and have a child, whom they name “Vincent.” I don’t think the couple has done anything wrong. The change in the intuition is in virtue of the fact that confidence in the outcome is lower. But some might still judge the couple’s choice to birth Vincent to be wrong. Such a judgment is independent of the outcome because it hasn’t yet been described. Thus, the judgment that what the couple has done is wrong ought to be stable regardless of whether the child is born (at a 25% chance) with the gene or without it (at 75% chance). That is, if the choice is wrong, then it’s wrong even when Vincent is born without the disease. But if the choice is wrong and Vincent is born without the disease, then the wrongness of the couple’s choice is straightforwardly not a matter of harming the child. Something else must ground the judgment. Change the case so that Vincent has a 50% chance of inheriting the disease (i.e., the father has the disease and there’s a 50/50 chance of passing it on). It seems to me that the couple deciding to not undergo germline screening is not wrong, but others might yet disagree. The above point still stands. If the couple’s foregoing screening is wrong, it is wrong even if the child doesn’t

34  Preventing Harm

inherit the disease. And if that’s true, then the wrongness can’t be a matter of violating a duty to non-maleficence toward future generations. The closer the case gets to certainty that the child will inherit the disease, the more wrong the couple’s choice seems. Make it a 95% chance that the child will inherit the disease, and it seems wrong for the couple to not undergo the screening, just as it does in the typical nonidentity cases in which the outcome is certain. But if what grounds the judgment about the couple at 25% or 50% chance inheritance is not that they violate a duty to non-maleficence, then presumably when the chance of inheritance is 95% we should the say the same thing: at 95% confidence in inheritance, the judgment that what the couple does is wrong is not a matter of violating a duty to non-maleficence. The same point applies to cases like depletion in which a community makes decisions that (a) affect the identities of the future persons and (b) cause them to suffer in some way. Suppose that in Parfit’s depletion case there’s a 50% chance that the people who live hundreds of years after the decision suffer from the decision, and 50% chance that they don’t. If it is wrong for the community to decide to deplete, then it is wrong even hundreds of years later when its ancestors are in the good case. But in that case, the wrongness cannot be in virtue of harm-based reasons. Or suppose that the case is more like the current scenario, one in which a community, such as one like the United States, decides to deplete rather than conserve resources, knowing that such depletion will not only affect procreative decisions of people in the community and elsewhere, but that there is also a very high chance, say, 80%, that those people will suffer because of the decision to deplete. It may seem wrong for the community like the United States to deplete. But if it is wrong, it’s also wrong in the 20% chance they don’t suffer, which means that the wrongness cannot be in virtue of violation of a duty to not harm future generations, since they aren’t harmed in any way by the decision. Judgments about these cases reveal two points. First, in cases of outcome uncertainty, judgments about nonidentity cases that an agent has wronged a future generation cannot be grounded by violation of duty to nonmaleficence. If what the couple has done is wrong, that wrongness must be grounded by something else. My proposal: what grounds the judgment that what the couple does is wrong is that they have a duty to protect their future offspring and that in deciding to forego the germline screening they violate this duty. Second, if in cases of outcome uncertainty the judgment of wrongness is grounded not by the duty to non-maleficence but by some other duty (such as the duty to protect), then plausibly it is also this duty that grounds judgment in all nonidentity cases, even those in which the outcome is certain. In other words, what grounds the wrongness in typical “Parfitian” nonidentity cases is not a violation of a duty to non-maleficence. That the outcome is

Preventing Harm 35

certain in these cases masks the more fundamental duty violation, the duty to protect. That the outcome is certain guarantees that this duty to protect is violated. The alternative is to claim that in nonidentity cases whether the wrongness is a matter of a duty to non-maleficence or some other duty (such as the duty to protect) is dependent upon the certainty of outcome: at 95% chance of inheritance, the couple’s wrong choice is due to violating a duty to protect, but at 100% the wrongness is due to violating a duty to non-maleficence. This is a plausible alternative. But in the actual case of climate change or some other catastrophe that is likely to both affect the identities of future people and cause them to suffer, there is no certainty that the suffering will result. That suffering is likely doesn’t mean it’s certain. Suppose that no one does anything about climate change: there is no reduction in emissions in any way and instead communities continue to emit more than ever. This decision will cause many people to reproduce differently than they otherwise would, affecting the identities of those future people. But there is no guarantee that they will suffer from the decision to do nothing. There is a slim chance all of the science is wrong, or at least that the predictions that the models make are inaccurate. There’s a slim chance that a single genius will take it upon themselves to develop a technology that solves the problem of global warming. There’s a slim chance of a bunch of different things that, collectively, bring down the confidence that future people will suffer. Let’s say that they bring it down to an 80% chance. Maybe that’s high; maybe it’s low. But it doesn’t really matter whether it’s accurate. As long as it’s not 100%, then my point stands. In the good case, however unlikely it may be, the duty to non-maleficence cannot ground the judgment that depletion wrongs future generations. Future persons’ suffering from climate change is preventable. If one can prevent suffering without sacrificing anything of comparable moral significance, then one ought to prevent that suffering. Presuming that the prevention of future persons’ suffering from climate change doesn’t require the sacrifice of anything of comparable moral significance, we therefore ought to prevent future persons’ suffering from climate change. The structure of this “ought” is not that it is a moral duty to not harm future generations, however. It must be a different duty. My proposal is that the duty that grounds the normative force of the “ought” is the duty to protect future generations.

Duty to Protect It seems uncontroversial that at least some people have a duty to protect others. The most obvious examples are familial obligations. Parents have a duty to protect their children when they are young. Children may also have an obligation to protect their parents when they are old. But there are others.

36  Preventing Harm

Ship captains have a duty to protect their passengers and crew; lawyers have a duty to protect their clients; boards of directors have a duty to protect their shareholders; wardens have a duty to protect their prisoners; the state has a duty to protect its citizens; current generations have a duty to protect future generations. The actions required to satisfy these duties vary. But to at least some extent, all of these protectors must shield the protectees from suffering. The clearest instance of the duty to protect is the duty that parents have to protect their children. The list of things that a parent must do to satisfy this duty is long. And at certain times of the child’s life satisfying the duty requires constant attention. Children are entitled to some degree of protection—to be shielded—from the dangers of the world, and it is the parent’s duty to be that shield. When the shield is let down, the child is typically not harmed. But that doesn’t mean that the parent has done nothing wrong, nor that no one has been wronged. When a parent fails in their duty to protect their child, regardless of whether the child is harmed, they wrong their child. Less clear is the obligation that a child has to protect their elderly parents. It’s less clear when this duty kicks in, if at all, and when adult children are excused from it. An elderly parent with dementia who lives with an adult child is entitled to some protection. An elderly parent who lives alone and lacks the physical capability to move around is also entitled to some protection, and in the absence of some other arrangement, an adult child seems obligated to provide it. The duty to protect holds outside of the family, as well. A ship’s captain has a duty to protect their passengers. They don’t have a duty to see to the passengers’ interests, nor is it necessary that the passengers are even identifiable. A captain’s duty to protect is often satisfied or violated prior to anyone even embarking, as they ensure the seaworthiness of their ship. Failing to make sure there are enough lifeboats is perhaps the most famous failure of a captain failing to satisfy their duty to protect their passengers. But failing to ensure that critical systems are functioning properly and that the provisions are consumable and in great enough supply are also violations of the duty. Other professionals have a duty to protect. Lawyers have a duty to protect their clients from legal liability. This duty is not violated in case the client is found guilty or liable—some targets are so easy that no shield can protect them. But the lawyer has a duty to put up the shield and make it as strong as possible. To a lesser degree, doctors must protect their patients. The most obvious instance of this duty is a doctor’s duty to protect the confidentiality of the patient. When a doctor fails in this duty, which happens frequently, the patient is typically not harmed. But the patient is nevertheless wronged by the inappropriate disclosure of health information. Institutions may also have a duty to protect their members. States have at least some limited duty to protect its citizens from invasion. During the

Preventing Harm 37

COVID-19 pandemic, many states protected their citizens from the disease, while others let it ravage them. Those states that allowed the disease to ravage them failed in their duty to protect, though the satisfaction of this duty did not require that no person be affected. Similarly, boards of directors of companies have a duty to protect their shareholders’ financial interests in the company. This duty may be satisfied even if unforeseeable circumstances, or foreseeable but not preventable circumstances, force financial losses. And it may be violated even if no member of the company takes a financial hit. Satisfying the duty to protect guides much of what we do. The duty to protect can be distinguished from other duties by several properties. First, its violation doesn’t require that a person be harmed. There are ways in which a person can be wronged but not harmed, for example, when one is entitled to protection from another person and that person fails to provide it. Second, a person may have a duty to protect, even if there is no identifiable, specific person who holds a claim against that person. Unlike the right to not be harmed, the right to be protected depends on our relationships and the special obligations those generate. Consider the captain of a ship. The captain has a duty to protect all her passengers, and passengers have a right to be protected. But once that relationship is over and there is no longer a special obligation holding between captain and passenger, there is no more duty to protect. When the passengers disembark, they no longer have a claim against the captain to protect them, but they still have the claim against the captain, and everyone else, that they not be harmed. Robert Goodin (1986) provides an account of special obligations, those obligations that one owes to some people but not others. According to his account, these obligations are derived from the vulnerability between two people. A  has a duty to protect B just in case B’s interests (broadly construed) are vulnerable to A’s actions. Passengers on a ship are vulnerable to the actions of the captain; a child’s interests are vulnerable to the actions of their parents. One person is vulnerable to another when the satisfaction of their material and psychological interests depends on the actions of another person. Goodin bases his argument on common-sense morality. We often recognize these obligations that hold between specific individuals but not others. Agent-neutral utilitarianism seems wrong because it ignores what seems to be obvious, namely that sometimes a person owes more to one person than they owe to others. Parents have duties to their own children that they don’t have toward the stranger down the street. While others find the source of these special obligations elsewhere, Goodin’s view is that these special obligations arise out of vulnerability. One virtue of his account is that it can explain both the scope of these obligations (i.e., who holds the obligations and to whom) as well as the content of the obligation (i.e., what its satisfaction requires of the person).

38  Preventing Harm

When a person has a duty to protect another person, they must protect the interests of that person. However, the degree to which one must do so can vary according to the degree of dependence and, as with all obligations, the degree to which one is able to protect those interests. Thus, the person nearest the child drowning in a shallow pond bears the greatest duty to protect that child, because the child is most vulnerable to that person’s actions. Children are highly vulnerable, and more so than they are to others, upon their parents. So, parents’ duty to protect their children is strong, perhaps strongest of all special obligations. Elderly parents are highly vulnerable to their adult children, especially emotionally so. Adult children are therefore obligated to protect them. Some of the interests of passengers on a ship are vulnerable to the captain, who therefore has a duty to protect those interests.

Duty to Protect in Nonidentity Cases One type of response to the nonidentity problem is to provide a noncomparable account of harm, such that it becomes possible to deny the premise that the “wrong” decision harms the future person by making them worse off than they otherwise would be. A threshold account of harm helps to resist the nonidentity problem. On such an account, harm is acting in such a way that a person falls below some defined threshold (Hanser, 1990; Shiffrin, 1999; Meyer, 2003; Rivera-López, 2009). Whether a person has been harmed does not depend, on such an account, on how the person otherwise might have been. In other words, whether an act harms another person is relative to some threshold line rather than some other way they might have been. In this way, a threshold account can deny the premise that if a person A harms person B, A makes B worse off than she otherwise would have been. Thus, a threshold account can preserve the notion that in nonidentity cases like depletion, the current generation harms the future generation, but avoid the problematic conclusion of the nonidentity problem that they don’t. Current generations have a duty to non-maleficence toward future generations, and this duty is to not cause them to fall below some threshold of well-being. Threshold accounts of harm, as solutions to the nonidentity problem, face significant challenges, most of them related to working out what constitutes the defined threshold. While it might seem reasonable to think that people who die young or have debilitating diseases are below some threshold for well-being, there are plenty of ways one person can harm another without causing them to follow some below the threshold. For example, suppose the threshold is set at a person’s basic needs. Then if A takes $10 from B, this would count as a harm, as in most circumstances losing $10 doesn’t make a difference as to whether basic needs are met. But intuitively stealing money from someone is harmful, even if it’s only $10. Some authors offer sophisticated accounts of how to settle this problem (Crisp et al., 2003; Benbaji,

Preventing Harm 39

2006). Once the threshold is defined, there is a further matter of how to distribute resources according to the threshold. This matter is particularly important instances of policy decisions and how they affect future generations. One promising account, sufficientarianism, is that resource priority should be determined by the threshold—priority is given to those below it according to the distance between their well-being and the threshold. Once their well-being reaches their threshold, they are not owed any resources. Threshold accounts of harm won’t help in nonidentity cases of outcome uncertainty, because the wrong action toward future generations in such cases is not a matter of harming them, but of failing to protect them. The duty to protect is a special obligation that holds when one person’s interests are vulnerable to the other’s actions/omissions. The degree of vulnerability, which can further depend on proximity in space and time between the two people, establishes the strength of the duty to protect. The duty is violated when the protector fails to shield the protectee from an unacceptably high risk of suffering. The protectee doesn’t have to actually suffer in any way for the duty to be violated. The parent storing bleach and knives in the lower cabinet still violates their duty to protect even if the kid doesn’t ever get into them. By neglecting to make sure there are enough lifeboats, the ship’s captain violates the duty to protect even if the ship makes it safely to port. However, the child and the passengers are in no way harmed—they are not worse off at all, relative to however they might otherwise have been or to some threshold of well-being. And because they haven’t suffered at all, there is no cause of suffering. While the protectees may not have been harmed, they have been wronged. Presumably, anytime a person’s claim is violated they have been wronged, regardless of whether they believe they have been wronged or they are aware that they are not getting that to which they are entitled. If it is correct that the child is wronged even if they don’t get into the knives and bleach and the passengers are wronged even if they make it to port, then a person can be wronged without being harmed. For the nonidentity argument to work, wrongness must be linked with harm. This strategy breaks this link. Accounting for the intuitions that the subjects’ actions are wrong is also straightforward. A person’s immediate future children are entitled to protection because their interests are vulnerable to their actions/omissions. In the standard nonidentity cases in which the outcome of the subjects’ decisions is certain, the people affected by it are vulnerable to the subjects. Lisa is vulnerable to Heather’s decisions regarding reproduction. In all the “Parfitian” cases, the people affected are vulnerable to the subjects. This is true even of Parfit’s “Risky Policy” case, in which the decisions of one population affect the interests of people living hundreds of years later. In all of these cases, the interests of those affected are vulnerable to the subjects. Thus, if vulnerability between two people establishes one’s duty to protect the other, then in these cases the subjects have a duty to protect the people affected.

40  Preventing Harm

When they act/omit, they fail to protect the people they are obligated to protect. What grounds the judgment that what the subjects do is wrong is their violation of this duty to protect those who are vulnerable to them. That the people are additionally harmed may add to the wrongness, but it doesn’t ground it. The duty to protect can account for the judgments in the standard nonidentity cases in which the outcome is certain. But it can also account for judgments in the cases in which the outcomes are less certain, such as the cases in which there is an even chance of an undesirable outcome, but that it never realizes. In those cases attributing the wrongness to harm-based reasons won’t work, because there was no harm and the resulting person doesn’t suffer at all for the decision. However, they were still vulnerable to the subjects’ decisions. And even though nothing undesirable happened, they were still wronged. They were wronged because they were entitled to protection, yet didn’t receive it. Grounding the duty to protect in vulnerability can account for why subjects in nonidentity cases are wrong, regardless of whether the people affected by those decisions are harmed or suffer or their interests are undermined in any way. Other strategies to resist the nonidentity problem that attribute the wrongness of the subjects to the harm that occurs cannot account for those cases in which there is no harm. Thus, while the strategy appealing to the duty to protect can explain why the subjects in all kinds of nonidentity cases are wrong, the strategy of appealing to harm-based reasons can only account for those cases in which the bad outcome occurs. However, while it is possible to identify cases that the harm-based strategy cannot account for, it is not possible to identify any cases that the duty to protect strategy cannot account for. If one person having the duty to protect another is a matter of the vulnerability relation between the two people, then any nonidentity case will involve a duty to protect. Nonidentity cases arise when a person’s actions and decisions affect the identity and interests of others. But these dependence conditions are just vulnerability conditions. Thus, it isn’t possible to show there are some nonidentity cases in which there is no duty to protect because to do so it must be possible to identify nonidentity cases in which there is no vulnerability relation. But no such cases are possible, because nonidentity cases necessarily incorporate one person’s vulnerability to another. The duty to protect arises out of vulnerability. While vulnerability is a relation, it is not necessary to identify both parties to it in order to say that it exists. That is, future people, whoever they happen to be, are vulnerable to the current decisions that current people make. We may not be able to identify now the people who will exist in the future, but for whoever they happen to be their vulnerability is current. The lack of identifiable protectee is not generally a problem for establishing the duty to protect, because it is triggered by vulnerability. Consider the ship’s captain who is checking the

Preventing Harm 41

engine and provisions before any passengers embark. It may not be possible for the captain or anyone else to identify who exactly is owed protection, but whoever they happen to be are vulnerable to her decisions as she is checking, or not, the seaworthiness of the ship. The duty to protect is triggered well before there is an identifiable protectee because the vulnerability is contemporary. There may be no one who is entitled to protection, but the absence of an identifiable protectee doesn’t mean that the ship’s captain lacks a duty to protect. What, then, creates the vulnerability when there is no identifiable protectee? The same thing as when there is an identifiable protectee: the causal dependence of some person’s welfare upon another. The “some person” here can refer to any person whose welfare may be causally dependent on the actions/omissions of another, the degree of this dependence corresponding to the strength of the duty to protect. And since future peoples’ welfare is causally dependent on the actions/omissions of current people, they can be vulnerable to us, even though we may not be able to say exactly who those people are. Future persons are highly vulnerable to our current behavior in response to collective risks. We therefore have an obligation to protect them. Doing nothing in response to these collective risks violates this duty. If it is in our power to do something that prevents catastrophe, we ought to prevent catastrophe. In what follows, I argue that the method of prevention—the way we satisfy our duty to protect future generations—is by enhancing our ability to respond to collective risk.

Chapter 3

An Epistemic Argument for Compulsory Moral Bioenhancement

In this book, I make the case that moral bioenhancement should be compulsory and covertly administered. Most of what follows establish that it should be administered covertly rather than overtly. Persson and Savulescu (2014b) argue on anthropological grounds that without compulsory moral enhancement the future of humanity is quite bleak. We continue to threaten ourselves with scientific and technological developments of which we collectively are incompetent stewards. But the greatest threat is perhaps climate change. Persson and Savulescu argue that our moral psychology, as it has evolved with its biases and aversions, is not equipped to prevent the widespread suffering that is likely to result from our poor stewardship. But while Persson and Savulescu may be right that the anthropological deck is stacked against a successful response to the various existential threats that contemporary humans face, the situation is rather worse. In this chapter, I offer a new argument that moral bioenhancement ought to be compulsory. But my argument has little to do with moral psychology. Instead, I argue the epistemology of decision-making suggests that it is very improbable that we can prevent the catastrophic and the subsequent suffering without enhancement. For us to prevent this catastrophe, there are epistemic obstacles that we must overcome, both as individuals and collectively. But these obstacles are insurmountable without cognitive or moral enhancement or both. My argument rests on one notion and a series of observations. The observations are those of how people play the public goods game, which is a game used to test social preferences and behaviors in the context of collective risk and benefit (van de Kragt et al., 1983; Palfrey & Rosenthal, 1984; Dawes et al., 1986). The observations do not engender optimism about our ability to respond to collective risks like climate change. The notion is that of an epistemic burden, which is an idea initially developed by my colleague, Scott Scheall, that we have discussed in a number of articles (Crutchfield & Scheall, 2019; Scheall & Crutchfield, 2020). An epistemic burden is simply that which one does not know but must, in order to achieve some goal.

An Epistemic Argument 43

Epistemic burdens function in decision-making in ways that make successfully responding to climate change, a pandemic, or any other large-scale collective action problem unlikely. There are many existential threats: weapons of mass destruction, an asteroid hitting the planet, some cosmic or quantum event, the death of the sun, and so on. The threat that seems most likely to result in significant harm to all or most, however, is climate change. The collective risk of climate change seems much greater than that of an asteroid strike or even nuclear war. Thus, in what follows I  focus on the prevention of harm from climate change. But the argument can be translated to other potential existential threats that require collective action, such as a pandemic. Even so, climate change may not be the threat it’s made out to be. And even if it is, the cause of climate change, the emission of fossil fuels, produces fantastic benefits for people around the world. If the threat is not as great as it is made out to be and if benefits could not be achieved in the absence of climate-changing levels of fossil fuel emissions, it would be considerably more difficult to argue that we should do anything about climate change. Furthermore, we may not even have an individual moral obligation to do anything about it (Sinnott-Armstrong, 2005). These factors don’t count against taking measures to mitigate the suffering associated with climate change, or more generally figuring out a way to prevent and mitigate suffering associated with other collective risks. In the case of climate change, probably the benefits that people obtain from the emission of fossil fuels can be obtained using other power sources. And in any case, fossil fuels are limited, so it’s not as though these benefits will continue in the absence of action. We may not have an individual moral obligation to not leisurely drive gas guzzlers, but that doesn’t mean that preventing suffering isn’t good (Sinnott-Armstrong, 2005). And there are of course other threats as well. A global pandemic even more lethal than the coronavirus responsible for COVID-19 may spread. A meteor might hit, wiping out most life. The sun will die one day billions of years from now. If humans, or some other group of persons, inhabit the planet at that time, they will die, most of them quite horribly. These are threats that we can imprecisely predict. There may be others that we can’t predict at all. This is all to say that even if we have reasons to not act on climate change, that doesn’t mean that we shouldn’t figure out how to respond to large-scale collective risks. The problem is that we lack the ability to so respond, given our significant epistemic limitations. In the next section, I review some of the observations others have made of how people respond to collective risks in the public goods game. In the section that follows, I  introduce the notion of an epistemic burden and explain how it functions in decision-making. I then argue that the epistemic burdens of responding to climate change are so great that in order to surmount them we either need to be biologically enhanced.

44  An Epistemic Argument

The Problem My argument relies on empirical research on how people behave in social preference games and concludes from these observations and other premises that we should boost our epistemic and moral capacities so that we can prevent significant suffering. Other authors have taken a similar strategic approach to enhancement (Anomaly, 2020; Anomaly & Jones, 2020; Bognar, 2012). Combining this approach with the notion of an epistemic burden recommends a more specific type of enhancement, one that may help humans respond to collective risks in a way that prevents or mitigates the suffering associated with various collective risks. A livable climate is a public good. A public good is a resource that exhibits non-excludability and non-rivalry. A resource is non-excludable when people can’t be excluded from its consumption. A resource is non-rival when one person’s use of it doesn’t affect another person’s use. A livable climate exhibits both non-rivalry and non-excludability because my consumption of it doesn’t affect your ability to consume it and it’s available to everyone; no one is excluded from its consumption. We could say the same thing about a planet that is free of meteor strikes or revolving around a star in a habitable zone. There are some collective risks the prevention of which is not a public good, though. A vaccine for a disease causing a global pandemic is not a public good, because people can quite easily, and necessarily, be excluded from its consumption. But herd immunity from the disease is a public good because it also exhibits non-rivalry and non-excludability. Public goods games are supposed to model social preferences and behavior in the achievement of public goods. The basic version of the game is played as follows: individual members of a group are endowed with some fund; they are given the option of contributing some portion of that fund to a pool that, after contributions are finished, will be multiplied by some factor (e.g., 0.5×, 2×, 5×); the product of that function is then redistributed equally among the players, who get to keep the proceeds from the pool and whatever is left of their private endowment. The game is useful for investigating the variables that factor into preferences and behaviors affecting groups, as the game can be changed in multiple ways to isolate variables. The public goods game is essentially an iterated prisoner’s dilemma. As such, it can measure a person or group’s disposition to cooperate in securing individual and collective advantages. Free-riding is common, and often because of its cooperation in achieving the good disintegrates. Bowles and Gintis (2011) develop a comprehensive account of the evolution of this cooperation. In reviewing the empirical observations of behaviors in social preference games, they show that humans want to cooperate to solve collective action problems. They write (p. 38): The most parsimonious and compelling proximate explanation of behavior in the ultimatum game, public goods game, and other social

An Epistemic Argument 45

dilemma experiments is that people think that cooperating is the right thing to do and enjoy doing it and that they dislike unfair treatment and enjoy punishing those who violate norms of fairness. People prefer to be a strong reciprocator or someone who will “sacrifice their own payoffs in order to cooperate with others, to reward the cooperation of others, and to punish free-riding, even when they cannot expect to gain from acting in this way” (p. 20). In public goods games, when enough people exhibit strong reciprocity, cooperation is sustained. At some threshold, enough strong reciprocity can eliminate the free-riding that would otherwise undermine the achievement of the public good. But a person is only able to behave with strong reciprocity if other conditions obtain. That is, in the absence of these other conditions, sustained cooperation to achieve the public good is unlikely. One is the opportunity for strong reciprocators to punish free riders beyond merely withdrawing their own cooperation. This punishment, in turn, sustainably induces free riders’ cooperation. That the opportunity to punish helps to sustain cooperation is something that is repeatedly observed in public goods games and under a variety of conditions (Dawes et al., 1986; Ostrom et  al., 1992; Fehr  & Gächter, 2000, 2002; Masclet et  al., 2003; Chaudhuri, 2011). Importantly, strong reciprocators will punish even when it is costly for themselves (Fehr & Gächter, 2000; Falk & Fischbacher, 2006; Fudenberg & Pathak, 2010). Thus, this punishment is altruistic, in that the strong reciprocators sacrifice their own earnings so that they can punish others for not contributing. Punishment in collective action only works if it induces free riders to contribute. Bowles and Gintis’ account is that what bridges punishment and cooperation is a person’s sense of shame. That is, shame turns punishment into sustained cooperation. Punishment plus the emotion of shame sends the free rider a message: “whatever you did, undo it if possible, and do not it again” (p. 192). Thus, in populations where the opportunity for altruistic punishment and shame are common, punishment will be particularly effective, thus used infrequently. But punishing free riders with shame-inducing opprobrium is also much less costly than punishing them with violence or withdrawal of cooperation, which positions that population to be more successful in repelling other threats. Thus, Bowles and Gintis suggest that altruistic punishment and shame contribute to a population’s evolutionary fitness. Whether punishment and the emotional response to it can sustain cooperation depend on a more fundamental condition: that the informational signals between individuals are reliable. They identify two dimensions along which informational signals may vary: they may be private or public and they may be perfect or imperfect. Most empirical work presumes that they are public and perfect—all players have all accurate information about other players and their contributions. But of course it is rarely the case the

46  An Epistemic Argument

flow of information among players is so good. Usually, the information is not entirely accurate, leading to other players having false beliefs, or if it is private, leading to ignorance throughout the group. Using their models, Bowles and Gintis show that in larger groups (i.e., up to 14 members) sustaining cooperation drops off dramatically as information deviates from the ideal of public and perfect (p. 66–67). Thus, ignorance and falsity of beliefs are significant epistemic obstacles—epistemic burdens to overcome—to sustain cooperation. They write (p. 187): If one contributes little and is not punished, one comes to consider these acts as unshameful. If, by contrast, one is punished when one has contributed generously, the emotional reaction may be spite toward the members of one’s group. This is one of the reasons why the “antisocial” punishment of high contributors in public goods experiments has such deleterious effects on the level of cooperation in a group. Having private or imperfect signals—having high epistemic burdens— makes it more likely that punishment will be inappropriately withheld from a non-contributor, which means they won’t feel shame and remediate their behavior. High epistemic burdens or poor-quality information also make it more likely that a generous contributor will be inappropriately punished, causing them to withdraw cooperation. Punishment and shame might be critical to sustaining cooperation in achieving a public good, but they can only work if the players have overcome the epistemic burdens associated with that cooperation and the information that flows among them is neither private nor significantly imperfect. Cooperation deteriorates when there is significant ignorance among the players, even, or especially, when the opportunity to punish and elicit shame facilitates cooperation. However, responding to threats like climate change and pandemic and asteroids requires averting the loss of something that the group already has, rather than securing some benefit that it doesn’t. And humans are biased toward loss aversion (Tversky & Kahneman, 1981). Thus, it could be that although cooperation in acquiring a public good requires punishment, shame, and good information to keep it altogether, because of our bias toward loss aversion, the requirements for success are looser when the public good is keeping what players already have. When the public goods game is used to investigate responses to the pressure of climate change or other similar collective risks, often a contribution threshold is introduced into the game. The threshold is set such that the public good is realized only if the group meets the threshold. For example, Milinski et al. (2006) used a game in which if the players failed to meet a threshold, not only was the public pool eliminated but so was what was left of the individuals’ private endowment. This is supposed to be a good model for climate change because we all stand to lose something, possibly

An Epistemic Argument 47

everything, from the effects of climate change. Imposing a threshold upon the players is similar to the thresholds that scientists frequently cite, such as the 2°C temperature threshold that, if met, portends catastrophe. But there may be other climate thresholds that, if met, represent an irreversible change in the environment that makes catastrophe highly likely. The results from these studies are not promising. If using the public goods game is a valid model, then we ought to be worried about our ability to respond to climate change in a way that prevents catastrophic suffering. Milinksi et al. studied distinct groups of six people. Each individual was initially endowed with €40. Each group then played ten rounds. In each round, an individual could choose to contribute €0, €2, or €4 to the public pool. If after ten rounds the public pool met a threshold of €120 the pool was divided evenly among the group and each individual got to keep both their cut as well as the remainder of their private endowment. However, if after ten rounds the public pool was less than €120, then investigators took not only the public pool but also each individual’s private endowment at various levels of chance. There were three conditions, with groups losing their private and public endowments at 10%, 50%, and 90% chances. For example, for the 90% condition, those groups that failed to meet the €120 threshold stood a 90% chance of losing both their public and private endowments. When the groups knew that there was a 90% chance of losing everything if they didn’t meet the threshold, still only 50% of the groups met the threshold. In other words, when they knew that there was an excellent chance of losing everything, and they had the means to prevent that loss, half of the groups failed to do so. The groups in the other conditions were even worse. For those groups who stood a 50% chance of losing everything by failing to meet the threshold, 85% of them failed. And more than 90% of the groups in the 10% condition failed to meet the threshold. These observations do not support any sort of optimism about our ability to respond to the pressures of climate change. Players had several advantages that the public does not. First, they could keep a running tally of how close they were to meeting the threshold. With respect to contributions to prevent the suffering from climate change, the public lacks that information. There is no transparency about who is doing what, especially at the level of the individual. Second, there were only six players, a setup which greatly simplifies collective action. In the case of responding to the pressures of climate change, there are billions of players, most of whom we cannot identify. Third, the players were certain of the odds of loss. But the people who are in a position to help prevent or mitigate the suffering that results from climate change—Americans and others in the developed nations that most contribute to climate change—are uncertain of the odds of loss. Some still think that climate change isn’t even a real phenomenon, while others maintain that it is not anthropogenic and therefore there’s nothing to be done about it. Even with these advantages, players in the game only reached

48  An Epistemic Argument

the threshold 50% of the time when they were 90% confident that failure would result in a total loss. Tavoni et  al. (2011) found similar results. Approximately 50% of the groups tested met the threshold. But groups in which coordination and communication were facilitated—subjects were permitted to “pledge” the amounts they planned on contributing—were more successful in meeting the threshold. Other actions increase success. Hauser et  al. (2014) found that when voting on how much to contribute to a public pool, the proceeds of which would go to future generations, implementing voting sustained the pool for future generations. That is, allowing the players to vote on an enforceable rule about how much should be contributed to sustain future generations significantly increased the overall contributions. Indeed, using voting to implement an equally enforceable contribution policy resulted in sustaining all future generations. Indeed, it is generally true (beyond threshold versions of the public goods game) that cooperation is better sustained when players remedy ignorance about the other players, what they are likely to contribute, and of likely outcomes of behaviors. Money-back guarantees and insurance against loss increase a person’s contributions (Croson  & Marks, 2000; Zhang et  al., 2015). Allowing or facilitating the players to punish each other helps to sustain contributions both when the punishment is monetary as well as it when it is nonmonetary, such as when players are permitted to expel others from the group (Fehr & Gächter, 2000; Masclet et al., 2003; Chaudhuri, 2011; Burton-Chellew et al., 2016). Communication both before and during play helps to sustain contributions. Knowing what others have contributed helps to sustain contributions (Pereda et al., 2019), but so does knowing the distribution of conditional cooperators (some of which are like strong reciprocators). That is, when groups know that every player in the group is a conditional cooperator, they contribute more and better sustain the public good than when they are ignorant of the fact that they are all conditional cooperators. This indicates that knowledge of how people are distributed also matters (Chaudhuri & Paichayontvijit, 2006; Chaudhuri, 2011). There is much to be said about using the public goods game to model responses to the pressures of climate change. The point important for the present purpose is that increasing the players’ knowledge (or repairing their ignorance) increases contributions, not only when the players themselves earn the payoff but also when future generations earn it. The study from Milinksi et  al. may be idealized, but the players in that study still lacked information about what other players were going to do. And as confidence in outcome decreases, so do contributions. Compare this behavior to that in the study from Tavoni et  al., where players contributed more when others told them their plans. The pledges were nonbinding and so there was still some uncertainty that others would do what they said they would do. But greater epistemic access to others’ contributions yielded

An Epistemic Argument 49

greater contributions. Finally, consider the study from Hauser et al., where an equally enforceable policy of contribution yielded success in sustaining the public good for future generations. In that study, players knew exactly what others would contribute—the enforceable policy provided epistemic access to what others were going to contribute. Conditional cooperators, those whose contributions are conditional on others’ behavior, contributed more when they knew their contributions wouldn’t be exploited (Hauser et al., 2014; Burton-Chellew et al., 2016). By repairing players’ ignorance about what is likely to happen and what others are likely to do, contributions to the public good increase. But these behaviors support pessimism for our ability to respond to climate change or other large-scale collective risks. If participating in collective action in ways that result in the reduction of collective risk is at least partly a matter of our epistemic position, then we should not be hopeful, as our epistemic position is not likely to improve. At a minimum, we would need to overcome things about which most of us are profoundly ignorant. First, we have no idea how most others behave in response to the pressures of climate change. We can see our friends and neighbors drive gasoline-powered vehicles and conclude that they are not contributing everything that they can contribute. But we can’t conclude that they are free-riding in the same way that players in the public goods game observe their fellow players contribute nothing. That is, outside of experimental economics, we don’t really know how other individuals are contributing. If knowing how others contribute increases one’s own contributions, then we should not expect many people to contribute significant resources into the public pool, certainly not enough to prevent or mitigate the suffering that results from climate change. Second, we don’t know which of our own actions count as contributions. Many know that they should drive less, fly less, bike and walk more, be warmer in the summer and cooler in the winter, reduce, reuse, recycle. It is also becoming more commonplace for people to recognize the devastating environmental effects of eating meat, especially beef and pork. But it is difficult to know what of our other actions contribute to climate change. Does buying produce from a chain grocery store that purchases it in bulk from a national wholesaler who buys from large farms using Roundup and other herbicides contribute to climate change? It might. But then to contribute to the public good the person needs to know how to pursue some alternative, which is difficult and may even require driving longer distances to find more local resources? Which alternative is better? My point is that it is very difficult for the person making such a decision to know. And if knowledge is required to contribute to the public good, we shouldn’t hold out much hope that the person will do anything other than ride free on others’ contributions. Cooperation is undermined by imperfect and private signals, by ignorance. My point is that this is the status quo in ordinary settings of cooperation aimed at averting or preventing collective risks.

50  An Epistemic Argument

Epistemic Burdens Results from the public goods games suggest that to successfully contribute to the public good one must repair a great deal of ignorance. This is to say that in order to purposefully contribute to the public good in a way that suffering is averted or that the good is sustained, there is much one must know but doesn’t; there are many epistemic burdens one must overcome. Scott Scheall (2019) defines the epistemic burden of some objective as “simply everything that the actor must know (that and how), which the actor does not already know, in order to realize the objective deliberately as a result of related actions directed to its realization, i.e., not spontaneously, or otherwise in virtue of luck or fortune.” Epistemic burdens thus encompass both propositional and nonpropositional knowledge (or, more exactly, ignorance). Epistemic burdens preconsciously shape our incentive structures or our preference rankings. Courses of action that seem to bear impossibly heavy epistemic burdens are typically not counted as options in the menu over which an actor deliberates and from which she eventually makes a conscious choice, while courses of action that seem to bear comparatively heavy epistemic burdens are systematically discounted in the actor’s menu relative to options that appear less epistemically burdensome. Ignorance shapes the psychology of decision-making. Furthermore, that ignorance shapes the psychology of decision-making in the way that Scheall and I posit means that epistemic considerations are logically prior to moral, prudential, and pecuniary considerations. In order to have moral, prudential, or pecuniary obligations, one must have options, and epistemic burdens—the nature and extent of one’s ignorance (that and how) with respect to various courses of action—both bracket one’s options and serve to determine where surviving options are ranked. In short, one cannot have obligations without doing some epistemic work, even if, as we posit, such work typically occurs preconsciously. We have used the concept of an epistemic burden and how it shapes decision-making to account for the moral authority of surrogate decision-making (2020, 2019) and to argue that epistemic considerations are logically prior to all other types of normative considerations (2020). For example, in the late 1980s and early 1990s, Bo Jackson was one of the premier athletes in the world, excelling as both an especially agile Major League Baseball centerfielder and the leading running back in the National Football League. Faced with a ten-foot wall blocking his path, Bo knew how to scale it. Confronted with a ten-foot wall, one of the options available to Bo was to climb over it and move on to the next athletic challenge or Nike commercial shoot. For most of us, however, climbing such a wall with only our hands, legs, and brute strength would not appear as an option. Climbing the wall would appear in Bo’s menu of options, but not in most of our menus of options, for example. More generally, climbing the wall

An Epistemic Argument 51

would be ranked in a different position in Bo’s menu of options than in the menus of anyone who believed themselves insufficiently athletic to scale it. Differences in the rank placement of this option across distinct individuals, we contend, are due to the different epistemic burdens that various individuals take themselves to face in pursuing this option (Scheall & Crutchfield, 2020). Here’s another example: Imagine you are visiting New York City. It’s your first time there and you are planning to see an old friend. She tells you to pick up some wine on the way to her apartment, something she’ll like. You are faced with two problems: knowing what she likes and knowing how to get it. If you are ignorant of her tastes, then arriving at her apartment with a bottle of wine she likes will be entirely accidental, a lucky guess. But you can easily overcome this ignorance: just ask her what kind of wine she likes. Ignorance of how to get the right wine is not so easily overcome. You know that the gift shop in the hotel lobby sells wine. But the selection is small and may not include what she likes. The alternative is to seek out a shop in the city with a wider variety. But this requires locating a wine shop in an unfamiliar city. Then you will have to find the shop from the hotel and your friend’s apartment from the wine shop. Faced with these additional burdens, you will probably just buy a bottle in the lobby and hope that it’s something she likes. Now imagine instead that you are familiar with New York, so familiar that you have a favorite wine shop. Likely you will go to the wine shop. Epistemic burdens shape our decision-making in two ways. First, they determine what initially appears on the menu of options. Second, they determine how the different menu options are ranked. In the next two sections, I argue that both of these avenues of influence have important implications for decisions regarding contributions to the public good, specifically the public good of preventing or mitigating the suffering that results from climate change.

What’s on the Menu The behaviors observed in the public goods games suggest that people need to overcome epistemic burdens to contribute to the public good. But these observations aren’t necessary to establish this point, as it is prima facie obvious that in order to perform some action deliberately, that action must appear as an option to them. The epistemic burdens associated with contributing to the mitigation or prevention of suffering that results from climate change are so significant that they constrain what will appear on a person’s menu of options. In other words, a person won’t be able to contribute to the mitigation or prevention of suffering that results from climate change if the options they need to pursue to achieve this goal don’t appear on the menu of

52  An Epistemic Argument

options. And those options that are overly epistemically burdensome won’t appear on the menu. The question is: are the options that one needs to choose in order to contribute to the public good in the class of options that will be absent from one’s menu of options? It is certainly true that there are some things about which individuals are so ignorant that options requiring such knowledge won’t appear on an individual’s menus of options. For many people, they may be so ignorant of the effects of eating meat and of possible alternatives that not eating meat doesn’t appear on their menu of options. It will never be an option for them to not eat meat because they are so ignorant of how to procure other foods and prepare them in ways that are both nutritious and satisfying (or just satisfying). The same could be said of many other behavior changes that must occur in order to mitigate or prevent the suffering that results from climate change. The “unknown unknowns” don’t appear on the menu. However, the case that the epistemic burdens associated with mitigating or preventing the suffering that results from climate change are insurmountable does not rest on how epistemic burdens shape what appears on the menu. Instead, the case rests on the fact that of those options that do appear on the menu, the epistemic burdens associated with pursuing those options are insurmountable, implying that such burdened individuals will not pursue them. So, while it’s true that for very many individuals the options one would need to pursue the public good don’t appear on their menus of options, it’s for the present purpose irrelevant. What is relevant is that of those that do appear, they are overly burdensome to appear high in a person’s preference ranking.

The Epistemic Burdens of Available Options Suppose that the people who are collectively in a position to contribute to the public good in such a way that their contributions prevent or mitigate the suffering that results from climate change have on their menu of options all of the things they could choose to do. So, for examples, they know that not eating meat is an alternative, that they could drive less and bike more, that they could differently cool and heat their homes, that they must stop buying so many products that are manufactured, shipped, and disposed of in ways that deplete natural resources. These options appear on the menu; they are not the unknown unknowns. However, even people who know that they need to behave in certain ways are deeply ignorant, and this ignorance undermines collective action to achieve the public good. That’s because they are ignorant of how to pursue these options, and the burdens of acquiring this know-how are so great that the options, as compared to the option of not changing behavior, are very low in the preference ranking. A person may know that they need to stop eating

An Epistemic Argument 53

meat, but not know how to go about feeding their family on plants alone. Or they may know that they need to drive less, but not know how to get a bus or bike safely on busy roads. But to change their behavior, they need to know how to do these things. Now consider how burdensome acquiring this knowledge is. I drive to work every day, about six miles one way. There are alternatives. I could bike, as I have done in the past. But, for now, at least, it is often snowy and icy, which would require that I develop the further know-how of safely biking on icy roads. There are also buses, but they are infrequent and to get where I want to go would require several transfers. I would need to know where the stops are, how long of a wait it will be at each one, and when they come and go so that I can plan on my kids getting to school. For me to change my driving behavior, I would need to know how to do something else, but the burdens of acquiring this knowledge are significant. The same is true of everyone else who behaves in ways that not only fail to contribute to the public good, but extract from it. The epistemic burdens of doing the things I would need to do to contribute to the public good are so great that they will never appear very high in my preference ranking. In principle, these epistemic burdens are not insurmountable. In some sense of could, I could learn how to ride a bike in snowy and icy conditions. I could learn how to take public transportation to work while also getting my kids to school. I’ve learned how to not eat meat. But we argue that epistemic burdens determine what one will decide to do when other things are equal between the available options. Of course, almost always things aren’t equal, and they aren’t in the case of behaving in ways that contribute to the public good. Other values weigh for and against the epistemic burdens of the options on the menu. The epistemic burdens of not driving to work are significant. For that option to be high in my preference ranking, the practical or moral value of that option would have to be sufficiently high to outweigh those burdens. Compare it to the alternative, which is associated with very few epistemic burdens: aside from the burden of checking the traffic a few seconds before I leave, I know very well how to get my kids to school and myself to work safely and timely. Driving is not only the least epistemically burdensome, but it also appears in my menu as having the much greater moral and practical value. Driving achieves several goals that not driving would not. I ensure that my children receive a predictable, quality education; I spend more time at work interacting with my colleagues, whom I enjoy; and because I do these things in a timely manner I spend more time with my family in the ways that I want to spend that time. These are of course privileges of my social position and many people don’t enjoy these. But the people who don’t enjoy them are also not the primary contributors to climate change. I and those like me (i.e., most of those of us in highly developed nations) are so responsible. My point is simply that

54  An Epistemic Argument

to prevent or mitigate the suffering that results from climate change, I and everyone like me would need to change behavior. So, my available options are to not change my behavior, which is associated with few epistemic burdens and high practical and moral value, or I could change my behavior, which is associated with significant epistemic burdens. Is the value associated with contributing to the public good by not driving high enough to move that option higher in my preference ranking? In one sense, the answer to this question is most assuredly “Yes.” The value of preventing or mitigating the suffering that results from climate change totally swamps any value I might derive from my work or my family. There is no comparison. But in another sense, the answer is, “No.” First, the value is not clearly going to be achieved, because there are so many free riders. So the expectation that it will be achieved is very low. Second, the value is distant, abstract, and cognitively unavailable. Deciding to not drive trades a certain epistemically easy high-value choice for an epistemically burdensome uncertain high-value choice. At least one of two things would need to happen for me to choose to change my behavior in a way that contributes to the public good. Relative to my other options, such as continuing to drive, either the subjective value of contributing to the public good would need to go up, or the epistemic burdens of contributing to the public good would need to go down. That option will move up in my preference ranking only if the ratio of epistemic burdens to subjective value is lower, as compared to the option of continuing to drive. Not only must the ratio of epistemic burden to subjective value be lower than other options, it must be lower for everyone. In order to prevent or mitigate the suffering that results from climate change, it must be true that for everyone, or at least for the majority of people who can democratically constrain the behavior of the minority, the ratio of epistemic burden to subjective value is lower than other alternative options. Thus, contributing to the public good requires either a widespread drastic reduction in epistemic burdens associated with such behavior or a widespread drastic increase in subjective value of behaving in ways that contribute to the public good.

From Epistemic Burdens to Enhancement Contributing to the public good requires either a widespread drastic reduction in epistemic burdens associated with such behavior or a widespread drastic increase in subjective value of behaving in ways that contribute to the public good. The evidence from the public goods games suggests that if these epistemic burdens can be overcome, then the goal of such contributions— preventing or mitigating the suffering that results from climate change—can be achieved. But we should be pessimistic that the epistemic burdens can be surmounted through traditional means of surmounting epistemic burdens.

An Epistemic Argument 55

A person will choose to surmount them in the first place only if so choosing is easier than the alternative options, presuming that the value of achieving the different goals are approximately equal. This means that the epistemic burdens of contributing to the public good have to be lowered prior to a person choosing to overcome them, otherwise the person is unlikely to choose that option in the first place. Instead, they will choose the much epistemically easier option of continuing to live how they have been living. It’s not that it is simply a matter of learning how to respond. To learn to contribute to the public good one must make the choice to surmount the epistemic burdens associated with doing so. And as long as the epistemic burdens associated with that choice are so high, the person simply will choose an alternative. So, it’s not simply that the burdens need to be overcome. That will only occur after it has become easier to do so. The epistemic burdens need to be less burdensome for many people to be in a position to choose to contribute to the public good. Above I noted that Bo Jackson in his heyday could scale a ten-foot wall with ease. I, like most normal people, have never been able to do that. Faced with the burden of scaling a ten-foot wall and the goal of getting over it, I have two options: I can either make the wall shorter or increase my ability. If I achieve either one, the burden of getting over the wall is lower. Lacking any tools or other materials, I’d better start doing some calisthenics. The same is true of any epistemic burdens a person may face. The degree of difficulty in surmounting an epistemic burden is a matter of the objective features of the obstacle relative to one’s ability to overcome it. The degree of difficulty of the epistemic burdens associated with contributing to the public good must be lower. Sometimes the degree of difficulty of overcoming epistemic burdens can be lowered by manipulating the objective features of the obstacle. Before the internet was widely available the epistemic burdens associated with doit-yourself home improvement were much greater. Now, if a person wants to install tile, fix a toilet, add a room, and so on, it is much easier to find instructions for doing so. Prior to the internet, the epistemic burdens were much greater, requiring knowledge of where to find that information and how to get it (e.g., know how to get to the library, look up in the card catalog a book’s location, find the book on the shelves, etc.). Instruction manuals of all kinds help to reduce epistemic burdens by increasing access to know-how. But increasing access to know-how in the case of contributing to the public good is not likely to significantly improve the chances of preventing or mitigating the suffering that results from climate change, because the biggest epistemic burden is knowing how to integrate these changed behaviors into a day-to-day living. Providing an instruction manual may decrease the epistemic burden of knowing what steps to take to eat no meat or to catch a bus. But it’s difficult to conceive of an informational shortcut to integrating

56  An Epistemic Argument

these behaviors into one’s life. Even if there were an informational shortcut to integrating these behaviors into one’s life, it remains that enough people do this such that (a) free riders who encourage further defection are minimized and (b) that the public good is achieved. This is to say that everyone or something close to it must integrate these behaviors into their life. It seems implausible that so many individuals will overcome their own epistemic burdens associated with contributing to the public good. Furthermore, this assessment ignores the epistemic burdens associated with collective action. For example, there are significant epistemic burdens associated with coming to know how likely catastrophic climate change is. And the evidence from public goods games indicates that this matters to the success a collection of individuals has in reaching a threshold goal (Milinksi et al., 2008). Something similar could be said about the epistemic burdens associated with knowing what others are likely to contribute. Knowing what others are likely to contribute to the public good affects one’s own contributions. Given that most of our behaviors are private, or at least not in full view of the other players, the only feasible way to overcome the epistemic burdens associated with collective action is by electing leaders who are able and willing to pass legislation that enforces contributions equally across the population, even for those people who would otherwise free ride, as the evidence from the public goods games suggests may in principle work. Such legislation would not lower the individual epistemic burdens significantly. People would still need to surmount the epistemic burdens associated with, for example, only taking public transportation. But this route to securing the public good requires overcoming an additional epistemic burden. The elected representatives must know how to pass this legislation. The epistemic burden of knowing how to pass such legislation is great. For it to be effective, the policy would have to curtail individual liberties in important ways. For example, it would have to place limits upon, or entirely prohibit, eating meat, flying, driving, and using fossil fuels to heat, cool, and power our homes and industries. The policy would be one of the most liberty-restricting policies that, at least in the United States, has ever been passed into law. Policymakers would not only have to know how to write such a policy but also how to manage the politics required for its passing. There is absolutely no reason to believe that contemporary policymakers are capable of acquiring this know-how. It seems implausible that the epistemic burdens that individuals or policymakers face can be reduced to the point that the ratio of epistemic burdens to subjective value of acting collectively tips the balance in favor of collective action. Either the epistemic burdens need to be easier to surmount or the subjective value of overcoming them needs to be higher than it is. Thus, there are two further options that might tip the balance in favor of collective action. Since lowering the burdens themselves will not work, one option is to increase individuals’ abilities to overcome the epistemic burdens; the

An Epistemic Argument 57

other is to increase the subjective value of collective action to secure the public good. Both options require bioenhancement. The argument that to improve individuals’ abilities to overcome the epistemic burdens associated with collective action to secure the public good requires bioenhancement is straightforward. The abilities to overcome epistemic burdens are generally abilities related to cognition and practical reasoning, like the abilities related to overcoming ten-foot walls are generally related to abilities to move the body in certain ways. The traditional means of enhancing cognition and practical reasoning is by education. The enhancement must occur across a large population. Using education to enhance abilities across a large population takes a lot of time, potentially decades, generations. Collective action on preventing or mitigating the suffering that results from climate change must happen sooner than enhancing cognition and practical reasoning across a large population allows. Thus, nontraditional means of enhancing cognition and practical reasoning must be pursued. The most feasible nontraditional means of enhancing cognition and practical reasoning is bioenhancement. Thus, to improve individuals’ abilities to overcome the epistemic burdens associated with collective action individuals’ cognition and practical reasoning ought to be biologically enhanced. I am claiming that people will not choose to overcome them unless either they have enhanced abilities to do so or the subjective value of overcoming them is greater than it is. The argument in the previous paragraph concludes that if abilities to overcome the epistemic burdens are to be enhanced, then that enhancement ought to be bioenhancement. But the alternative is that people may choose to overcome the epistemic burdens associated with collective action if the subjective value is great enough. We should be skeptical, however, that the subjective value of contributing to the public good will increase to the point that it makes choosing to overcome the associated epistemic burdens the choice that is highest in a person’s preference ranking. Earlier I mention that alternative choices also have high subjective value but comparatively lower epistemic burdens. Thus, the increase in subjective value would need to be significant for the choice to contribute to the public good to rise above the alternative of not doing contributing to the public good. Persson and Savulescu make the case for why we haven’t evolved in a way that makes it possible to take collective action in the way that we would need to avert the suffering from climate change (or, for them, all the other ways in which we may be hoisted on our own petard). Essentially, we are too narrowly focused on our own social groups, biased away from distant threats, and not sufficiently altruistic. I endorse their argument. An implication of their position, and the additional premise that the subjective value of collective action must significantly increase in order to outweigh the significant epistemic burdens of that choice, is that so increasing it

58  An Epistemic Argument

requires bioenhancement. While biologically enhancing a person’s abilities in cognition and practical reasoning requires cognitive enhancement, using biological enhancement to increase the subjective value of contributing to the public good is a moral bioenhancement, because the values that such action promotes are moral values. That is, the values that individuals must promote are those associated with reducing widespread harm and suffering and promoting utility, liberty, equality, and pleasure, to name a few. Though it may be convenient to bifurcate bioenhancement into enhancements that improve cognition and practical reasoning and those that improve sensitivity to moral values, or to suggest that they are distinct interventions, it may be impossible to disentangle them. It may not be possible to enhance cognition without simultaneously enhancing moral reasoning. Cognitive enhancement may imply moral enhancement (Harris, 2011; Rakić, 2014a, 2014b; Persson & Savulescu, 2008; Douglas, 2013)) because a component of moral behavior is moral reasoning and cognitive enhancement would improve it. Similarly, practical reasoning involves consideration of moral values, so enhancing practical reasoning may imply moral enhancement. What is really needed is an enhancement that causes a person to be more sensitive to and motivated by the moral values present in consideration of collectively acting to contribute to the public good and an enhancement that causes the person to be better at reasoning practically about them. Anomaly and Jones (2020) argue that enhancing intelligence also boosts a person or group’s ability to be more cooperative. If this is true, then enhancing intelligence just is moral enhancement, insofar as the relevant moral capacities that need to be enhanced are those associated with cooperation. We may not be able to disentangle cognitive and moral bioenhancement, which is good because we need both to make it so that the epistemic burdens of contributing to the public good are more balanced with the subjective value of doing so, which may push that option higher in individuals’ preference rankings. The epistemic burdens of contributing to the public good by collective action are so great, and the subjective value of doing so so low, that compared to other options choosing to contribute to the public good will be ranked lower in a person’s preference ranking. For the option of contributing to the public good to move up in a person’s preference ranking, the epistemic burdens must decrease or the subjective value increase or both. I  have argued that for this to happen the individual’s ability to overcome these epistemic burdens must be improved, that the subjective value must be increased, and that, either way, bioenhancement is necessary.

Bioliberal Moral Psychology The claim that we ought to use moral bioenhancement to improve our ability to respond to collective risks is a “bioliberal” recommendation. The

An Epistemic Argument 59

most notable bioliberal position on moral bioenhancement is from Persson and Savulescu, who argue that our moral psychology has evolved such that we are strongly biased toward those who are in own group. Because of our evolved moral psychology, our altruism doesn’t extend to other social groups. Our moral psychology is rather exclusivist, but to overcome many of the problems that confront current and future humans we must be much more inclusivist, which requires, argue Persson and Savulescu, moral bioenhancement such as a boost of our empathy. Powell and Buchanan (2016) contend that Persson and Savulescu’s claims fall short. Buchanan and Powell agree that humans are biased toward their own group. But instead of viewing this bias as immutably biological, something to be counteracted by moral bioenhancement, they view the predisposition toward exclusion as a byproduct of modifiable social and cultural practices. Or, instead of counteracting our predisposition with bioenhancement, we can change the conditions that give rise to the predisposition in the first place. According to Buchanan and Powell, there are two significant problems with using moral bioenhancement to resolve the collective action problems that, in the absence of a solution, are going to harm us. The first problem is that it won’t be effective. The collective action problems we face arise out of social and epistemic practices: “Much of the skepticism seems to be enmeshed in a web of morally exclusivist beliefs, with scientific communities being perceived as a liberal threat to in-group identity” (p. 251). Buchanan and Powell doubt that moral bioenhancement will be effective at changing these webs of belief and social epistemic practices. For example, making everyone more empathetic by administering oxytocin won’t help resolve the predisposition for exclusion in a way that resolves these social epistemic obstacles, because oxytocin doesn’t resolve intergroup conflict. Instead, biologically induced empathy may instead make intergroup conflict even worse (De Dreu et al., 2011). There are other good reasons to doubt that we’d be better off if everyone had more empathy for those outside of their own groups. Since empathy is psychologically tied to parochialism, if we want to break the chains of in-group bias, we should try to eliminate empathy, so as to adopt a more general moral point of view (Powell & Buchanan, 2016, p. 252; Prinz, 2011). The first problem with the bioliberal position on moral bioenhancement is that it won’t be effective at resolving what it is supposed to resolve. The second problem is that there are other means of addressing the predisposition to exclusivity, so it isn’t necessary either, according to Powell and Buchanan. Instead, they claim that we should change our social institutions (e.g., educational, political, economic, criminal justice) so that they offer greater economic and physical security, reduce the incidence of infectious disease, and protect the continued humanization of out-groups from the invasion of harmful social epistemic practices such as propaganda and

60  An Epistemic Argument

disinformation. Since exclusionary behavior arises out of these conditions, according to Buchanan and Powell, by manipulating these institutions, the predisposition for exclusion cannot so easily arise and go on to undermine our moral progress. Powell and Buchanan’s argument does not undermine my position that we should engage in moral bioehancement. If anything, their argument bolsters mine. Persson and Savulescu’s argument rests on our psychological proclivities. My argument rests on our epistemic abilities. Powell and Buchanan rightly claim that the obstacles to moral progress, or the mitigation and prevention of collective risks, are epistemic. The discussion so far in this chapter shows that their view is entirely compatible with mine. This compatibility, therefore, constitutes an advantage of my argument over Persson and Savulescu’s argument. My argument is immune to Buchanan and Powell’s criticism, whereas Persson and Savulescu’s argument is susceptible to it. However, Buchanan and Powell’s contention that changing institutions can induce moral progress doesn’t appreciate how deeply our collective epistemic incompetence penetrates. Changing institutions is an enormously epistemically burdensome project. While it’s true that changing educational or political or criminal justice systems may foster greater inclusion and efficient collective action, these institutional changes are themselves problems of collective action. The obstacle to mitigating or preventing suffering from collective risk is that in large groups we are ineffective cooperators. But Buchanan and Powell’s proposed solution—manipulating the institutions that give rise to exclusivity—requires that we are competent, effective cooperators. It requires successfully deploying the very capacities that are the problem in the first place. Their solution presumes that we and our leaders know how to successfully pursue these aims to change institutions. But we don’t know how, and neither do our leaders. That is, the epistemic burdens that we confront in making such significant changes to our institutions are insurmountable, as insurmountable as those of more directly addressing collective risks. If we could figure out how to overcome them to make significant institutional changes, then we could figure out how to overcome them to more directly address the threatening collective risks, obviating the institutional change that Buchanan and Powell recommend. For example, Buchanan and Powell claim that the exclusion and intergroup conflict arises out of the prevalence of infectious diseases transmitting between groups (p.  254). If we could have better infection control, we could manipulate one of the sources of the problematic predisposition to exclusion. But as the COVID-19 pandemic illustrates, control of infectious diseases is itself a problem of collective action. The solution to our inability to mitigate or prevent collective risk can’t be to rely on our ability to mitigate or prevent collective risk.

An Epistemic Argument 61

Compulsory Moral Bioenhancement Bioenhancement is necessary for people to choose to contribute to the public good and, thus, for securing the public good of preventing or mitigating the suffering that results from climate change or other collective risks. Since it is not likely possible to disentangle cognitive bioenhancement from moral bioenhancement, and the primary goal of the bioenhancement is to secure a moral value (e.g., the prevention of suffering), moral bioenhancement is necessary for large segments of the population to choose to overcome the epistemic burdens of contributing to the public good. Suppose we could disentangle cognitive enhancement from moral enhancement in a way that allowed for the enhancement of cognitive capacities but left a person’s (insufficient) moral capacities intact. This might position a person to overcome the epistemic burdens associated with contributing to the public good. But there may be other obstacles to them contributing. For instance, they may lack the motivation to do much about it. That is, a person might know how to do the things that are necessary for contributing to the public good, but simply not be motivated to act on that know-how. In that case, mere cognitive enhancement would be insufficient—the motivation to contribute to the public good would need to be introduced by way of moral enhancement. Cognitive enhancement implies moral enhancement, or it doesn’t. If it does, moral enhancement is necessary, because cognitive enhancement is necessary. If it doesn’t, moral enhancement is still necessary to boost a person’s motivation, even if they have the requisite knowledge. Either way, moral enhancement is necessary. Moral bioenhancement is necessary for anyone for whom the option of contributing to the public good, as compared to mutually exclusive alternatives, is not the highest ranked option in their preference ranking. This is likely the vast majority of people in those countries that most contribute to climate change. This is the population that must not only improve their sensitivity to the values that climate change is likely to undermine but also in their ability to reason practically about them. The membership in this population is in the billions of people. It’s possible that in some nations, such as China and the United States, the world’s two biggest contributors to climate change, not all people would need it. There may be people who, while not likely to choose to contribute to the public good, are also not contributing to climate change. For example, there may be regions where people do not use fossil fuels or farm livestock. But for everyone else in the nations that contribute to climate change, moral bioenhancement is necessary. Because so many people need moral bioenhancement, it must be compulsory. This is a critical premise in the argument that I  make in this book. I can’t make the argument unless the necessity of widespread moral

62  An Epistemic Argument

bioenhancement entails that it ought to be compulsory. The premise is easy to defend, however. The only way moral bioenhancement could achieve widespread administration is if it is compulsory. Voluntary moral bioenhancement, as others have proposed (Rakić, 2014a, 2014b, 2015, 2017a, 2017b), is insufficient. First, the people who need moral bioenhancement are those who are not sufficiently sensitive to or motivated by the values that climate change undermines. To make moral bioenhancement available but not compulsory relies on their wanting to improve their moral capacities, which itself is a moral capacity. And this is a group that has already demonstrated moral deficiencies. Thus, making moral bioenhancement merely voluntary requires that morally deficient people overcome their moral deficiencies so that they can improve their moral capacities. But if they could do that by will alone, then they would be in less need of moral bioenhancement. In other words, the reason they need moral bioenhancement is that they are morally deficient, which moral bioenhancement is intended to correct. If such people could reliably self-improve, then they wouldn’t need the moral bioenhancement. The problem is that they can’t reliably self-improve their moral capacities. It would be like requiring that a person has overcome addiction in order to access a drug intended to treat addiction. The problem and the proposed solution generalize beyond climate change or even collective risk. Though most everyone is morally deficient, the people who need moral bioenhancement the most are the most morally deficient among us. Because of these deficiencies, they are also highly unlikely to seek out moral bioenhancement. Second, if moral bioenhancement is voluntary, there will simply not be enough people who volunteer for it. Voting is voluntary, and in the United States, only half of the eligible voters volunteer their vote. Though some, maybe many, people would volunteer themselves and their children with the aim of being better people, it is unlikely that this number would be sufficient to make collective action successful. There would have to be enough strong reciprocators coupled with people sufficiently susceptible to shame in order to eliminate significant free-riding and provide assurance that an individual’s contributions won’t go to waste. The moral bioenhancement wouldn’t have to get to every single person for this to happen, just a very high percentage of us, much more than the percentage of us that would volunteer for it. For these two reasons, moral bioenhancement should be compulsory. That it should be compulsory presumes that the public good in preventing or mitigating the suffering associated with climate change or other collective risks is a greater good than not being compelled to receive moral bioenhancement. In Chapter 9, I argue that compulsory moral bioenhancement is compatible with even the most protective form of libertarianism. Third, other successful widespread public health interventions are compulsory. Vaccines are compulsory, and even then a dangerously high number of people fail to get vaccinated. Seatbelts are generally compulsory; not

An Epistemic Argument 63

wearing one can earn a person a fine. And even still people fail to wear them. That public health interventions can be compelled in order to counteract the collective risk of harm was the basis of the famous Buck v Bell decision. And although the decision itself (“Three generations of imbeciles is enough”) is often derided, the moral reasoning that grounds it is sound (Anomaly, 2018). And it aligns with a basic reading of Mill’s Harm Principle, which is that the only time it is permissible to limit a person’s liberty is when the exercise of that liberty harms others. It is to be expected that, even if compulsory, moral bioenhancement won’t reach everyone. But if it’s merely voluntary it will reach only the people who need it the least, which are the people already disposed to contribute to the public good. The only way for moral bioenhancement to do what it needs to do is for it to be compulsory.

Conclusion Others have argued that moral bioenhancement ought to be compulsory. The argument in this chapter draws on that work. But the notion of epistemic burdens and their role in decision-making helps to clarify why moral bioenhancement ought to be compulsory. The problem of acting in ways that contribute to the public good is not purely a moral problem. It is also an epistemic problem, an epistemic problem that won’t be overcome without the aid of moral bioenhancement. Not only are we not good enough people to act collectively to secure a public good, we are also not knowledgeable enough.

Chapter 4

A Moral Argument for Compulsory Moral Bioenhancement

In the previous chapter, I claimed that human cognition and behavior in response to collective risk isn’t up to the task of averting the suffering associated with catastrophes like climate change and pandemics. We aren’t smart enough, and we aren’t good enough to work in large groups to prevent this suffering. We must become smarter and better than we naturally are. But it’s plausible that what makes a person a person is their cognitive and moral capacities. If this is right, then it’s plausible that these may improve so much that the enhanced person becomes transhuman, a moral post-person. Moral post-persons are so morally superior to others that they may have a different moral status than the rest of us. The gap in moral capacities between most individuals is so small that, to the extent our moral capacities influence our moral status, we all have the same moral status. But moral post-persons’ moral capacities would be such that the gap between their capacities and ours is much bigger, so big that we may think of them having a different moral status. In this chapter, I argue that we should bring these moral post-persons about. To do so is to promote our and our descendants’ individual and collective interests. Upon concluding that we ought to bring about moral post-persons with moral enhancement, I argue further that enhancement ought to be compulsory. Whereas I argued in the previous chapter on epistemic grounds that moral bioenhancement ought to be compulsory, I argue in this chapter on moral grounds for the same conclusion.

Agar on Moral Post-persons A prominent argument against bringing about moral post-persons is Nicholas Agar’s (2013). He claims that moral post-persons, having a different moral status than that of mere persons, would have opportunities to sacrifice our interests so that they can pursue their own, much in the way that we sacrifice the interests of animals so that we can pursue ours. If we could expect moral post-persons to behave toward mere persons in the way that mere persons behave toward animals, then it would likely be contrary to

A Moral Argument 65

our interests to bring moral post-persons about. Mere persons treat animals quite horribly. We don’t want that same treatment. It is true by definition of “moral post-person” that if moral bioenhancement can bring about moral post-persons, then moral bioenhancement can influence the moral status of those receiving the enhancement. Some authors have claimed that moral post-persons are not possible. Buchanan (2009) argues that moral status is a threshold concept such that those beings that have crossed the threshold share a moral status. If this is true, then moral post-persons are indeed not possible, for they would have crossed the relevant threshold along with mere persons like us. Buchanan’s claim, if true, should ease any worry one might have about moral post-persons: they aren’t possible, so there is need to concern ourselves with them. Even if Buchanan is correct that moral status is a threshold concept, we may still justifiably worry about bringing moral post-persons into existence. There might be more than one threshold (DeGrazia, 2012), one that we don’t meet but that moral post-persons do. Or even if there is only one threshold that mere persons and moral post-persons both meet, that’s no guarantee that moral post-persons wouldn’t exploit and sacrifice mere persons. Some groups of mere persons aim to exploit and sacrifice other groups of mere persons, even though we are all of the same moral status. What’s more, there are perfectly good alternatives to Buchanan’s account of moral status. Regardless of what account of moral status is correct, so long as it is possible that moral bioenhancement can bring about beings who are morally superior to us, we should bring them about (Rakić, 2015). Agar (2013) claims that not only are they possible, but, given the development and administration of cognitive or moral enhancement, are likely. He thinks we should not bring about these beings; I think we should. Agar’s argument that we should not bring about moral post-persons starts by noting that in supreme emergencies, it is permissible to sacrifice some persons for the sake of other persons (McMahan, 2009). The COVID-19 pandemic, among other emergencies, has made obvious that sacrificing some people so that others avoid harm is morally permissible. People may disagree about which of those affected by the pandemic ought to be sacrificed (e.g., those in the restaurant industry or the elderly) or which harm is worse, but sacrificing one group so that another avoids harm is, and has been, inevitable. But what drives Agar’s argument is concern about what moral post-persons might do not to prevent harm to others, but to secure benefit for themselves. Sacrificing others’ interests to prevent harm to some other group is different from sacrificing others’ interests so that one may benefit. One significant difference is that in supreme emergencies it is permissible to sacrifice beings with full moral status to prevent harm to some other persons with full moral status. But it may not be similarly permissible to sacrifice others’ interests so that some other group, or even oneself, might benefit, when those people whose interests are sacrificed have full moral status. Consider biomedical

66  A Moral Argument

research: the point of such research is to benefit persons; but it is the interests of beings with less-than-full moral status, such as mice or rats or rabbits or dogs, that are sacrificed to secure this benefit. Thus, there is an asymmetry between what sort of interest sacrificing is acceptable to prevent harm in supreme emergencies and what is acceptable to secure a benefit. The source of this asymmetry, according to Agar, is the relative moral statuses of those benefitting and those sacrificed. Having the highest moral status privileges persons such that it becomes permissible to sacrifice beings with a lower moral status. The worry, according to Agar, is that by creating moral post-persons, mere persons, being of a lower moral status, would be eligible for sacrifice for the benefit of moral post-persons in the same way that animals are eligible for sacrifice so that mere persons may benefit. Agar writes that “the act of creating post-persons would, in itself, dislodge mere persons from the role of having the highest moral status. Their demotion would render them eligible for sacrifice to provide significant benefits for post-persons” (p. 72). Suppose that Agar is right that moral post-persons are permitted to sacrifice mere persons’ interests so that they may benefit. It still doesn’t mean that we ought not to bring moral post-persons about. For this permission to sacrifice to be problematic, it also has to be true that moral post-persons would do so. Being permitted to behave in a certain way doesn’t imply that the person so permitted will act in that way. And we have good reasons to think that moral post-persons would not sacrifice mere persons’ interests so that they may benefit. Agar thinks that moral post-persons will sacrifice mere persons because mere persons will be good models for research, from which moral postpersons can derive a benefit for themselves. Mere persons of course do the same with animals with lower moral status. Because it would be permissible for moral post-persons to do the same to us, and we can expect them to do so, we shouldn’t bring moral post-persons about. But his argument presumes (a) that the moral post-persons’ interests would be in conflict with mere persons’ interests and (b) that when they are in conflict they would pursue their own at the expense of mere persons’. But I think we have good reasons to reject both of these assumptions. What makes moral post-persons who they are is that they promote others’ interests first. If they didn’t, they wouldn’t be moral post-persons.

Moral Exemplars Moral post-persons are more capable of acting morally than mere persons. It’s in virtue of those greater capacities that they are moral post-persons. The most morally capable mere person will still not be as morally capable as a moral post-person. But of all mere persons, the best of us will share moral capacities with moral post-persons to the greatest degree. So, if we want to

A Moral Argument 67

know what moral post-persons are like, the best of mere persons is a good place to start. Moral exemplars are those mere persons who we may regard as the best of us. When we look at moral exemplars, we see that they are highly capable of promoting others’ interests. Linda Zagzebski (2017) offers an account of moral exemplars. She claims that moral exemplars are “are those persons whom we see, on close observation and with reflection, to be admirable in all or most of their acquired traits” (p. 65). Reflective admiration is the act of admiring a person’s traits after scrutinizing them. Moral exemplars, the best of us mere persons, are those people whose traits we most admire after reflection. There are several sources of our admiration, according to Zagzebski. One is that we admire the psychology behind the admirable behavior. Another is that we admire acquired psychological properties more than we admire psychological properties that one has in virtue of their inheritance. We admire those psychological traits that ground admirable behavior more when they were developed than when they simply arise out of the person’s natural condition. A third source of our admiration is that we admire “motives of concern for others more than motives of self-interest” (p. 64). Zagzebski discusses two recent examples of moral exemplars. One was Leopold Socha, who was a Polish sewer inspector who helped Jews escape Nazis liquidating their Polish ghetto in May 1943. In escaping Nazis murdering others in their community, some Jews sought the sewers of Lvov, which is where Socha encountered them. He helped them survive by bringing them materials for survival and religious observance until they could escape more than a year later. When they finally were able to leave the sewers, Socha had arranged housing for them. And though Socha was formerly a thief and was initially paid, he took saving them as his mission. Socha, upon reflection, seems admirable in the ways moral exemplars are supposed to be. His motivation to promote others’ interests was acquired, which seems to meet the conditions for being a moral exemplar. Zagzebski’s second example is Jean Vanier and his L’Arche communities. Vanier was a philosophy professor who, over an extended period of training in the academy and the ministry, developed a calling to improve the lives of those with mental disabilities. At first, he invited two men with disabilities to live with him. He assisted them in all aspects of their lives, living together as a family. Over time, more and more communities were developed. In the communities, volunteers live as families with the disabled, promoting their members’ interests. They assist in their day-to-day lives, striving to help those rejected by society attain independence. The volunteers report to be motivated by factors similar to those of Holocaust rescuers, such as factors relating to the closeness to and the welfare of others. Also like Socha, Vanier and the volunteers in L’Arche communities seem to pass the test of exemplarity.

68  A Moral Argument

One doesn’t have to agree with Zagzebski’s entire moral theory that rests upon reflective admiration to recognize that Socha and other Holocaust rescuers and Jean Vanier and the L’Arche communities are moral exemplars. What makes them so is their acquired motivation to promote others’ interests. If they are moral exemplars, and what makes them so is this acquired motivation to promote others’ interests, then moral post-persons will exhibit this capacity to an even greater degree. Moral post-persons are likely to be highly motivated to promote others’ interests, even if, as is the case with Socha and Vanier, others’ interests conflict with their self-interests. Maybe moral post-persons’ interests and others’ interests won’t conflict, but if they do we can expect moral post-persons to, like moral exemplars, promote others’ interests.

We Should Engender Moral Post-persons Agar’s argument relies on the assumption that moral post-persons will promote their own interests at the expense of mere persons’ interests. But instead, we can expect moral post-persons to promote mere persons’ interests, even if doing so conflicts with promoting their own. Because an assumption upon which Agar’s argument rests is likely false, his conclusion that we shouldn’t bring about moral post-persons is cast into doubt. But this doesn’t yet establish that we should bring them about. To establish this conclusion, more is needed. Here, as elsewhere, I rely on Singer’s principle that if we can prevent something bad from happening without sacrificing something of comparable moral significance, we ought to do so (Singer, 1972). Mere persons are in a poor position to prevent harm from collective risks. Moral exemplars are in a better position to prevent harm from collective risks. Moral post-persons will be in a much better position, given that their capacities for promoting others’ interests will be so much greater than even moral exemplars’ that they have a different moral status. Singer’s principle is reasonable and intuitive because it doesn’t require that one sacrifice something comparably significant to prevent harm to others. But some authors think that administering moral bioenhancement undermines a person’s freedom (Harris, 2011).1 And since a person’s freedom may be of comparable moral significance to the interests of others, one might think that using moral bioenhancement to save ourselves undermines the application of Singer’s principle. I’m skeptical of the idea that moral bioenhancement undermines freedom, for all of the reasons that others have discussed. But whether moral bioenhancement undermines freedom is irrelevant, because even if it does undermine freedom it still doesn’t follow that we should not bring about moral post-persons. Consider Socha’s recollection, as quoted by Zagzebski (p. 71, italics mine): “ ‘When I squeezed through the shaft, into the little cellar, you were sitting there with Krysia and Pawel under each arm. Like

A Moral Argument 69

a mother kite and her chicks. It was at that moment, when you were sitting there with the children, at that moment I decided to save you’.” Socha claims he decided to save the Jews. There is no indication that he instead was compelled to do so. But suppose that he was compelled to save them. Even if this were true, it isn’t obvious that being compelled to save someone undermines freedom. But if it does, it still doesn’t follow that it being so undermined is of comparable moral significance. If Socha being compelled to rescue the Jews is of comparable moral significance to saving them, it should be difficult to say which of the following two alternatives is better: Socha deciding to do nothing and then allowing the Jews to die in the sewer or sacrificing his freedom but saving the Jews. It seems obvious the second state of affairs is better, which is a good indication that they are not two alternatives of comparable moral significance. The same would be true of moral post-persons. They could either preserve their freedom and decide to not save us from catastrophe or they could sacrifice their freedom and save us. If freedom is of comparable moral significance to saving humanity, then it should be difficult to say which of the alternative is better. But it isn’t—the latter is clearly better. Maybe the moral post-persons that result from enhancement would have to sacrifice their freedom, and maybe they wouldn’t. Whether it is of comparable moral significance is not up to us—it is up to them. Moral postpersons have greater moral capacities. Even if in our judgment sacrificing freedom is of comparable moral significance to promoting others’ interests, our judgments are those of mere persons, with our limited, and incompetent, moral capacities. Moral post-persons are likely to come to a different judgment about the relative significance of freedom and promoting others’ interests. Or, whether moral post-persons’ freedom is of comparable moral significance is not for us to decide. Given that moral post-persons are likely to find promotion of others’ interests far more important and motivating than such considerations are for mere persons, it is reasonable for mere persons to suspend judgment about what moral post-persons would judge. That moral post-persons may be in a good position to save us from ourselves supports the notion that we should bring them about. It is in our collective and individual interests to do so. But recall that moral exemplars are reflectively admirable not only because they promote others’ interests by preventing harm from coming to them, but also by positioning others to flourish. Jean Vanier (1998, p. 5) writes: This book is about the liberation of the human heart from the tentacles of chaos and loneliness, and from those fears that provoke us to exclude and reject others. It is a liberation that opens us up and leads us to the discovery of our common humanity. . . . The discovery of our common humanity liberates us from self-centered compulsions and inner hurts; it is the discovery that ultimately finds its fulfillment in the forgiveness and

70  A Moral Argument

in loving those who are our enemies. It is the process of truly becoming human. Although “liberation” may bring to mind saving someone, the above passage makes it clear that Vanier was trying to position those in his communities to flourish by giving those with mental disabilities the environment, care, and relationships needed to allow their “transformation.” Just as we can expect moral post-persons to promote mere persons’ interests by preventing harm, we can expect them to also do so by positioning mere persons to flourish. Many mere persons’ lives are good, even more are bad. But as good as mere persons’ live are, most of them could be better still—all of our lives could be better than they are. Our collective moral incompetence, rife selfishness, and cognitive incompetence mean that we are poorer, less efficient, sicker, and live shorter lives than we could. Just as Vanier and other moral exemplars position others to flourish, and in so doing promote others’ interests, we can expect moral post-persons to do the same. We can expect them to give us the care, environment, and relationships that will allow us to flourish. It is in our own self-interests to bring about such people. We may promote our own interests in lots of ways. One way would be to administer moral bioenhancement to engender moral post-persons so that they can join us in that pursuit. Although Agar’s argument doesn’t establish that we shouldn’t bring about moral post-persons, and I have offered arguments that we should, one may still object to the conclusion on the grounds that I’ve got moral post-persons wrong—they wouldn’t necessarily be concerned with promoting others’ interests. They might have different concerns altogether, and they might have different capacities. This objection doesn’t work, though. Moral exemplars are good, but moral post-persons will be even better. If moral post-persons and moral exemplars are different in kind, having different moral capacities, then we’re wrong about who the best mere persons are and, consequently, what moral capacities make the best of mere persons so good. Of all mere persons, the best mere persons will necessarily share to the greatest degree moral capacities with moral post-persons. If the capacities of the mere persons whom we regard as exemplars are different in kind than those of moral post-persons— if we are wrong about what is in the shared set of capacities—then we are wrong about what counts as moral behavior. If we are so wrong about what constitutes moral behavior, then we need moral enhancement even more.

From Moral Post-persons to Compulsory Moral Enhancement We are obligated to engender moral post-persons. That we have this obligation supports a second claim: moral bioenhancement ought to be

A Moral Argument 71

compulsory. In this section, I  offer an argument from the claim that we ought to engender moral post-persons to the conclusion that moral bioenhancement ought to be compulsory. Moral post-persons will necessarily be in a better position to maximize value, whatever that value happens to be. So long as there is some notion of value according to which that value is independent of any person’s attitude toward it, moral post-persons will be better at maximizing it. One might think that the greatest value is achieved by being virtuous or by forming virtuous characters. Others might think that the greatest value is achieved by behaving in accordance with, and from, one’s moral duty, such as a moral duty that is derived from a categorical imperative. Others might still think that the greatest value is achieved by producing the most utility, such as by acting in ways that maximize pleasures and minimize pains. Whatever the right objective notion of value is, moral post-persons will necessarily be in a better position to achieve it. Having such an ability is what makes them moral post-persons. Given that moral post-persons will be better at maximizing objective value, whatever that happens to be, more moral post-persons means more value generation. This is to say that value is maximized when everyone is a moral post-person. Or, at least more value is achieved by everyone being a moral post-person than not. If the primary value to be achieved is virtue, this is what moral post-persons will maximize. If the primary value is acting in accordance with and from moral duty, this is what moral post-persons will maximize. If the primary value is experiencing pleasure and avoiding pain, this is what moral post-persons will maximize. Or, value is maximized when everyone is a moral post-person. One may object to this claim by holding that some combination of moral post-persons and mere persons achieves more value than everyone being a moral-post person. It’s not clear what could possibly justify such a claim, however. It would require that there are some moral post-persons who achieve less value than their mere person counterparts. But such an arrangement contradicts the very notion of a moral post-person. Alternatively, it may be that a mix of moral post-persons and mere persons facilitates certain relationships, or allows for a certain arrangement of relationships, that wouldn’t exist if everyone were a moral post-person. If such an arrangement indeed maximized value, then that would be a reason to reject the premise that value is maximized when everyone is a moral post-person. But if maximizing value is something that is achieved by maintaining particular relationships, moral post-persons are in a better position to maintain those relationships than mere persons. Relative to any achievement of value, if mere persons can do it, moral post-persons can do it better. So, it seems plausible that value is maximized by everyone by being a moral post-person. It’s also plausible that one ought to maximize value. If it’s true that one ought to maximize value, and value is maximized by everyone becoming

72  A Moral Argument

a moral post-person, then it follows one ought to become a moral postperson. Since it one can’t plausibly reject the claim that value is maximized by everyone becoming a moral post-person, rejecting this argument requires claiming that it is permissible to not maximize value. This goes beyond rejecting the claim that it is permissible to forego utility. The claim requires that one hold that it is permissible to also forego virtue or duty satisfaction, which is necessarily less plausible than claiming that it is permissible to forego utility. But the notion that it is permissible to forego utility, or “satisficing” consequentialism, provides a useful structure of the idea. Satisficing consequentialism stems from the idea that maximizing consequences entail obligations that are too demanding for a person to satisfy. Thus, goes the reasoning, it is permissible for a person to achieve utility that is merely “good enough,” but not maximized. So, if it’s false that one ought to maximize value, and it’s still true that one has some obligation to achieve something of value, then satisficing is what lies in the middle. The claim would then be that it is “good enough” to be a mere person, foregoing value but achieving enough. It may be better if one maximizes value, but it’s not obligatory. This strategy faces significant obstacles. The first is that it is unmotivated. Satisficing consequentialism is motivated by the idea that maximizing utility is too demanding for a person. But in the case of becoming a moral postperson by way of enhancement, it isn’t all that demanding. One needs only to receive the enhancement. That’s it. Then one’s new moral capacities will take it from there. So, it’s not too hard to maximize value, if doing so involves receiving an enhancement to become a moral post-person. Not only is satisficing not motivated in this case, but it’s also false, in the sense that being a mere person is not good enough. Being a mere person is the entire problem. Our moral flaws in particular endanger not only the health and well-being of everyone currently alive, but also that of future generations. Being a mere person isn’t good enough, because it’s going to be the cause of a great deal of preventable suffering. So, whatever one might say about mere persons being good enough—that they meet some threshold of value achievement, for example—the response is simply that, no, they don’t. Perhaps there are instances in which it is permissible to forego some value and instead do something that is merely good enough. But foregoing the value associated with becoming a moral post-person is not satisficient. Since value is maximized when everyone is a moral post-person, and one ought to maximize value, everyone ought to be a moral post-person. This is to say that for all persons, one ought to become a moral post-person. This argument generates the moral obligation to become a moral post-person, but doesn’t yet generate the conclusion that moral enhancement ought to be compulsory. To go from the claim that everyone ought to become a moral post-person to the conclusion that moral enhancement ought to be compulsory, it is

A Moral Argument 73

sufficient that compulsory moral enhancement is necessary to satisfy the obligation that everyone ought to become a moral post-person. That moral enhancement must be compulsory for everyone to become a moral postperson is defensible. First, no one will become a moral post-person without moral enhancement of some kind. Second, if it is not compulsory, many people will not receive the enhancement. Their failure to receive it will be for a variety of reasons, many of them having to do with their deficient moral capacities, deficiencies which necessitate moral enhancement in the first place. And if many people fail to receive the enhancement, then the obligation that everyone becomes a moral post-person goes unmet. In some cases, satisfying an obligation requires that one perform some act that not only has negative value in and of itself, but that also has more negative value than the satisfaction of the obligation generates in positive value. In such cases, the obligation that necessitates the value-negative act ought to go unmet. But that’s not the case here. Compelling the enhancement is not worse than the value achieved by everyone becoming a moral post-person. Much of what follows supports this claim. The central idea is this: the value achieved by compelling moral enhancement is enormous; we and our descendants get to continue to live and do so in, overall, less painful circumstances. Receiving a moral enhancement is not overly burdensome. This is especially true if the compulsory moral enhancement is administered covertly, without the knowledge of those receiving it. Of course, it may not be possible to engender moral post-persons with moral bioenhancement; the improvement of mere persons’ moral capacities may be the ceiling. If true, then there is no obligation to engender moral post-persons. But the impossibility of moral post-persons would not imply that we shouldn’t engage in compulsory moral bioenhancement. What justifies compulsory moral bioenhancement is that it gives us the best chance of mitigating the collective risk of suffering from preventable global catastrophe. Moral post-persons would be best; improved mere persons may be good enough. We’d still be much better off if we were all as good as our current moral exemplars. But unenhanced mere persons are not good enough. If we remain as the people we are, there ought to be no hope.

Note 1. See DeGrazia (2014) for an argument that moral enhancement doesn’t undermine freedom and even if it does it is not so valuable to defeat the permissibility of moral enhancement. For discussion of Harris’ worry about freedom and some replies to it, see Persson and Savulescu (2013), Savulescu and Persson (2012), and Douglas (2013).

Chapter 5

The Proposal

I argue that moral bioenhancement should be administered to everyone without them knowing it. This may seem crazy. And my goal of this book is to make this idea seem not all that crazy. One reason to think that covert compulsory moral bioenhancement is not crazy is that our society already accepts practices of withholding information and deception. In particular, when withholding information or deception is aimed at improving the lives of others, we often find it acceptable. This is not to say that withholding information from others for whom the information may be relevant or that deception is ethical, unethical, or even morally unproblematic. It is simply to say that we readily accept these practices, regardless of whether we are right to do so. Often accepted practices of information withholding are informal. Withholding information from loved ones seems a thread in the fabric of a wellfunctioning family. We withhold information from our spouses, partners, siblings, parents, and children, often when the withheld information is relevant to them. Suppose one parent discovers their teenager smoking cigarettes. This parent knows that if the other parent has this information, they will react very negatively, not only toward the teenager, but in such a way that their own well-being will be undermined. The parent who discovered the smoking knows that they can handle it on their own and guide the teenager in a way that will get them to stop smoking. This parent can tell the other parent or not. In this case, it seems totally acceptable for the parent to withhold the information from the other parent. Or consider a case in which a dying mother discloses to one adult child that she was always her favorite child. We would readily accept that child withholding that information from her brothers and sisters. Disclosing this information would be relevant to the siblings, but would likely undermine the function of the family. A child who witnesses a parent die may tell their sibling that they died in peace, when in fact they did not. Examples like this are not difficult to come by. In general, we readily accept the withholding of information, often to protect family members’ well-being. This protection is sometimes directed at the well-being of the individual from

The Proposal 75

whom information is withheld, as in when the favorite child withholds that information from her less-liked siblings. Other times information is withheld so that others are shielded, such as when one parent neglects to reveal a cigarette-smoking teen to another parent. But information withholding and deception are even more basic to social practices. Suppose an already stressed partner asks, “How are you doing?” and one responds “I’m doing well,” but in fact one is not doing well. Generally, we accept this lie as permissible, typically uttered to protect one’s partner from further diminished well-being. Again, none of this is to say that this is morally permissible or impermissible. It is just to point out that we readily accept significant deception and information withholding in some of the most important parts of our lives. Occasionally, this ignorance is willful—we don’t want to know or we want to be lied to. One partner might have a hunch that the other is cheating but turn a blind eye to the evidence for the sake of their own well-being. A  parent might subtly invite their unmarried child to lie to them about living with their partner. Instances of willful ignorance are not difficult to come up with. They show that not only do we accept withholding information and deception, we sometimes invite it. Sometimes we want to not know; and sometimes we want to be lied to. Furthermore, this deception and information withholding is not incidental to decisions we might otherwise make in how to conduct our lives. It is centrally relevant to how we view ourselves as mothers, fathers, sisters, brothers, sons, and daughters. Social institutions have also formalized information withholding and deception. In medicine, there are well-known exceptions to the duty to obtain informed consent from a patient prior to administering a treatment. The most common of these is the emergency exception, which is that one need not obtain informed consent from a patient emergently needing a treatment intended to save life or limb. We also allow a person to waive the right to informed consent, which is usually because they either don’t want to know a diagnosis or don’t want to make the decision about how to treat it. I  have argued (Crutchfield et  al., 2018) that there is no obligation to obtain informed consent when doing so would undermine the achievement of something objectively valuable, such as when a placebo is an indicated treatment. This exception is a corollary to the exception most relevant to the present discussion, therapeutic privilege. Therapeutic privilege is permissibly executed when a physician withholds information from a patient in order to prevent some further harm coming to them. For example, a person may find a new diagnosis to be so distressing that knowing about it will exacerbate their mental health in a way that harms them. In such a case, the conveying of information causes harm. In that case, it is permissible for the physician to withhold the diagnosis. The commitment to preventing harm overrides the patient’s right to know and make an informed decision. Therapeutic privilege is legal and often

76  The Proposal

explicitly invoked, but there are other circumstances in medicine when withholding information from patients or family is permissible. A physician may withhold information about the manner of death, for instance, whether the patient was suffering, from the family when they disclose death to them. Like the role of information withholding and deception in the fabric of the family, such practices, when intended to prevent harm, are readily accepted in society, in this case by policy and practice. In medicine therapeutic privilege is intended to prevent harm to the indi­vidual who is being deceived or incompletely informed. But in human ­subjects research, it is standard practice to allow some information withholding and deception. And in such research it is not to ­prevent harm to a particular individual but so that society may benefit from a greater fund of knowledge. Information withholding in research is critical to producing reliable results. Telling a patient that may be getting a placebo or that they may be getting the experimental treatment is a version of information withholding. There is a fact of the matter of what the subject will receive. That fact is withheld from the subject. Human subjects’ research couldn’t reliably conducted without placebos, so it couldn’t be reliably conducted without withholding information. It is a practice fundamental to human subjects’ research. Deception in research is more controversial. Deception is more frequently used in social and behavioral sciences, though it still has a role in medical research, such as in research on surgical outcomes (compared to sham surgeries). Much has been written on the ethics of deception in research, and I don’t try to adjudicate this debate. Generally, there are several ways researchers may deceive subjects. They may deceive subjects as to what the task is measuring or of what the experimental hypotheses are. They may deceive the subjects as to what their performance in the research says about them. Or they may deceive the subject as to the identity of the experimenters. The use of confederates in psychology and behavioral science is common. One of the alleged worries about deception in research is that it undermines the “dignity” of the subject. I have yet to encounter an account of human dignity that is usefully distinct from the notions of autonomy or rational agency. So this worry amounts to a worry that deception undermines a person’s autonomy, and that it does so in a way that harms the subject. But there is little evidence that, even if deception undermines autonomy, it does so in a way that harms the subject. Boynton et al. (2013) experimentally investigated the self-reported harms of being deceived in research and found that task obfuscation and false feedback (telling the subject something false about what the research says about them) did not harm the subject. Standard practice is to debrief subjects in order to mitigate any potential harm. But if one is concerned about the potential harms of deception in research, they

The Proposal 77

ought to be concerned about these debriefings. Knowing that one has been lied to might harm the subject as much as being lied to in the first place. The subject may be better off never learning about the deception. The wrong of deception—if it’s wrong at all—is not made better by harming the subject further with the knowledge that they have been lied to. Deception in research is obviously not intended to benefit the subject. Researchers deceive subjects to benefit society. The benefit to society from any individual study is almost always very small, a tiny drop in the bucket. There are of course some studies that have been highly impactful, such as the Milgram experiment. But most of the research that involves subject deception has much less benefit. Yet we justify the deception by this potential benefit. At the political level, we also endorse information withholding and deception. That is, we don’t want the government to disclose all information, especially when to do so would be to risk national security or strategic advantage. Redaction is information withholding, and we readily accept redaction, at least in principle. We might dispute whether a given document is justifiably redacted, but in principle we endorse redaction. Similarly, we endorse governmental deception when that deception is intended to maintain the safety and security of citizens, such as in matters of military action or espionage. None of the above discussion is intended to demonstrate that we ought to withhold information or deceive in these contexts. It is to demonstrate that we do, as a matter of standard practice, withhold information and deceive, and that for the most part, we are accepting of this practice. Information withholding and deception are not only widely accepted, but also appear to be functionally important in national security, research, medicine (a physician couldn’t satisfy the duty to not harm if she were required to disclose harmful information), and in how a family interacts. It’s false that, in principle, we do not or cannot accept deception and information withholding. Rather, the opposite is true: in principle, we accept information withholding and deception in ordinary life, especially when it is intended to benefit others. In what follows, I argue that moral bioenhancement ought to be covertly administered to everyone. To administer such a program, the fact of its administration must be withheld from those receiving it, which is everyone (or very near everyone). The reason for its administration is that it is necessary to prevent the significant suffering that is likely to result from, for example, catastrophic climate change. Not having to undergo this suffering, or not dying, is a significant benefit, certainly greater than any benefit from social or behavioral research or from not being told you were your mother’s second least favorite child. If benefits can ever justify information withholding or deception, the benefits we receive from not suffering and dying can justify information withholding and deception.

78  The Proposal

There are several ways moral bioenhancement might be administered covertly. The two that are most feasible are administration by way of the public water supply and by packaging it with compulsory vaccines. Already some chemicals are distributed through the public water supply. Flouride is added to public water supplies that reach two-thirds of people in the United States. Of this two-thirds, some significant percentage of people install filters on their residential fixtures or otherwise drinks only filtered water. It is reasonable to suppose that water filters would also screen out whatever the moral bioenhancement is. Thus, at least one-third of the United States’ population would fail to get the moral bioenhancement, if it were administered covertly through the public water supply. Administering moral bioenhancement along with other compulsory vaccines is another route of administration. In the United States, vaccination rates vary by state and vaccination, but are generally around 95% (“Vaccination Trends,” Health Testing Centers). Although this is much lower than the percentage in 2001, it means that far more people receive compulsory vaccines than drink fluoridated water. There are other advantages to packaging moral bioenhancement in vaccines. Importantly, it better promotes fairness. Water filters can be expensive. They are more difficult to access for poorer people. Because of this, people who can afford water filters can more easily avoid the moral bioenhancement, and this difference in opportunity is unfair. Vaccines are compulsory and receiving them less financially burdensome. They are not necessarily easier to avoid for rich people. Packaging moral bioenhancement with vaccines also ensures intervention earlier in life, prior to or during the development of moral capacities. To the extent that moral bioenhancement interferes with a person’s moral capacities, the interference may be less noticeable to the person the earlier in the development process in which it is administered. In fact, if the intervention partly determines the person’s moral capacities in the first place, then it may be incorrect to describe it as an interference at all. It wouldn’t interfere with person’s moral capacities any more so than any other factor that goes into their development, such as the child’s environment or inheritance. It may also be easier to conceal moral bioenhancement from recipients if it is packaged with vaccines than it would be by putting it in the public water supply. The chain of custody of vaccines may be simpler, involving many fewer people. Most vaccines are produced by just a handful of companies and end up in the hands of doctors and nurses. The chain from these companies to the nurse can potentially be both tightly regulated and involve very few people. Relatedly, since the supply is more tightly controlled, it may be more difficult for third parties to detect the moral bioenhancement in samples and disclose the intervention. Testing of water supplies is both common and public, but this is not the case for vaccines. Finally, packaging moral bioenhancement with vaccines may drastically reduce the number of people involved. The more people involved in the administration of moral bioenhancement, the more difficult its administration is to keep covert. If

The Proposal 79

it were packaged with vaccines, such as at the time of production, very few people in the supply chain would need to know about it. There is a significant disadvantage of packaging moral bioenhancement with vaccines. Collective action to prevent or mitigate catastrophe must occur soon. Children vaccinated now won’t be in a position to contribute to collective action for quite some time, likely decades. Thus, only administering moral bioenhancement with vaccines won’t help, so long as vaccines like the flu shot are not compulsory for adults. Because of this, moral bioenhancement will need to be administered in several ways for it to reach the entire population. It will likely need to be packaged with vaccines and included in the public water supply and by way of other methods that are likely to reach many people. Some may doubt whether it is feasible for such a program to be implemented. It’s difficult to counter such skepticism because any examples I might offer are necessarily bad examples. If a program can be used as an example for the present purpose, then it was not covertly administered and therefore not sufficiently similar to what I am suggesting. If something like what I am suggesting has been implemented successfully, then I and most readers are similarly ignorant of it. But the sheer number of people involved in a conspiracy to covertly enhance may make maintenance of the secrecy difficult, if not impossible. One way to minimize the number of people involved is to intervene at the point of production of other products, such as vaccines or fluoride supplements. If the production itself is limited to just a handful of firms, the enhancement could plausibly be included in the production of other things without many people knowing about it. Involving fewer people makes disclosure less likely. But even if there are many people involved, it would be difficult to predict that the administration of moral bioenhancement would be disclosed: those people who know about the intervention would presumably also have received the moral bioenhancement. Being enhanced, they are more likely to promote the maintenance of secrecy so that suffering from catastrophe can be averted. The motivation to disclose of those involved in the administration of the moral bioenhancement would be lower than it otherwise would be because they would themselves have enhanced moral capacities. This notion alone undermines the expectation that the administration of the moral bioenhancement would involve so many people that disclosure would be inevitable. It wouldn’t be inevitable, so long as administrators are enhanced. In earlier chapters, I argued that moral bioenhancement should be compulsory. In the next several chapters, I  argue that moral bioenhancement should be covert, if it is to be compulsory. I  have suggested that moral bioenhancement be administered by way of compulsory vaccines and the public water supply. What a covert administration of moral bioenhancement involves is open, but the above considerations indicate that it may be feasible.

Chapter 6

The Epistemology of Moral Bioenhancement

I argue in this chapter that there are epistemological difficulties with the administration of a moral bioenhancement program. These difficulties, I argue, would render a program ineffective unless the program is administered covertly. There is no point in administering widespread moral bioenhancement if the enhancement if that administration is overt—that the people receiving the intervention know they are receiving it. I outline five problems with an overt program that making the program covert solves. The five epistemic problems arise from the basic facts that we form beliefs upon reasons, that we know this is happening, and that these beliefs affect how we behave.

The Reflection Problem The Reflection Problem is grounded in the fact that humans are generally aware of what they believe. Though one could be wrong about what they believe (Williamson, 2002), a person’s beliefs or other attitudes are transparent to them. A person has special, though not infallible, access to their own mental states. It’s also true that an administration of moral bioenhancement can only be effective if it changes a person’s mental states. Combine this obvious point with the fact of transparency, and the Reflection Problem results: people will be aware of not only their new moral beliefs but also they will be aware that their beliefs have changed. It takes very little reflective ability to recognize such a change in one’s own beliefs. But the epistemological problem is that although the person who has received the intervention will be aware of a change in their moral beliefs, unless the moral bioenhancement also intervenes on the evidence for that new belief, the person won’t be aware of any new evidence. Thus, the person receiving the enhancement will be aware that they have new moral beliefs, but also be aware that they have no new evidence. It is not a great leap to anticipate that such a person would come to understand that they have an unjustified moral belief. But because knowing that one’s beliefs are unjustified makes it much less likely a person will act on those beliefs, when

The Epistemology of Moral Bioenhancement 81

this epistemic arrangement of new beliefs without new evidence arises, the intervention is unlikely to induce the targeted behavior, whatever that happens to be. There are several ways one might resist the Reflection Problem. One is to claim that people don’t actually reflect in the way I propose. For example, maybe people just don’t reflect on their own beliefs as much as the Reflection Problem requires of them. I think this is false—we may not have infallible, perfect access to our own mental states, but it is quite common to take a step back and consider what one believes, how that belief came to be, and whether that belief grounds behavior. Moreover, if people are not reflective and considerate of their reasons, then it’s hard to see how any type of intervention that affects behavior by intervening on mental states will work. Another way to resist the Reflection Problem is to claim that I overestimate the degree to which justification influences behavior. Rather, whether a person’s belief is justified for them has little bearing on whether they behave in any particular way. But if this is true, then moral bioenhancement faces a bigger problem. When a person behaves a certain way, it is typically in pursuit of some goal. When it is, there must be some connection between the behavior and the goal such that person has a reason to behave that way, rather than others, in pursuit of that goal. This is just to say that for the person to behave in a certain way, they must have justification to believe that behaving in that way has an acceptably high likelihood of achieving the goal. But if instead there is a weak relation between justification and behavior, then there is no expectation that people can deliberately behave in the targeted ways. Maybe they would behave in the desired way, but it would only be by accident, which makes the failure of the moral bioenhancement very likely. The Reflection Problem arises because the beliefs a person has are likely to be incongruent with their evidence, which undermines the performance of the targeted behavior. One way to resolve this problem is to intervene on the person’s beliefs, but to do so by intervening on the person’s evidence for those beliefs. The problem could be resolved if the person who receives the moral bioenhancement has not only the targeted belief but also the evidence that supports it. Evidence justifies beliefs for a person. If that evidence is itself another belief, then that belief requires further evidence, which, if it is yet again a belief, requires more evidence. From there, it’s turtles all the way down. Thus, for a moral bioenhancement to be up to epistemic snuff and resolve the Reflection Problem, it must introduce non-doxastic evidence for the targeted beliefs. That non-doxastic evidence provides justification for beliefs is now a common view in epistemology (Pollock  & Cruz, 1999; Pryor, 2000). Typically, the types of nonbelief states that can provide justification for beliefs are perceptual states, which are representational but not necessarily doxastic. Some authors hold that moral intuitions are like this, too

82  The Epistemology of Moral Bioenhancement

(Huemer, 2008). Moral intuitions are “seemings,” or representational states that are not necessarily beliefs. As such, they provide prima facie justification for moral beliefs. Other authors go even further and claim that perception can represent moral content (Werner, 2016, 2020) just as it can represent color content. To solve the Reflection Problem, an administration of moral bioenhancement would have to intervene at the level of non-doxastic evidence—it would have to induce in the person evidence for their moral beliefs. Or, in other words, it would have to induce moral intuitions.

The Moral Hallucination Problem The Reflection Problem can be resolved by intervening on a person’s nondoxastic moral representations, those representations that constitute evidence, and therefore provide justification, for the targeted moral beliefs. But if the administration of such an intervention is overt, a second problem arises. The problem is that knowing that one’s moral intuitions have been enhanced is akin to knowing that one’s perceptual representations have been manipulated. For example, one might have the moral intuition that they ought to stop eating meat. But if they also know, or even simply believe, that that intuition is a result of a pill they swallowed or an injection they received, their justification for the resulting belief would be undermined in the same way that knowing one is hallucinating bats swooping overhead undermines the justification for the belief that bats are swooping overhead. It’s of course possible to have justification to believe the contents of hallucinations. Justification is not a matter of truth or falsity, but of what one has reason to believe. Visual representations of bats swooping overhead, in the absence of a reason to think otherwise, provide such justification. But knowing that one is hallucinating provides one with defeating reason, undermining any justification one might otherwise have. The same is true for moral intuitions and the beliefs that they provide evidence for. Knowing that one is “hallucinating” the intuition that eating meat is wrong defeats the justification one has to believe that eating meat is wrong, which in turn makes it less likely that one will refrain from eating meat, rendering the moral bioenhancement ineffective. The only way to resolve the Moral Hallucination Problem is to administer the moral bioenhancement in such a way that the person receiving it doesn’t have reason to believe that their moral intuitions result from the intervention. The only way for this to occur is for the moral bioenhancement to be administered covertly, without the person receiving it believing that they are receiving it. Resolving the Reflection Problem requires that the moral bioenhancement intervene upon non-doxastic moral intuitions. If this intervention occurs overtly, then the Moral Hallucination Problem arises, which can only be resolved with a covert administration of the moral bioenhancement. The case for an overt administration gets even worse, however, as there are

The Epistemology of Moral Bioenhancement 83

three additional problems, the only solutions to which are to administer the moral bioenhancement covertly.

The Trustworthiness Problem The next problem that an overt administration of compulsory moral bioenhancement faces is the Trustworthiness Problem. This is the simple problem that the people who receive the enhancement are unlikely to trust the source of the intervention. While it is false that there are no moral experts and that moral disagreement is widespread, the belief that these are true is widespread. Most people wrongly believe that there are no moral experts, and they wrongly believe that genuine moral disagreement is widespread. That these beliefs are widespread is so obvious it is not worth defending. The burden of evidence is instead upon the person who thinks that people do believe that there are moral experts (outside of their own moral community) and that people do believe that there is widespread moral agreement. The beliefs that there are no moral experts and that moral disagreement is widespread undermine the trust one may have in the people administering the moral bioenhancement. Unless one agrees with the moral positions of the administrators, one is likely to mistrust the authority and role of the administrators of the moral bioenhancement. They are instead likely to judge the intervention as coming from an unreliable or erroneous source. Whether such attitudes toward the administrators are justified is irrelevant; what matters is that people will have them and that these attitudes undermine the person’s own justification for their newfound moral beliefs. And, as above, when one lacks justification for their moral beliefs, however, those beliefs were generated, the beliefs themselves are unlikely to result in the target behavior. Indeed, the mistrust may even result in them doing the opposite. One type of moral bioenhancement that may avoid the Trustworthiness Problem is the sort of intervention that induces an undesirable biological response to the disposition to do something wrong. For example, a drug may be administered that causes the person to vomit in response to neurochemical precursors to aggression. When a person is about to behave aggressively, one vomits. Since the person would generally want to avert vomiting, they won’t behave aggressively. Whether the person mistrusts the person or people administering the moral bioenhancement may be irrelevant since the behavior is not a result of moral reasoning or motivation. An intervention that conditions behavior in this way may avoid the epistemological problems associated with compulsory overt moral bioenhancement. But it may not be effective at preventing or mitigating the suffering that results from the great threats to humanity. First, unless everyone is so conditioned, then such an intervention may make a person vulnerable. Aggression is sometimes morally appropriate, and occasionally morally

84  The Epistemology of Moral Bioenhancement

obligatory. If a person is attacking my children, I have a moral obligation to aggressively protect them from that attack. If that aggression is cut short by a classically conditioned aversion to aggression, then the intervention prevents me from satisfying my moral obligation. It downgrades my moral capacities. For every moral capacity that might be conditioned, either with positive or negative reinforcement, there are circumstances in which that capacity is better exercised in atypical circumstances. Sometimes aggression is obligatory. Sometimes giving to charity is impermissible (such as when it would go to a bad cause). Enhancements that condition behaviors will make people vulnerable to these moral violations, given that the conditioning won’t be sensitive enough to discriminate between the times in which aggression is called for and the times when it isn’t. Unless the enhancement program can also manipulate all of the external circumstances which may dictate the atypical exercise of the conditioned capacity, then manipulating capacities in this way will make people vulnerable to these moral violations. This is not to say that enhancements that use classical conditioning are ineffective at manipulating some moral capacities, just that they may be better used for treating individual pathologies. For example, a better use of biologically conditioning moral capacities may be to administer the drug that induces vomiting in response to predicted aggression to people with antisocial personality disorder. Enhancements that pass through the process of moral reasoning and motivation make better interventions for widespread use, because they may better empower the person to navigate the nuances of proper moral behavior. An enhancement that causes a person to be more cooperative and less likely to defect doesn’t also entail that they will never be aggressive in the face of threats. Classical conditioning would. Just as making a compulsory moral bioenhancement program covert resolves the Moral Hallucination Problem, it also resolves the Trustworthiness Problem. If people don’t believe that they are receiving a moral enhancement, they won’t be in a position to mistrust the source of their moral intuitions and moral beliefs. In turn, the epistemology of these intuitions and beliefs would be preserved, making the performance of the targeted behavior much more likely. The same is true of the next two problems.

The Rational Irrationality Problem Drawing on Downs (1957) and Caplan (2001, 2003), Huemer (2010) claims that the most significant social problem facing humanity is the problem of rational irrationality, which is the idea that it can be instrumentally rational to be epistemically irrational. It may be in a person’s individual interests to hold epistemically irrational beliefs. Given that we have some limited control over our beliefs, it can be in a person’s interests to hold certain beliefs about current affairs and political problems and they may do things to pursue these beliefs. However, to hold epistemically rational beliefs about

The Epistemology of Moral Bioenhancement 85

these things, they must incur a cost, such as that associated with finding and weighing countervailing evidence. It takes time and effort to hold epistemically rational beliefs. It takes time and effort to hold true beliefs about climate change or the spread of infectious disease and how to prevent it. It may be in one’s practical interests to instead allow their cognitive biases to take over and simply believe that which is more convenient. Doing so consumes less of one’s time and effort, and it may make living in one’s social group easier and more productive. It is far easier to believe that climate change isn’t a significant problem or that masks don’t prevent the spread of the coronavirus, which implies that some of one’s practical interests lie in holding these beliefs. Huemer’s claim that our tendency to be rationally irrational is the most significant social problem is supported further by the idea that it undermines our ability to solve any big problems. We can’t expect to solve social problems without having true and, ideally, epistemically rational beliefs about them. This requires that it is instrumentally rational to hold true beliefs. The problem is that it is often not instrumentally rational to do so. Resolving problems like climate change and pandemics requires solving political disagreements. Administering a compulsory moral bioenhancement would also require solving political disagreements if its administration were overt. Thus, administering overt compulsory moral bioenhancement program would require overcoming the problem of rational irrationality. For many people, it may be instrumentally rational to hold epistemically irrational beliefs about moral bioenhancement and its administration. It may be instrumentally rational for some to believe that the administrators are nefarious agents of a shadow government of elites aiming to secure a permanent and total grip on power and control over others. Or it may be instrumentally rational to believe that one’s moral capacities are fine, in no need of improvement rather than hold the epistemically rational belief that, like everyone, one’s moral capacities are not fine. The problem of rational irrationality is particularly severe for the administration of moral bioenhancement, because the need for moral bioenhancement stems in part from our tendency to be rationally irrational. We need moral bioenhancement because we can’t solve social problems otherwise, because we are too often rationally irrational. Indeed, the problem of rational and irrationality and the need for moral bioenhancement may even have a common cause, namely our tendency to be extremely selfish, causing our individual practical interests to be at odds with the practical interests of others. If moral bioenhancement could be administered throughout the population, it may even mitigate the problem or rational irrationality. Moral post-persons are unlikely to be disposed to rational irrationality. But moral bioenhancement can’t achieve widespread administration until the problem of rational irrationality is resolved. As long as it is practically rational to hold epistemically irrational beliefs about moral bioenhancement,

86  The Epistemology of Moral Bioenhancement

it can’t be effectively administered. Maybe there are some solutions to the problem of rational irrationality that can be achieved prior to the administration of moral bioenhancement, such as incentives significant enough to make it practically rational to hold epistemically rational beliefs about moral bioenhancement. But even those solutions are likely to succumb to the very problem they aim to solve, as there would be significant political disagreement about those policies. Instead, one obvious solution to the problem of rational irrationality is to avoid it entirely and administer the moral bioenhancement covertly. On the one hand, if people don’t know it’s happening, then they can’t deliberately pursue their practical interests by avoiding it. Under a covert administration of moral bioenhancement, there is no opportunity for a person’s instrumental rationality to come into conflict with their epistemic rationality. On the other hand, the problem of rational irrationality would undermine an overt administration, rendering the administration ineffective in achieving its aim, which is to solve significant social problems.

The Free Rider Problem In Chapter 3, I discuss the social preferences associated with mitigating collective risk and how moral bioenhancement can intervene upon them such that we may be in a position to avert the harms associated with these risks. One of the central difficulties of collective action is figuring out how to solve the problem of free riders. Recall that free riders not only fail to contribute to public goods but also undermine others’ assurance that their contributions will not go to waste. This, in turn, encourages those people otherwise willing to contribute to instead refrain from contributing any further. From there, cooperation disintegrates, and the public good—aversion of collective risk—fails to be achieved. The Free Rider Problem may undermine the aversion of collective risk, but it also is likely to undermine the effectiveness of an overt administration of compulsory moral bioenhancement. I discuss this in more detail in the next chapter, but if compulsory moral bioenhancement is overt, people will simply fail to receive the intervention. They will do things to avoid getting it, in the way that people now avoid things like vaccinations, fluoride in the water, and masks to prevent the spread of the coronavirus. They will leave their own moral capacities alone, and allow others to bear whatever the burdens of receiving the moral bioenhancement are. In turn, those otherwise disposed to receive the intervention will see this free riding and, reasoning that receiving the intervention is unlikely to avert risk, withhold their own cooperation. In other words, if the administration is overt, people will free ride, which will undermine others’ assurance that their own contributions will be effective and they will subsequently withhold them. One might look to vaccines and see the relatively low rate of free riding and infer that we could expect similar rates of acceptance of moral

The Epistemology of Moral Bioenhancement 87

bioenhancement. But there are some crucial differences between vaccines and moral bioenhancement. One is that one may have greater incentive to have their immune system enhanced than they do to have their moral capacities enhanced, as the former enhancement protects the individual as well as others and the latter enhancement primarily protects others. A second difference is that vaccines, especially those that are compulsory, have been around for decades and have long track records of success, and have been integrated into the standard of medical care. Such is not the case with moral bioenhancement. A third difference is that vaccines are recommended very strongly by people who the public recognizes as experts. As I mentioned earlier, it is unlikely that the public will recognize anyone as an expert, not to mention come to an agreement over who the experts are and why they are so. So, while we might find the rate of acceptance of compulsory vaccines encouraging, there are significant differences between vaccines and moral bioenhancement that provide good reasons to think that the public is unlikely to accept the latter at a rate anywhere close to that of the former. I argued earlier that moral bioenhancement is necessary to solve the problems associated with collective action to avert collective risk. But acceptance of moral bioenhancement—even if it is compulsory—itself suffers from the same collective action problems. The solution in this case, however, is to make the administration of the moral bioenhancement covert. If it were administered covertly, the Free Rider Problem would simply go away.

Recommendation I have argued that the epistemology of moral bioenhancement recommends that moral bioenhancement programs be covert for them to induce the desired moral behavior in enough people to meet the pressing moral demands that necessitate the institution of the program. To be clear: the epistemology of moral bioenhancement requires that a compulsory moral bioenhancement program be administered covertly. An overt program will not work. One objection to covert moral bioenhancement is that it undermines autonomy. I  discuss the objection in detail in the next chapter, but Alex Zambrano claims that the epistemology of covert enhancement undermines autonomy. He claims that “under an overt program in which people are aware that they are being morally enhanced, people will be put in an epistemic position to combat the effects of the manipulation, and then either embrace or reject their newfound desires, values, and beliefs” (Zambrano, 2019, p.  4). But if the administration were covert, the enhanced would not be in this position. Overt and covert programs thus imply different epistemic positions for the enhanced. That they are in different epistemic positions is supposed to imply further that those being covertly enhanced

88  The Epistemology of Moral Bioenhancement

are less independent than those that are overtly enhanced. This lack of independence undermines autonomy, according to Zambrano. I have argued in this chapter that there is no such thing as an effective compulsory overt moral bioenhancement. The only way to make compulsory moral bioenhancement effective, because of our epistemic positions and dispositions, is for it to be administered covertly. Zambrano says this is beside the point, but it isn’t if we are to compare side by side covert and overt programs. For all the reasons I  discuss in this chapter, the epistemic position of those in an overt program as Zambrano conceives it would render the moral bioenhancement ineffective. If the moral bioenhancement program is not effective, then there it can’t be ethically justified. Effectiveness is intertwined with the enhanced population’s epistemic position. As the population’s epistemic position gets stronger—they are more able to recognize and manipulate their own reasons for belief, the less effective an overt program will be. The inverse is also true. As the population’s epistemic position improves, the Reflection Problem undermines the effectiveness of an overt moral bioenhancement program. Suppose the population is highly reflective, able to introspect not only their beliefs and the reasons that ground them but also their higher order belief and those grounding reasons. They are aware of the evidential support of these reasons and constantly adjusting their confidence levels so that they fit with their evidence. For such a population in such a strong epistemic position, the Reflection Problem will easily undermine the effectiveness of an overt moral bioenhancement. They won’t fail to notice a change in belief without a corresponding change in evidence. And even a change in evidence might not go unnoticed, as they may introspect that they have not had any experiences that would warrant a change in evidence. They would notice any attitude that results from this enhancement and notice its lack of justification. But as the population’s epistemic position weakens, so does the threat of the Reflection Problem. Consider a population that barely even has the concept of a mental state, whose members are mostly unaware of their own beliefs and other attitudes and the reasons that ground them. It’s not that they’re skeptical, just ignorant about their mental lives. For such a population, the Reflection problem might not be a problem at all. Members of this population wouldn’t notice at all the attitudes or capacities that an overt moral bioenhancement program brings about. For such a population, an overt moral bioenhancement program might be effective. But in fact people are not like this. We are much more similar to the other idealized population, the population that is in the strong epistemic position. We do so imperfectly and often irrationally, but we monitor and adjust our mental states according to our evidence. We are aware of our own thoughts and their reasons. Indeed, for a population that is not capable of having higher order awareness of their own mental lives, preventing

The Epistemology of Moral Bioenhancement 89

catastrophe may not even be a moral concern, as one prominent theory of consciousness requires such higher order awareness for conscious experience. Thus, lacking such an ability might undermine the population’s ability to experience pains and pleasures. And it is the prevention of pains and promotion of pleasures that ultimately justifies the prevention of catastrophe. We are clearly not like that. Our epistemic position may not be perfect, but it is strong enough that the Reflection problem would undermine the effectiveness of an overt moral bioenhancement program. Furthermore, a population in such a weak epistemic position would arguably lack the ability to have autonomous thought. The independence that seems required for autonomy seems to require at least some ability to have access to and the ability to manipulate one’s own mental states. Thus, if the overt moral bioenhancement is presumed to be effective, as it must be if we are to compare side by side covert and overt moral bioenhancement programs, then the population’s epistemic position must be much weaker than Zambrano imagines. Specifically, for an overt program to be effective, it must also be the case that the population has very limited access to and control of their own mental states, certainly more limited than what Zambrano needs to support his claim that people could embrace or reject their newfound beliefs. Because of the Reflection Problem, a person or population cannot both be effectively enhanced by an overt moral bioenhancement program and be in the epistemic position to introspect and manipulate their own mental states. It’s one or the other. In elucidating the epistemological challenges of compulsory moral bioenhancement, I assume, perhaps wrongly, that justification is an internal matter, something determined by what’s in a person’s head, rather than whatever external circumstances they happen to be in. One might think instead that justification is a matter of those external circumstances, such as whether one’s beliefs result from a reliable process. Reliabilism, however, will not save overt moral bioenhancement. The problem with overt moral bioenhancement is that knowing that one is being enhanced is plausibly interpreted as a defeater for justification. The Trustworthiness Problem undermines trust in the process; the Reflection Problem gives a person a reason to think that the belief resulting from the enhancement is false. Even externalist theories of justification have a “No Defeater” condition, which requires that for a belief to be justified one can’t have a defeater for the belief. Compulsory overt moral bioenhancement ensures that they do. Thus, whether justification is an internal or external matter, compulsory overt moral bioenhancement will be epistemologically problematic, so much so that it won’t bring about the targeted behaviors.

Chapter 7

Covert Moral Bioenhancement

I have argued that moral bioenhancement should be compulsory. In the previous chapter, I argued that the only way for a moral bioenhancement program to be effective is if the intervention is administered covertly, without the knowledge of the people receiving the enhancement. That was an epistemic argument. I argue for a similar conclusion in this chapter. I argue that if moral bioenhancement ought to be compulsory, it ought to be covert. But the “ought” in “ought to be covert” is a moral “ought.” In the previous chapter, the “ought” denotes a practical obligation: if we want moral bioenhancement to do what we want it to do, then it ought to be covert. Thus, there are multiple argumentative routes to the conclusion that moral bioenhancement ought to be covert. We have both a practical obligation as well as a moral obligation to covertly enhance the moral capacities of all of those people who are in a position to bring about the significant harm associated with climate change, the proliferation of weapons of mass destruction, and other existential threats. The argument in this chapter rests on the claim that moral bioenhancement is a matter of public health. Since it is a matter of public health, we should look to public health ethics to determine whether and how it should be administered. When we do that, I argue, we find that a covert compulsory moral bioenhancement program is more ethical than an overt program. After supporting this conclusion, I address a range of objections, including those related to the logistical difficulty of implementing such a program and to the alleged threat of moral bioenhancement to autonomy.

Compulsory Moral Bioenhancement Is a Matter of Public Health The case that a safe, effective, and necessary moral bioenhancement is a matter of public health is rather straightforward. Moral bioenhancement is a health intervention. The main aim of any health intervention is to improve well-being, but sometimes the intervention aims to improve the well-being of an individual and the individual’s close social circle, while other times

Covert Moral Bioenhancement 91

the intervention aims to improve the well-being of populations and only secondarily the well-being of a particular individual. One significant ethical difference between interventions aimed at individuals and interventions aimed at populations is the range of values that one considers in whether to administer the intervention. In the case of an intervention on an individual’s health, these considerations include the patient’s preferences, the promotion of what’s good for the patient and the prevention of what’s harmful, and the personal relationships the patient has with others. In the case of interventions on a population’s health, the values of concern are those that apply to the whole population, such as whether the intervention is administered fairly, promotes liberty or opportunity, uses resources appropriately, and engenders a net gain of well-being. The aims of a compulsory moral bioenhancement program are to prevent suffering associated with large-scale collective risks, the sort of suffering concomitant with ultimate harm. Ultimate harm is a condition of large groups of people. Compulsory moral bioenhancement is a health intervention aimed at preventing large groups of people from being harmed, an intervention aimed at the public’s health. Therefore, whether a compulsory moral bioenhancement program ought to be instituted is a matter of public health, and its ethical permissibility ought to be determined by the ethical frameworks of public health interventions. The ethics of using moral bioenhancement to prevent suffering associated with large-scale collective risks do not resemble the ethics of a terminal patient’s choice to die on one’s own terms, or the decision of a pregnant woman to abort a fetus, or an infertile couple’s decision to use new technologies to reproduce, or even an athlete’s choice to use performance-enhancing drugs. The primary difference is that the effects of moral bioenhancement are much more widespread than the effects of an individual’s medical treatment. Thus, the frameworks of biomedical ethical reasoning that have been developed to investigate the ethics of decisions that are limited to the individual or the individual’s close social group should not be expected to also adequately inform the ethics of issues that are of such great public concern. The move to public health ethics is significant because the values considered in administering the intervention are different from those of interventions aimed at individuals. In particular, for public health ethics, the restriction of individual liberty is more permissible than it would be if the public’s health weren’t at risk. This fact is obvious when we consider the ethical permissibility of quarantining a patient with a highly infectious disease and the ethical permissibility of quarantining a patient with a disease that causes equal individual harm but is not infectious at all. In the former case, the quarantine is ethically permissible; in the latter, it isn’t. The restriction in liberty is justified by the greater expected utility of preventing the spread of infection. Other examples of the permissibility of liberty restrictions for the sake of the population’s health are the fluoridation of

92  Covert Moral Bioenhancement

public water supplies and mandating the vaccination of children if they are to attend public schools.

Public Health Ethics Frameworks The development of frameworks for public health ethics is relatively recent. Several distinct frameworks have been developed, advocating for the consideration of a wide variety of values that ought to be considered when implementing a public health program. Given that public health interventions are intertwined with liberty interests and potentially competing interests in utility, it is not surprising that the frameworks all call for the consideration of these two values. Ross Upshur (2002) developed a framework that proceeds in the way similar to other discussions in bioethics, via principles of action. He claims there are four of these: that liberty restrictions can only be justified by the prevention of anticipated harm, that the least liberty-restricting means must be used, that public health officials must reciprocate for compliance with the program, and that the program should be transparent. Other authors formulate their frameworks as a way to balance a range of potentially competing values. Childress et  al. (2002) suggest that the values that should be considered include maximizing benefits and minimizing harms (utility), distributing benefits and burdens fairly, respecting autonomous choices and liberty of action, protecting privacy, keeping promises, being transparent, and building and maintaining trust. Like Upshur, they also give several conditions for infringing on one of these values, or how to balance the different values. In fact, the conditions are similar in content to Upshur’s principles. The conditions are that the infringement be necessary, effective in achieving the desired public health benefit, that benefits proportionally outweigh the infringement, that it be the least restrictive means, and that it be publicly justified. The frameworks of Upshur and Childress et al. place the balance of utility and liberty as the central feature of determining the ethics of a public health program. Others broaden the range of values even further. For examples, Selgelid (2009) argues that rather than adhere to the principles above, the ethics of a public health program should be determined by the balance of utility, liberty, and equality, as it is possible that sometimes the promotion of equality outweighs potential liberty restrictions (it’s possible that each of the three values outweighs the other two). And Grill and Dawson (2015) argue that a wider range of values ought to be considered. They write that “some values that we would ourselves include in many public health decisions are individual health, population health, health equality, individual liberty, solidarity, social trust, and material wellbeing,” as well as the honoring of duties, the noninfringement of rights, and the expression of virtues. They go on to provide a method of making decisions about public health programs that is neutral between these values.

Covert Moral Bioenhancement 93

Collecting the different frameworks, we have several principles or conditions, and a wide range of values to consider. The conditions are that it’s necessary, effective, proportional, publicly justified, and the least restrictive means. The values considered are liberty, utility, equality, fairness, transparency, promotion of individual and population health, trust, solidarity, respectful of autonomy, and honoring of duties.

Covert Moral Bioenhancement The present issue is not whether the public health program of administering moral bioenhancement ought to occur; it’s a matter of how it should occur. Let us suppose that if it were to occur overtly, it would occur similarly to vaccination programs for children: at the age where the moral bioenhancement is safe and effective, children would receive the moral bioenhancement from their pediatrician or family physician or community health department, and that would be that. That information would then go on their health records, and they’d go on with their more moral lives. Let us also suppose that if the program were administered covertly it would be conducted in a similar fashion. When children are scheduled to receive vaccinations, they are at the same time given the moral bioenhancement, but neither the children nor their parents or guardians are told about the moral bioenhancement and it doesn’t go in their health records. The administration of it could be double- or even triple-blinded, so that only a few individuals are aware of the moral bioenhancement. Everyone would go on with their lives unaware of the moral bioenhancement. The question is: which is the most ethically desirable scenario? I argue it is the second scenario in which the moral bioenhancement is administered covertly. Values in Public Health Ethics

The task is to compare the two methods of administering the compulsory moral bioenhancement and see which one is the most ethically permissible. Consider first the fact that as compared to a covert moral bioenhancement program that is blind to everyone except a few, an overt program would reduce the expected utility of the program. It would reduce the expected utility of the moral bioenhancement program because if people knew that they were being morally bioenhanced, at least some of them would fail to receive the bioenhancement. They would request exemptions from the policy on the grounds that it conflicts with their religion or their personal convictions, or they would falsely believe that the moral bioenhancement leads to various disorders or diseases unrelated to the intervention. People would slip through. Some would slip through because of failing to pay attention, while others would outright refuse the intervention. That this

94  Covert Moral Bioenhancement

would happen is obvious when we consider policies on vaccination or quarantine: people refuse vaccines or otherwise fail to get them, and people slip through quarantines and other methods of isolation. If the moral bioenhancement were overt, the expected utility would be less than it would be if it were covert. It’s not that the utility of preventing ultimate harm is less; it’s that the expectation that the moral bioenhancement will succeed in preventing it is lower. The more people that slip through the compulsory moral bioenhancement, the lower the expectation that ultimate harm will be prevented. If the program were covert, people would be unaware of the intervention, and so would not be in a position to avoid it, resulting in many fewer people failing to receive the intervention. Both overt as well as covert compulsory moral bioenhancement programs would restrict the range of moral attitudes, dispositions, and behaviors of its participants. The range of moral attitudes, dispositions, and behaviors that would be restricted would be the same for both types of program, as it is the intervention upon these that is presumably necessary to prevent ultimate harm. So the extent to which the interventions themselves are liberty-restricting, the liberty restrictions will be equal between a covert and an overt program. But for overt compulsory moral bioenhancement programs, participants would also know that their moral attitudes, dispositions, and behaviors are being intervened upon. For some of these people who know that their moral capacities are being restricted, they will desire to not be so restricted. Thus, the desires of these people will be frustrated, which results in suffering. If the program were covert, the people who desire to not have their moral capacities restricted wouldn’t be aware of any restriction, so from their perspective, the desire to not be restricted wouldn’t be frustrated, which means they wouldn’t suffer from knowing that they are participating in a compulsory moral bioenhancement program. This point rests on the idea that, all things considered, there is disutility in not just having desires frustrated, but also in knowing or believing that they are frustrated.1 The same point could also apply to other public health programs, such as those that require people be vaccinated. Some people desire to not be vaccinated. When these people knowingly receive a vaccination— to attend school, for example—their desires are frustrated, and this frustration causes suffering. If it were possible to achieve all of the benefits of vaccination without having to cause the suffering that results from believing that one is vaccinated, then that would be preferable to actual vaccination procedures. To put it another way, in comparing an overt program with a covert program, the disutility of an overt program is greater, because in addition to the disutility of the fact of desires being frustrated, there’s also the further disutility of believing that they are frustrated. This latter disutility is absent in the case of a covert program.

Covert Moral Bioenhancement 95

The frameworks of public health ethics require balancing the expected utility with the promotion or demotion of other values. One of these values is liberty. A covert compulsory moral bioenhancement program is less liberty-restricting than a similar overt program is. The discussion above points out that if the moral bioenhancement program were overt, inevitably some people would refuse or otherwise fail to receive the intervention. Because the program is compulsory, however, policies would be required to compel such people to undergo the intervention. These policies would take the form of isolation (e.g., preventing dissenters from fully participating in society), taxes or fees as penalties, or, in severe cases, imprisonment. All of these methods of compulsion restrict liberties. Moreover, given that the expectation of preventing ultimate harm is lower for an overt program, the potential for more significant liberty restrictions is greater, as our liberties may be more likely to be restricted by our harsher environments that result from having undergone ultimate harm. And upon one’s death from ultimate harm, one’s liberties are fully restricted—dead people have no liberties. These liberty restrictions should not be overlooked. The world in which ultimate harm has occurred is a world in which people have very few liberties. Compared to a covert program, an overt program makes this world more likely. If the program were covert, however, enforcement of liberty restrictions would be unnecessary, as people would be unaware of the intervention in the first place and so there would be no need for such policies to compel participation. And the potential for the liberty restrictions that ensue from having undergone ultimate harm is also lower. Some may think that utility and liberty are the most important values to consider in implementing a public health program. Balancing the liberty and utility of a covert program and an overt program, the better balance is that of a covert program, because it has greater expected utility and is less restrictive of participants’ liberties. But even when other values are balanced with the utility of preventing ultimate harm, a covert program is preferable. A covert program better promotes equality, because by keeping the program covert to everyone, the program ensures that all participants are treated equally. It is totally impartial. In an overt program, it would remain open that some populations are in a better position to avoid the intervention, such as those that could easily afford the penalties imposed for refusing, or those that do not rely on public health clinics. Another potential source of unequal treatment is that likely many physicians would disagree with the policy, putting them in a better position to refuse to administer the moral bioenhancement. Based on this variance of attitudes within physicians, it is likely that the treatments would be administered unequally.2

96  Covert Moral Bioenhancement

Similarly, a covert program would be fairer than an overt program. Because everyone would receive the moral bioenhancement, there is no population that would be forced to bear a disproportionate burden. Some populations may bear a greater burden, such as psychopaths whose moral psychology must change to a greater degree than others.’ For these populations, the burden of the moral bioenhancement would be greater (similar to a higher tax), but it is not a disproportionate burden. An overt program, however, may encourage others to find ways to avoid receiving the enhancement, meaning that they wouldn’t be required to bear any burden, which is unfair.3 A covert program would also better promote population health than an overt program. As discussed above, the expectation that ultimate harm will be prevented is lower in an overt program, so the threat to population health is greater. For the same reason, an overt program may not promote individual health as much as a covert program would. Otherwise, the two programs promote population and individual health equally. The promotion of other values may also turn out to be equal between overt and covert programs. Solidarity would presumably be the same between the two types of programs, though it is possible that it would be greater in an overt program: the administration of the program could induce people to join each other in protesting it. And, if revealed, a covert program would undermine social trust. Indeed, the disclosure of a covert program may undermine social trust so much that it could destabilize the society and its government. In this respect, the potential for such destabilization is much lower in an overt program. However, in the event of ultimate harm, society is also destabilized. I claim above that an overt program is less likely to prevent ultimate harm. So, although in one respect an overt program makes destabilization less likely, in another respect it makes it more likely. It’s therefore not clear that an overt promotes social stability more than a covert program does, though it may better promote social trust. Another value is that of transparency. Obviously, an overt program will better promote this value. Comparing the two programs, a covert program wins on the values of utility, liberty, equality, and fairness. They are approximately equal in promoting health. An overt program wins on the values of promoting transparency and solidarity and potentially trust. What about the values of honoring duties and respecting autonomy? I assume that many readers are uncomfortable with the notion of a compulsory covert moral bioenhancement program and that the source of this discomfort is the judgment that such a program undermines a person’s autonomy, which we have a duty to promote. This, I  assume, constitutes the primary reason that one would object to the proposal of making

Covert Moral Bioenhancement 97

the compulsory moral bioenhancement covert. I  address this objection shortly. Conditions in Public Health Ethics

In addition to the values discussed above, there are plausible conditions that need to be satisfied for a liberty-restricting public health program to be justifiably implemented. One of the two foundational assumptions is that a moral bioenhancement program is necessary. But the assumption does not distinguish between a covert program and an overt program. Thus, it is best to say that both types of programs would satisfy the condition that the program be necessary. Another assumption is that the compulsory moral bioenhancement is effective and, again, the assumption makes no distinction between the effectiveness of a covert program versus the effectiveness of an overt program. So both satisfy the condition that the intervention be effective. A third condition is that the program be proportional to the threat. Whether a covert or overt program is more proportional to the threat depends on the burdens the programs require subjects bear. As I argue above and below, overt programs require that those receiving the enhancement bear greater burdens. Thus, if I am right about these other values and conditions, then a covert program is more proportional to the threat of ultimate harm than is a covert program because it burdens subjects less. The covert and overt programs diverge with the next two conditions: that the program be the least restrictive means and that it be transparent. As argued above, a covert program restricts liberties less than an overt program does. So an overt program doesn’t satisfy that condition. But it does satisfy the condition that the program be transparent, whereas the covert program doesn’t. Thus, as measured by the collection of values and conditions that public health ethics frameworks propose, the two types of programs are similar. But the differences are key. A covert program better promotes or preserves utility, liberty, equality, and fairness. It fails to be transparent, promote trust, or promote solidarity. Whether one thinks that a covert program is preferable to an overt program depends on whether one thinks that the values of utility, liberty, equality, and fairness are more valuable than transparency and the promotion of trust and solidarity. I believe they are, and for this reason, a covert program is preferable to an overt program. I offer no criteria for ranking these values. But utility, liberty, equality, and fairness have enjoyed a more central role in moral and political philosophy than have the values of transparency, trust, and solidarity. So those wishing to claim that an overt program is ethically preferable have the burden of establishing that transparency, trust and solidarity are more valuable than utility, liberty, equality, and fairness.

98  Covert Moral Bioenhancement

Implementation

I have argued that a covert moral enhancement program is preferable to an overt program, and have done so on ethical grounds. Though the argument is valuable even if it is impossible for a covert program to be implemented, the practical implications are much greater if such a program can be implemented. It is admittedly difficult to provide examples of other public health interventions of the sort a covert moral enhancement program would be. If I  know about them, then they weren’t very covert, and so wouldn’t be good examples of how a covert moral enhancement program should be administered. But there may be ways to administer the moral enhancement to the relevant populations, without those populations knowing about the enhancement. The method of delivery would depend on the mechanism of action of the enhancement, but one possible way of distributing it to the relevant populations is by way of the public water supply. Another way it could be distributed is by packaging it with various vaccines, while eliminating most exceptions. Or perhaps it could be distributed through forced air systems in public buildings, or some combination of these. There are of course members of the population that don’t use public water, don’t get vaccinated, and don’t visit public buildings. For example, there are many people in rural areas who get all of their water from wells drilled into natural aquifers. These people wouldn’t regularly come into contact with the public water supply. But they would still spend time in public places or get vaccinated, both opportunities to receive the enhancement. There are still individuals who never use public water, never get vaccinated, and never go out in public. And if the enhancement were only administered through these channels, such individuals would never receive the enhancement. The threat of ultimate harm is greater now than it has been because people now have greater access to information and materials that are capable of triggering ultimate harm, such as advancing technologies or widespread use of fossil fuels. People who refrain from public services may not have widespread access to the information and materials that make ultimate harm a threat. There are large segments of the human population who lack access to utilities, vaccinations, or public services of any kind. And if these are the channels of distribution, then these segments of the population will not receive the enhancement. But these segments of the population aren’t the ones who increase the threat of ultimate harm—they lack access not only to public services but also to the information and materials that make ultimate harm a threat.4 In some cases, individuals in these segments could travel to places where the information and materials are accessible, but in so doing they would be traveling to a place with, and very likely use, public services, and would receive the

Covert Moral Bioenhancement 99

enhancement. It’s the bad actor with access to a published recipe for bird flu and the intelligence and infrastructure to produce it or the person who consumes the resources which contribute to climate change that we have to worry about, rather than a member of a recently discovered Amazonian tribe or someone who never engages with the public. Though the methods of distribution described above occur by way of public utilities or public spaces, there is no implication that the administration of a covert moral enhancement must be a matter of public policy. Governments and bureaucrats are possible administrators, but the argument doesn’t require that a covert program be carried out by any particular individual or group. A covert program would indeed require excellent coordination and control to not only maximize the number of people who receive enhanced moral capacities but also to maintain secrecy. Such coordination is not logically impossible, not metaphysically impossible, and not even practically impossible. It may even be easier to administer than other attempts at secrecy, if the administrators themselves are being enhanced, as their moral motivation to maintain secrecy may be stronger than it otherwise would be. Keeping a covert program covert would be a challenging obstacle. But just because it would be a challenging obstacle doesn’t mean that my argument is unsound. Further, if one is concerned about the difficulty of implementing such a program, and for this reason doubts my argument, consider that it is also no easy task to cobble together a range of public programs which individually are aimed at preventing one potential avenue of ultimate harm but collectively aimed at preventing ultimate harm. It’s not clear that would be any easier to implement. Such a collection would need to include programs aimed at stopping or reversing climate change, programs preventing the proliferation of nuclear arms and other devastating weapons along with the information used to manufacture them, programs protecting democratic institutions from electing powerful bad actors, programs preventing dictators from becoming powerful bad actors, programs preventing scientists from developing, even accidentally, threatening materials or organisms, programs that prevent amok AI, programs that allow humans to colonize space before the death of the sun, and so on. It seems reasonable to me that a covert program can be implemented. Others may not be convinced. But this disagreement isn’t about whether it’s true that if moral bioenhancement ought to be compulsory, it ought to be covert. One could instead argue on the basis of a covert program’s impossibility that it’s false that a covert, and therefore compulsory, moral bioenhancement ought to be carried out. This argument requires that it be true that ought implies can. The ought implies can principle is that one ought to do something, only if one is able to do that thing. It’s controversial that the principle is true. But if it is true, and if a covert program is impossible, then it’s false that moral bioenhancement ought to be covert (by the ought

100  Covert Moral Bioenhancement

implies can principle). And if it’s false that moral bioenhancement ought to be covert, then it’s false that it ought to be compulsory (by the truth of the conditional I argue for). So, the impossibility of covert moral bioenhancement doesn’t refute my argument that if moral bioenhancement ought to be compulsory, then it ought to be covert (because the falsity of a consequent doesn’t entail the falsity of the conditional). But it could refute that moral bioenhancement ought to be compulsory, by way of the principle that ought implies can. In the third chapter, I argue that the epistemic burdens associated with mitigating significant collective risk are so overwhelming that enhancement is necessary to overcome them. If this is true, and it’s true that a compulsory covert moral bioenhancement program is impossible, then we are doomed. We won’t be able to mitigate the risk. If we are to have any hope at all, either I must be wrong that the epistemic burdens are overwhelming or my objector must be wrong that a covert program is impossible. We might also compare the relative epistemic burdens associated with mitigating the collective risk without the aid of moral enhancement with the epistemic burdens associated with implementing a covert compulsory moral bioenhancement program. The total epistemic burdens associated with mitigating the collective risk are much greater than that of a covert enhancement program. First, to mitigate collective risk without the aid of enhancement, the epistemic burdens of billions of people need to be overcome. And for any given individual these are not particularly easy burdens to overcome. People need to overcome the epistemic burdens associated with reducing their consumption and emissions and how to cooperate, among others. Second, for a covert enhancement program to be implemented, very few people will bear epistemic burdens. Granted, the particular epistemic burdens this small group of people will need to bear are daunting. These include the epistemic burdens associated with developing the enhancement, including its testing for safety and efficacy, along with the epistemic burdens Table  Scorecard of Overt versus Covert Overt Effective Necessary Liberty Utility Equality Fairness Transparent Trust Solidarity Autonomy

X

X X X

Covert X X X X X X

X

Covert Moral Bioenhancement 101

associated with its covert implementation. But the fact that these significant burdens are shared among a much smaller group of people suggests that the total epistemic burden of implementing a compulsory covert moral bioenhancement program is much lower than the total epistemic burden of mitigating collective risk without the aid of enhancement. However, if both are impossible, then there is no hope in preventing the harm associated with climate change, pandemic, or some other significant threat to our collective well-being.

Objections It’s Unrealistic

Above I warn that some readers may find the assumptions to be so demanding that they push the argument into the unreal. What is far more likely is that the intervention is not fully safe or fully effective. What happens when we can only be 80% sure that it will be effective, or when 10% of those receiving the enhancement will suffer a terrible side effect? Though the certainty of whether the intervention will be safe or effective impacts how the different values are balanced, I don’t think the impact is significant enough to drastically alter the balance. Suppose there’s an equally low chance that an overt and covert program is effective in preventing ultimate harm, say 10%. In that case, as long as it is necessary to prevent ultimate harm, then a covert program is preferable. If there’s a 1% chance, or even substantially lower, of it being effective, then ultimate harm may be overdetermined, in which case no program should be instituted. If an overt program were considerably more likely to be effective, or considerably safer, then it would be preferable. But if the safety and effectiveness of an overt program are similar to that of a covert program, and a moral bioenhancement program is necessary to prevent ultimate harm, then a covert program is preferable. Additionally, some may find the possibility of ultimate harm remote, and if not remote, at least not a threat that will emerge in the near future. Ultimate harm is not inevitable—humans may die away slowly and quietly, without experiencing the significant suffering and loss the prevention of which justifies moral bioenhancement. But there are many potential avenues of ultimate harm: bioterror, nuclear war, devastating climate change, meteor strikes, and geologic disasters are just a few possibilities. At the very least, if the human species (or its evolutionary descendants) survives to the death of the sun, ultimate harm is likely, unless we colonize space, prevent the death of the sun, or become disembodied. Given the difficulty in determining how likely ultimate harm is, it’s difficult to say whether a covert moral bioenhancement program will ever need to be administered. What likelihood of ultimate harm is sufficient to trigger

102  Covert Moral Bioenhancement

the program? A 50% chance of ultimate harm seems sufficient to trigger it, but a one-tenth of 1% chance seems insufficient, though humans often go to great lengths to prevent unlikely events, even when the costs associated with those events are comparatively low. It’s Dishonest

If preventing ultimate harm requires lying, even by omission, then lie we must. Even if the dishonesty is not necessary and it is unambiguously wrong to be so dishonest, the wrongness of the dishonesty does not outweigh the utility of preventing ultimate harm. To argue that it is not necessary to be dishonest about the moral bioenhancement is nothing more than the claim that a covert moral bioenhancement is not necessary for the prevention of ultimate harm. However, I am not claiming that a covert moral bioenhancement program is necessary. I am arguing that it is preferable to an overt program.5 To measure this preference, I have appealed to frameworks of public health ethics. And it seems according to these frameworks that a covert program is preferable. One may also think that there is disutility in dishonesty. Thus, when do our balancing of values, we must include this disutility that comes along with a covert moral bioenhancement program. The alleged disutility of the dishonesty would have to outweigh the risk of ultimate harm. It seems obvious to me that it doesn’t, that it is more akin to a little white lie when compared to the potential risk of being honest. But others may think that it’s worth the greater risk to the existence of humanity to be honest. It’s Unlike Other Public Health Programs

Compulsory moral bioenhancement is a matter of public health. According to this objection, other public health programs, in particular those that require participation, are more precisely targeted at those who need them. For example, only children are required to get vaccinated, because they are the people who need the intervention. Since not everyone needs moral bioenhancement, its implementation is disanalogous to other public health programs. While it’s true a covert program would be unlike anything we know about (though it could be like programs we don’t know about), it’s not true that other compulsory public health programs only target those who need the intervention. Not all children need vaccinations for their own health— some might never get the flu or chicken pox or polio in the absence of the vaccine. But making the vaccine compulsory, even in the absence of individual need, protects the public. Quarantines routinely isolate people who have no indication of disease. People who brush their teeth regularly and abstain from sugary foods don’t need fluoridated water. Some people may

Covert Moral Bioenhancement 103

not need moral bioenhancement (though I suspect this number is very low, as most of us contribute to various potential catalysts of ultimate harm). An individual’s lack of need for a public health intervention does not imply that requiring that individual’s participation in the intervention is unjustified. Zambrano (2019) has criticized my argument on the grounds that it fails to satisfy the conditions for permissible public health liberty restriction. He first identifies three conditions for a public health intervention to permissibly restrict liberty. These conditions are (a) that the harm is specific and identifiable; (b) that the mechanism by which the harm will occur is identifiable; and (c) that there is direct empirical evidence that the threat of harm is immediate or imminent or that it would be an immediate threat if there were no intervention. For example, according to Zambrano, quarantine satisfies these conditions: the spread of harmful disease is a specific and identifiable threat; we generally know how this spread occurs; and we have lots of evidence that these diseases can spread without proper isolation procedures. Pertussis vaccines also satisfy the conditions. The harms associated with pertussis are specific and identifiable; the mechanism by which it spreads is identifiable; and we have direct empirical evidence that this threat was immediate or imminent when “many children failed to become vaccinated.” However, according to Zambrano, the interventions that aim to prevent ultimate harm don’t satisfy these conditions. The threats themselves are not specific or identifiable; we don’t know their mechanism of action; and there is poor evidence that they are immediate or imminent. Since the problem moral bioenhancement doesn’t satisfy Zambrano’s conditions, it fails to be a justified public health intervention. In an earlier article (Crutchfield, 2019), I grant that moral bioenhancement fails to meet Zambrano’s conditions. Instead, I undermine the conditions themselves by showing that there are clearly justified public health interventions that fail to satisfy the conditions. The conditions are too strong, for they would prohibit interventions such as taxes, fluoridated water, and the MMR vaccine. But granting that moral bioenhancement fails to meet his conditions was merely for the sake of argument. I don’t think it’s true. Even if Zambrano’s conditions were necessary for permissible libertyrestricting public health interventions, the public health threats that warrant covert compulsory moral bioenhancement would meet them. It may be true that “ultimate harm” doesn’t entail any identifiable threat, but climate change does. Oceans will rise; cities will flood; people will drown or migrate and suffer the pains associated with migration. Food and freshwater will be in shorter supply; arable land will become arid; famine will set in; states will become destabilized; security and safety will be compromised; and so on. These are all identifiable threats. The same could be said for nuclear war or a weapon of mass destruction. An antisocial leader of a state or terrorist group will decide that he has been slighted and retaliate with a nuclear weapon, starting a chain of nuclear strikes. One person or a few

104  Covert Moral Bioenhancement

people will break quarantine, travel quickly and easily to another nation, and set off a pandemic. These are specific, identifiable threats. We also know very well the mechanism by which these occur. For climate change, it is humans’ innate inability to do much about collective risk. Our propensity to defect causes the suffering that results. For nuclear war, the mechanism is the failure to protect nuclear weapons or other weapons of mass destruction. For spreading novel diseases, there is the biological mechanism, for example how a disease moves from an animal population to a human population, but also the mechanism of communication (international travel and physical contact, for example). Finally, there is empirical evidence that if we don’t do anything we will suffer. If we don’t intervene in any way, there is likely to be catastrophic suffering from climate change. This is supported by empirical evidence. The same is true of novel diseases, such as the coronavirus. I have argued that covert compulsory moral bioenhancement is an intervention necessary to prevent the collective risk that these threats present. Zambrano’s conditions are not necessary for permissible liberty-restricting public health interventions. But even if they were, covert compulsory moral bioenhancement would meet them. It’s Paternalistic

The prevention of ultimate harm is of far greater value than the prevention of paternalistic policies. And in any case, whether the compulsory program is covert or overt is independent of whether it is also paternalistic. In other words, an overt program is equally paternalistic. I  discuss paternalism in detail in the next chapter. But the main point is that paternalistic policies not only determine a person’s interests and coerce their pursuit. But neither overt nor covert moral bioenhancement is intended to pursue the interests of the person interfered with. They are intended to pursue the interests of the neighbor of the person interfered with. They are interventions for the good of other people, not for the good of the person interfered with. It Diminishes Autonomy

Some may think that by intervening upon a person’s moral attitudes, dispositions, and behaviors the moral bioenhancement program is preventing a person from expressing their autonomy. It prevents him or her from acting contrary to the intervention. And if that’s what the person wants to do, then the intervention prevents the person from doing what he or she wants to do, when, in the absence of the intervention he or she could. This diminishes the person’s autonomy. And since we have a duty to respect autonomy, a covert compulsory moral bioenhancement is not ethically permissible.

Covert Moral Bioenhancement 105

However, if the compulsory moral bioenhancement program were overt, at least then the person would be in a position to accept or reject the intervention. Or so the objection goes. The notion of autonomy that I have in mind says to be autonomous is to be in a position to self-govern in accordance with one’s desires, values, or other attitudes. This may not capture every conception of autonomy, but it at least seems central to the different variations (Christman, 2020). With this conception of autonomy in hand, it may be that a covert moral bioenhancement program does not diminish a person’s autonomy at all. But if it does, it at least diminishes it less so than an overt program. If to be autonomous is to be in a position to self-govern in accordance with one’s desires, values, and other attitudes, then one’s autonomy could be diminished, if one is out of position to self-govern. Thus, to be autonomous one must have the appropriate power to reason and choose, but must also be able to reason and choose independent of external manipulation, and one’s desires, values, and attitudes must be authentic. If this is right, then the objection that a covert moral bioenhancement program violates a person’s autonomy amounts to the claim that it violates either one’s independence from external influence or the authenticity of one’s desires, values, and other attitudes. Consider first the independence condition, the condition that one’s reasoning and decisions be free of external influence. It is common that our reasoning and decisions are dependent upon external influence. For example, experimental evidence suggests that a person’s moral judgment about which life-saving public health program to implement in response to a disease outbreak is dependent on the words used to describe the programs (Tversky  & Kahneman, 1981). This sort of framing effect on our moral judgments is common (Sinnott-Armstrong, 2008). A slightly different type of influence is that of one’s environment. In one experiment, subjects finding a dime in a phone booth were much more likely to mail a seemingly misplaced addressed envelope than those who found no dime (Isen & Levin, 1972). And in another study subjects were more likely to help an injured man if the ambient noise was normal than they were if a loud motor was running (Mathews & Canon, 1975). In these experiments, and presumably all of the instances to which they generalize, the subjects arrive at a moral judgment induced by others without awareness of the influence. Intuitively, these types of external influences do not violate the independence condition. If they do violate the condition, then many, if not most, of our moral judgments are not autonomous, and moral bioenhancement doesn’t constitute a unique threat to autonomy. As long as the moral bioenhancement exerts a similar type of influence, the fact that a moral bioenhancement program exerts external influence makes it no different from these other ways our judgments are influenced.6

106  Covert Moral Bioenhancement

Even if a compulsory moral bioenhancement program does undermine our independence from external influence in an autonomy-violating way, it’s the compulsory nature of the intervention that does it, because it’s the intervention that is the external influence. This intervention, in a compulsory moral bioenhancement program, occurs whether the program is overt or covert. So, if a covert moral bioenhancement program undermines a person’s independence from external influence, then so does an overt program. And if they violate it (or don’t) equally, then autonomy violation is not a reason to prefer one program over another. However, an overt program and covert program are not equal when it comes to the potential violation of the authenticity condition, or the condition that one embraces one’s desires, values, and other attitudes as one’s own. In this case, a compulsory moral bioenhancement program does violate autonomy, but only if the program is overt. If a person is compelled to participate in a moral bioenhancement program, and the person believes that the new moral capacities—including the new desires, values, and other attitudes—are caused by the enhancement, it is much more difficult to see how the person would embrace these capacities as their own. The knowledge that some of one’s moral capacities are the result of manipulation by another agent undermines trust in their authenticity. Thus, an overt program is likely to violate the authenticity condition. If the moral bioenhancement is covert, the person is in a much better position to embrace the new capacities as one’s own. Though the new capacities are in fact not one’s own, there are fewer obstacles to embracing them as one’s own, such as the knowledge that they are not. If from the person’s perspective he or she cannot notice any difference between one’s own capacities, and the new, enhanced capacities, then from their perspective there would be no reason to not embrace the new capacities as one’s own. So, as long as the enhancement is covert and the person doesn’t believe he or she has been enhanced, the program introduces no additional threat to the embrace of one’s capacities as one’s own. But an overt program does. So, if a moral bioenhancement is compulsory, to best preserve authenticity, it is preferable for the program to be covert. Even if a moral bioenhancement program does diminish a person’s autonomy—I don’t think it does—there is no implication that to do so is wrong. Quarantine diminishes the isolated person’s autonomy, and we don’t think that quarantine is wrong. The same goes for forced treatment of people with mental health disorders disposing them to be harmful to themselves or others. I am arguing that we should treat those that are disposed to engender ultimate harm as hazards to the public’s health. So, if moral bioenhancement of this population is going to be compulsory like other public health interventions, it ought to be done covertly, as this best preserves or promotes the values and conditions of frameworks of public health ethics.

Covert Moral Bioenhancement 107

It’s Inegalitarian

Robert Sparrow (2014) has argued that widespread moral enhancement is inegalitarian. This conclusion rests on several claims (Persson & Savulescu, 2014a). One is that, in principle, moral enhancement requires classifying some people (i.e., the enhanced) as better people than others. To do so is inegalitarian, and it risks another threat to egalitarianism, that the enhanced may have a greater claim to representation in government. A  suppressed premise is that we should promote egalitarianism. Granting for the moment that we should promote egalitarian policies, covert compulsory moral bioenhancement would not problematically conflict with egalitarianism. Because the enhancement would be covert, there would be no possibility of overrepresentation in government, at least intentionally. No one would know to claim a greater entitlement to representation. Relatedly, if everyone is enhanced, then there is no overrepresentation of the enhanced, as compared to the unenhanced. The same could be said about classifying people. The fact that the enhancement is covert means that no one would know to make distinctions between people, at least distinctions based on whether has superior moral qualities than others. And the fact that the enhancement is compulsory means that the distinctions themselves won’t be possible. People make distinctions between people when there are perceived differences between them. But because the moral bioenhancement is covert, there is no such perception; and because it’s compulsory there are no differences. Instead, covert compulsory moral bioenhancement, if anything, will increase equality. To the extent that social inequalities result in unequal moral capacities (which may be significant) is the extent to which covert compulsory moral bioenhancement will diminish social inequalities. The CEO who makes a fortune by polluting his community will be less likely to do so, thereby diminishing the social inequalities between him and the other community members. It is especially true that covert compulsory moral bioenhancement will diminish inequality if egalitarianism is the best moral theory of how resources should be distributed. If egalitarianism is right, then widespread moral enhancement, especially compulsory and covert moral enhancement, should result in greater equality. People, maybe even moral post-persons, with their improved moral capacities, should be better at doing the right things. If the right thing is to prioritize equality, then the enhanced will prioritize equality. Thus, confident egalitarians should enthusiastically endorse widespread moral enhancement. However, suppose that covert compulsory moral bioenhancement was incompatible with egalitarianism. This incompatibility wouldn’t undermine covert compulsory moral bioenhancement. Rather, it would undermine egalitarianism. Covert compulsory moral bioenhancement is necessary to prevent the catastrophic pain and suffering that is likely to result from

108  Covert Moral Bioenhancement

climate change or nuclear proliferation or disease, and so on. If achieving the preservation of humans and preventing this suffering is incompatible with egalitarianism, then so much the worse for egalitarianism. This would be an extreme version of the leveling down problem for egalitarianism. The leveling down problem is that when equality is prioritized, then those who might otherwise have more get less for the sake of equality: it’s better for two people to have two widgets each than for one person to have two widgets and the other person to have three. If egalitarianism is incompatible with covert compulsory moral bioenhancement, and covert compulsory moral bioenhancement is necessary to prevent the significant pain and suffering we can expect, then we should reject egalitarianism. Insisting on the moral priority of equality over continued existence is not a defensible position. Others have suggested that to promote equality, bioenhancement should be withheld from those who possess unequal shares of opportunity, relative to others, and that, instead, bioenhancement opportunities should only go to those who are less well off. This would not only entail the same leveling down problem but also for moral bioenhancement intended to avert catastrophe, its outcome will be no different. Withholding moral bioenhancement from those who are well off, while perhaps increasing equality, is likely to result in significant defection in the collective mitigation of collective risk. For the risk to be mitigated, everybody needs to contribute. Indeed, those who are most well off are likely to be those who most contribute to the risk in the first place, as they are likely to consume more of the resources that contribute to climate change. Withholding moral bioenhancement from them may be effective in improving equality initially, but it will especially do so if everyone is equally dead or suffering. Rather than undermining equality or exacerbating or promoting inequality, compulsory public health programs often promote equality. Compulsory vaccines are enhancements that promote everyone’s immune system. People aren’t left to a good diet, sleep, exercise, and luck to be protected from communicable diseases. No matter one’s social or economic opportunities, having the capacity to resist these diseases is an opportunity afforded to all. Similarly, fluoridated water systems protect everyone from the world’s most prevalent disease. For people who can afford fluoridated toothpaste and regular trips to the dentist, fluoridated water is no great improvement upon their health. But for people who can’t afford those things, fluoridated protects their teeth from dental caries. Thus, fluoridated water helps equality. The same can be said for many other compulsory public health interventions. Covert compulsory moral bioenhancement would be no different. Incompetent Administrators

Like any other policy, the administration of a covert compulsory moral bioenhancement program would require human implementation and maintenance. As such, one might legitimately worry that there aren’t people with

Covert Moral Bioenhancement 109

the competence to implement and administer such a program. They may lack the knowledge or the temperament or the moral capacities to ensure that the program achieves its aim. I endorse this worry. If the administration of a compulsory covert moral bioenhancement program required ordinary policymakers, or even ordinary executives, then there would not be much hope that the program could succeed. I discuss this in greater detail in the next chapter, but the administrators would have to pursue the interests of others much more so than policymakers typically do. They could not be people who are primarily self-interested. They would have to be much moral like moral exemplars, or moral post-persons, than ordinary policymakers or executives. The administrators of the program would need to be some of the very first people to receive the enhancement. Before the program administers the moral bioenhancement to anyone, indeed, before the policies and procedures governing its administration are developed, the administrators would have to be enhanced. They wouldn’t have to be the very first people to receive it, but they would certainly need to be in the very early stages so that they could competently pursue its widespread administration. Perhaps the only group of people who would need to receive the enhancement prior to the administrators are those people who select the administrators. That way, all decisions regarding who is best suited to administer the program and how it is best to do so are all made by people who themselves have enhanced moral capacities. If the administrators are themselves enhanced, then the worry that they, whoever they happen to be, lack the competence necessary to successfully enhance others should go away.

Conclusion I have argued in this chapter that if moral bioenhancement ought to be compulsory, then it ought to be administered covertly. Since it is true that moral bioenhancement ought to be compulsory, it is true that it ought to be administered covertly. Such a program is justified on epistemic, moral, and practical grounds. The main remaining tension between covert compulsory moral bioenhancement and its implementation is political. It seems that covert compulsory moral bioenhancement may be at odds with individual liberty and democracy. In the next two chapters, I relieve this tension and argue that covert compulsory moral bioenhancement is compatible with both libertarianism, and by implication other political theories, as well as democracy.

Notes 1. This is not to say that having unconscious desires be unconsciously frustrated doesn’t matter. I  am merely pointing out that in an overt program; there is an additional awareness of desire frustration.

110  Covert Moral Bioenhancement 2. A potential source of inequality in a compulsory program is between the people administering the program and the people receiving it. But the potential for this inequality is due to the program being compulsory, not to it being covert. Overt programs have the same potential to introduce this kind of inequality. 3. Some authors have complained that moral bioenhancement would be inegalitarian (e.g., Sparrow (2014) A covert program would not be inegalitarian). 4. For example, access to “dual-use” research, which is research that can be used for public good or public harm. In some cases, access only requires internet access. 5. If moral bioenhancement is necessary, the program can be either overt or covert. 6. Whether it exerts a similar influence would depend on the mechanism of action. One may think that even when one’s moral judgment is influenced by the situation one finds oneself in, one still has the ability to reject the influenced moral judgment, but that in the case of moral bioenhancement one would not be in such a position. This misses the lesson of framing effects—they take us out of the position to accept or reject the framed moral judgments.

Chapter 8

Transparency

The administration of covert compulsory moral bioenhancement is a public health intervention. But it is obviously not transparent, which, for some, implies that it is not a morally justifiable public health intervention. For example, Childress et al. claim that public justification of an intervention is a necessary condition for that intervention’s moral permissibility. In this chapter, I challenge the idea that transparency is required for permissibility. With the need for transparency out of the way, the scope of which public health interventions are permissible widens. The scope widens to the point where it may be permissible for a small group of people to intervene on the public’s health, without the subjects’ knowledge. But such an administration may seem paternalistic to some, and some believe that paternalism is prima facie wrong. I have four aims in this chapter. The first aim is to undermine transparency as a necessary condition for the moral permissibility of public health interventions. But not only is transparency not necessary, it may not even be a moral good. Part of my argument depends on the idea developed by Hibbing and Theiss-Morse (2002) that people don’t particularly want transparency, that they’d rather be kept in the dark. People want “stealth democracy.” What people really want out of their policymakers is that they not involve them in the decision-making process, but that the policymakers themselves be empathetic, non-self-interested decision-makers (ENSIDs). Hibbing and Theiss-Morse are skeptical that it is feasible for ENSIDs to become a reality. But if moral bioenhancement is possible, this skepticism is unwarranted. Thus, my second aim is to show that ENSIDs are feasible and that with them stealth democracy may be possible. That is, covert compulsory moral bioenhancement may not only be necessary to avert the significant suffering that is likely to result from inevitable catastrophe, it may also be exactly what people want. The notion of ENSIDs (i.e., moral post-persons or moral exemplars) making decisions about moral bioenhancement might seem to some to be paternalist, so my third aim is to show that it is not so. The administration of covert compulsory moral bioenhancement is not paternalistic. With

112 Transparency

concerns about transparency and paternalism out of the way, my fourth aim is to conclude that private institutions such as corporations could permissibly administer covert compulsory moral bioenhancement.

What Justifies Transparency? Many people think that transparency is not only an important value in considering whether a public health intervention is permissible, but others think that it is a necessary condition for permissibility. Childress et al. (2002) provide five conditions for a public health intervention to be ethically permissible. The fifth of these is that there is public justification for the intervention. They write (p. 173): In the context of what we called “political public”, public health agents should offer public justification for policies in terms that fit the overall social contract in a liberal, pluralistic democracy. This transparency stems in part from the requirement to treat citizens as equals and with respect by offering moral reasons, which in principle they could find acceptable, for policies that infringe general moral considerations. Transparency is also essential to creating and maintaining public trust; and it crucial to establishing accountability. Childress et al. offer three distinct reasons for thinking that transparency is a requirement of all permissible public health interventions. One reason is that to be transparent is to demonstrate respect for persons. The second reason is that transparency engenders trust. A third reason is that it is important in the public’s ability to hold responsible parties accountable. I think that none of these reasons is good enough to establish transparency as a necessary condition for permissible public health intervention. Respect for Persons

The first reason Childress et al. offer for thinking that transparency is necessary for the permissibility of public health interventions is that to be transparent is to demonstrate respect for persons. They say that transparency “stems in part from the requirement to treat citizens as equals and with respect by offering moral reasons, which in principle they could find acceptable, for policies that infringe general moral considerations.” This claim does not make clear why stemming from respect for persons makes transparency a requirement. But one plausible interpretation of the claim is that respect for persons is more fundamental to morality than are public health interventions such that things that enable respect for persons are required for the permissibility of those interventions. Since transparency enables respect for persons,

Transparency 113

it enables a value that is more fundamental to right action than the public health intervention and therefore necessary for the intervention to be permissible. Transparency, according to this line of thinking, enables respect for persons because it respects the person’s ability to engage moral reasoning. It’s hard to make sense of the claim if this interpretation is incorrect. If respect for persons is not a more morally fundamental value than whatever value the public health intervention promotes, then there would be no grounds for respect for persons to bind public health intervention. The claim that transparency stems from respect for persons and therefore is necessary for public health interventions presumes that several other propositions are true: (i) that respect for persons is more morally fundamental than the value that the public health intervention promotes; (ii) that respect for persons is achieved by engaging the person’s moral reasoning; (iii) and that transparency in public health engages a person’s moral reasoning. There are good reasons to doubt each of these propositions. That Respect for Person Is More Fundamental Than the Value the Public Health Intervention Promotes

In order for respect for persons to restrict which public health interventions are permissible, respect for persons must be more fundamental than whatever value the intervention promotes. To say that a given public health intervention is impermissible because it is not transparent because it does not appropriately respect persons, which is what Childress et al. appear to be claiming, is to say that there is greater value in respecting persons than there is in promoting the value that the public health intervention promotes. Suppose the state wants to aerially spray an insecticide to control a spreading mosquito-borne disease, but will not allow property owners to opt out of the intervention. The value the intervention promotes is the prevention of suffering from a mosquito-borne disease, but the promotion of this value is secondary, according to Childress et al., to the respect for persons that making the justification public promotes. Without such public justification, it would be impermissible to spray. There’s no other way of making sense of the claim that transparency is necessary for the permissibility of public health interventions because it stems from respect for persons. But there are good reasons to be skeptical of the claim that respect for persons is more fundamental to morally right action than is whatever value a given public health intervention promotes. Respect for persons is a matter of moral status; if a thing doesn’t have the moral status of a person, then there is no obligation to have respect for it qua person. But if a thing does have the moral status of a person, then there is such an obligation. It is common to think that the moral status of a being is at least partly a matter of the being’s capacity to reason or be a rational agent. But there are other

114 Transparency

common accounts of what makes a being a person, and not some other being lacking such a moral status. Suppose that being a person is a matter of having interests, the most fundamental of which are the experience of pleasure and the avoidance of suffering. According to this view, a being’s moral status is a matter of the pains and pleasures that the being experiences (or can experience). Any being with interests is a being with moral status due at least some respect. Those beings whose satisfied interests produce the most pleasure and whose dissatisfied interests produce the most suffering would certainly count as persons, but other beings might, also. Such an account could be attached to a threshold account of moral status, according to which only those beings whose satisfied/dissatisfied interests meet some threshold of pleasure/pain. Or it could be attached to a scalar account of moral status, according to which beings with greater capacity to experience pleasure and pain have a higher moral status and are therefore due more respect than other beings (see Chapter 4, Buchanan, 2009; DeGrazia, 2012). The view that moral status is a matter of a being’s interests being satisfied or dissatisfied is an attractive alternative to the view that moral status is a matter of rational agency. I have no intention of adjudicating between the two accounts here. But if moral status, and thus respect due to persons, is a matter of interests rather than rational agency, then it is not necessary that respect for persons is more morally fundamental than the values that a given public health intervention promotes. Public health interventions almost always aim at promoting an individual’s interests, typically those interests in avoiding pain and suffering, but often those interests in experiencing pleasure. If to respect a person is to respect their interests in experiencing pleasure and avoiding pain, then the values that most public health interventions promote are the very same values that ground respect for persons. That is, respect for persons is not more morally fundamental than the values that public health interventions promote. Respecting persons and the values public health interventions promote are both a matter of interest promotion. For example, it may be that it is more respectful of a person to intervene on their suffering at the expense of transparency, than it is to allow them to experience the preventable suffering but be transparent in that allowance. Childress et al. claim that the absence of transparency makes public health interventions impermissible. This claim relies on the further claim that lack of transparency undermines respect for persons. The only way this claim can be true is if respect for persons is more morally fundamental than whatever value a given public health intervention promotes. But there are good accounts of moral status, and so good accounts of respect for persons, according to which this claim is false: the values that ground moral status and the values that public health interventions promote are equally fundamental because they are the very same values.

Transparency 115

That Respect for Persons Is Achieved by Engaging the Person’s Moral Reasoning

Childress et al. claim that transparency respects persons because it engages the person’s moral reasoning. As the above discussion shows, respect for persons need not be a matter of the person’s moral reasoning. If moral status is instead a matter of a person’s interests being promoted, then unless people derive significant pleasure from public health officials being transparent with them (or not having information withheld causes them to suffer), transparency doesn’t necessarily alter whether a given intervention respects them as a person. In fact, engaging a person’s moral reasoning may be contrary to promoting their interests in avoiding suffering and experiencing pleasure, given that so many, if not most, people are incompetent at reasoning about what best promotes their own interests. I am not arguing that the interests-based view of moral status and respect for persons is true. I am arguing that it is a reasonable alternative to the view that Childress et al. use to ground their argument that transparency is necessary for a public health intervention to be morally permissible. That there is such an alternative casts doubt on the notion that transparency respects persons in any way that determines whether an intervention is permissible. But suppose that the interests-based view of respect for person is wrong and that persons are respected not by promoting their interests but by engaging their moral reasoning. That Transparency in Public Health Engages a Person’s Moral Reasoning

Even if it is true both that to respect a person is to engage their moral reasoning and that respect for persons is more fundamental than the value that a given public health intervention promotes, for it to be true that transparency in public health respects the person it must also be true that transparency engages the person’s moral reasoning. There are good reasons to think it doesn’t. Consider, for example, compulsory public health interventions. Vaccines are compulsory in many communities. In some places, one can opt out for personal reasons, after overcoming various logistical obstacles. In other places, the only exemptions allowed are those that are for medical reasons. In the United State in 2019, there was an outbreak of Eastern equine encephalitis (EEE), which is transmitted by infected mosquitos can that kills about 50% of those infected. Massachusetts generally allows citizens to request an exemption from aerial mosquito spray. But in the regions where EEE was prevalent, they allowed no such exemption. Michigan, however, experienced a similar epidemic but allowed citizens to opt out, which resulted in enough people opting out that the state canceled EEE control by aerial

116 Transparency

insecticide in some locations. Quarantine and other isolation of infectious individuals are also compulsory. All of these compulsory interventions are administered transparently. But it’s not clear that this transparency engages the moral reasoning of those upon whom the intervention is compelled. When a person with Ebola is quarantined in the hospital, being told that they are being quarantined due to them having Ebola doesn’t clearly engage with the infected’s moral reasoning, given that they can’t do anything about the quarantine. They may reason that they do not want to be quarantined and that they would rather risk the lives of other people and conclude that they ought not to be quarantined. But if moral reasoning is so important that it can ground something fundamental like respect for persons, it can only do so if it positions a being to express moral agency. And agency is only possible with choices. A  person who is quarantined has no choice in the matter. To meaningfully engage with a person’s moral reasoning, a public health intervention must engage with it in a way that enables action. But for truly compulsory interventions such as quarantine and Massachusetts’ waiver of the insecticide exemption, citizens have no choice in the matter. When public health officials in Massachusetts tell citizens that their property will be sprayed with insecticide whether they want it or not, doing so doesn’t seem to engage with their moral reasoning in the same way that, for example, informing citizens about the dangers of tobacco use or the importance of proper nutrition and exercise do. Childress et al. write as though with this moral reasoning people may find the intervention acceptable. But whether they find it acceptable has no bearing on whether they receive the intervention. To be compelled to behave in ways one doesn’t want to behave is an insult to a person’s moral agency. If they are going to get the intervention one way or the other, isn’t better for them to not be insulted? Transparency in public health interventions may sometimes engage a person’s moral reasoning, in many cases in ways that enable them to express their moral agency. But transparency about compulsory public health interventions doesn’t engage moral reasoning in the same way, because people are not in a position to refuse, and so not in a position to accept it, either. The transparency in such cases doesn’t enable moral agency. However, compulsory public health interventions can be morally justified. Childress et al. claim that transparency is necessary for public health interventions to be ethically permissible because to be transparent is to respect persons. But this claim relies on three further claims, none of which they defend. Furthermore, I have offered in this section reasons to doubt all three claims. It is reasonable, therefore, to doubt that transparency in public health respects the person. Trust and Transparency

Transparency in public health may not respect persons, but Childress et al. offer three reasons to think that transparency is necessary for public health

Transparency 117

interventions to be morally permissible. The second reason they offer is that transparency is necessary because it is “essential to creating and maintaining public trust” (p. 173). Transparency is necessary only if both (i) it is essential to creating and maintaining public trust and (ii) public trust is necessary for a public health intervention to be morally permissible. If trust is not necessary for public health interventions to be morally permissible, then it’s of course false that public health officials are morally obligated to be transparent. Once again, there are good reasons to doubt both of these claims. Transparency Is Essential to Creating and Maintaining Public Trust

That transparency is essential to creating and maintaining public trust is an empirical claim, and needs empirical support. There is scant empirical research, but what research there is suggests that transparency is not only unnecessary for public trust but also detrimental to it. De Fine Licht (2011) and Grimmelikhuijsen et  al. (2013) provided information to subjects in South Korea and the Netherlands regarding governmental decision-making, policy information, and policy outcome. They then tested those subjects’ perceptions of the governments’ competence, benevolence, and honesty. Compared to a control group, the increase in information subjects received reduced their perceptions of competence, benevolence, and honesty. This negative effect was more pronounced for the South Korean subjects, but still present among the Dutch subjects. Though societies differ in many respects, including in how the public relates to its government, in at least of some of these societies greater transparency evidently diminishes public trust. In a different study, Grimmelikhuijsen and Meijer (2014) found that transparency affected trust only when the subject had a low baseline level of knowledge and low predisposition for trust in government. Among those subjects, transparency was associated with an increase in trust. But for other groups of subjects, such as those with high baseline knowledge and predisposition for trust, there was no association between transparency and trust, which suggests that knowledge moderates the relation, if any, between transparency and trust. Similar research from Tolbert and Mossberger (2006) found that information on a government website only increased trust in local government and only for people who were already visiting local government websites. More directly, De Fine Licht (2011) found that transparency in decision-making about public health interventions was associated with decreased trust in the government administering the intervention. Specifically, she tested subjects’ trust in the Swedish government after reading vignettes about public health interventions regarding allocation of funding to various initiatives. When subjects were merely informed of the policy, they had higher levels of trust in the Swedish government than when they were informed of both the policy and who made and how they made it. She found that the people in

118 Transparency

the low transparency group had higher levels of trust. Transparency again appears to decrease trust. There are limitations to the research. And other research (Cook et  al., 2010) indicates that transparency may increase a person’s confidence in specific government policies (rather than, for example, a global evaluation of a government’s trustworthiness). But the present point is not that the research conclusively shows one thing or another. Childress et al. justify the moral requirement of transparency on the grounds that transparency is essential to trust. This is an empirical claim, but the empirical evidence suggests, at a minimum, that transparency may not only be unnecessary for public trust, but that it may actually diminish public trust. It may be an intuitive assumption that transparency is essential to trust, and in 2002 they may have been entitled to it. But it is not an assumption one is entitled to any longer, because the empirical evidence suggests that it’s false: transparency is not essential to creating or maintaining trust. The function of the empirical evidence is not to conclusively show this. The function is rather to chip away at the often asserted but rarely defended proposition that transparency is not only valuable in public health but also necessary. It blocks the assumption that transparency is good. The proponent of transparency must then bear the burden of providing countervailing evidence and a normative argument for why we should aim for transparency. For it to be true that transparency is a necessary condition for morally permissible public health interventions because transparency is essential to creating and maintaining trust, it must also be true that transparency actually creates and maintains trust. But it probably doesn’t. Thus, it can’t also be true that transparency is a necessary condition for morally permissible public health interventions. Public Trust Is Necessary for a Public Health Intervention to Be Morally Permissible

However, it must also be true that public trust is necessary for a public health intervention to be morally permissible. But this claim is also false. Public trust is quite obviously not necessary, and it doesn’t take much of an imagination to see why. A person infected with Ebola may not trust at all their government. But it is morally permissible to quarantine them regardless of whether they trust the government. In Flint, Michigan, where the government failed the public in their inability to deliver lead-free water and in their disclosure that the public’s water was contaminated with lead, public trust is very low, maybe so low that it will never be recovered, especially with respect to interventions on the public water supply. But there is a multitude of interventions that are morally permissible in spite of this lack of trust. For example, the city may intervene on the supply such that the water is no longer contaminated with lead. This intervention would obviously be

Transparency 119

morally permissible—it’s obligatory—even though the citizens of Flint have little trust in the government. In this section and the preceding section, I have argued that the first two reasons Childress et al. use to justify the claim that transparency is necessary for morally permissible public health interventions don’t stand up to critical scrutiny. Transparency doesn’t clearly respect persons, and it is not essential to creating and maintaining public trust. Thus, if transparency is necessary for morally permissible public health interventions, it can only be so in virtue of their third reason, that transparency is crucial to establishing accountability. Accountability

The entirety of the justification for the necessity of transparency rests on the value of accountability. The other reasons Childress et al. offer are inadequate. For accountability to ground the necessity of transparency, it must be true that the value of accountability is greater than the value any given public health intervention promotes. It has to be that accountability is always more valuable than whatever value is achieved by intervening. The value of accountability can be achieved in a couple of ways. First, accountability can simply be responsibility. But if accountability is simply responsibility, then transparency is irrelevant. There will always be some person or group of people that is responsible for a public health intervention, regardless of whether other people know about it. That is, there will always be someone who is blameworthy or praiseworthy for a public health intervention whether other people know it or not. This can’t be the type of accountability for which transparency is crucial. Second, if accountability isn’t a matter of blameworthiness or praiseworthiness, it must be a matter of actually blaming or praising, or at least having the opportunity to blame or praise. Transparency is crucial to establishing this opportunity. A person or the public can’t pass praise or blame without knowing something about an intervention, who did it, why they did it, when they did it, or that an intervention even occurred. For accountability to justify the necessity of transparency, it must be that the value of having the opportunity to blame or praise outweighs the value of any public health intervention. It seems wrong that the most important public health interventions would have been impermissible if the public lacked the opportunity to blame or praise. For most public health interventions, the public has the opportunity to blame or praise people for this intervention. Whether the people who are blamed or praised are in fact blameworthy or praiseworthy is beside the point: they may not be the same people, but it doesn’t matter anyway because what’s important is the opportunity to blame or praise. It if is the opportunity to blame or praise that is so valuable, it must be in virtue of something

120 Transparency

that it does for the subject of the blame or praise since the opportunity to blame or praise is distinct from blameworthiness or praiseworthiness. Having the opportunity to blame or praise can do for the subject a few things. One is that it can create solidarity among those who share in the opportunity. Another is that it can feel good to blame or praise someone. Suppose the value is achieved by increasing solidarity and the good feeling for passing blame or praise. Then it has to be true that this value outweighs the value that is achieved by the intervention. And it seems obvious that it does not because not dying of heart disease or whooping cough or the flu seems like it is much more valuable. Does the positive value of being one with one’s country people and praising its administrators outweigh the negative value of contracting whooping cough, transmitting it to your infant, and then having that infant die? No, it does not. And since it does not, it can’t be that the opportunity to blame or praise is always more valuable than the value achieved by public health interventions. And since transparency is only relevant to accountability if accountability is the opportunity to blame or praise, the value of accountability does not outweigh the value of a given public health intervention. This means that accountability can’t justify the requirement that public health interventions be transparent. I’m not arguing that accountability is not valuable. And I’m not arguing that it should be excluded from consideration when planning public health interventions. I am arguing that it is not valuable enough to make transparency necessary for a public health intervention to be morally permissible. Childress et  al. claim that transparency is necessary, and their reason for this necessity is that it establishes accountability. But this reason isn’t good enough. Their other reasons also fail in establishing the necessity of transparency. Thus, we shouldn’t think that transparency is necessary for a public health intervention to be morally permissible.

Public Justification Childress et al. say that public justification (in the form of transparency) is necessary for the moral permissibility of a public health intervention. But there is another sense of “public justification” that is relevant to the permissibility of covert compulsory MBE. Where Childress et al. are concerned with public justification as a matter of decisions and their justification being made openly, this other sense provides the conditions of when coercion is permissible. More specifically, this other notion of public justification is captured by the Public Justification Principle: A coercive law L is justified in a public P if and only if each member i of P has sufficient reason(s) Ri to endorse  L (Vallier, 2018). The conditions for public justification aren’t a matter of openness or publicity, but about individuals’ reasons to endorse or accept a coercive law. This notion of public justification is not widely held as, say, the need for transparency, so the requirement that covert compulsory moral

Transparency 121

bioenhancement be compatible with the principle is looser. But my interest is in showing that covert compulsory moral bioenhancement—not necessarily a law but nonetheless a form of coercion—is not especially problematic, regardless of one’s theory. The Public Justification Principle appears restrictive at first: there are not clearly many laws such that every individual has sufficient reason to endorse it. But much of the principle’s restrictiveness depends on how we interpret “sufficient,” “reason,” and “endorse.” There are plausible interpretations of these terms according to which covert compulsory moral bioenhancement is compatible with the Public Justification Principle. The details of how best to interpret the Public Justification Principle are largely internal to public reason liberalism, and exploring them goes far beyond the present purpose. It suffices to demonstrate that there are interpretations that are compatible with covert compulsory moral bioenhancement. First, only on some interpretations of public justification does the principle even apply to fine-grained acts of coercion, such as covert compulsory moral bioenhancement, applying instead only to coarse-grained laws, such as those in a constitution (Vallier, 2018). So, according to some views, the Principle of Public Justification would not even apply to covert compulsory MBE. But supposing, as others do, that the principle does apply to fine-grained laws, the principle is still compatible with covert compulsory moral bioenhancement. Kevin Vallier writes that the appropriate way to interpret the principle is that the notion of endorsing a law can take two forms, one counterfactual and one rational. Counterfactual endorsement is when an individual would endorse the law through some deliberate act of will (requiring specification of the antecedent conditions). The other notion of endorsement is that a law is counted as endorsed when rationality requires an individual to endorse it; covert compulsory moral bioenhancement is compatible with this second notion of endorsement. If it’s true that a coercive law counts as endorsed when a person is rationally committed to it, then covert compulsory moral bioenhancement is plausibly endorsed by each member of the public. The project of this book is to demonstrate the rational requirement of covert compulsory moral bioenhancement, but this account of endorsement doesn’t require that a person actually consider or entertain the particular propositions to which they are rationally committed. It merely must be the case that rationality demands certain commitments to propositions. I am arguing that a person is rationally committed to endorse covert compulsory moral bioenhancement (but for them to be so committed it is unnecessary that they’ve read the book!). The case for covert compulsory moral bioenhancement rests ultimately on the idea that we should prevent very bad things from happening when the benefits of doing so outweigh whatever sacrifices we might have to make. It rests on the obligation to try to prevent preventable harms. If one accepts this notion, which presumably most people do, then one is rationally committed to covert compulsory moral bioenhancement. My aim is to establish

122 Transparency

this consequent, but whether one is aware of these reasons is not a condition on permissible coercive laws. There is thus space for the public reason liberal, the sort of liberal who holds the Principle of Public Justification, to accept covert compulsory moral bioenhancement. Covert compulsory moral bioenhancement is a public health intervention and justified in the same way as other public health interventions. If the public reason liberal doesn’t accept covert compulsory moral bioenhancement, then they must also say why covert compulsory moral bioenhancement is different from other public health interventions, such as compulsory vaccines or quarantine. On the face of it, if covert compulsory moral bioenhancement is incompatible with public justification, then so is compulsory vaccination. Compulsory vaccination coerces people to receive injections, and some people at least deny that they have sufficient reason to endorse such laws, and do so vociferously. If these protests imply that the conditions of the Principle of Public Justification fail to be satisfied, then laws compelling vaccination are impermissible, which counts against public justification.

Stealth Democracy I have argued that transparency is not necessary for public health interventions to be morally permissible. But there are good reasons to think that it is not even valuable in policymaking, including in making those policies governing public health interventions. Since it’s not necessary and maybe not valuable, concerns over the lack of transparency of covert compulsory moral bioenhancement should not undermine its potential administration. The reasons we have to think that transparency is not significantly valuable come from John Hibbing and Elizabeth Theiss-Morse (2002). They argue that what people really want most of all from their policymakers is “stealth democracy.” They write (p. 2): Rather than wanting a more active, participatory democracy, a remarkable number of people want what we call stealth democracy. Stealth aircraft such as B-2 bombers are difficult to see with standard radar techniques, yet everyone knows they exist. Similarly, the people want democratic procedures to exist but not to be visible on a routine basis. But how can people in a stealth democracy hold government accountable for its policy decisions? The focus of this question is actually off the mark. The people as a whole tend to be quite indifferent to policies and therefore are not eager to hold government accountable for the policies it produces. This does not mean people think no mechanism for government accountability is necessary; they just do not want the mechanism to come into play except in unusual circumstances. The people want to be able to make democracy visible and accountable on those rare occasions when they are motivated to be involved. They want to know that

Transparency 123

the opportunity will be there for them even though they probably have no current intention of getting involved in government or even of paying attention to it. Just as stealth bombers can be made to show up on radar when desired, the people want to know that their government will become visible, accountable, and representative should they decide such traits are warranted. Until that time, however, most people prefer not to be involved and therefore desire unobtrusive accountability. People want their policies made in the dark, but to also have the opportunity to shine a light on them, if desired. Now, this wouldn’t be possible in the case of covert compulsory moral bioenhancement: shining a light on its administration would undermine its benefit. But it does show that transparency and accountability are not as valuable as perhaps Childress et al. think, and certainly not valuable enough to outweigh the benefit of covert compulsory moral bioenhancement, namely averting significant suffering that may result from pandemic, climate change, bioterrorism, rogue artificial intelligence, and so on. If Hibbing and Theiss-Morse are correct, then transparency in public health should ideally be of little concern. Hibbing and Theiss-Morse support the claim that people want their policies made in the dark with results from a series of surveys. They found that more than 93% of people agreed with at least one of the following prompts: (1) elected officials would help the country more if they would stop talking and just take action on important problems; (2) what people call compromise in politics is really just selling out on one’s principles; (3) our government would run better if decisions were left up to nonelected, independent experts rather than politicians or the people, or that our government would run better if decisions were left up to successful business people. Forty-two percent of people agreed with exactly two of these statements, and 27% agreed with all three. They take this as evidence that people want stealth democracy, as it appears to be the best explanation for these responses. If people wanted policy decisions to be made transparently, or to participate in the decision-making process, they wouldn’t respond in this way—they’d disagree with these statements. The ideal situation for many people is one in which they are not involved in the decision-making process, a process that doesn’t even occur out in the open. But people nevertheless call for transparency and accountability, and even to actively participate in decision-making. Hibbing and Theiss-Morse reconcile people’s attitudes and this behavior by noting that people don’t believe that decision-makers are qualified to make decisions in the dark, mostly due to the fact that they are beholden to special interests which promote their own self-interests. Hibbing and Theiss-Morse write (p. 158): People are not sure what policies they want, but they know what processes they do not want—and those are processes in which people who

124 Transparency

are making political decisions are able to feather their own nests. People’s political involvement, such as it is, is often driven by the perception that politicians and their special-interest ventriloquists are taking advantage of ordinary people. People don’t want to be involved in the decision-making process. But they want more to not be taken advantage of by self-interested policymakers. So they get involved and demand transparency. It would be better to not have to be involved, but they want more to not be exploited. Instead of self-interested policymakers who use their positions to “fluff their own nest” Hibbing and Theiss-Morse claim that people want “caring, other-regarding, common-good-oriented decision makers. They want empathetic, non-self-interested decision makers (ENSIDs)” (p. 161). Thus, ideally ENSIDs would make policy decisions in the dark, leaving people out of the process until interests in participation or accountability arise, which occur rarely. There are many ways in which a public health intervention or policy might be transparent. Publishing data on rates of death of infectious diseases is one form; involving marginalized populations in the policymaking is another. There are many others. Transparency varies not only by type but also by degree. Promoting and broadcasting on television the entirety of a policy debate is very transparent; offering the opportunity to request partially redacted documents by mail is less so. Transparency is not only one thing, and it’s not all or nothing. The type and degree of transparency that people want will vary. The point is not that these preferences are wrong or that they should go unsatisfied, but rather that the moral value of satisfying these preferences, to the extent that they exist, is less than what is typically assumed in discussions of the permissibility of a public health intervention. Hibbing and Theiss-Morse offer a controversial thesis that people want stealth democracy. But it may also be true that people don’t want democracy of any kind—that they want whoever is making and executing policy to simply advance their interests, but are indifferent to how that person or group came to be in that position. The present point is not that people want democracy—they may not—but rather that there are good reasons to think that the governed don’t necessarily care much about policymaking so long as it does something for them. That opacity in policymaking is permissible not only has empirical support, it is also an implication of utilitarianism. Henry Sidgwick (1907/1981, p. 489, italics mine) writes: [O]n Utilitarian principles, it may be right to do and privately recommend, under certain circumstances, what it would not be right to advocate openly; it may be right to teach openly to one set of persons what it would be wrong to teach to others; it may be conceivably right to

Transparency 125

do, if it can be done with comparative secrecy, what it would be wrong to do in the face of the world; and even, if perfect secrecy can be reasonably expected, what it would be wrong to recommend by private advice and example. Sidgwick’s claims about “esoteric morality” are contested by a range of prominent moral philosophers, including Bernard Williams (1985), who calls it “Government House utilitarianism,” but we might as well call it “stealth democracy.” Katarzyna De Lazari-Radek and Peter Singer (2010) argue that Sidgwick’s claims are just an implication of utilitarianism. You can’t adopt utilitarianism without also at least implicitly endorsing esoteric morality. They argue on utilitarian grounds that “There are acts which are right only if no one—or virtually no one—will get to know about them. The rightness of an act, in other words, may depend on its secrecy. This can have implications for how often, and in what circumstances, such an act may be done” (p. 35). Covert compulsory moral bioenhancement is one of these acts, as I argue in previous chapters. de Lazari-Radek and Singer defend Sidgwick’s claim using the following example (p. 39). Suppose a soldier is brought before a general, charged with desertion in battle, for which the mandatory penalty is a long term of imprisonment. The soldier admits that he did desert, but begs for pardon, saying that he does not want his two small children to grow up without a father. Only one or two other people, who the general feels he can trust, know that the soldier deserted. The general assigns the soldier to duties behind the front line, telling him he must never say a word to anyone about deserting, or being charged with desertion. He tells his administrative officers to destroy all records of the charge and forget all about it. Although telling the world about the soldier’s desertion and pardon would have terrible consequences, if the general can conceal the pardon, then the benefits of pardoning the soldier appear to outweigh the harms. He won’t go to prison; his children will grow up with a father; and further hardship will not be imposed upon his wife. Given this scenario, the utilitarian seems committed to the claim that the general ought to pardon the soldier, if he can conceal it from the world. But if he can’t conceal it, then the act of pardoning him is wrong. Thus, there are circumstances in which the rightness of act depends on its secrecy, by utilitarian standards. The same is true of covert compulsory moral bioenhancement. The act of compelling moral bioenhancement is conceivably right to do, but only if it is done in secrecy. Or, alternatively, the morality of the act of compelling moral bioenhancement is dependent upon whether it can be done in secret. Covert compulsory moral bioenhancement is thus not only rooted in contemporary political science but also can be seen as an implication of garden-variety

126 Transparency

utilitarianism going back more than a century. Once the facts of the benefits and costs are fixed, we should accept compulsory moral bioenhancement, if it can be conducted in secret.

Covert Compulsory Moral Bioenhancement and Stealth Democracy In practice, however, transparency may be valuable, though not necessary. It may be valuable because decision-makers are self-interested. Given the option of participating and monitoring policymaking by self-interested individuals and the option of not doing so, people prefer the former. But if the options were between participating and monitoring policymaking by self-interested individuals and not participating or monitoring policymaking by ENSIDS, people prefer the latter. But policies, including public health policies, are not generally made by ENSIDs. Hibbing and Theiss-Morse are rightly skeptical about the prospects of realizing stealth democracy, due to the fact that decision-makers are almost never ENSIDs, and it’s difficult to see how they could come to be so. But if ENSIDs were possible, then it might be possible for people to get the government they want, a government in which policy decisions are made in the dark by other-regarding, non-self-interested experts. In Chapters 3 and 4, I argued that we should make moral bioenhancement compulsory, because we’d end up with a bunch of moral post-persons, or at least moral exemplars. In particular, we’d end up with people who first and foremost put the interests of others prior to their own. This is to say that the people that receive moral bioenhancement would end up as otherregarding and non-self-interested. They’d be ENSIDs. Transparency is valuable in part because decision makers are not ENSIDs. But if the decision makers were ENSIDS, as they could be if they received moral bioenhancement, then there’d be less value in transparency. Engendering ENSIDs by way of moral bioenhancement would not only undermine the value of transparency but also give people the government they want, if Hibbing and Theiss-Morse are right. While it’s true that people would still be unaware of ENSIDs and their role in making decisions regarding moral bioenhancement, the people’s desire to have ENSIDs make policy in the dark would be satisfied. That is, they’d get the government they want, but not necessarily know that they are getting what they want. When people hear the main claim of this book, that moral bioenhancement should be compulsory and covert, they may initially have the intuition that such a program would not be transparent and for this reason, among others, would be wrong. But transparency is not only unnecessary for the moral permissibility of public health interventions, in the right conditions people prefer the absence of transparency. Moral bioenhancement can bring about those conditions. Thus, the intuition that a lack of transparency

Transparency 127

undermines the moral permissibility of covert compulsory moral bioenhancement is false, though understandable given the apparent value of transparency in other contexts of political decision-making. But transparency is not necessary for the moral permissibility of public health interventions, and in the particular case of covert compulsory moral bioenhancement, not even valuable. The lack of transparency doesn’t undermine the moral permissibility of covert compulsory moral bioenhancement, and may instead count in its favor.

Paternalism The lack of transparency should not be concerning to those who doubt that we should administer compulsory covert moral bioenhancement, but there are related concerns. I address them in a later chapter, but one that is relevant to concerns about many public health interventions is that covert compulsory moral bioenhancement is overly paternalistic. And since generally we should avoid paternalism, covert compulsory moral bioenhancement is not morally permissible. I take no stance on whether paternalism, if ever, is appropriate. I don’t need to take such a stance because covert compulsory moral bioenhancement is not paternalistic, so it is irrelevant whether the conditions for appropriate paternalism obtain. Gerald Dworkin writes that “Paternalism is the interference of a state or an individual with another person, against their will, and defended or motivated by a claim that the person interfered with will be better off or protected from harm” (2019). Most public health interventions are not defended or motivated by a claim that the person interfered with will be better off or protected for harm. They are defended and motivated by the claim that the person interfered with neighbor or colleague or fellow human will be better off. The interference is for the sake of another person’s well-being, not for the sake of the well-being of the person interfered with. Covert compulsory moral bioenhancement is no different. The defense of covert compulsory moral bioenhancement isn’t that you will be better off and so you should be enhanced. The defense is that your neighbors, your descendants, your fellow humans will be better off. The interference in your moral behavior is defended on the grounds that it is necessary to promote the interests of other people, though you may benefit, as well. Your benefit does not justify the interference, though. Covert compulsory moral bioenhancement is unlike other public health interventions on a person’s choices, such as “nudges” (Thaler & Sunstein, 2009) that compose libertarian paternalism. Nudges manipulate a person’s choice architecture in ways that make them more likely to make choices that are in line with their best interests. For example, placing healthier foods in the places a person is most likely to see may make it more likely that they select that healthier food, which promotes their interest in health. Nudges

128 Transparency

are aimed at promoting the interests of the person interfered with. Covert compulsory moral bioenhancement is aimed at promoting the interests of the person interfered with the community. It would be absurd to assert that interventions aimed at the benefit of others are paternalistic and therefore wrong. Such a view would imply that many public health measures (quarantine, mask-wearing, restriction of large gatherings) are in principle immoral. But they are not immoral. Thus, it’s false that interventions aimed at the benefit of others are paternalistic and therefore wrong. Finally, covert compulsory moral bioenhancement doesn’t disrespect the person. If dignity is tied to respect for persons, then it doesn’t insult a person’s dignity. I argue in Chapter 6 that it doesn’t undermine a person’s autonomy. It’s not transparent, of course, but that’s not a problem, as I have argued in this chapter. It’s not paternalistic. In the last chapter, I argue that covert compulsory moral bioenhancement is also compatible with a strain of libertarianism highly protective of individual liberties. So, it respects the person (and their dignity), doesn’t undermine autonomy or liberty, and has the greatest benefit.

Private Institutions in Public Health Public health interventions administered by the government are supposedly good because they (a) allow the public to participate in decisions regarding public health; (b) it is the only way to institute such interventions transparently; and (c) that interventions often involve enforcement, which is purely a function of the state. But covert compulsory moral bioenhancement wouldn’t require state enforcement. And the lack of transparency is not especially problematic. And it’s not clear that people really do want to participate in the decision-making process. Thus, very little value is lost if the administration of covert compulsory moral bioenhancement is conducted privately. If a private institution, such as a corporation or foundation, administered covert compulsory moral bioenhancement rather than the state, it would be no great loss. Covert compulsory moral bioenhancement is a public health intervention. In the next chapter, I  argue that it doesn’t constitute a significant infringement upon a person’s liberties. But even if it did, it would be permissible for the state to carry it out. Other authors agree that there are instances in which the state can permissibly restrict a person’s freedom so that a public health intervention in the form of enhancement may be carried out. For example, Buchanan et al. (2001) claim that it is in principle permissible for the state to require some genetic enhancements which would limit reproductive freedom, though they recognized that at the time the circumstances surrounding the science did not support such an intervention.

Transparency 129

It is a natural thought that when public health interventions aim at securing a public good the government is the actor that should carry out that intervention. But it is not necessary that the government carry out covert compulsory moral bioenhancement—private institutions may permissibly administer it. Anomaly (2011) has argued that public health interventions ought to be limited to those health interventions that aim at securing pure public goods (those that are non-rival and non-excludable) and that the government is often in the best position to secure public goods (independent of whether the good is health-related) (2015). The pursuit of public goods encourages free-riding, which then undermines cooperators’ assurance that their own contributions will pay off, causing them to withdraw. Thus, often the market will fail in securing this public good. The government, with its monopoly on the permissible use of force and its ability to compel contributions, can step in and do what the market can’t and prevent free-riding and assure others that their contributions won’t go to waste. Supposing it’s true that the government’s role in securing public goods is to repair market failure, the government could permissibly administer covert compulsory moral bioenhancement. But as Anomaly (2015) points out, even though the government may secure a public good, it is not always necessary that it, rather than a private institution, does so. If the goal of government action is to compensate for market failure, and private institutions can also compensate for that market failure, then those private institutions can also permissibly provide the public good. If the goal of a public health intervention is not to merely provide pure public goods but to also compensate for market failure (Horne, 2019), and a private institution can do that, then it may also intervene. In fact, the government’s role in administering covert compulsory moral bioenhancement may be superfluous. That’s because in such a program there is not likely to be significant market failure in providing the intervention. There is not likely to be significant market failure, because there won’t be a market at all. The people who are intervened upon won’t know it’s happening, so they won’t be in a position to free-ride or need to be assured that their contributions won’t go to waste. People won’t be in a position to intentionally withdraw from cooperation, and people won’t be able to intentionally ride on the contributions of others. To the extent that market failure depends on participants’ awareness of their participation, a covert program would resolve the market failure. Without a need to compensate for market failure, there is little need for the government, rather than a private institution, to administer covert compulsory moral bioenhancement. To claim that only governments can administer such an intervention requires some other moral argument (Anomaly, 2015), such as that from Childress et al., but one that is sound. In the absence of an additional argument, private institutions may provide covert compulsory moral bioenhancement, even if they are for-profit. So long as the people administering

130 Transparency

it have themselves been enhanced, then they could permissibly administer it. For example, suppose Jeff Bezos and Amazon recognize that their longterm interests require lots of people having expendable income and that a devastated climate or a population ravaged by infectious disease or meteor impact would significantly undermine those interests. So long as Bezos and company have been themselves enhanced, they could permissibly administer the intervention. Usually, such arrangements imply a conflict of interests. But in this case the aims of the intervention and the fact of the administrators having been enhanced themselves ensure an alignment of interests.1 Of course, if the compulsory moral bioenhancement program were overt rather than covert, the only possibility for its success is if the government administers it. In an overt program, there would be significant market failure, and the only way to compensate for it would be by use of force.2

Notes 1 . See Chapter 4. 2. See Chapter 5.

Chapter 9

Libertarian Covert Compulsory Moral Bioenhancement

I have argued that covert compulsory moral bioenhancement is epistemically and morally justified. In this chapter, I  offer another argument that covert compulsory moral bioenhancement ought to be administered. In the previous chapter, I argued that covert compulsory moral bioenhancement does not threaten liberty in any way that defeats its moral justification. But in this chapter, I push this claim. Of all of the political theories, the one that would seem to be most incompatible with covert compulsory moral bioenhancement is libertarianism. My aim in this chapter is to demonstrate that covert compulsory moral bioenhancement is compatible with libertarianism. If libertarianism is compatible with covert compulsory moral bioenhancement, then presumably other political theories will be as well. If the libertarian can accept that one ought to be given a moral enhancer through the water system without their knowledge, then presumably so can the contractarian or the egalitarian. But I do not argue for this further claim. Libertarianism is the test case. My argument requires the empirical premise that in the absence of covert compulsory moral bioenhancement the people who would have been enhanced are likely to experience significant liberty restrictions. Consider not even the worst-case scenario, one in which climate change merely exerts too much pressure on governments, undermining their ability to uphold the laws that democracies have agreed to, such as laws prohibiting taking another person’s stuff. In such a scenario one’s scheme of liberties might be as extensive as it could possibly be. But in that case life is solitary, poor, nasty, brutish, and short (Hobbes, 1668/1994), all conditions that make it impossible to exercise those liberties and that themselves place significant limits upon one’s self-determination. Or consider this more mundane but highly likely scenario. Currently, those who are most likely to necessitate covert compulsory moral bioenhancement enjoy an extensive scheme of liberties. We can fly on jets to almost any place we would like to go; we can cool our homes; we can eat beef, build new wood houses near the coasts, and drive vehicles. Climate change, even if not catastrophic, will limit our liberty to engage in these and

132  Libertarian Moral Bioenhancement

many other behaviors. Beef, air travel, air conditioning, and internal combustion engines may all become impossible. All that is required for my argument to work is that the scope of liberties that people currently enjoy narrows due to climate change or nuclear war or amok artificial intelligence or synthetic bird flu or whatever else. As living conditions become worse, our ability to use whatever liberties we have is undermined. For my argument, it is important to distinguish between being unable to use the liberty and not having the liberty itself. I don’t need the premise that the effects of climate change or nuclear war or whatever implies that we won’t have the liberties we currently enjoy. Indeed, if I needed this premise I’d be begging the question against the libertarian. Rather, I need the premise that we won’t be able to enjoy or use these liberties. The scope of liberties we have may not narrow. The scope of liberties we can enjoy will. Libertarians must accept that some intrusions on liberties are permissible. Otherwise, my neighbor’s lawn mowing or evening grilling would be impermissible intrusions on my property. In order to protect his liberty to use his property in the way he wants, the mainmast of the libertarian ship, we must allow some intrusions of my property. My argument is roughly that covert compulsory moral bioenhancement is like my neighbor’s lawn mowing or grilling. In order to protect the future use of liberty, we must (unknowingly) accept the intrusion that is moral bioenhancement.

Liberty and Ultimate Harm There are many possible collective risks that bring with them significant suffering. Among these collective risks, climate change seems most likely to engender the most suffering, perhaps not for those living, but for near-future generations. Changing climate may result in total disruption of agriculture worldwide, leading to famine. Warmer temperatures mean that mosquitoborne illnesses and other diseases more easily thrive. Rising oceans flood coastlines. Rivers dry up. It’s a familiar story by now. Life will be much less pleasant. Often the focus of the effects of climate change is on the disutility of it. Often ignored are the effects climate change has on our liberty. If climate change restores our living conditions back to a state of (spoiled) nature, we may have a more extensive scheme of liberties. But we won’t be able to use them. Everyone may have the liberty to sew wheat or corn or rice, but if the land is too arid, or the water too polluted, that liberty can’t be used. Everyone may have the liberty to build shelter on the coast, but that liberty can’t be used when what counts as the coast keeps moving inland. All people may have equally the liberty to be vegan, but when the only things available to eat are things with legs or eyes, that liberty is not usable. Everyone may have the liberty to protect themselves from assault. But when hunger and thirst have sapped one’s energy, that liberty is much more difficult to use.

Libertarian Moral Bioenhancement 133

That climate change will limit humans’ ability to use liberties doesn’t rely on the assumption that it will push us back into a state of nature. There are many more mundane liberties that we will lose the ability to use. No grass lawns means no ability to use the liberty to mow it, fertilize it, water it, and gaze at it. We may be allowed to cool our homes using electricity or natural gas, but it may be so cost-prohibitive that doing so becomes untenable. We may be allowed to step outside at dusk or dawn, but not be able to use that liberty for fear of contracting a mosquito-borne disease made more prevalent by later frosts. We may have the liberty to live in Phoenix or New Orleans or the Maldives, but not be able to use those because they are too hot or flooded. It’s possible the climate change may expand both the liberties we have and the liberties we can use, but this seems highly unlikely. Climate change will reduce the total available resources. These resources are required to use whatever liberties we have. So it is intuitive that such a significant reduction in resources will result in a significant reduction in ability to use liberties. The same could be said of the other possible causes of ultimate harm. If they occur, they will only make it more difficult to exercise whatever liberties we have, even if the range of liberties we have widens. There are many varieties of libertarianism. These varieties generally share the core notions that people have extensive and weighty rights against others intervening on their property and how they want to use it, where the property is understood as encompassing the body. Because the rights of people to not be interfered with are extensive and strong, the range of things the state can do to permissibly coerce behavior is narrow. Matt Zwolinksi (n.d.) describes what libertarians share: Libertarians are committed to the belief that individuals, and not states or groups of any other kind, are both ontologically and normatively primary; that individuals have rights against certain kinds of forcible interference on the part of others; that liberty, understood as noninterference, is the only thing that can be legitimately demanded of others as a matter of legal or political right; that robust property rights and the economic liberty that follows from their consistent recognition are of central importance in respecting individual liberty; that social order is not at odds with but develops out of individual liberty; that the only proper use of coercion is defensive or to rectify an error. An implication of this family of commitments is that the state cannot permissibly coerce behavior on the grounds that the intervention is for the person’s own good. Libertarianism is strongly anti-paternalistic. The state can’t permissibly prevent me from vaping nicotine, for example, on the grounds that I’m better off not vaping. The state doesn’t get to say what makes me better or worse off—only I can make that judgment.

134  Libertarian Moral Bioenhancement

This is not to say that the state can’t interfere with or coerce a person’s behavior. But what morality prohibits far outstrips what the state may permissibly prohibit. Jason Brennan explains (2018, pp. 39–40): For instance, it would be wrong for me to, for no reason at all, forbid my parents from ever again seeing their grandchildren. It would be wrong for Harry to, just out of cruelty, jilt Sally at the alter. It would be wrong to write a book advocating genocide. Nevertheless, libertarians and most liberals would conclude that the state may not enforce morality in these cases; in these cases, it may not use coercion to make individuals do the right thing. However, libertarians (and liberals in general) hold that some moral duties are enforceable. The state cannot use coercion to make me let my parents visit their grandchildren, but it can coerce me into feeding my kids. It may not coerce Harry to marry Sally, but it may coerce Harry to stop him from attacking Sally. It may not coerce me to stop me from advocating genocide, but may coerce me to stop me from actually engaging in genocide. The state may also coerce behavior in order to protect the rights of others. The rights of others are protected in Brennan’s examples above. As Zwolinksi points out, it can only do so if it is to protect or defend a right or rectify an error. Defense or protection of a right includes defense or protection of the ability to use it. The state may permissibly enforce limitations on the height of my neighbor’s fence, for example, so that my right to use my property in the way I want to use it is protected. I wouldn’t be able to use my property in the way I wanted to if my neighbor were permitted to build a fence a hundred feet high. There is nothing about property easements that is necessarily at odds with libertarianism. The point that liberty protection includes protection of liberty use is important in demonstrating that covert compulsory moral bioenhancement is compatible with libertarianism, as I argue that covert compulsory moral bioenhancement protects liberty use.

Clean Hands I am claiming that covert compulsory moral bioenhancement is compatible with libertarianism. Covert compulsory moral bioenhancement is a public health intervention aimed at preventing widespread significant suffering from, for example, the effects of climate change. Libertarianism is of course compatible with a wide range of public health interventions, such as taxes on tobacco products and even quarantine, the forced isolation of individuals who may transmit infectious diseases. Recently, Jason Brennan has argued that libertarianism is compatible with compulsory vaccinations. Vaccinations present a greater challenge for the libertarian than, say, quarantine, because one’s negative right to bodily integrity is at odds with the welfare of others

Libertarian Moral Bioenhancement 135

but the risk of transmission, and thus risk of harm, is still very low even if a person fails to be vaccinated. Requiring a person to be vaccinated requires a forced intrusion of their skin with a needle so that they and the people around them don’t get sick. This sacrifices the right to bodily integrity for welfare of others. But the forced intrusion is also for the good of the person being vaccinated, which may be overly paternalistic for the libertarian. Jason Brennan (2018) has argued that libertarians should not only accept compulsory vaccinations, but that they should endorse them. His argument rests on the clean hands principle, which is that there is an enforceable moral obligation to not participate in collectively harmful behaviors. To motivate the clean hands principle, he first considers the following case, Firing Squad (p. 40): A band of 10 sharpshooters is about to kill an innocent child. They have been trained to shoot in such a way that each shot will hit the child at the same time, and each shot would be fatal on its own. You can’t stop them from killing the child. They ask you if you’d like to join in and take the 11th shot. Brennan claims that it is wrong to accept the invitation to be the 11th shooter and that it would be permissible to enforce this prohibition. But if it is wrong to be the 11th shooter, and the state can enforce this prohibition, then the state can also compel vaccination, because to do so is simply to prevent a person from participating in a similarly harmful activity. The clean hands principle permits the prevention of a person participating in a collectively harmful activity. But according to Brennan it also permits the prevention of collectively risky activity. He considers a case more analogous to those who fail to be vaccinated (p. 41): Elon Musk has just invented instantaneous interplanetary teleportation, and the technology is widely available. Suppose a group of privatelyfunded astronauts plans to visit a newly discovered planet, a planet that, for all they know, contains a wide range of deadly bacteria and viruses. When they arrive, they drink the water, without sanitizing it. They also give the possibly contaminated water to their children. When they arrive back home a day later, they refuse quarantine. Some of them visit Disneyland, while others immediately place their (for all they know, infected) children in daycare centers or schools. They could have taken steps to sanitize the water samples and to prevent themselves from contracting any alien diseases, but they decided not to do so, because they get their health advice from Jenny McCarthy. This case is analogous to those adults who refuse vaccinations for themselves and their children. The clean hands principle explains why we judge

136  Libertarian Moral Bioenhancement

the astronauts as having done something wrong. Since the clean hands principle permits the state to require the returning astronauts to be isolated, it is similarly permissible to require vaccination. According to Brennan, the clean hands principle permits compulsory vaccinations. He guards against objections from the libertarian by noting that the libertarian ought to endorse the clean hands principle, for if she doesn’t then impermissible state interference can be permitted simply by ensuring that collectives, rather than individuals, interfere. He imagines a fascist saying (p. 41): We’ll engage in war, censorship, mass murder, and impose high taxes, mass regulation of the economy, and ramp up the war on drugs, all things which libertarians hate. But, to make sure we don’t run afoul of libertarianism, we’ll always impose these action and rules through collective activities. For instance, when we shut down libertarian websites, we’ll train a team of censors to do so simultaneously, such that the actions are always overdetermined, as in the Firing Squad case! According to Brennan, what allows the libertarian to reject the above fascist’s actions is the clean hand principle. Thus, the libertarian ought to accept the clean hands principle and endorse compulsory vaccination.

Clean Hands and Covert Compulsory Moral Bioenhancement Preventing the suffering from climate change is similar in many ways to preventing suffering from measles or polio. The risks associated with climate change and those associated with infectious diseases are both collective— everyone is at risk, though some more than others. But the actions needed to mitigate the risk and prevent the suffering are also collective. Everyone must contribute in order for the risk prevention to work. If enough people fail to be vaccinated, then measles or mumps or whooping cough will spread. Similarly, barring any technological development that lets us all off the hook, everyone will have to contribute to any intervention aimed at preventing suffering from climate change. At the level of the nation, this is certainly true. It makes no sense for Europe to work diligently at preventing suffering from climate change if the U.S. is going to do nothing. Collective cooperation is required. Brennan argues that compulsory vaccinations are compatible with libertarianism by the clean hands principle. I argued earlier that if we are to prevent the suffering associated with climate change, then covert compulsory moral bioenhancement is necessary. If vaccines and preventing the suffering from climate change are analogous, then the clean hands principle should also work to make covert compulsory moral bioenhancement

Libertarian Moral Bioenhancement 137

compatible with libertarianism. Consider a modified version of Brennan’s case: A group of people drives vehicles that emit harmful chemicals into the atmosphere. They heat and cool their homes using energy that emit these same chemicals. They eat foods the production of which emit different though equally harmful chemicals. They buy other luxury products that are manufactured and delivered by processes that entail the emission of other chemicals. Individually these behaviors have a negligible impact, but collectively these behaviors suffice to warm the planet, which in turn undermines the ecosystems that enable human life and causes significant suffering. This group of people could use alternative methods of transportation, heating, and cooling without loss of anything of moral significance. They could eat other foods without sacrificing any health benefits. They could forego the luxury products. But they choose not to do any of these things. Whether the group consists of one member or a billion members is irrelevant. As long as their behavior suffices to engender the risk to others, and that risk is unacceptably high, the clean hands principle warrants interference in the behavior of these individuals. If it is true that covert compulsory moral bioenhancement is necessary to prevent the suffering associated with climate change in the way that vaccines are necessary to prevent the spread of measles or polio or whooping cough, and there is libertarian justification for requiring vaccinations, then it should also justify covert compulsory moral bioenhancement on libertarian grounds. The contention will be whether the risks associated with climate change are unacceptably high. In the case of compulsory vaccinations, Brennan adopts Sven Hansson’s account of acceptable risk (2018). Hansson claims that exposing others to risk is acceptable “if and only if this exposure is part of an equitable social system of risk-taking that works to her advantage.” To be acceptably imposed on another person, the risk must be associated with a net advantage to the individual. Imposing the risks associated with driving is acceptable, because it is to everyone’s advantage. But, according to Brennan, imposing the risks associated with infectious diseases is not acceptable, because for the person upon whom the risk is imposed the risk is not to her advantage. We could say the same thing about imposing risks associated with climate change upon others. When the people in the above case impose the risk of having to move my residence or fight harder for scarce resources or be infected with a mosquito-borne illness, that is not to my advantage or to any other person upon whom those risks are imposed. Thus, the risks of climate change are unacceptable, on Hansson’s account. So long as the collective behavior suffices to engender this unacceptably high risk, the clean

138  Libertarian Moral Bioenhancement

hands principle warrants interference. If covert compulsory moral bioenhancement is the only feasible intervention, then it would be seem to be justified. And because that justification rests on the clean hands principle and Hansson’s account of acceptable risk, both of which are compatible with libertarianism, covert compulsory moral bioenhancement appears to be compatible with libertarianism. Whether covert compulsory moral bioenhancement is justified by the clean hands principle along with Hansson’s account of acceptable risk depends ultimately on whether covert compulsory moral bioenhancement or the suffering from climate change is more advantageous to the person. That is, if what justifies on libertarian grounds the idea that moral bioenhancement should be compulsory and covert is the principle that people have an enforceable obligation to not participate in collectively harmful activities, and the notion that it is not to the net advantage of those upon whom the risk of harm from climate change is imposed, then whether covert compulsory moral bioenhancement is justified on such grounds is dependent on covert compulsory moral bioenhancement being a greater net advantage. What matters is the welfare of the individual upon whom the risks are imposed. If they are at a net advantage by having covert compulsory moral bioenhancement rather than risk suffering from climate change, then the clean hands principle can justify covert compulsory moral bioenhancement. But it’s possible that they’re worse off by having covert compulsory moral bioenhancement than they would be if they suffered the harms associated with climate change. I find it implausible that a person will be worse off by having covert compulsory moral bioenhancement than they would be if they suffered the harms associated with climate change. If it’s welfare we’re concerned about, covert compulsory moral bioenhancement doesn’t even need to be balanced by harms, as covert compulsory moral bioenhancement can only make a person better off. In any case, as the harms associated with climate change become more severe such that they constitute ultimate harm, so does the implausibility of the claim that a person is worse off from covert compulsory moral bioenhancement than they are from the harms associated with climate change. I also find Brennan’s argument compelling. But for the present purposes, it doesn’t work. The purpose of the present chapter is to demonstrate that covert compulsory moral bioenhancement is compatible with the even most protective libertarianism, because if it is compatible with the most protective libertarianism, then it is likely compatible with most other, less protective political theories. The most protective libertarianism does not accept welfare arguments as providing reasons to coerce or interfere with a person (Bernstein, 2017; Blunden, 2019). If proponents of protective strains of libertarianism accepted welfare arguments, then they would accept that covert compulsory moral bioenhancement is (morally, epistemically, politically)

Libertarian Moral Bioenhancement 139

justified, because the arguments for covert compulsory moral bioenhancement are already welfare arguments. Brennan’s argument for compulsory vaccination is at its foundation a welfare argument, so it won’t work to extend that argument to covert compulsory moral bioenhancement to demonstrate compatibility with a more protective form of libertarianism. My strategy instead is to show that covert compulsory moral bioenhancement is compatible with the sort of libertarianism that holds that the only permissible political interventions on a person’s life are those that protect others’ negative rights. In particular, some of these permissible interventions are those that protect not only the maintenance of others’ negative rights but also are those that protect others’ liberty to use those negative rights. Covert compulsory moral bioenhancement is compatible with libertarianism because covert compulsory moral bioenhancement is necessary to protect others’ ability to use their liberties, and libertarianism allows such protective interventions. In the next section, I set up the argument with discussion of an objection to libertarianism from Peter Railton and David Sobel. That claim is, roughly, that if libertarianism takes seriously the idea that individual liberty is a normative fundamental, then it must prohibit even minor intrusions of these liberties. But if it so prohibits, we wouldn’t be able to get around in the world. One libertarian response to this objection is that some minor intrusions are permissible. Intrusions may be permissible so long as the person intruded upon is adequately compensated (Nozick, 2013) or so long as the intrusion serves to protect the ability of others to use their liberties (Mack, 2015). In the subsequent section I argue that if covert compulsory moral bioenhancement is an intrusion at all, it is merely a minor intrusion.

Minor Intrusions Peter Railton (2009) and David Sobel (2012) have argued that a libertarianism protective of individual liberties is challenged by the fact that strong protections of liberties makes the exercise of those liberties more difficult. Eric Mack summarizes the problems (pp. 194–195): Both of these papers maintain that the rights of self-ownership and of private property in extra-personal holdings . . . are much more restrictive than their advocates generally recognize. The key thought is that libertarian rights demand that all agents desist from even the most minor unprovoked and non-consensual physical intrusions upon rightholders; in addition, these rights at least seem to demand that all agents desist from imposing on others even modest risks of such intrusions. Yet, compliance with these demands requires that individuals forego all or almost all of the ordinary exercises of rights by which individuals maintain their lives and achieve their ends.

140  Libertarian Moral Bioenhancement

The problem is that strong protections of individual liberties “hog-tie” the behavior of right-holders. If my neighbor has a strong property right, then when I  grill in my backyard and some smoke drifts onto her property, if she hasn’t consented to that intrusion, then my grilling violates her property right. Thus, by libertarianism, I must not grill in my backyard in a way that causes smoke to drift onto her property. But then this hog-ties my behavior. My use of my property rights (such as those to do with my property what I will) becomes morally impermissible. The challenge for the libertarian is to find a way to maintain the protection of a person’s liberties while also allowing their use. That is, the libertarian must find a way to balance the permissibility of using one’s liberty with the impermissibility of intruding on others’ liberties. One suggestion (Epstein, 1979; Sobel, 2012) is that intrusions of one’s property are tempered or made more palatable if one is adequately compensated. There are several ways one could articulate this idea. Richard Epstein claims that the compensation is in-kind. Intrusions of one’s property are tempered by that person enjoying the similar opportunity to intrude upon others. Mack explains (p. 201): Minor intrusions are to be allowed primarily because the general allowance of such intrusions provides each individual who is subject of these intrusions with in-kind compensation; they are to be allowed even though these compensated for intrusions seem to remain somewhat morally tainted. Presumably more extensive intrusions are also to be allowed if those subject to them receive the due in-kind or out-of-kind compensation. Compensation for an intrusion makes that intrusion allowable, on this account. The hog-tying problem is solved because the right is protected but also able to be used, so long as the possessor of the right upon which one intrudes is adequately compensated. Mack finds this solution to the hog-tying problem unsatisfactory. David Schmitz has claimed that the core of a property right is the “right to say no to the proposed terms of exchange” (Schmidtz, 2010, p. 79). Mack’s problem with this hog-tying solution is (p. 202): One’s right to say no is not honored by others infringing upon that right while providing one with the payment one would have accepted if one had not said no. One’s right to say no is not honored if the treatment one objects to is imposed upon one—even if it is true that, had one deliberated well, one would have waived one’s right against the treatment in exchange for the payment that is now offered as compensation. Mack’s strategy is to develop an alternative path between protecting the possession of liberties and protecting the use of them. He argues for the

Libertarian Moral Bioenhancement 141

“elbow room postulate,” which is the idea that “a reasonable delineation of basic moral rights must be such that the claim-rights that are ascribed to individuals do not systematically preclude people from exercising the liberty-rights that the claim-rights are supposed to protect” (p. 197). Solving the hog-tying problem by allowing intrusions so long as the person intruded upon is adequately compensated, according to Mack, weakens the notion of the rights. If we disallow the person the opportunity to say no to the terms of the proposed exchange, then the right to say no was not as strong as the libertarian wants it to be. Instead of weakening the notion of a right, he proposes to redraw the boundaries of what is allowable. Mack’s claim (p. 212) is that as long as a minor intrusion upon one’s property is not wanton and not malicious, then that intrusion does not count as an impermissible boundary crossing. If we are to use our person and property in the way that we will, Mack claims, such “elbow room” for minor intrusions suffices to solve the hog-tying problem. Mack writes that those intrusions that are wanton or malicious are those that “done for or on the verge of being done for their intrusiveness.” It’s not a precise definition but it’s good enough. Non-wanton and non-malicious intrusions are those in which intrusion itself is not among the aims of the intruder. Minor intrusions, according to Mack, are those intrusion which are “incidental to the agent’s decisions about how to deploy his person or property we reasonably view these deployments as fundamentally exercises of that agent’s rights” (p. 212). If an intrusion upon a right is incidental to the agent’s decisions about how to use that right, then that intrusion is a minor one. There are thus two conditions for an intrusion of one’s property or person to be allowable. An intrusion upon A’s property is allowable if it is (a) not done for its intrusiveness and (b) incidental to A’s decisions about how to use that person or property that she has the right to use. Allowing such intrusions solves the hog-tying problem and draws the libertarian boundaries for which intrusions are allowable and which are not. Mack’s elbow room postulate may solve the hog-tying problem. But it still might be too restrictive. When my neighbor barbecues in the evening and that smoke drifts into my open windows, that intrusion is not incidental to how I decide to use my property. I shut the windows because of the smoke—her outdoor cooking causes my shutting the window. But her outdoor cooking seems like it’s an allowable intrusion. The in-kind compensation solution to the hog-tying problem may more easily handle this case. I’m compensated in-kind by the opportunity to barbecue on my own property, the smoke of which may drift through her windows, and it’s this in-kind compensation that permits the intrusion. Thus, for the elbow room postulate to work, it requires a very permissive notion of what’s incidental and what isn’t. That Mack’s account requires a permissive notion of what’s incidental is no problem for the present purpose, however, because the aim in this

142  Libertarian Moral Bioenhancement

chapter is to demonstrate that covert compulsory moral bioenhancement is compatible with libertarianism. In the next section, I make this case. My argument is that if covert compulsory moral bioenhancement is indeed an intrusion (which it may not be), then adequate (not in-kind) compensation reasoning as well as elbow room reasoning allows it. Covert compulsory moral bioenhancement is a minor intrusion akin to other intrusions that we know nothing about, such as environmental pollutants or the neighbor who always picks up after their dog trained to use your lawn as its toilet. If covert compulsory moral bioenhancement is not allowed, then libertarianism ends up with another sort of hog-tying problem—climate change (nuclear war or bioterrorism, etc.) undermines our ability to use our person and property in the way we will.

Compatibility Between Covert Compulsory Moral Bioenhancement and Libertarianism There are at least two ways one might argue that covert compulsory moral bioenhancement is compatible with libertarianism. Brennan’s clean hands principle, in conjunction with some account of what risk is acceptable, may justify covert compulsory moral bioenhancement in order to prevent the suffering that results from climate change or some other way in which humans experience ultimate harm or something close to it. In principle, Brennan’s argument suffices to make covert compulsory moral bioenhancement compatible, so long as the clean hands principle is compatible with libertarianism. The argument relies on considering others’ welfare as a reason to compel behaviors. But if one is already disposed to take seriously such welfare arguments, then there is not likely a need for Brennan’s clean hands principle, for the moral argument supporting covert compulsory moral bioenhancement is itself a welfare argument. Another way to show that covert compulsory moral bioenhancement is compatible with libertarianism is to argue that the interventions involved in covert compulsory moral bioenhancement are minor intrusions that must be allowed so that others can exercise their rights to use their property and person—covert compulsory moral bioenhancement protects liberty use. There are two ways to make this case. One is to adopt one solution to the hog-tying problem and argue that the interventions involved in covert compulsory moral bioenhancement are allowable intrusions so long as those who receive them are adequately compensated. The other is to claim that the interventions involved in covert compulsory moral bioenhancement fall into the “elbow room” that libertarianism is supposed to be allowed. I take these on in order, but ultimately adopt the second solution. The solution to the hog-tying problem that allows minor intrusions when they are adequately compensated comes in two types. One type is found in Nozick (2013) and consists of the simple claim that minor intrusions of one’s property and person are allowable so long as they are adequately

Libertarian Moral Bioenhancement 143

compensated. Another type is that minor intrusions are allowable when they are compensated in-kind, or compensated with the opportunity to similarly intrude upon others’ property and person. These notions of compensation seem like a poor fit for demonstrating compatibility between libertarianism and covert compulsory moral bioenhancement. In-kind compensation, in particular, will not work. The interventions involved in covert compulsory moral bioenhancement are intrusions (if they are intrusions at all) upon one’s psychological states. Inkind compensation would require that those who receive the interventions involved with covert compulsory moral bioenhancement are offered the opportunity to similarly intrude on others’ psychological states. But in the case of covert compulsory moral bioenhancement, they will be offered no such opportunity, but not because it would be wrong for them to be so offered. Rather, to offer such an opportunity would be to offer an opportunity that could not be feasibly exercised. We already have opportunity to intrude upon one’s psychological states. I’m doing so now with these words. When the driver of the vehicle behind me honks their horn, they are doing so to me. Expanding the opportunity to intrude upon others’ psychological states beyond current opportunity is not feasible. Furthermore, in-kind compensation would seem to require that the person who is compensated knows they are receiving that compensation. It’s difficult to take advantage of opportunities we don’t know exist. Not informing a person about in-kind compensation they are entitled to is like the government putting a tax break into the tax code but not telling anyone. For the person receiving the intervention involved in covert compulsory moral bioenhancement, they would need to know they are receiving inkind compensation for that intrusion. But to know that one is being compensated for an intrusion is to know that an intrusion has taken place, which undermines the covert nature of covert compulsory moral bioenhancement. Something similar could be said about the other type of compensation, which is simply that minor intrusions are allowable when they are adequately compensated. On one hand, when a person receives compensation, they typically know that they are receiving compensation and thus they know that some exchange took place to warrant that compensation. Such knowledge would undermine covert compulsory moral bioenhancement. On the other hand, covert compulsory moral bioenhancement is intended to prevent the suffering that is inevitable in the absence of such an intervention. The compensation for the intrusion into one’s psychological states will not be some other thing lumped on top of what a person already has. Instead, the compensation will be that they get to keep what they already have, or at least more of it, where if they were to not receive the intervention they would lose it. The compensation is the prevention of otherwise inevitable loss, including the prevention of the inevitable loss of their ability to use their liberties. Whereas compensating someone by giving them something is likely to be overt, undermining any attempt at conversion,

144  Libertarian Moral Bioenhancement

preventing loss can be conducted covertly. Interventions to prevent loss are common and frequently unacknowledged by those whose property and person are protected from loss. For example, the administration of fluoride in the public water supply prevents the loss of tooth enamel, but many are unaware of this intervention aimed at preventing loss. Presumably fluoride could be administered covertly. Allowing minor intrusions when they are adequately compensated may make covert compulsory moral bioenhancement compatible with libertarianism if (i) the libertarian can accept this general strategy in solving the hog-tying problem and (ii) protecting the ability to use one’s liberties (and exercise rights) that they would otherwise lose is adequate compensation for the intrusion. I don’t address (i), but some have argued that the libertarian ought not to accept this strategy (e.g., Sobel, 2012; Mack, 2015). I address (ii) below. Another way of demonstrating that covert compulsory moral bioenhancement is compatible with libertarianism is by showing that it falls within the “elbow room” that Mack claims libertarianism can use to allow for minor intrusions. Recall that on his view the satisfaction of two conditions is sufficient for falling within the elbow room of libertarianism. An intrusion upon A’s property is allowable if it is (a) not done for its intrusiveness and (b) incidental to A’s decisions about how to use that person or property that she has the right to use. Given the protection of liberty use is a core component of any strain of libertarianism, I  assume that the prevention of the loss of liberty use that would occur from climate change is adequate compensation for minor intrusions. Thus, if covert compulsory moral bioenhancement satisfied Mack’s two conditions, not only would covert compulsory moral bioenhancement be compatible with libertarian elbow room, but also with the notion that minor intrusions are allowable in exchange for adequate compensation. That is, if covert compulsory moral bioenhancement satisfies (a) and (b), then it also satisfies (ii) above. And if covert compulsory moral bioenhancement can do all that, then there is good reason to think that covert compulsory moral bioenhancement is an intrusion the libertarian can allow regardless of whether they solve the hog-tying problem by invoking the notion of adequate compensation or by invoking elbow room. Whether preserving the ability to use liberties is adequate compensation for covert compulsory moral bioenhancement depends on how intrusive covert compulsory moral bioenhancement is.

Covert Compulsory Moral Bioenhancement Is a Minor Intrusion It is straightforward that covert compulsory moral bioenhancement satisfies (a) above, that an intrusion not be done for its intrusiveness. Covert

Libertarian Moral Bioenhancement 145

compulsory moral bioenhancement is done for the sake of liberty use protection, not for the sake of intrusion. This is to say that covert compulsory moral bioenhancement is neither wonton nor malicious. At most, the intrusiveness of the interventions involved with covert compulsory moral bioenhancement is an unintended but necessary double effect of the primary intention, which is to preserve the ability of others to use their liberties. The greater challenge is arguing that covert compulsory moral bioenhancement satisfies (b), that the intrusion be incidental to A’s decisions about how to use that person or property that she has the right to use. The foundation of the compatibility between covert compulsory moral bioenhancement and libertarianism is that covert compulsory moral bioenhancement is necessary to protect the ability of others to use their liberties, an ability that climate change would otherwise undermine. This foundation presumes that others are entitled to use these liberties. Thus, whether covert compulsory moral bioenhancement counts as a minor intrusion depends on whether the interventions involved in covert compulsory moral bioenhancement are incidental to a person’s decision about how to use their property or person. It may seem as though covert compulsory moral bioenhancement is not incidental to a person’s decisions about how to use the property or person upon which covert compulsory moral bioenhancement intervenes. The intervention aims to manipulate moral capacities and subsequent moral behavior. The whole point of covert compulsory moral bioenhancement seems as though it is not incidental to how a person makes decisions about transportation, food, living conditions, policies, democratic representation, and so on. However, covert compulsory moral bioenhancement is incidental to a person’s decisions about how to use the property or person that the intervention intrudes upon. It must be incidental, for if it is not then very many ordinary intrusions into a person’s mental states and subsequent behavior are similarly non-incidental, and therefore not allowable intrusions. Because ordinary intrusions would count as non-incidental and not allowable, libertarianism would be faced with a new hog-tying problem, one that prohibits the ordinary use of the mind to protect from non-incidentally intruding upon others’ minds and behaviors. In other words, the interventions involved in covert compulsory moral bioenhancement must be incidental, or else libertarianism has a new hog-tying problem. Loud, irritating noises affect moral judgment, and potentially behavior (Mathews & Canon, 1975; Seidel & Prinz, 2013) The same might be said of finding money, however trivial (Isen & Levin, 1972). Suppose a person is walking and finds a crumpled on the sidewalk a $10 bill. After searching for someone who might have dropped it, she puts it in her pocket. Finding this money affects her psychology in some way. A  few minutes later she witnesses someone drop a bag of groceries, and she runs over to help. Research suggests that she may be more likely to help, given that she found

146  Libertarian Moral Bioenhancement

the money. Had she not found the money, she would not have helped. Whoever dropped that $10 bill intruded upon the person’s mental state and subsequent moral behavior. The empirical research demonstrating the various factors that influence moral judgment and behavior is vast. Or consider the more mundane example that others’ moral behavior partly affects an individual’s moral behavior. The public goods game is a game in behavioral economics that is used to investigate how people exchange resources when incentivized by acquiring or risking a public good. In such circumstances, how one behaves—the amount a person contributes to the collective pot—depends in part on how much others contribute. If a person sees another player start to keep their money and not contribute (defect) they are more likely to do the same. Thus, how one behaves in ways that affect others’ well-being is partly affected by how others behave. This means that the other person’s behavior intrudes upon one’s own moral behavior. In both of the above examples, something intrudes on the person’s moral behavior, and presumably also upon their moral psychology. In the first case, it is finding a dropped $10 bill and in the second case it is another person’s behavior. If covert compulsory moral bioenhancement is non-incidental to moral judgments and behavior, then neither are either of these intrusions. One might object that in these two cases the found money and the other person’s behavior do not constitute intrusions, because they are not intended as such, but that covert compulsory moral bioenhancement is intended as an intrusion. But this hardly matters to whether something is an intrusion. Intent may matter in determining whether a given intrusion is allowable, but it is irrelevant to determining whether it is an intrusion in the first place. If in the process of erecting a new fence around his property my neighbor inadvertently puts a portion of the fence on my property, he has intruded on my property. It is irrelevant that he did not intend to do so. He may have even put reasonable effort into not putting his fence on my property by leaving a margin between the fence and what he thinks is property line or by getting a survey. But if his fence is nevertheless on my property, it is an intrusion. Intent is irrelevant to whether something is an intrusion, but it is also irrelevant to whether an intrusion is incidental to a decision. Whether an intrusion is incidental or not is purely a matter of the causal relation between the intrusion and the decision. If the intrusion closes off other decisions that one otherwise might have made, or if it simply makes it more likely that one decides as one does, then it seems as though the intrusion is not incidental to the decision. But if in the case of moral behavior such intrusions are counted as non-incidental, and being non-incidental grounds it being disallowed, then libertarianism will disallow many ordinary intrusions. For two examples, dropping $10 bills that might later be found would not be allowed, nor would be certain behaviors involved in acquiring or risking public goods. One couldn’t run a loud motor, as it may non-incidentally

Libertarian Moral Bioenhancement 147

affect another’s moral behavior. And so on. The range of things that intrude upon a person’s moral psychology and affect subsequent moral behavior is wide. All of these things would be prohibited on libertarianism. This prohibition creates another hog-tying problem, though. If running a loud motor is not allowed because it non-incidentally intrudes upon my neighbor’s moral behavior (e.g., makes him more likely to kick his dog when it barks at my loud motor), then I am prohibited from using my property in the way that I  want. This problem generalizes to any behavior that nonincidentally intrudes upon others’ moral psychology and behavior, which, given the range of behaviors for which this is true, is very many. This hogties a person’s ability to use their liberties to mow their lawn, to withhold resources from public goods, and to refrain from securing their cash in a wallet or purse, and so on. It is better for libertarianism for the interventions involved in covert compulsory moral bioenhancement to be considered incidental. They are also not done for their intrusiveness. Since being incidental to the decision and not done for their intrusiveness suffices for an intrusion to fall into the elbow room of intrusions that libertarianism can allow, covert compulsory moral bioenhancement is allowable on libertarian grounds, if the libertarian endorses the elbow room solution to the hog-tying problem. But the libertarian may instead endorse the adequate compensation solution. In such a case the fact that the interventions involved in covert compulsory moral bioenhancement are incidental and not wanton or malicious suggests that compensating for the intrusion by preserving and protecting the ability to use liberties that would otherwise be lost is more than adequate. Thus, covert compulsory moral bioenhancement is compatible with libertarianism.

What Difference Does Covert Make? To be compatible with libertarianism, it is important that the moral bioenhancement be covert. Libertarianism and covert compulsory moral bioenhancement are compatible because retaining the ability to use important liberties that otherwise would be lost is adequate compensation for the intrusion of covert compulsory moral bioenhancement and that covert compulsory moral bioenhancement falls into the elbow room that libertarianism makes for itself. If the compulsory moral bioenhancement were overt rather than covert, then it is more difficult to make the case that either of these things is true. Consider first that the compensation for overt compulsory moral bioenhancement would have to be much higher than that of covert compulsory moral bioenhancement to be considered adequate. The compensation would not only have to compensate for the intrusion, but also for a person’s knowledge of being intruded upon. The conjunction of being intruded upon and knowing it is much worse than being intruded upon and not

148  Libertarian Moral Bioenhancement

knowing it. Suppose that a virtual reality technology was developed for use in prisoners. When a prisoner is put in his cell he dons the virtual reality, which functions in a way such that he doesn’t know he’s in prison. Instead, it recreates his life and environment in a way that is perceptually seamless with this life prior to prison. He believes he’s still in his familiar environment with his normal experiences and exposures, but in fact he is in his cot in his cell. Intuitively, when the warden removes him from the virtual reality and he realizes that he is in prison rather than his normal environment this realization adds to the intrusion. This example isn’t exactly analogous to the compulsory moral bioenhancement, because the intrusion of moral bioenhancement is not comparable to the intrusion of prison. But it does show that knowing about an intrusion is itself an intrusion. Thus, if the way that libertarianism solves the hog-tying problem is by suggesting that minor intrusions are allowed if they are adequately compensated, then overt compulsory moral bioenhancement would need greater compensation than covert compulsory moral bioenhancement. Alternatively, making compulsory moral bioenhancement overt would be so much more intrusive than covert compulsory moral bioenhancement that it would no longer be a minor intrusion, and therefore not compatible with libertarianism. So, for the adequate compensation solution to the hog-tying problem, it is important that compulsory moral bioenhancement be covert. The same could be said for the elbow room solution. Knowing that one is being compelled to be morally enhanced might take the intrusion from the incidental to the non-incidental. Consider, for example, fluoridated water. Many people are ignorant of the fact that most public water supplies are supplemented with fluoride. But of those who know that the public water supply is supplemented choose to avoid the fluoride by buying only bottled, fluoride-free water or by installing filters in their homes or by filtering it directly from the tap. If they were ignorant of the presence of the supplement, they wouldn’t behave in this way. This fact suggests that knowing about the supplement is non-incidental to their decisions about using their liberty to consume public water. If this is right, then knowing about an intrusion may move the intrusion from the incidental to the non-incidental, which, for the elbow room solution to the hog-tying problem, moves the intrusion from an allowable minor one to a non-allowable one. Recall that Jason Brennan reconciles compulsory vaccination with libertarianism by invoking the clean hands principle. I have argued that covert compulsory moral bioenhancement is compatible with libertarianism on the grounds that the interventions involved are minor intrusions. But notice that this strategy won’t work to make compulsory vaccines compatible with libertarianism, because vaccines are plausibly not a minor intrusion, due to the fact that knowing that one is compelled to be poked with a needle and

Libertarian Moral Bioenhancement 149

injected with a solution is not incidental to one’s decisions about health and prevention of disease. Thus, it won’t fall into the libertarian elbow room and may not be something that can be allowed simply by being adequately compensated. However, if vaccines were administered covertly, they may count as a minor intrusion. With respect to their compatibility with libertarianism, the difference between compulsory vaccines and covert compulsory moral bioenhancement lies with the fact that one is overt and one is covert. This difference necessitates a different strategy to make them compatible with libertarianism. The clean hands principle may not work to make covert compulsory moral bioenhancement compatible with libertarianism; and the minor intrusion strategy that I use may not make compulsory vaccination compatible with libertarianism.

Right to Privacy Above I argue that the interventions involved in covert compulsory moral bioenhancement are incidental to one’s decision about using their property or person that they have a right to use. The reason it is better to say that these interventions are incidental is that if they are not incidental and so disallowed, then very many ordinary intrusions upon another person’s moral psychology would be similarly non-incidental. The external world, and especially our perception of other moral agents, bombards our moral psychology, affecting our moral behavior. If we have an obligation to protect against this bombardment, then our moral behaviors are hog-tied. We couldn’t do anything that intrudes upon another person’s moral psychology, which is to say that we couldn’t do much at all. Since this state of affairs would significantly limit the permissibility of using one’s psychology in the way one wants, we shouldn’t say that we have an obligation to protect our moral psychology from intrusion.1 This is simply to say that we have no right to mental privacy—or at least that’s what the libertarian should say. We have a right to privacy of our property. We don’t have a right to privacy of our mind. This difference between property and mind lies in the fact that the range of things that intrudes upon one’s property in a way that significantly affects what they do with their property is much narrower than the range of things that intrudes upon one’s mind in a way that significantly affects what they do with their mind. Because the range of intrusions is narrower, these intrusions can be more easily limited without hog-tying behavior. It’s possible to throw the bathwater out yet keep the baby. But in the case of intrusions upon the mind, the baby goes, too. My neighbor’s loud lawn mower barely affects what I do with my property. But if it causes me to kick the dog, then it significantly affects my moral psychology. If I have a right to mental privacy, then that loud motor intrudes upon that right, which means that, by parity of reasoning, everything else that has a similar effect does, too.

150  Libertarian Moral Bioenhancement

Christoph Bublitz (2019) argues that we have a right to mental privacy, which is rooted in the freedom to form opinions of our own, without interference. He finds that moral bioenhancements violate this right (p. 28): Exposing other persons to such interventions may violate the right to freedom of thought, conscience and religion which protects the forum internum, the formation of opinions and beliefs in moral (and other) matters and which guarantees the freedom to entertain any thought or moral conviction and to remain free from severe external influence, especially psychoactive drugs. This right ought to be so strongly protected that violations of it ought to be considered a crime, according to Bublitz (p. 29): Of great importance is the fact this right is widely considered to be inviolable, i.e. interferences cannot be justified by appeal to other rights, interests of the common good or any benefit and harm calculation. Even more, I suggest that direct interventions into other minds through novel technologies with the aim to change moral beliefs ought to be considered a serious offence, a crime against the mind. It is not plausible that we have an inviolable right to mental privacy and that violations of this right should be considered criminal. Loud noises interfere with one’s moral decision-making. Suppose my neighbor knows this and runs his lawn mower when he thinks I’m considering whether to donate to charity. The noises from the lawn mower cause changes in my brain which result in the decision to not donate to charity. If I  have an inviolable right to mental privacy and intentional violations of this right should be criminally enforceable, then my neighbor has committed a crime, of which I am a victim. This seems absurd. The fact that I  decided to not donate to charity is irrelevant. Suppose instead that my neighbor has discovered that his lawn mower can make pleasant sounds and he modifies the engine so that it does so all the time it is running. He knows that I am deciding whether to donate to charity and runs his motor. The noises from the lawn mower cause changes in my brain which result in the decision to donate to charity. Again, if Bublitz is right my neighbor has committed a crime. That the interference is by way of auditory channels is also irrelevant. Something stimulates the brain; the brain processes that stimulus; the person decides. That the stimulus comes through the ears rather than the eyes or mouth is irrelevant to whether it is a violation of a right to mental privacy. This is just to say that if the interference were visual it would be no more or less permissible or criminal. The same can be said for oral, olfactory, or haptic stimuli.

Libertarian Moral Bioenhancement 151

Egalitarianism Although this chapter is about the compatibility between libertarianism and covert compulsory moral bioenhancement, it is useful to show that such a program is also compatible with other theories that don’t prioritize liberty as much. Some authors (Sparrow, 2014, 2019) have expressed concern that moral bioenhancement is inegalitarian and for this reason should be rejected. Starting from the obvious fact that there are already immoral inequalities of opportunity and inequalities of condition, the worry is that moral bioenhancement will make these worse. As moral bioenhancement is introduced, it will at first be so costly that only those who already have disproportionate opportunities can access it. But the condition of such people is also disproportionately better than that of others. Thus, they will be able to use their disproportionate opportunity to widen the gap in condition between themselves and others. Those concerned with maintaining or promoting equality can readily accept covert compulsory moral bioenhancement. Such a program would promote both equality of opportunity as well as equality of condition. First, if promoting equality, whether it is equality of opportunity, equality of treatment, or both, is a fundamental moral value, then it will be more likely pursued by morally enhanced individuals. If egalitarianism is the best moral theory and that which is most justifiably pursued, then people with improved moral capacities are more likely to arrange the distribution of resources according to the theory. Egalitarians confident that they are right should invite covert compulsory moral bioenhancement, because moral exemplars and moral post-persons may be counted on to pursue it, if it is the right moral theory. Second, the administration of a covert compulsory moral bioenhancement program will reduce inequality of opportunity. Because it is covert and compulsory, the opportunity to receive it will be mostly equal, so long as its administration rides on the back of some other resource that is equally accessible such as vaccines or public water. Third, the enhanced are more likely to promote others’ interests, even at the expense of their own, which suggests that inequality of condition will decrease. Fourth, inequalities of opportunity and inequalities of condition are interdependent: as one’s condition improves, one’s opportunities increase in frequency and value. These greater opportunities then allow greater improvement in condition, and so on. It works the other way, too. The closing of opportunities leads to a worse condition, which further closes off opportunities. A covert compulsory moral bioenhancement may not break the cycle of interdependence, but by improving conditions or opportunity for everyone, the gap between the people for whom this cycle is advantageous and those for whom it is not will decrease. Fifth, and more generally, inequalities arise partly from our suboptimal moral capacities. If our moral capacities are optimized, then

152  Libertarian Moral Bioenhancement

inequalities will not be as frequent or severe. Egalitarians not only have nothing to worry about, if equality is morally fundamental, they should want covert compulsory moral bioenhancement, because it will promote this value.

Conclusion I have argued that covert compulsory moral bioenhancement is compatible with a protective strain of libertarianism. In order to protect and preserve the ability to use liberties, covert compulsory moral bioenhancement is an allowable minor intrusion. In whole, the argument is not significantly different from Hobbes’ claim (1668/1994) that to escape the state of nature we must subordinate some of our liberties to others. For his argument, the starting position is the state of nature and this subordination is required so that people are more able to use their liberties. For my argument, the starting position is a state of prosperity with abundant ability to use liberties. To continue this state of prosperity and prevent return to a state of nature, or something relevantly similar, we must subordinate some liberties to others’ abilities to use their liberties. This is how we protect our own ability to use those liberties we have. The account I have offered is not a welfare argument. Nor does it require paternalism. The justification for covert compulsory moral bioenhancement is that it is good for others, not that it is good for the person. If libertarianism didn’t allow enforceable intrusions for the good of others, then it would be hog-tied into incoherence. The issue is rather where to draw line between those intrusions that are allowable and those that aren’t. One way to draw this line is to allow minor intrusions. Covert compulsory moral bioenhancement is a minor intrusion, or else an array of ordinary intrusions would be disallowed, again hog-tying the libertarian. My argument is concerned with the compatibility between libertarianism and covert compulsory moral bioenhancement. If this compatibility is not acceptable to the libertarian, the only option left is to deny that covert compulsory moral bioenhancement is necessary for liberty protection. This strategy might help with the symptoms, but it won’t resolve the underlying issue. That covert compulsory moral bioenhancement is necessary for liberty protection is an empirical premise. The adequacy of libertarianism cannot rise or fall based on empirical premises. If it can, then it is much too weak a theory. In short, to deny that covert compulsory moral bioenhancement is necessary for liberty protection is to deny the antecedent of the following conditional: If covert compulsory moral bioenhancement is necessary for liberty protection, then covert compulsory moral bioenhancement is compatible with libertarianism. And that’s no reason to think that either the consequent or the conditional itself is false.

Libertarian Moral Bioenhancement 153

One might think covert compulsory moral bioenhancement is so unpalatable that there cannot be any moral justification for it. If that’s true, so much the worse for libertarianism. I have argued here that they are compatible. Irrelevant to this compatibility is whether covert compulsory moral bioenhancement is otherwise justified. One might also wonder how a family of political theories that descend from the common notion that the individual has inviolable domain over their property and person can allow secret state intrusions into a person’s moral psychology. This is a fair concern, but not one that I take up here. It’s rather a family matter. If compatibility with covert compulsory moral bioenhancement disqualifies a political theory from consideration as the best political theory, then libertarianism ought to be disqualified. But then all the others would have to also be disqualified, because if the theory that most protects individual liberty allows covert compulsory moral bioenhancement, then probably so do all the others.

Note 1. The right to privacy, if we have it, protects not only against intrusion, but also from access. For example, if we have a right to mental privacy, then some technologies that collect data about our brains may violate this right in the absence of strong consent to do so. The point is simply that the right to privacy is much broader than the right not be intruded upon. See Persson and Savulescu (2019).

Chapter 10

Conclusion

Objections I have discussed a range of objections to the view that moral bioenhancement should be compulsory and covert. However, there are several objections to the view that are common, and, to many, count decisively against covert compulsory moral bioenhancement. In this chapter, I address these three objections. Objection 1: “No. Just no.” One of the more memorable quips from contemporary philosophy comes from David Lewis about objections to modal realism: you can’t refute an incredulous stare. Objecting to the argument here by simply saying, “No. Just no,” is even worse than objecting by staring incredulously. At least an incredulous stare suggests that the objector is thinking. If it were so easy to undermine ideas and argument with thoughtless dismissals, philosophy would be much simpler. We might be deceived by an evil demon. No. Just no. The presence of suffering shows that God doesn’t exist. No. Just no. Induction is not justified. No. Just no. It’s as easy as that! Some philosophers object to covert compulsory moral bioenhancement this way. Those who find themselves both in a philosophical line of work as well as tempted to offer these sorts of objections are in the wrong line of work. There is no doubt that the argument here is imperfect. But it’s an introductory concept in philosophy that there are only two ways to rebut an argument: show that a premise is false or that the conclusion doesn’t follow logically from the premises. The only alternative is to not object. If a person is incapable of mastering day one of philosophy 101, they shouldn’t be doing philosophy amateurly, let alone professionally. The “No. Just no” objection is the same type of objection as the similarly thoughtless ad hominem ones: “You’re a Nazi;” “You’re a communist;” “You’re crazy;” and so on. These are fallacious, deleterious to discourse, and demonstrative of the dialectical incompetence of the objector.

Conclusion 155

Objection 2: Moral bioenhancement will change who we are so much that we will be different people I address this idea at length in another article (Crutchfield, 2018). Others have discussed moral enhancement and identity (DeGrazia, 2012; Douglas, 2008). The objection is substantive and significant: widespread moral bioenhancement may so significantly intervene on our psychology that it alters our personal identity. Maybe the enhancement will simply change our personalities a little bit. Or maybe it will cause us to be a different person altogether, killing who we were prior to the enhancement. The worry rests on the idea that our moral capacities are central to who we are. At a minimum, our moral capacities shape our personalities. At a maximum, our moral capacities determine our numerical identity. A decisive reason to not enhance everyone’s moral capacities is that in doing so it will cause them to be numerically distinct from who they were prior to the enhancement. It is less of a worry that a person’s personality change. That’s in fact the aim of the enhancement. If a person’s personality is wrapped up in their disposition to defect when it comes to making decisions about whether to contribute to the public good, the goal of enhancement is for that personality trait to change. If someone’s personality is in part that they are disposed to behave antisocially, perhaps by disregarding others when they have a novel communicable disease, the goal is for that disposition to change. If a person’s personality is that they are a bad person, the goal is for the enhancement to make them a good person. So it’s not overly worrisome that a person’s personality might change. If personality is partly a matter of one’s moral capacities, then not only should we expect an effective enhancement to change that personality, we should want it; that’s the goal. But undermining a person’s numerical identity is different. Killing people is often wrong. If moral enhancement were to kill people by so significantly changing their moral capacities, we probably shouldn’t engage in compulsory covert moral bioenhancement. Strohminger and Nichols (2014) offer some evidence that moral traits constitute personal identity. They conducted surveys with laypeople and discovered that, overall, laypeople thought that moral traits at least partly constitute personal identity. We shouldn’t be worried that moral enhancement can so significantly alter a person’s moral traits such that they become a different person, even if it’s true that moral traits constitute personal identity. Among a person’s moral traits are their higher order beliefs about their moral traits. A  person’s moral judgment that eating meat is wrong might be new to them, but they would likely retain continuity between their old moral judgment and their new moral judgment, especially if, as I recommend in Chapter 6, that the enhancement proceeds by way of providing justificatory evidence for the new judgment. We revise our beliefs, moral and otherwise, all the

156 Conclusion

time, and when we do we can usually hold continuous higher order beliefs about that change. We can continuously recognize the revision of our own mental states by taking a higher order stance toward them. So long as there is continuity among higher order states (i.e., our beliefs about our beliefs), throughout the change in moral capacities, there is no need to worry that the enhancement will affect our identities. For the intervention to change our identities, it would have to immediately and abruptly change all of our moral traits, severing any continuity between moral traits prior to the enhancement and those that result from the enhancement. If the change is gradual, then continuity, and so identity, is preserved. There is no reason to think that a moral bioenhancement can act so quickly as to sever all continuity between moral traits. And this even presumes that moral traits entirely constitute personal identity. This claim goes beyond even what Strohminger and Nichols offer empirical support for. Their surveys don’t support the claim that moral traits entirely constitute identity. Moreover, there’s a decent chance that the folk are wrong about personal identity. We don’t accept folk scientific theories, to the exclusion of those theories that the scientific method confirms. Why should we accept folk theories of personal identity? So, for moral bioenhancement to be a threat to our personal identities, the folk would have to be right (which they probably aren’t), identity would have to be entirely a matter of one’s moral traits (which it isn’t), and the intervention would have to immediately and abruptly sever any continuity of higher order moral traits (which it won’t). In other words, moral bioenhancement is not a threat to personal identity. Objection 3: Covert compulsory moral bioenhancement imposes particular values upon those enhanced, and it is wrong to do so This objection makes two assertions, both of which need to be true for the objection to have any force. However, both assertions are false. Consider the first assertion, that covert compulsory moral bioenhancement imposes particular values upon those enhanced. The assertion fundamentally misunderstands what moral bioenhancement would do. It’s not as though there would be a “utilitarian” pill and everyone would become an act-utilitarian after taking it. Moral bioenhancement, if it is to be safe and effective, would intervene on psychological states that are upstream of the judgment that you should push the fat man onto the tracks to save the five people in the trolley. It would intervene on cognition or reasoning or sensitivity to suffering or agency or all of the above. It may intervene on a person’s motivation to behave in ways that they already judge to be permissible or obligatory. More specifically, it would intervene on those psychological states that lead to effective collective action. So, it would increase sensitivity to the pains and pleasures of people outside of one’s own social group, such as those of other regions, countries, political ideologies, generations,

Conclusion 157

and even those who do not yet exist. And it would increase motivation to behave in cooperative ways that promote the well-being of these people. It wouldn’t impose on people the value of veganism, for example; it would impose upon people the ability to collectively respond to existential threats. If that response entails veganism, that’s not to impose the value of veganism on the enhanced. One might reply that even to impose abilities upon the enhanced imposes values on them; it imposes whatever value is attached to those abilities. Some may not value the ability to be sensitive to the pains and pleasures of others or value more ignoring suffering or promoting it further. It may be true of some people that they are unable to appreciate the suffering of the child drowning in the shallow pond, but that the absence of this ability is valuable to them. In such a case, imposing on them the ability to be sensitive to that suffering and the motivation and ability to avert is to impose values upon them. For this objection to work, it has to be true that by imposing abilities on a person covert compulsory moral bioenhancement imposes values and that it is wrong to do so. The above paragraph demonstrates that even if covert compulsory moral bioenhancement imposes values by way of imposing abilities, it is not wrong to do so. That’s because it is not wrong to impose upon a person sensitivity to others’ suffering. We already do this all the time. Blame and praise function in part to discourage some behaviors and encourage others. By blaming the person who ignores the drowning child, society discourages that behavior. That is to impose upon him sensitivity to the suffering of others or, even more directly, the value of helping others when it requires no sacrifice of comparable moral significance. Outside of praise and blame, we impose values on children all the time, as well. We impose upon them mistrust of strangers, sharing, intellectual independence, honesty, caring, individual responsibility, respect for others, and a thousand other values. It’s a version of the naturalistic fallacy to say that because we do it all the time it is thereby morally permissible. But if it is impermissible to impose values on a person, then it is impermissible to impose these values on anyone, including developing children. This objection only works if it is true both that covert compulsory moral bioenhancement imposes values and that it is wrong to impose values. It is not wrong to impose values upon others, for if it is then methods of steering social behaviors such as blame, praise, and childhood education are wrong. The person who registers this objection must therefore be prepared to hold the absurd view that it is impermissible to praise or blame or educate children. Not only is such a view obviously wrong but also incoherent: purposely withholding value imposition is to impose value-neutrality, which is itself a value. So, it’s not even a position one can coherently maintain. Objection 4: Whoever administers the covert compulsory moral bioenhancement will impose their own values upon the enhanced

158 Conclusion

I claim above that it is not generally wrong to impose values upon others. The presumption that it is wrong to do so grounds this objection. But also motivating the objection is that some people wonder who will end up administering the covert compulsory moral bioenhancement. What appears to go along with this curiosity is the gut feeling that whoever it is, they aren’t qualified, especially if it ends up being government bureaucrats. Who are these people, and why them? The eventual administrators of the covert compulsory moral bioenhancement need not be members of the government; they could be associated with private institutions, or no institution at all. The justification for covert compulsory moral bioenhancement is that it is necessary to avert terrible suffering. Whether the agent or agents who engender this aversion are agents of a state, a private institution, or some other loosely associated group of individuals, or simply one individual, is irrelevant to the justification for covert compulsory moral bioenhancement. The identity of the administrators is open. Likely it will involve scientists such as pharmacologists, physicians, neuro and cognitive scientists, political scientists, anthropologists, economists, epidemiologists, and experts in moral psychology. But it might not. Perhaps it will be administered by government officials. Or maybe it will be administered by an executive of a tech company. Whoever it is, it is irrelevant to its justification. Since its administration is likely to spring from scholarly and scientific inquiry, it would be natural to think that those involved in its development would be involved in its administration, but it is not necessary that they be so involved. What is important is that the people who administer it are themselves enhanced. Ideally they would be so enhanced that they would be among the first moral post-persons. But at a minimum they ought to be some of the first enhanced, even if their moral capacities are not so much greater that they have a different moral status. But they ought to be among the most morally capable individuals, and this will require that they have received the enhancement themselves. This requirement introduces a problem, however. I argue in Chapter 6 that the psychology and epistemology of moral bioenhancement require that it be administered covertly for it to be effective. But if the administrators of the covert compulsory moral bioenhancement are required to have been enhanced, then it seems impossible for them to have done so without their knowledge. The epistemological problems associated with overt moral bioenhancement are that the person receiving the intervention will not only fail to trust the administrators and that the lack of justification will undermine the performance of the targeted behavior. The administrators will therefore have to fully trust that the intervention works and recognize but set aside their lack of justification and instead behave in ways compatible with their new moral capacities. For such people, it may not be necessary that the moral

Conclusion 159

bioenhancement is covert. Given that the administrators will need to fully trust the intervention, its safety, and its efficacy, it is, again, likely that those involved in its development will be its administrators. The profile of the administrators is that they are involved in its development, trusting of it, and themselves the first of those enhanced, perhaps even moral post-persons. This seems like the profile of a person or group of people that can be trusted to act morally. Most importantly, they will be more morally capable than most everyone else. Their judgments about what is right and what is wrong would be better, along with their ability to pursue those actions. They would be more morally trustworthy than even ourselves, in that they would prioritize everyone else’s interests before their own and act in ways that promote those interests. So it’s not as though those administering the covert compulsory moral bioenhancement would be illmotivated and nefarious. They would be the most morally capable humans there are. Although it is not generally wrong to impose values on others, we should want such people to impose their values on ours, because to want those values is to want to be better people. Objection 5: Covert compulsory moral bioenhancement is tantamount to totalitarianism, and totalitarianism is wrong Like the first objection, both of these assertions need to be true for the objection to have any force. Also, like the first objection, both assertions are false. Consider the first assertion, that covert compulsory moral bioenhancement is tantamount to totalitarianism. Usually when people lodge this objection they have in mind dictators like Hitler or Mussolini or Stalin or others who executed grand social programs that ended up committing horrible atrocities. There is only one point of similarity between the horrible totalitarians in human history and the administrators of covert compulsory moral bioenhancement. That point of similarity is this: both sets of individuals thought or think they were doing the right thing (and even this is arguably false as it pertains to the totalitarians). But thinking that one is doing the right thing is not constitutive of doing the right thing, otherwise an implausible version of relativism is true. The difference between the worst totalitarians and the administrators of covert compulsory moral bioenhancement is the biggest difference there could be: totalitarians thought they were doing the right thing, but were wrong; the administrators would think they are doing the right thing, and would be correct. The difference between our historically horrible totalitarians and the administrators of covert compulsory moral bioenhancement is the moral difference between the worst humans and the best humans. No two individuals could be more morally different than Hitler and an administrator of covert compulsory moral bioenhancement. Indeed, most mere persons are more similar to Hitler than

160 Conclusion

would be an administrator of covert compulsory moral bioenhancement, at least until we all get the moral bioenhancement. Furthermore, the administrators of covert compulsory moral bioenhancement, unlike totalitarians, wouldn’t decide public policy, except for the obvious one that every person gets moral bioenhancement without their knowledge. The policymaking process could be exactly as it is now, though different policies would likely be made, such as policies that promote the aversion of suffering from catastrophe. But they would be made because citizens and their representatives want them, not because the administrators of covert compulsory moral bioenhancement want them. Univocal policymaking and enforcement is the mark of a totalitarian, and nothing about the administrators of covert compulsory moral bioenhancement would resemble this. Now consider the second assertion, that totalitarianism is wrong. If what I say in the previous two paragraphs is right, then the administration of covert compulsory moral bioenhancement is not totalitarian. But it is undemocratic. And one might think that because it is not democratic, it is wrong. But such a view is surely incorrect, at least when compared to the alternative of experiencing ultimate harm or extinction. To claim that undemocratic policy is wrong is to prioritize the value of democracy over the absence of suffering. One problem with this view is that it prioritizes something that is purely instrumentally valuable—whatever value democracy has, it is merely in virtue of the fact that it promotes other values such as autonomy—for something that is not only inherently valuable—the absence of suffering— but also perhaps the most valuable thing there is. Another problem is that if democracy is so valuable that it outweighs the absence of suffering, then when we face existential threats like climate change and these threats undermine our democratic institutions, we’d be losing something of great value. Just like we need covert compulsory moral bioenhancement to preserve the use of liberties, we need covert compulsory moral bioenhancement to preserve democratic institutions. Dead people don’t have any liberties, including those positive liberties to vote. If someone wants to maintain the objection that the administration of covert compulsory moral bioenhancement is wrong because it is undemocratic, they must hold that democracy is more valuable than the absence of suffering, or that the suffering is worth what is gained. But this position is untenable, because the suffering that would occur doesn’t gain anything, let alone the preservation of democracy. Immense suffering goes hand-in-hand with the loss of democracy. So if it’s democracy one wants, they should also want covert compulsory moral bioenhancement, because that is what will preserve it. Objection 6: It’s wrong to say these things

Conclusion 161

This objection has been lodged by many people, some of which are apparently also self-respecting ethicists. They say that in these highly divided and politicized times my proposal could scare some people and that this fear might cause them to behave some or way or other. This type of objection is raised against other controversial claims. And aside from noting that this type of objection is not only counterproductive and in violation of norms of inquiry, in the case of my claim, it is self-defeating. The objection presumes that it is in some cases permissible to keep information from people so that they may experience a benefit from that secrecy, namely that information they object to. On the one hand, they claim that my argument ought to be withheld for the sake of others. But on the other hand, what they object to is the withholding of information for the sake of others. Clap the hands together and you get a contradiction. They can’t consistently object to my argument on these grounds. Objection 7: Enhancement is fine, but for it to be virtuous one must do it oneself Developing virtues takes practice. But to refuse a shortcut to being a better person, so that one can develop virtues in the Aristotelian way is extremely selfish. When the cost of not being enhanced is so high for current and future people, to opt out so that one can become a better person by practice and habit development is to sacrifice others’ interests for the sake of self-discovery. It’s the equivalent of leaving one’s small children at home to fend for themselves while one hikes the Pacific Crest Trail on a journey of self-discovery. If it’s virtue one wants, practicing extreme selfishness won’t help to achieve it. Objection 8: Your proposal is straight out of dystopian fiction The relations between science, society, and fiction are extremely interesting. One can make the case that such fiction, such as Brave New World or Gattaca, influences both the public’s attitudes toward scientific discovery as well as the practice of science itself. But that’s not even close to what the objection is aiming to do. The objection rather seems to take the existence of works of fiction as evidence that outcomes similar to those in the work of fiction await us, if we carry out interventions similar to those in the works of fiction. For example, if we are overly liberal with what gene editing we allow, we are likely to end up in a situation similar to the one in Brave New World, which is apparently a bad outcome. In other words, that a practice leads to a particular outcome in a work of fiction provides us reasons to think that that practice actually leads to that outcome. This claim is obviously false. That a relation obtains in a work of fiction does not provide evidence that that relation actually obtains, even if

162 Conclusion

life imitates art. We don’t think that actually throwing gold rings into fire vanquishes evil because that’s what happens when Frodo tosses the ring into Mount Doom. We shouldn’t form similar attitudes based on the plots of works of fiction. Moreover, in Brave New World, the people being given soma aren’t facing societal collapse. If they were, and soma prevented it, we might feel differently about it.

The Covert Enhancement Hypothesis If I  am right that covert compulsory moral bioenhancement should be administered, and it’s true that there is currently no substance that can do what such an intervention would need to do, where does that leave us? What are the next steps? At a minimum, there is good reason for researchers to study and potentially develop a moral enhancer and for funding agencies to fund such research. But premises of my argument are currently testable. In Chapter 6, I argue that a moral enhancer must be administered covertly in order for it to be effective. The epistemology and psychology of moral enhancement are such that telling people they are getting a moral enhancer undermines the achievement of the intended outcome. This claim—the Covert Enhancement Hypothesis—is a testable claim. In the Introduction, I mentioned that oxytocin is often offered as a moral enhancer but that it is not suitable for widespread use as such. But its prosocial effects in certain circumstances, such as those in which the social behaviors are among others in the same social group, are well documented. So there are at least circumstances in which moral bioenhancement occurs. It is possible to test the Covert Enhancement Hypothesis in these circumstances. Furthermore, often the studies on oxytocin use games like the public goods game to test social behavior. So, there is already a methodological model available. The model just needs to be modified slightly. The basic idea is the same, though. Give some people oxytocin, some people a placebo, and then see if there are any differences in how they play the public goods game. But to test the Covert Enhancement Hypothesis, the required slight modification is that some of the people who receive the oxytocin need to believe they are not receiving it, and some of the people who aren’t receiving oxytocin need to believe they are receiving it. By introducing these two additional groups, it will be possible to test (a) whether overt administration of a moral enhancer undermines the achievement of the intended effect and (b) whether contributions to the public good can be promoted by the placebo effect. Both findings would be significant advancements in our understanding of moral enhancement. It’s also possible that the Covert Enhancement Hypothesis is not only false, but that its converse is true. It’s possible that knowing that one is getting a moral enhancer amplifies the achievement of the intended effect. If that were true, then there would be little support for

Conclusion 163

covert compulsory moral bioenhancement. Instead, an overt compulsory moral bioenhancement program would be justified.

The City Walls Even if my argument for covert compulsory moral bioenhancement is good, still people are not likely to accept it. I mentioned in the preface that the project initially began as a way of testing the limits of bioethics. I’ll count it a success if my argument moves someone off of the position of outright rejection. Perhaps instead of outright rejection my arguments move someone to not outright acceptance or endorsement of covert compulsory moral bioenhancement, but to indifference or mere consideration. It might be that covert compulsory moral bioenhancement, even if supported by good arguments, is not something a person can have any attitude toward other than outright rejection. Someone might think that covert compulsory moral bioenhancement is beyond the limits of what bioethics can accept, or even consider. In that case, covert compulsory moral bioenhancement and the arguments for it can act as a theoretical lever to move other, allowable positions in bioethics. For example, my argument suggests that transparency in public health and utility, liberty, and equality promotion may be at odds. If covert compulsory moral bioenhancement is not an allowable position, then it looks like one may be committed to the notion that transparency in public health is more valuable than utility, liberty, and equality. Or, for another example, given that covert compulsory moral bioenhancement is compatible with libertarianism, the libertarian may want to rethink their position regarding minor intrusions. What if bioethics can accept the central claim? If bioethics can accept that it is permissible to dose the entire population with a substance that alters their moral capacities, what could the field ever find unacceptable? If there is no such thing, then what does that say about bioethics, other than that utilitarianism may run deeper than is usually recognized? What does bioethics do with well supported but otherwise unacceptable conclusions? By overtly keeping a well-supported position outside of the city walls, it shows us what the field really cares about. Maybe it is transparency and solidarity and not utility, liberty, equality, or concern for future generations. Or maybe not allowing covert compulsory moral bioenhancement shows more about the intuitions a person might have and their relation to reasons—the reasons are good but it “feels” wrong, a feeling which suffices to define the boundaries of the field. In that case, this book finds the limits of bioethics.

References

Agar, N. (2013). Why is it possible to enhance moral status and why doing so is wrong? Journal of Medical Ethics, 39(2), 67–74. https://doi.org/10.1136/medethics-2012-100597 Agar, N. (2014). A question about defining moral bioenhancement. Journal of Medical Ethics, 40(6), 369–370. https://doi.org/10.1136/medethics-2012-101153 Alcorn, J. L., Rathnayaka, N., Swann, A. C., Moeller, F. G., & Lane, S. D. (2015). Effects of intranasal oxytocin on aggressive responding in antisocial personality disorder. The Psychological Record, 65(4), 691–703. https://doi.org/10.1007/s40732-015-0139-y Anomaly, J. (2011). Public health and public goods. Public Health Ethics, 4(3), 251–259. https://doi.org/10.1093/phe/phr027 Anomaly, J. (2015). Public goods and government action. Politics, Philosophy and Economics, 14(2), 109–128. https://doi.org/10.1177/1470594X13505414 Anomaly, J. (2018). Defending eugenics: From cryptic choice to conscious selection. Monash Bioethics Review, 35(1–4), 24–35. https://doi.org/10.1007/s40592-0180081-2 Anomaly, J. (2020). Creating future people: The ethics of genetic enhancement. Routledge. Anomaly, J.,  & Jones, G. (2020). Cognitive enhancement and network effects: How individual prosperity depends on group traits. Philosophia (United States). https://doi. org/10.1007/s11406-020-00189-3 Benbaji, Y. (2006). Sufficiency or priority? European Journal of Philosophy, 14(3), 327– 348. https://doi.org/10.1111/j.1468-0378.2006.00228.x Bernstein, J. (2017). The case against libertarian arguments for compulsory vaccination. Journal of Medical Ethics, 43(11), 792–796. https://doi.org/10.1136/medethics2016-103857 Blunden, C. T. (2019). Libertarianism and collective action: Is there a libertarian case for mandatory vaccination? Journal of Medical Ethics, 45(1), 71–74. https://doi. org/10.1136/medethics-2018-104752 Bognar, G. (2012). Enhancement and equality. Ethical Perspectives, 19(1), 11–32. https:// doi.org/10.2143/EP.19.1.2152677 Boorse, C. (1976). What a theory of mental health should be. Journal for the Theory of Social Behaviour, 6(1), 61–84. https://doi.org/10.1111/j.1468-5914.1976.tb00359.x Boorse, C. (1977). Health as a theoretical concept. Philosophy of Science, 61(4), 579–591. https://doi.org/10.1086/288913 Boorse, C. (1997). A  rebuttal on health. In J. M. Humer  & R. F. Almeder (Eds.), What is disease? (pp.  1–134). Totowa, NJ: Humana Press. https://doi. org/10.1007/978-1-59259-451-1_1

References 165 Boorse, C. (2014). A second rebuttal on health. Journal of Medicine and Philosophy (United Kingdom), 39(6), 683–724. https://doi.org/10.1093/jmp/jhu035 Bostrom, N., & Ord, T. (2006). The reversal test eliminating status quo bias in applied ethics. Ethics, 116(4), 656–679. https://doi.org/10.1086/505233 Bowles, S., & Gintis, H. (2011). A cooperative species: Human reciprocity and its evolution. Princeton University Press. Boynton, M. H., Portnoy, D. B.,  & Johnson, B. T. (2013). Exploring the ethics and psychological impact of deception in psychological research.  IRB: Ethics  & Human Research, 35(2), 7. Brennan, J. (2018). A libertarian case for mandatory vaccination. Journal of Medical Ethics, 44(1), 37–43. https://doi.org/10.1136/medethics-2016-103486 Broadbent, A. (2019). Health as a secondary property. The British Journal for the Philosophy of Science, 70(2), 609–627. Bublitz, J. C. (2019). Saving the world through sacrificing liberties? A critique of some normative arguments in unfit for the future. Neuroethics, 12(1), 23–34. https://doi. org/10.1007/s12152-016-9265-8 Buchanan, A. E. (2009). Moral status and human enhancement. Philosophy  & Public Affairs, 37(4), 346–381. https://doi.org/10.1111/j.1088-4963.2009.01166.x Buchanan, A. E. (2011). Beyond humanity?: The ethics of biomedical enhancement. Oxford University Press. Buchanan, A. E., Brock, D. W., Daniels, N., & Wikler, D. (2001). From chance to choice: Genetics and justice. Cambridge University Press. Burton-Chellew, M. N., El Mouden, C., & West, S. A. (2016). Conditional cooperation and confusion in public-goods experiments. Proceedings of the National Academy of Sciences, 113(5), 1291–1296. https://doi.org/10.1073/pnas.1509740113 Caplan, B. (2001). Rational ignorance versus rational irrationality. Kyklos, 54, 3–26. Caplan, B. (2003). The logic of collective belief. Rationality and Society, 15(2), 218–242. Chaudhuri, A. (2011). Sustaining cooperation in laboratory public goods experiments: A selective survey of the literature. Experimental Economics, 14(1), 47–83. Chaudhuri, A.,  & Paichayontvijit, T. (2006). Conditional cooperation and voluntary contributions to a public good. Economics Bulletin, 3(8), 1–14. Childress, J. F., Faden, R. R., Gaare, R. D., Gostin, L. O., Kahn, J., Bonnie, R. J., Kass, N. E., Mastroianni, A. C., Moreno, J. D., & Nieburg, P. (2002). Public health ethics: Mapping the terrain. The Journal of Law, Medicine & Ethics, 30(2), 170–178. https:// doi.org/10.1111/j.1748-720X.2002.tb00384.x Christman, J. (2020). Autonomy in moral and political philosophy. In E. N. Zalta (Ed.), The Stanford encyclopedia of philosophy (Fall 2020 Edition). https://plato.stanford. edu/archives/fall2020/entries/autonomy-moral/ Cook, F. L., Jacobs, L. R., & Kim, D. (2010). Trusting what you know: Information, knowledge, and confidence in social security. Journal of Politics, 72(2), 397–412. https://doi.org/10.1017/S0022381610000034 Cooper, R. (2002). Disease. Studies in History and Philosophy of Science Part C: Studies in History and Philosophy of Biological and Biomedical Sciences, 33(2), 263–282. https://doi. org/10.1016/S0039-3681(02)00018-3 Crisp, R., Casal, P., Cohen, J., Draper, K., Dunlop, F., Griffin, J., . . . Tasioulas, J. (2003, July). Equality, priority, and compassion*. Ethics, 113, 745–763. Crockett, M. J. (2014). Moral bioenhancement: A  neuroscientific perspective. Journal of Medical Ethics, 40(6), 370–371. https://doi.org/10.1136/medethics-2012-101096

166 References Croson, R. T. A.,  & Marks, M. B. (2000). Step returns in threshold public goods: A meta- and experimental analysis. Experimental Economics, 2(3), 239–259. Crutchfield, P. (2016). The epistemology of moral bioenhancement. Bioethics, 30(5), 389–396. Crutchfield, P. (2018). Moral enhancement can kill. Journal of Medicine and Philosophy, 43(5), 568–584. Crutchfield, P. (2019). Compulsory moral bioenhancement should be covert. Bioethics, 33(1), 112–121. Crutchfield, P. (2020a). Engendering moral post-persons: A novel self-help strategy. Bioethics, 34(7), 679–686. Crutchfield, P. (2020b). It is better to be ignorant of our moral enhancement: A reply to Zambrano. Bioethics, 34(2), 190–194. Crutchfield, P., Gibb, T.,  & Redinger, M. (2018). Placebos and a new exception to informed consent. AJOB Neuroscience, 9(3), 200–202. Crutchfield, P.,  & Scheall, S. (2019). Epistemic burdens and the incentives of surrogate decision-makers. Medicine, Health Care and Philosophy. https://doi.org/10.1007/ s11019-019-09899-2 Dawes, R. M., Orbell, J. M., Simmons, R. T., & Kragt, A. J. C. Van de. (1986). Organizing groups for collective action. The American Political Science Review, 80(4), 1171– 1185. De Dreu, C. K. W., Greer, L. L., Van Kleef, G. A., Shalvi, S., & Handgraaf, M. J. J. (2011). Oxytocin promotes human ethnocentrism. Proceedings of the National Academy of Sciences, 108(4), 1262–1266. https://doi.org/10.1073/pnas.1015316108 De Dreu, C. K. W.,  & Kret, M. E. (2016). Oxytocin conditions intergroup relations through upregulated in-group empathy, cooperation, conformity, and defense. Biological Psychiatry, 79(3), 165–173. https://doi.org/10.1016/j.biopsych.2015.03.020 De Fine Licht, J. (2011). Do we really want to know? The potentially negative effect of transparency in decision making on perceived legitimacy. Scandinavian Political Studies, 34(3), 183–201. https://doi.org/10.1111/j.1467-9477.2011.00268.x DeGrazia, D. (2012). Genetic enhancement, post-persons and moral status: A  reply to Buchanan. Journal of Medical Ethics, 38(3), 135–139. https://doi.org/10.1136/ medethics-2011-100126 DeGrazia, D. (2014). Moral enhancement, freedom, and what we (should) value in moral behaviour. Journal of Medical Ethics, 40(6), 361–368. https://doi.org/10.1136/ medethics-2012-101157 De Lazari-Radek, K., & Singer, P. (2010). Secrecy in consequentialism: A defence of esoteric morality. Ratio, 23(1), 34–58. Douglas, T. (2008). Moral enhancement. Journal of Applied Philosophy, 25(3), 228–245. https://doi.org/10.1111/j.1468-5930.2008.00412.x Douglas, T. (2013). Human enhancement and supra-personal moral status. Philosophical Studies, 162(3), 473–497. https://doi.org/10.1007/s11098-011-9778-2 Downs, A. (1957). An economic theory of democracy. Harper. Dworkin, G. (2019, Fall). Paternalism. In E. N. Zalta (Ed.), The Stanford encyclopedia of philosophy. https://plato.stanford.edu/archives/fall2020/entries/paternalism/ Epstein, R. A. (1979). Nuisance law: Corrective justice and its utilitarian constraints. The Journal of Legal Studies, 8(1), 49–102. https://doi.org/10.1086/467602 Fabiano, J. (2020). The fragility of moral traits to technological interventions. Neuroethics. https://doi.org/10.1007/s12152-020-09452-6

References 167 Faden, R., Bernstein, J., & Shebay, S. (2020, Fall). Public health ethics. In E. N. Zalta (Ed.), The Stanford encyclopedia of philosophy. https://plato.stanford.edu/archives/ fall2020/entries/publichealth-ethics/ Falk, A., & Fischbacher, U. (2006). A theory of reciprocity. Games and Economic Behavior, 54(2), 293–315. Fehr, E., & Gächter, S. (2000). Cooperation and punishment in public goods experiments. American Economic Review, 90(4), 980–994. Fehr, E., & Gächter, S. (2002). Altruistic punishment in humans. Nature, 415, 137–140. Frick, J. (2017). On the survival of humanity. Canadian Journal of Philosophy, 47(2–3), 344–367. https://doi.org/10.1080/00455091.2017.1301764 Fudenberg, D.,  & Pathak, P. A. (2010). Unobserved punishment supports cooperation. Journal of Public Economics, 94(1–2), 78–86. Glackin, S. N. (2019). Grounded disease: Constructing the social from the biological in medicine. Philosophical Quarterly, 69(275), 258–276. https://doi.org/10.1093/pq/ pqy063 Goodin, R. E. (1986).  Protecting the vulnerable: A  re-analysis of our social responsibilities. University of Chicago Press. Grill, K., & Dawson, A. (2015). Ethical frameworks in public health decision-making: Defending a value-based and pluralist approach. Health Care Analysis. https://doi. org/10.1007/s10728-015-0299-6 Grimmelikhuijsen, S., Bijhouwerstraat, G., Utrecht, Z. C., Porumbescu, G., Hong, B.,  & Im, T. (2013). The effect of transparency in trust in government : A  crossnational comparative experiment. Public Administration Review, 73(4), 575–586. Grimmelikhuijsen, S. G.,  & Meijer, A. J. (2014). Effects of transparency on the perceived trustworthiness of a government organization: Evidence from an online experiment. Journal of Public Administration Research and Theory, 24(1), 137–157. Hanser, M. (1990). Harming future people. Philosophy & Public Affairs, 19(1), 47–70. Hansson, S. (2018, Fall). Risk. In E. N. Zalta (Ed.), The Stanford encyclopedia of philosophy. https://plato.stanford.edu/archives/fall2018/entries/risk/ Harman, E. (2009). Harming as causing harm. In M. A. Roberts & D. T. Wasserman (Eds.), Harming future persons. Springer Verlag. Harris, J. (2011). Moral enhancement and freedom. Bioethics, 25(2), 102–111. https:// doi.org/10.1111/j.1467-8519.2010.01854.x Hauser, O. P., Rand, D. G., Peysakhovich, A., & Nowak, M. A. (2014). Cooperating with the future. Nature, 511(7508), 220–223. https://doi.org/10.1038/nature13530 Hibbing, J. R., Theiss-Morse, E., & Kuklinski, J. H. (2002). Stealth democracy: Americans’ beliefs about how government should work. Cambridge University Press. https://books. google.com/books?id=roOhDUnlulsC Hobbes, T. (1668/1994) Leviathan (E. Curley, Ed.). Hackett Publishing. Horne, L. C. (2019). Public health, public goods, and market failure. Public Health Ethics, 12(3), 287–292. https://doi.org/10.1093/phe/phz004 Huemer, M. (2008). Ethical intuitionism. Palgrave Macmillan. https://books.google.com/ books?id=wXKlGwAACAAJ Huemer, M. (2010). Why people are irrational about politics. Politics, III, 1–16. Isen, A. M.,  & Levin, P. F. (1972). Effect of feeling good on helping: Cookies and kindness. Journal of Personality and Social Psychology, 21(3), 384–388. https://doi. org/10.1037/h0032317 Jebari, K. (2014). What to enhance: Behaviour, emotion or disposition? Neuroethics, 7(3).

168 References Kavka, G. S. (1982). The paradox of future individuals. Philosophy and Public Affairs, 11(2). Mack, E. (2015). Elbow room for rights. Oxford Studies in Political Philosophy, 1, 194– 221. https://doi.org/10.1093/acprof:oso/9780199669530.003.0009 Masclet, D., Noussair, C., Tucker, S.,  & Villeval, M. C. (2003). Monetary and nonmonetary punishment in the voluntary contributions mechanism. American Economic Review, 93(1). Mathews, K. E., & Canon, L. K. (1975). Environmental noise level as a determinant of helping behavior. Journal of Personality and Social Psychology, 32(4), 571–577. https:// doi.org/10.1037/0022-3514.32.4.571 McMahan, J. (2009). Cognitive disability and cognitive enhancement. Metaphilosophy, 40(3–4), 582–605. https://doi.org/10.1111/j.1467-9973.2009.01612.x Meyer, L. (2003). Past and future. In L. H. Meyer, S. L. Paulson, & T. W. Pogge (Eds.), Rights, culture and the law: Themes from the legal and political philosophy of Joseph Raz. Oxford University Press. Milinski, M., Semmann, D., Krambeck, H. J.,  & Marotzke, J. (2006). Stabilizing the earth’s climate is not a losing game: Supporting evidence from public goods experiments. Proceedings of the National Academy of Sciences, 103(11), 3994–3998. https://doi. org/10.1073/pnas.0504902103 Milinski, M., Sommerfeld, R. D., Krambeck, H.-J., Reed, F. A., & Marotzke, J. (2008). The collective-risk social dilemma and the prevention of simulated dangerous clzimate change. Proceedings of the National Academy of Sciences, 105(7), 2291–2294. https://doi. org/10.1073/pnas.0709546105 Nozick, R. (2013). Anarchy, state, and utopia. Basic Books. https://books.google.com/ books?id=fVITAAAAQBAJ Ostrom, E., Walker, J., & Gardner, R. (1992). Covenants with and without a sword: Selfgovernance is possible. American Political Science Review, 86(2), 404–417. Palfrey, T. R., & Rosenthal, H. (1984). Participation and the provision of discrete public goods: A strategic analysis. Journal of Public Economics, 24(2), 171–193. Parfit, D. (1984). Reasons and persons. Oxford University Press. Pereda, M., Tamarit, I., Antonioni, A., Cuesta, J. A., Hernández, P.,  & Sánchez, A. (2019). Large scale and information effects on cooperation in public good games. Scientific Reports, 9(1), 1–10. Persson, I., & Savulescu, J. (2008). The perils of cognitive enhancement and the urgent imperative to enhance the moral character of humanity. Journal of Applied Philosophy, 25(3), 162–177. https://doi.org/10.1111/j.1468-5930.2008.00410.x Persson, I.,  & Savulescu, J. (2013). Getting moral enhancement right: The desirability of moral bioenhancement. Bioethics, 27(3), 124–131. https://doi. org/10.1111/j.1467-8519.2011.01907.x Persson, I., & Savulescu, J. (2014a). Against fetishism about egalitarianism and in defense of cautious moral bioenhancement. The American Journal of Bioethics: AJOB, 14(4), 39–42. https://doi.org/10.1080/15265161.2014.889248 Persson, I., & Savulescu, J. (2014b). Unfit for the future: The need for moral enhancement. Oxford University Press. http://books.google.com/books?id=EGHengEACAAJ Persson, I., & Savulescu, J. (2019). The irrelevance of a moral right to privacy for biomedical moral enhancement. Neuroethics, 12(1), 35–37. https://doi.org/10.1007/ s12152-017-9343-6

References 169 Pokorny, T., Preller, K. H., Kometer, M., Dziobek, I., & Vollenweider, F. X. (2017). Effect of psilocybin on empathy and moral decision-making. International Journal of Neuropsychopharmacology, 20(9), 747–757. https://doi.org/10.1093/ijnp/pyx047 Pollock, J. L., & Cruz, J. (1999). Contemporary theories of knowledge (Vol. 35). Rowman & Littlefield. Powell, R., & Buchanan, A. (2016). The evolution of moral enhancement. In S. Clarke, J. Savulescu, C. A. J. Coady, A. Giubilini,  & S. Sanyal (Eds.), The ethics of human enhancement: Understanding the debate. Oxford University Press. Prinz, J. (2011). Against empathy. The Southern Journal of Philosophy, 49, 214–233. Pryor, J. (2000). The skeptic and the dogmatist. Noûs, 34(4), 517–549. https://doi. org/10.1111/0029-4624.00277 Railton, P. (2009). Locke, stock, and peril: Natural property rights, pollution, and risk (1985). Facts, Values, and Norms, 187–225. https://doi.org/10.1017/ cbo9780511613982.008 Rakić, V. (2014a). Voluntary moral bioenhancement is a solution to Sparrow’s concerns. The American Journal of Bioethics: AJOB, 14(4), 37–38. https://doi.org/10.1080/1526 5161.2014.889249 Rakić, V. (2014b). Voluntary moral enhancement and the survival-at-any-cost bias. Journal of Medical Ethics, 40(4), 246–250. https://doi.org/10.1136/medethics-2012-100700 Rakić, V. (2015). How moral is (moral) enhancement? We must create beings with moral standing superior to our own. Cambridge Quarterly of Healthcare Ethics, 24, 58–65. https://doi.org/10.1017/S0963180114000309 Rakić, V. (2017a). Compulsory administration of oxytocin does not result in genuine moral enhancement. Medicine, Health Care and Philosophy, 20(3), 291–297. https:// doi.org/10.1007/s11019-017-9762-5 Rakić, V. (2017b). The issues of freedom and happiness in moral bioenhancement: Continuing the debate with a reply to Harris Wiseman. Journal of Bioethical Inquiry, 14(4), 469–474. https://doi.org/10.1007/s11673-017-9805-x Raus, K., Focquaert, F., Schermer, M., Specker, J., & Sterckx, S. (2014). On defining moral enhancement: A  clarificatory taxonomy. Neuroethics, 7(3), 263–273. https:// doi.org/10.1007/s12152-014-9205-4 Reznek, L. (1987). The nature of disease. Routledge & Kegan Paul. Rivera-López, E. (2009). Individual procreative responsibility and the nonidentity problem. Pacific Philosophical Quarterly, 90(3), 336–363. https://doi. org/10.1111/j.1468-0114.2009.01344.x Roberts, M. A. (2019, Winter). The nonidentity problem. In E. N. Zalta (Ed.), The Stanford encyclopedia of philosophy. https://plato.stanford.edu/archives/win2020/ entries/nonidentity-problem/ Rysiew, P. (2016, Winter). Epistemic contextualism. In E. N. Zalta (Ed.), The Stanford encyclopedia of philosophy. https://plato.stanford.edu/archives/win2020/entries/ contextualism-epistemology/ Savulescu, J., & Persson, I. (2012). Moral enhancement, freedom and the god machine. The Monist, 95(3). Scheall, S. (2019). Ignorance and the incentive structure confronting policymakers. Cosmos + Taxis, 7(1–2), 39–51. Scheall, S.,  & Crutchfield, P. (2020). The priority of the epistemic. Episteme, 1–12. https://doi.org/10.1017/epi.2019.56

170 References Schmidtz, D. (2010). Property and justice. Social Philosophy and Policy, 27(1), 79–100. https://doi.org/10.1017/S0265052509990045 Schroeder, S. A. (2013). Rethinking health: Healthy or healthier than? British Journal for the Philosophy of Science, 64(1), 131–159. https://doi.org/10.1093/bjps/axs006 Seidel, A., & Prinz, J. (2013). Sound morality: Irritating and icky noises amplify judgments in divergent moral domains. Cognition, 127(1), 1–5. https://doi.org/10.1016/j. cognition.2012.11.004 Selgelid, M. J. (2009). A moderate pluralist approach to public health policy and ethics. Public Health Ethics, 2(2), 195–205. https://doi.org/10.1093/phe/php018 Shiffrin, S. V. (1999). Wrongful life, procreative responsibility, and the significance of harm. Legal Theory, 5(2), 117–148. https://doi.org/10.1017/S1352325299052015 Sidgwick, H. (1907/1981). The methods of ethics. Hackett. Simon, J. (2007). Beyond naturalism and normativism: Reconceiving the “disease” debate. Philosophical Papers, 36(3), 343–370. Singer, P. (1972). Famine, affluence, and morality. Philosophy  & Public Affairs, 1(3), 229–243. Sinnott-Armstrong, W. (2005). It’s not my fault: Global warming and individual moral obligations. Advances in the Economics of Environmental Resources, 5, 285–307. Sinnott-Armstrong, W. (2008). Framing moral intuitions. In Moral psychology. Vol. 2: The cognitive science of morality: Intuition and diversity. MIT Press. Sobel, D. (2012). Backing away from libertarian self-ownership. Ethics, 123(1), 32–60. https://doi.org/10.1086/667863 Sparrow, R. (2014). Egalitarianism and moral bioenhancement. The American Journal of Bioethics: AJOB, 14(4), 20–28. https://doi.org/10.1080/15265161.2014.889241 Sparrow, R. (2019). Yesterday’s child: How gene editing for enhancement will produce obsolescence—and why it matters. The American Journal of Bioethics, 19(7), 6–15. https://doi.org/10.1080/15265161.2019.1618943 Stanley, J. (2005). Knowledge and practical interests. Clarendon Press. Strohminger, N., & Nichols, S. (2014). The essential moral self. Cognition, 131(1), 159– 171. https://doi.org/10.1016/j.cognition.2013.12.005 Tavoni, A., Dannenberg, A., Kallis, G., & Loschel, A. (2011). Inequality, communication, and the avoidance of disastrous climate change in a public goods game. Proceedings of the National Academy of Sciences, 108(29), 11825–11829. https://doi.org/10.1073/ pnas.1102493108 Thaler, R. H., & Sunstein, C. R. (2009). Nudge: Improving decisions about health, wealth, and happiness. Penguin. https://books.google.com/books?id=bt6sPxiYdfkC Timmermann, M., Jeung, H., Schmitt, R., Boll, S., Freitag, C. M., Bertsch, K.,  & Herpertz, S. C. (2017). Oxytocin improves facial emotion recognition in young adults with antisocial personality disorder. Psychoneuroendocrinology, 85, 158–164. https://doi. org/10.1016/j.psyneuen.2017.07.483 Tolbert, C. J., & Mossberger, K. (2006). The effects of e-government on trust and confidence in government. Public Administration Review, 66(3), 354–369. https://doi. org/10.1111/j.1540-6210.2006.00594.x Tversky, A., & Kahneman, D. (1981). The framing of decisions and the psychology of choice. Science, New Series, 211(4481), 453–458. http://links.jstor.org/sici?sici=00368075%2819810130%293%3A211%3A4481%3C453%3ATFODAT%3E2.0.CO %3B2-3

References 171 Upshur, R. E. G. (2002). Principles for the justification of public health intervention. Canadian Journal of Public Health, 93(2), 101–103. http://journal.cpha.ca/index.php/ cjph/article/view/217 Vallier, K. (2018, Spring). Public justification. In E. N. Zalta (Ed.), The Stanford encyclopedia of philosophy. https://plato.stanford.edu/archives/spr2018/entries/justification-public/ van de Kragt, A. J. C., Orbell, J. M., & Dawes, R. M. (1983). The minimal contributing set as a solution to public goods problems. American Political Science Review, 77(1), 112–122. Vanier, J. (1998). Becoming human. Paulist Press. Wakefield, J. C. (1992). The concept of mental disorder on the boundary between biological facts and social values. American Psychologist, 373–388. Werner, P. J. (2016). Moral perception and the contents of experience. Journal of Moral Philosophy, 13(3), 294–317. https://doi.org/10.1163/17455243-4681063 Werner, P. J. (2020). Which moral properties are eligible for perceptual awareness? Journal of Moral Philosophy, 17(3). Williams, B. (1985). Ethics and the limits of philosophy. Fontana Press. Williamson, T. (2002). Knowledge and its limits. Oxford University Press. Wiseman, H. (2016). The myth of the moral brain: The limits of moral enhancement. MIT Press. https://books.google.com/books?id=_TuQCwAAQBAJ Zagzebski, L. (2017). Exemplarist moral theory. Oxford University Press. https://books. google.com/books?id=AuQwDgAAQBAJ Zambrano, A. (2019). Covert moral bioenhancement, public health, and autonomy. Bioethics, 2–5. https://doi.org/10.1111/bioe.12567 Zhang, J., Zhang, C., & Cao, M. (2015). How insurance affects altruistic provision in threshold public goods games. Scientific Reports, 5, 1–7. Zwolinksi, M. (n.d.). Libertarianism. In Internet encyclopedia of philosophy. Retrieved October 22, 2019, from www.iep.utm.edu/libertar/

Index

altruism 3, 45, 57, 59 artificial intelligence 29, 99, 132 autonomy 76, 160; and enhancement 2, 87 – 90, 96, 100, 104 – 106, 128; in public health 93, 100 belief 2, 46, 59, 80 – 89, 133, 150, 155 – 156 bioethics 92, 163 Brave New World 161 – 162 categorical imperative 71 certainty 32 – 35, 39, 48, 101 cognition: bias 14, 54, 85; capacities 3, 5, 27, 61, 64, 70; enhancement of 3, 5, 7, 13, 14, 42, 58, 61, 65 collective action 1, 3, 26 – 28, 62, 79, 136, 156; epistemology of 5, 43 – 45, 47 – 49, 52, 56 – 60, 86 – 87; and future generations 41 collective risk: averting 1 – 4, 7 – 8, 60 – 63, 68, 73, 86, 91, 132, 135 – 138; behavioral response to 27 – 28, 41 – 44, 46, 49, 58, 64, 73, 87, 100 – 101, 104, 108, 146 compensation: adequate 142 – 144, 147 – 148; in-kind 140 – 141, 143 competence 60, 70, 109, 117, 154 conceptual analysis 11 – 12, 21 consequentialism 72 contextualism 7, 11 – 12, 20 – 21, 23 – 27 contribution to public good 61 – 63, 79, 99, 108, 136, 155, 162; behavioral economics of 44 – 58, 86, 129, 146 conversational context 7, 19, 21 – 26

cooperation 5, 7, 84, 100, 136, 157; conditions for 44 – 46, 48 – 49, 58, 60, 86, 100, 129 coronavirus 43, 85 – 86, 104 Covert Enhancement Hypothesis 162 COVID-19 37, 43, 60, 65 culture 1, 16 deception 7, 74 – 77 decision-making 150; epistemology of 42 – 43, 49, 51, 63; and politics 111, 117, 123 – 124, 127 – 128 defeat 73n1, 82, 89, 131, 161 democracy 109, 112, 160; stealth 111, 122 – 126 desires 16, 94, 105 – 106, 109n1 dual-use research 110n4 dystopian fiction 5, 161 – 162 education and enhancement 3, 53, 57, 59, 60, 157 effectiveness and safety of enhancement 4 – 5, 7, 90, 93, 100 – 101, 156, 159 egalitarianism 13, 107 – 108, 110n3, 151 – 152 elbow room 141 – 142, 144, 147 – 149 enforcement 56, 95, 128, 134 – 135, 150, 160 epistemic: burden 8, 42 – 43, 46, 50 – 58, 60 – 61, 63, 100 – 101; justification 81 – 83, 88 – 89; position 49, 52, 55, 61, 84, 87 – 89, 105 – 106, 129 equality 2, 13, 58, 92 – 93, 95 – 97, 100, 107 – 108, 110, 151 – 152, 163 esoteric morality 125 evolution 16, 25, 42, 45, 57, 59, 101 extinction 1, 28 – 30, 160

Index 173 fairness 45, 78, 91 – 93, 96 – 97, 100 family 31, 36, 53 – 54, 67, 74, 76 – 77, 153 feasibility of enhancement 6 – 7, 56 – 57, 77, 79, 138, 143 fluoride 78 – 79, 86, 144, 148 framework 91 – 93, 95, 97, 102, 106 framing effects 105, 110n6 freedom 2, 68 – 69, 73, 128, 150, 166 – 167, 169 free-riding 45, 49, 54, 56, 62, 86 – 87, 129 function 10 – 14, 18 – 19, 21 – 23, 26 – 27; natural 16, 24 – 25; normal 13, 15, 16, 24; proper 15, 25, 36 future generations 8, 29, 31 – 32, 34 – 36, 38 – 39, 41, 48 – 49, 59, 72, 161 Gattaca 161 genetic enhancement 5 – 6, 128 globalization 1, 29 government and enhancement 85, 96, 99, 107, 130 hallucination 82, 84 Hobbes, Thomas 131, 152 hog-tie 140 – 142, 144 – 145, 147 – 149, 152 identity: of administrators 83, 85, 99, 109, 158 – 159; personal 2, 155 – 156, 158 ignorance 8, 46, 48 – 52, 75, 79, 88, 148 imposing values 157 incredulous stare 154 inequality 13, 107, 110, 151 infectious disease 4, 29, 60, 85, 91, 116, 124, 130, 134, 136 – 137 justification see epistemic, justification liberty 56, 58, 63, 131 – 134, 144, 148, 151 – 153; and enhancement 109, 128, 142, 144 – 145; intrusions upon 139 – 141; in public health 91 – 97, 100, 103 – 104, 163 market failure 129 – 130 minor intrusion 139 – 145, 148 – 149, 152, 163 moral exemplars 66 – 70, 73, 109, 111, 126, 151 moral post-person: characteristics of 8, 64 – 69, 85, 107, 109, 151; and

enhancement 70 – 73, 111, 126, 158 – 159 moral psychology 8, 67, 96, 158, 162; and evolution 42, 58 – 59; intrusions upon 145 – 147, 149, 153 moral status 2, 8, 64 – 66, 68, 113 – 115, 158 motivation 2, 61, 67 – 68, 79, 83 – 84, 99, 156 – 157 national security 77 network effects 5 nonidentity problem 8, 31 – 35, 38 – 40 non-maleficence 28, 33 – 35, 38 options 50 – 58, 126 oxytocin 3 – 4, 59, 162 pandemic 1, 4, 29, 37, 43 – 44, 46, 60, 64 – 65, 85, 101, 104, 123; see also collective risk; COVID-19 paternalism 104, 111 – 112, 127, 152 personality 3, 14, 84, 155 pollution 18, 107, 132, 142 practical interests 26, 53, 85 – 86 Principle of Public Justification 120 – 122 privacy 92, 149 – 150, 153 property 113, 116, 132 – 134, 139 – 147, 149, 153 psilocybin 3 – 4, 7 psychopaths 96 public goods: definition of 44; game 42 – 51, 54, 56, 146, 162 quarantine 91, 94, 102 – 104, 106, 116, 118, 122, 128, 134 – 135 Rapa Nui 1 rationality 84 – 86, 88, 121 reciprocity 45 reliabilism 89 respect for persons 92 – 93, 96, 105, 112 – 114, 115 – 116, 119, 128 Reversal Test 14 – 15 rights 92, 133 – 134, 139 – 142, 144, 150 risk 5 – 6, 17, 39, 77, 102, 107, 116, 135, 137 – 139, 142, 146; see also collective risk satisficing 72 self-interest 9, 67 – 68, 70, 109, 111, 123 – 124, 126 shame 5, 7, 45 – 46, 62

174 Index skepticism 20 – 21, 27n1, 57, 59, 79, 88 Socha, Leopold 67 – 69 societal collapse 1, 6, 29, 162 solidarity 92 – 93, 96 – 97, 100, 120, 163 special obligation 37 – 39 state of nature 132 – 133, 152 sufficientarianism 39 totalitarianism 159 – 160 transhuman 64; see also moral post-person transparency: epistemic 80; in experiments 47; in public health 9, 92, 96 – 97, 100, 111 – 116, 128; in public policy 9, 117 – 119, 122 – 127 treatment/enhancement distinction 6 – 7, 13 – 14, 20, 26 – 27 trust: and enhancement 8, 83 – 84, 89, 106, 157 – 159; and public health 92 – 93, 96 – 97, 100, 112, 116 – 119

ultimate harm 28 – 30, 69, 91, 94 – 99, 101 – 104, 106, 132 – 133, 138, 142, 160 utilitarianism 37, 124 – 126, 156, 163 utility 58, 71 – 72, 91 – 97, 100, 102, 132, 163 vaccine 7, 29, 62, 78 – 79, 86 – 87, 92 – 94, 102 – 103, 108, 115; as enhancement 13 – 14; and justice 14, 44, 98, 151; and liberty 122, 134 – 139, 148 – 149 Vanier, Jean 67 – 70 virtue ethics 71 – 72, 92, 119, 161 voluntary moral enhancement 62 – 63 vulnerability 37 – 41, 85 water 7, 78 – 79, 86, 92, 98, 102 – 103, 108, 118, 131 – 133, 144, 148, 151 withholding information 7, 74 – 77, 157, 161