Ethics in the Gray Area: A Gradualist Theory of Right and Wrong 9781009336789, 1009336789

What should morally conscientious agents do if they must choose among options that are somewhat right and somewhat wrong

197 27 2MB

English Pages 236 [228] Year 2023

Report DMCA / Copyright

DOWNLOAD PDF FILE

Recommend Papers

Ethics in the Gray Area: A Gradualist Theory of Right and Wrong
 9781009336789, 1009336789

  • 0 0 0
  • Like this paper and download? You can publish your own PDF file online for free in a few minutes! Sign Up
File loading please wait...
Citation preview

ETHICS IN THE GRAY AREA

What should morally conscientious agents do if they must choose among options that are somewhat right and somewhat wrong? Should you select an option that is right to the highest degree, or would it perhaps be more rational to choose randomly among all somewhat right options? And how should lawmakers and courts address behavior that is neither entirely right nor entirely wrong? In this first book-length discussion of the “gray area” in ethics, Martin Peterson challenges the assumption that rightness and wrongness are binary properties. He argues that some acts are neither entirely right nor entirely wrong, but rather a bit of both. Including discussions of white lies and the permissibility of abortion, Peterson’s book presents a gradualist theory of right and wrong designed to answer pressing practical questions about the gray area in ethics. M a rt i n Pet er son  holds the Sue G. and Harry E. Bovay, Jr.  Chair in the Department of Philosophy at Texas A&M University. He is the author of The Dimensions of Consequentialism (Cambridge University Press, 2013), An Introduction to Decision Theory (Cambridge University Press, 2nd ed., 2017), The Ethics of Technology (Oxford University Press, 2017), and Ethics for Engineers (Oxford University Press, 2019).

Published online by Cambridge University Press

Published online by Cambridge University Press

E T H IC S I N T H E G R AY A R E A A Gradualist Theory of Right and Wrong M A RT I N PET ER SON Texas A & M University

Published online by Cambridge University Press

Shaftesbury Road, Cambridge CB2 8EA, United Kingdom One Liberty Plaza, 20th Floor, New York, NY 10006, USA 477 Williamstown Road, Port Melbourne, VIC 3207, Australia 314–321, 3rd Floor, Plot 3, Splendor Forum, Jasola District Centre, New Delhi – 110025, India 103 Penang Road, #05–06/07, Visioncrest Commercial, Singapore 238467 Cambridge University Press is part of Cambridge University Press & Assessment, a department of the University of Cambridge. We share the University’s mission to contribute to society through the pursuit of education, learning and research at the highest international levels of excellence. www.cambridge.org Information on this title: www.cambridge.org/9781009336789 DOI: 10.1017/9781009336772 © Martin Peterson 2023 This publication is in copyright. Subject to statutory exception and to the provisions of relevant collective licensing agreements, no reproduction of any part may take place without the written permission of Cambridge University Press & Assessment. First published 2023 A catalogue record for this publication is available from the British Library. A Cataloging-in-Publication data record for this book is available from the Library of Congress. ISBN 978-1-009-33678-9 Hardback Cambridge University Press & Assessment has no responsibility for the persistence or accuracy of URLs for external or third-party internet websites referred to in this publication and does not guarantee that any content on such websites is, or will remain, accurate or appropriate.

Published online by Cambridge University Press

Contents

List of Figures page vi List of Tables vii Preface ix Introduction

1

1 Meaning Tracks Use

18

2

37

Conflicting Reasons

3 Conflicting Sources of Normativity

59

4 The Binary Theory

80

5 Moral Indeterminacy and Vagueness

101

6 Normative Ethics for Gradualists

120

7 Rational Choice for Gradualists

144

8 Indeterminate and Vague Laws

166

9 Depolarization

184



204

Conclusions

References208 Index216

v

Published online by Cambridge University Press

Figures

1.1 A classic multidimensional scaling of Table 1.6 page 30 1.2 A three-dimensional metric multidimensional scaling of Table 1.6 32 1.3 A two-dimensional nonmetric multidimensional scaling of Table 1.6 33 3.1 A Voronoi diagram with five seed-points 71 3.2 Overlapping sources of normativity 74 3.3 Conflicting sources of normativity 75 3.4 Hampton’s S-shaped function 76 3.5 Gray areas induced by conflicting sources of normativity 77 6.1 The repugnant conclusion 134 6.2 The mere addition paradox 134 7.1 A sequential choice problem 150 7.2 A counterexample to the average and maximin criteria 152 7.3 A sequential choice under risk 154 7.4 The rightmost squares represent choices between options that are right to the same degree 156 9.1 The original Hegselmann–Krause model 194 9.2 The Bayesian model 195 9.3 The extended Hegselmann–Krause model 197 9.4 The extended Bayesian model 198 9.5 The same simulation as in Figure 9.4 but with a slightly higher confidence interval 199 9.6 The extended Bayesian model with three hypotheses and five peers 200 9.7 The extended Bayesian model with four hypotheses and five peers 201

vi

Published online by Cambridge University Press

Tables

0.1 1.1 1.2 1.3 1.4 1.5 1.6 2.1

Multidimensional consequentialism page 9 A Mann–Whitney U-test21 Four ordinal levels of agreement 22 Disagreement does not explain gradualist responses 23 Relative frequencies for C3–C6 27 Pairwise chi-square tests for C1–C6 28 Average degree of dissimilarity 29 A conflict between two verdictive reasons: wellbeing and equality 56 6.1 One- versus multidimensional consequentialism 123 6.2 Degrees of moral rightness w.r.t. total wellbeing and equality 124 6.3 Small changes matter 125 6.4 Portmore’s objection 129 6.5 Implications of Portmore’s objection 129 7.1 Does Up-Down dominate Down? 157 9.1 An example with three gradualist moral hypotheses 193

vii

Published online by Cambridge University Press

Published online by Cambridge University Press

Preface

This book explores the idea that some acts are neither entirely right nor entirely wrong, but rather a bit of both. We can think of such acts as falling in a moral gray area. Philosophical theories of indeterminacy and vagueness (where vagueness is understood as indeterminacy with sorites susceptibility) can shed light on some aspects of the gray area, but they also leave important questions open. It is, for instance, not clear what morally conscientious agents should do if forced to choose among options that are somewhat right and somewhat wrong. Should one select an option that is more right than wrong, or at least as right as every alternative, or would it perhaps be more rational to choose randomly among all somewhat right options? And how should lawmakers and courts address behavior that is neither entirely right nor entirely wrong? This work has been in the making for over a decade. Several individuals and organizations deserve my sincere thanks for invaluable help and support. Between 2009 and 2011, the Swedish Research Council generously funded a project on gradualism in ethics, which enabled me to set aside time to work on the core idea of this monograph. The project also enabled me to hire two post-docs, Nicolas Espinoza and Anna Bergqvist, who helped me organize a series of workshops. At the end of the project, Nicolas and I published an op-ed in a Swedish newspaper in which we argued that some abortions fall in a gray area (Dagens Nyheter, April 17, 2011). A cabinet minister wrote a response in which she made it very clear that she disagreed with our analysis. In her view, abortion is a binary moral issue. The minister’s argument was that there is no gray area in the legislation about abortion, so therefore there cannot be any moral gray area. This struck me as a weak argument. Existing laws may not always reflect what we ought to think about an issue. However, this exchange made me realize that abstract philosophical questions about degrees of moral rightness can have important implications not just for moral theory but also for society. ix

https://doi.org/10.1017/9781009336772.001 Published online by Cambridge University Press

x

Preface

Nicolas Espinoza and Anna Bergqvist provided helpful feedback in the early phases of the project, as did numerous colleagues at TU Eindhoven, in particular Rosemary Lowry and Philip Nickel. In 2013, Vuko Andric and Attila Tanyi hosted a workshop at the University of Konstanz on my book The Dimensions of Consequentialism, in which one of the chapters is devoted to the idea that rightness and wrongness come in degrees. I am very grateful to all workshop participants for their helpful and stimulating comments (later published in a special issue of Ethical Theory and Moral Practice), which helped me develop my gradualist theory further. I hereby wish to thank all participants: Campbell Brown, Joanna Burch-Brown, Roger Crisp, Frances Howard-Snyder, Vuko Andric, Attila Tanyi, and Jan Gertken. Since moving to Texas A&M University in 2014, I have benefitted from stimulating discussions with my wonderful colleagues in the department of philosophy, in particular Nathan Howard, who has offered insightful feedback on many chapters, as well as the members of the Group for Analytic Metaphysics and Epistemology, who gave me good feedback on Chapter 7. Finally, I would like to thank my research assistant Sean Conte for helping me prepare the final version of the manuscript. Six of the nine chapters have been written from scratch. Chapter 1 is a slightly revised version of “Moral Rightness Comes in Degrees,” Journal of the American Philosophical Association, 2022, and Chapter 3 is a revised version of “Radical Evaluative Ignorance,” in R. Peels (ed.), Perspectives on Ignorance from Moral and Social Philosophy, Routledge, 2017. Chapter 5 is partly based on “Multidimensional Consequentialism and Population Ethics,” in C. Seidel (ed.), Consequentialism: New Directions, New Problems?, Oxford University Press, 2018, and “The Dimensions of Consequentialism: Reply to Schmidt, Brown, Howard-Snyder, Crisp, Andric and Tanyi, and Gertken” in Ethical Theory and Moral Practice, 2016. A few sections also draw on “Abortion Is Neither Right Nor Wrong,” in Journal of Value Inquiry, 2022.

https://doi.org/10.1017/9781009336772.001 Published online by Cambridge University Press

Introduction

Murder is wrong and so is speeding on the highway, but murder is more wrong than speeding. From a moral point of view, the death of an innocent human being matters more than traffic violations. It is likewise more right to eradicate world poverty than to complete a book review on time. This is because the stakes are higher when people are starving.1 Be that as it may. There is also another, philosophically more interesting sense in which moral rightness and wrongness seem to come in degrees. According to the gradualist hypothesis, some acts are somewhat right and somewhat wrong. On this view, certain acts fall in a moral gray area, meaning that they are neither entirely right nor entirely wrong. Compare this to the observation that murder is more wrong than speeding. The latter comparison does not challenge the idea that every act is either right or wrong simpliciter, but the gradualist hypothesis is designed to do precisely that. Many traditional moral theories, in so far as they make claims about rightness and wrongness, are binary. In every choice situation, every act is judged to be either right or wrong but never a bit of both.The binary theory does not rule out the possibility of moral dilemmas in which all alternatives are wrong, but it rejects the possibility of dilemmas in which acts are both right and wrong. The gradualist hypothesis (I call it a “hypothesis” because it is novel and untested) and the binary theory (it is a “theory” because it is an established view) do not exhaust logical space. A third option is to refrain from 1

Hurka (2019) argues that both rightness and wrongness come in degrees in a sense that depends on how much moral value is at stake. He uses the terms “more seriously wrong” and “more importantly right” for characterizing these gradualist notions. Other authors have proposed similar ideas. For instance, Wedgwood (2014: 7) notes that, “[i]t might seem that in a sense the objective wrongness of acts comes in degrees: some acts are only slightly wrong, while other acts are very wrong, and so on,” but he eventually rejects this proposal. Wedgwood never suggests that rightness comes in degrees. See also Eriksson (1997) and Gustafsson (2016) for discussions of utilitarianism and degrees of wrongness.

1

https://doi.org/10.1017/9781009336772.002 Published online by Cambridge University Press

2

Ethics in the Gray Area

ascribing deontic properties to acts altogether.2,3 Inspired by Aristotle, G. E. M. Anscombe argues that modern moral philosophy rests on a mistake. She writes that “the concepts of obligation, and duty – moral obligation and moral duty, that is to say – and of what is morally right and wrong, and of the moral sense of ‘ought,’ ought to be jettisoned if this is psychologically possible.”4 Anscombe’s position is an example of what I call deontic nihilism. According to the nihilist, it is a mistake to characterize acts as right or wrong (or permissible or impermissible, or just or unjust) either because such deontic classifications are not central to moral theory, or because they are misleading. Another version of deontic nihilism, proposed by Alastair Norcross, holds that some acts are better or worse than their alternatives, but never right or wrong, or right or wrong to a degree.5 If we believe that some acts are somewhat right and somewhat wrong in the sense articulated by the gradualist hypothesis, this may have profound implications for how we think about critical real-world issues. Consider, for instance, the teenage girl Ana. At the age of seventeen, she got pregnant but eventually decided to have an abortion. In an interview six years later, she described the circumstances that influenced her decision as follows: New York, 2007 Unlike many Latinos, we’re not religious. My parents are progressive and always said I needed an education. It was my senior year of high school. My boyfriend was homeless. I bought a pregnancy test at Duane Reade and went to the bathroom in the middle of class. I sort of panicked but also thought, Let me get back to this tomorrow. On the train going home, I saw a sign. In my daze, all I saw was abortion. It was one of those places where they convince you to keep the baby. They showed me the ultrasound, but I wasn’t falling for that. Later, I went to see a counselor, and she made an appointment at Planned Parenthood. I had it on a Friday so I could recover for school. On Monday, I found a note on my bed—my boyfriend had left for California. When I got pregnant later that year, I was in Argentina. Abortion’s illegal there. I drove around with a doctor looking for someone who would do it. I can’t even say why I decided to keep the baby. I didn’t want an illegal abortion. And I was in love, I guess. I didn’t think I could go to college with a kid, but I’m graduating this year.6 2

3 4 5 6

By “deontic property” I mean an act’s property of being morally right, wrong, or somewhat right and somewhat wrong. Examples of influential views committed to the binary theory are Ross (1930), Nozick (1974), Scanlon (2000), Dancy (2004), and Kamm (2008), as well as many versions of consequentialism. There are several additional, logically possible but less plausible views. One example is the suggestion that all acts have one and the same deontic property (say, right). Anscombe (1958: 1). See Norcross (2006, 2020). In Winter (2013).

https://doi.org/10.1017/9781009336772.002 Published online by Cambridge University Press

Introduction

3

Did Ana act wrongly? The debate over abortion is deeply polarized. For quite some time moral philosophers and members of the public have debated, often in a very combative manner, whether abortion can ever be morally right. Some believe it is always wrong for a pregnant woman to abort her fetus, while others vehemently reject this conclusion. Ana’s story illustrates how abstract philosophical claims about the structure of moral judgments can alter the way we think about important ethical issues. Instead of concluding that Ana’s act was wrong, or right, the gradualist hypothesis brings to light a third possibility: some abortions might be somewhat right and somewhat wrong. If so, the choice between pro-life and pro-choice might be too blunt. We could, it seems, depolarize the debate by articulating a third, more nuanced alternative. According to this gradualist analysis, the moral complexity of the abortion issue warrants a verdict that does justice to all relevant moral considerations. Such a gradualist verdict may eventually enable us to find common ground that would otherwise be beyond reach. The gradualist hypothesis can, of course, be applied to other moral controversies as well. Binary disagreements over nuclear power, capital punishment, human cloning, military drones, and censorship on the Internet can be softened by introducing gradualist verdicts. The wide scope of the gradualist hypothesis is a strength, not a weakness. Moreover, in so far as philosophical theories can influence political discussions, the gradualist hypothesis might also be instrumental for depolarizing contentious political debates. Deep partisan divides in the United States and elsewhere have created a polarized political environment which the gradualist hypothesis could help defuse by introducing middle ground positions that may appeal to many voters.

Gradualism about Right and Wrong By definition, a gradualist is someone who accepts the gradualist hypothesis. Before we discuss the pros and cons of this hypothesis, it is helpful to clarify what gradualists believe and do not believe. Gradualists make a semantic claim about the meaning of certain words, but they also believe that some acts are a bit right and a bit wrong. The latter point is a first-order normative claim. Consider the following analogy: UNICORN (capitalized to indicate that we are talking about the concept and not the entity it refers to) means “horse with a single straight horn,” but this does not entail that there are any unicorns in the world. If no horse is equipped with the right kind of horn, then unicorns do not exist. This suggests that the gradualist hypothesis is best understood as a conjunction

https://doi.org/10.1017/9781009336772.002 Published online by Cambridge University Press

4

Ethics in the Gray Area

of two claims. The first is the semantic claim that deontic concepts such as RIGHT and WRONG are gradable, meaning that our moral language includes, and should include, concepts such as SOMEWHAT RIGHT, A BIT WRONG, and ALMOST ENTIRELY RIGHT. The second claim is the firstorder normative claim that some acts do in fact have deontic properties singled out by gradable deontic concepts. Deontic concepts are concepts that tell us what to do, whereas evaluative (or axiological) concepts such as GOOD and BAD assess the worth of things.7 It is widely agreed that evaluative concepts are gradable, but according to the gradualist hypothesis, this is true for deontic concepts as well. Consider the following definition of the gradualist hypothesis: The gradualist hypothesis =df Our moral language includes, and should include, gradable deontic concepts such as SOMEWHAT RIGHT, A BIT WRONG, and ALMOST ENTIRELY RIGHT, and some acts we perform can be accurately described by these and other gradable deontic concepts.

It is best to not say anything at this point about how many gradable deontic concepts there are. Some gradualists may insist that there are infinitely many shades of moral gray, while others might think that the number is finite. It is also worth keeping in mind that words and concepts are, of course, different things. The words “somewhat right” and “a bit right” are not the same, but SOMEWHAT RIGHT and A BIT RIGHT could nevertheless be the same concept. Meaning tracks use, and as indicated by data reported in Chapter 1, “somewhat right” is often used interchangeably with “a bit right.” This indicates that these phrases are likely to express synonymous concepts, so I will use both expressions for referring to the same idea. However, note that nothing important hinges on this. If further investigations reveal that SOMEWHAT RIGHT and A BIT RIGHT have different meanings, the gradualist should of course stop treating them as synonyms. Another potentially contentious issue concerns the deontic concepts PERMISSIBLE and IMPERMISSIBLE. In some fields, such as deontic logic and philosophy of law, it is common to distinguish between permissible and impermissible acts rather than right and wrong ones, but even though the words “permissible” and “right” may have slightly different connotations, I will nonetheless accept the common stipulation that an act is permissible if and only if it is right, impermissible if and only if it is 7

For a helpful overview of the distinction between deontic and evaluative concepts and properties, see Tappolett (2013).

https://doi.org/10.1017/9781009336772.002 Published online by Cambridge University Press

Introduction

5

wrong, and obligatory if and only if it is the only right option.8 My reason for this is pragmatic as this significantly simplifies the presentation. So, to say that an act is somewhat permissible is equivalent to saying that it is somewhat right, and so on. Advocates of the binary theory deny the semantic part of the gradualist hypothesis, and hence also the normative claim that some acts are somewhat right and somewhat wrong. According to the binary theory, RIGHT and WRONG (and PERMISSIBLE and IMPERMISSIBLE) do not permit of degrees. This is, of course, consistent with the idea that some acts may lack deontic properties altogether. The choice between using your left or right hand to open a door is, perhaps, not a moral choice. If so, neither option is right or wrong. This indicates that it might be a mistake to define the binary theory as the claim that every act is either right or wrong. The following definition is designed to be sensitive to this concern, without taking any definitive stand on its merits. (Mainstream consequentialists would of course insist that every act is either right or wrong because all acts have outcomes that can be ranked.9) The binary theory =df Deontic concepts such as RIGHT and WRONG do not permit of degrees, meaning that no act could be somewhat right and somewhat wrong. Every act, if it has any deontic property at all, is either right or wrong.

This definition clarifies the disagreement between gradualists and those who defend the binary theory. I have less to say about deontic nihilism. This is partly because the nihilist approach to ethics is so fundamentally different. Consider, for instance, the virtue ethicist who believes that it is a mistake to evaluate acts as right or wrong because the central question in ethics is to identify character traits that make us flourish.10 If I could show this to be false, then that would be worthy of more attention than the dispute between gradualists and binary theorists. For this reason, I leave it to others to discuss the pros and cons of deontic nihilism. However, I will mention some versions of deontic nihilism when that is helpful for understanding the gradualist hypothesis. For instance, the scalar consequentialist view proposed by Alasdair Norcross holds that acts should be ranked as 8 9 10

See, for example, Tännsjö (1998: 51 ff). As explained on p. 8, this holds true even if the ranking is incomplete. While I take this to be the dominant view in the literature, it is worth keeping in mind that some modern virtue ethicists disagree. For instance, Hursthouse (1999: 18) claims that “An action is right iff it is what a virtuous agent would characteristically do in the circumstances,” and Slote (2001) presents an alternative criterion of rightness for virtue ethicists.

https://doi.org/10.1017/9781009336772.002 Published online by Cambridge University Press

6

Ethics in the Gray Area

better or worse in accordance with their consequences, but never as right or wrong simpliciter. This view shares some ground with the gradualist hypothesis, which makes it worth discussing.11 It is also worth keeping in mind that deontic nihilism is a big tent with somewhat fuzzy boundaries. Consider, for instance, J. L. Mackie’s error theory.12 The error theorist is willing to admit that people say that acts are right or wrong, but insists that all moral statements are false. Hence, “this act is right,” “this act is wrong,” and “this act is somewhat right and somewhat wrong” are all false, no matter what features the act has. The problem is that the same is true of the nihilist statement “this act has no deontic status.” Because the error theory rejects its central claim, it would be odd to classify the error theory as a version of deontic nihilism. In fact, the error theory rejects the central claims of all major positions in this debate, that is, the binary theory, the gradualist hypothesis, and the nihilist theory. A possible way out could be to treat the error theory as a separate position that makes no claim about the deontic status of any act.

Three Central Questions Depolarizing ethical and political debates by introducing gradualist positions is desirable for obvious instrumental reasons, which I shall not dwell on here, but first and foremost because the gradualist hypothesis is interesting in its own right. I will focus on three questions: What does it mean to say that RIGHT and WRONG permit of degrees? What are the best reasons for accepting (or rejecting) the gradualist hypothesis? And what should a morally conscientious agent do if forced to choose among alternatives that are somewhat right and somewhat wrong?13 A natural answer to the first question is that we should interpret the gradualist hypothesis as a claim about moral indeterminacy or vagueness. Vagueness is a subspecies of indeterminacy: a concept is vague if is indeterminate and is susceptible to a sorites series, that is, if there exists a sequence of similar cases in which there is no sharp boundary between correct and incorrect applications of the concept. Philosophers have argued

11 12 13

See p. 98 ff. Mackie (1977). A morally conscientious agent is motivated entirely by moral concerns and always performs acts that are optimal means to the end of acting morally.

https://doi.org/10.1017/9781009336772.002 Published online by Cambridge University Press

Introduction

7

that numerous deontic concepts, including RIGHT, WRONG, PERMISSIBLE, and IMPERMISSIBLE are vague in this sense.14 For instance, Tom Dougherty asks us to consider how late a noisy party may go on. He suggests that it is determinately not wrong to make noise at 8 pm, and determinately wrong to do so at 3 am, but indeterminately wrong to make noise at 12 am. He also notes that this type of moral vagueness is problematic because it leaves us in a deliberative limbo.15 Would it, for instance, be wrong for a neighbor to complain about the noise at 12 am? Dougherty notes that we might sometimes be able to eliminate moral vagueness by introducing new social conventions: “If the residents universally come to accept the 12 am parties, and common knowledge arises about this universal acceptance, then these parties do actually become morally acceptable.”16 However, not all instances of moral vagueness can be eliminated in this manner, as illustrated by the debate over abortion. It would be overly optimistic to think that it is possible to broker peace between pro-choicers and pro-lifers by simply introducing new social conventions. Their disagreement is much deeper than so. Miriam Schoenfield notes that, “we can create a sorites series, admitting of borderline cases of permissibility, out of a series of abortions in which the fetus’ age differ by a day (or a minute, or a second).”17 Interestingly, she claims that “the only satisfactory account of moral vagueness is an ontic account.”18 In her view, moral vagueness would thus persist even if we were omniscient and spoke a perfect language. However, advocates of the gradualist hypothesis do not have to make any claim about moral vagueness. Rightness could vary in degrees in ways that do not require any vagueness. To account for such cases, we could simply add a moral concept between RIGHT and WRONG, which we could call SOMEWHAT RIGHT AND SOMEWHAT WRONG. Just like vagueness, this kind of moral indeterminacy could be ontic, epistemic, or semantic. The key difference would be that the transition from RIGHT to SOMEWHAT RIGHT AND SOMEWHAT WRONG would be discrete, meaning that there is no sorites series. In Chapter 5, I discuss these and other forms of moral indeterminacy and vagueness in detail.

14 15 16 17 18

See, for example, Constantinescu (2014), Dougherty (2014, 2016), Schoenfield (2016), and Williams (2017). Dougherty (2016: 448). Ibid., p. 459. Schoenfield (2016: 263). Ibid.

https://doi.org/10.1017/9781009336772.002 Published online by Cambridge University Press

8

Ethics in the Gray Area

The gradualist hypothesis differs in important respects from the familiar idea that moral values come in degrees. Everyone agrees that some good entities are better than other, less good entities, and that some entities are both a bit good and a bit bad at the same time. For instance, Mozart’s opera The Magic Flute is good (because the music is great) but it is also somewhat bad (because the libretto is weak), but it is not as good as The Marriage of Figaro (because both its music and libretto are better). Another example is Castro’s transformation of Cuba into a nondemocratic communist state. Castro’s revolution was bad to a high degree (because human rights were violated on countless occasions), but the revolution was also good to some degree (because it led to significant improvements in healthcare and education for the poor). The observation that evaluative concepts permit of degrees does not support the gradualist hypothesis.19 This is so even if we believe that some moral values are on a par or incomparable. (The notion of parity I have in mind is that proposed by Ruth Chang.20) Imagine, for instance, that it is right to perform an act just in case the total balance of moral value produced by the act is not determinately less than that of any alternative act. If so, all non-defeated alternatives would be right simpliciter, regardless of whether the relevant values are on a par, incomparable, or fully comparable. If we instead believe that it is wrong simpliciter to perform every act that is not determinately at least as good as every alternative, this would offer no more support for the gradualist hypothesis. Both these analyses are compatible with traditional binary accounts of RIGHT and WRONG and do not support the gradualist hypothesis.

Origins of the Gradualist Hypothesis The gradualist hypothesis is of recent origin. John Stuart Mill writes that “The creed which accepts as the foundation of morals, Utility, or the Greatest Happiness Principle, holds that acts are right in proportion as they tend to promote happiness, wrong as they tend to produce the reverse of happiness.”21 If we read this literally, it follows that acts that 19 20

21

For a recent attempt to define a binary notion of “ought” in terms of “better,” see chapter eight in Lassiter (2017). Lassiter does not discuss the gradualist hypothesis. Chang (2002) suggests that two items are on a par if and only if they are commensurable but not equally good, and neither item is better nor worse than the other. That two items are incommensurable means that no positive value relation holds between them, that is, it is false that one is better than the other, and it is also false that they are equally good or on a par. Mill (1863: 9), my emphasis.

https://doi.org/10.1017/9781009336772.002 Published online by Cambridge University Press

Introduction

9

Table 0.1  Multidimensional consequentialists believe that B is more right than A and C, but not entirely right Act

Alice’s wellbeing

Bob’s wellbeing

A B C

99 50 0

1 49 0

produce half as much happiness as the optimal alternative(s) are halfright, and so on. However, Mill’s gradualism seems to have been a slip of the tongue. In the rest of his writings, he never discusses this implicit notion of degrees of rightness, and he never uses the idea that rightness comes in degrees for supporting other claims, or for rebutting objections to his utilitarian theory. The standard view among contemporary consequentialists is that an act is right just in case no alternative brings about better consequences, and that every act that is not right is wrong. This criterion is incompatible with the gradualist hypothesis. No matter what the consequences are, every act will either be right or wrong simpliciter. This holds true even if the best consequences are incomparable or on a par, because in that case the consequences are not worse than those of any alternative, meaning that each such acts is right. That said, some nontraditional consequentialist theories do allow for gradable notions of RIGHT and WRONG. According to the multidimensional account of consequentialism defended in my Dimensions of Consequentialism, an act’s rightness or wrongness depends on several irreducible aspects, for example, the total wellbeing produced by the act and how wellbeing is distributed in society.22 According to the multidimensional theory, moral rightness comes in degrees whenever no act is optimal with respect to all moral aspects. Suppose, for instance, that total wellbeing and equality are two separate moral aspects.23 In the example in Table 0.1, act A is optimal with respect to total wellbeing but scores poorly with respect to equality.24 The opposite is true of C. Act B is not optimal with respect to any of the two aspects, but scores quite well with respect to both. Multidimensional consequentialists believe that all three alternatives are somewhat right and somewhat wrong, but B is 22 23 24

See Peterson (2013) for a detailed discussion of this proposal. For a definition of the term “moral aspect,” see chapter one in Peterson (2013). Ibid., p. 13.

https://doi.org/10.1017/9781009336772.002 Published online by Cambridge University Press

10

Ethics in the Gray Area

right to a higher degree than A and C. Multidimensional consequentialists might, for instance, conclude that B is almost entirely right (right to degree 0.99 on a scale from 0 to 1) while A and C are, roughly speaking, half right and half wrong (right to degree 0.5). Some deontologists also accept the idea that RIGHT and WRONG permit of degrees. In Moral Uncertainty and Its Consequences, Ted Lockhart claims that “actions come in varying degrees of moral rightness between ‘right’ and ‘wrong’.”25 He argues that this view is particularly attractive for those who accept W.D. Ross’ theory of prima facie duties: Even for some deontological theories, we have no great difficulty entertaining a many-valued concept of moral rightness. A prima facie duties theory, for example, may readily regard actions that, ceteris paribus, accord with more prima facie duties than others as having greater moral rightness. Moreover, for some prima facie duties, it seems that we can conform to or violate them to greater or lesser degrees. Examples of such duties in W. D. Ross’ theory include beneficence, nonmaleficence, and justice.26

Lockhart’s proposal is a revisionary extension of Ross’ theory, which Ross himself would not accept. Ross is committed to what Lawlor calls a “threshold account” of rightness, just as mainstream consequentialists: Acts that have the right kind of properties make it across the threshold, and those acts are right (or permissible) simpliciter.27 All other acts are wrong. For mainstream consequentialists, the threshold property is that of having optimal consequences, and for Ross the threshold property is to be fitting to perform in light of all applicable prima facie duties. Somewhat surprisingly, Lockhart’s own theory is a hybrid view that collapses into a threshold account when only two alternatives are available: x has greater degree of moral rightness than y in situation S for agent A just in case, if x and y were the only alternatives open to A in S, then x would be morally right for A in S and y would be morally wrong for A in S.28

Lockhart’s thesis is compatible with Hurka’s common sense version of gradualism mentioned in footnote one, as well as with Mill’s gradualist account of utilitarianism. However, as pointed out by Campbell Brown, Lockhart’s reductive analysis is unable to compare degrees of rightness across different situations: “a person who uses her mobile phone while 25 26 27 28

Lockhart (2000: 75). Ibid., p. 80 Lawlor (2009). Lockhart (2000: 81).

https://doi.org/10.1017/9781009336772.002 Published online by Cambridge University Press

Introduction

11

watching a movie in a public cinema may be acting wrongly, but surely not as wrongly as one who does the same while driving a heavy lorry on a motorway thereby causing a major accident’.”29 The problem for Lockhart is that the acts described by Brown are performed in different situations and can therefore not be compared according to his criterion.

Why Accept the Gradualist Hypothesis? I will try to show that the gradualist hypothesis is supported by three arguments: the meaning-tracks-use argument (discussed in Chapter 1), the argument from conflicting reasons (see Chapter 2), and the argument from conflicting sources of normativity (see Chapter 3). I do not claim that these arguments are conclusive, but together they give significant support to the semantic claim that RIGHT and WRONG permit of degrees, as well as the first-order normative claim that some acts are somewhat right and somewhat wrong. The meaning-tracks-use argument starts from a version of Wittgenstein’s thesis that meaning is intimately connected with use.30 However, the argument is not premised on the controversial assumption that meaning is use in a strict semantic sense. The meaning-tracks-use argument merely assumes that use is a reliable guide to meaning. We can leave it open what meaning is. The key premise is that meaning tracks use: if the vast majority of speakers persistently and sincerely use RIGHT and WRONG as gradable concepts, then this supports the conclusion that RIGHT and WRONG are gradable concepts. To test the empirical part of the meaning-tracks-use argument, Chapter 1 presents data from three surveys. These surveys indicate that respondents use RIGHT and WRONG as gradable concepts to approximately the same extent as color concepts, suggesting that rightness and wrongness come in degrees just as much as colors do. Almost no respondents persistently use right and wrong as non-gradable concepts. Moreover, gradualist responses are common even if no gradualist terms or phrases appear in the stimuli, which reduces the risk of acquiescence bias. The meaning-tracks-use argument does not by itself offer enough warrant for the gradualist hypothesis. What is missing is an account of why 29 30

Brown (2016: 27). In paragraph 43 of his (1953), Wittgenstein writes that “For a large class of cases – though not for all – in which we employ the word ‘meaning’ it can be defined thus: the meaning of a word is its use in the language.”

https://doi.org/10.1017/9781009336772.002 Published online by Cambridge University Press

12

Ethics in the Gray Area

we should use RIGHT and WRONG as gradable concepts. The way ordinary people use moral concepts is a good starting point for moral theorizing, but it is of course possible that our current usage ought to be modified. Another way to express this worry is to point out that it is not clear if moral theories are, or should be, concerned with folk notions of RIGHT and WRONG. Although the gradualist hypothesis may offer an accurate description of how deontic concepts are used by the folk, it could nevertheless be preferable to use RIGHT and WRONG as binary concepts for theoretical reasons. The best response to this objection is to explain why moral philosophers, if they are willing to ignore the folk notions of RIGHT and WRONG, may nevertheless have reason to accept the gradualist hypothesis. What gradualists need to offer is a clear account of what is gained by introducing gradable versions of RIGHT and WRONG in moral philosophy. The argument from conflicting reasons, discussed in Chapter 2, seeks to do this by showing that the gradualist hypothesis is crucial for understanding moral conflicts. At the beginning of Republic, Plato asks us to consider the following example of a moral conflict: “Suppose that a friend when in his right mind has deposited arms with me and he asks for them when he is not in his right mind, ought I to give them back to him?”31 Socrates points out to Cephalus that under those circumstances, it would be wrong to return the arms because the friend is not in his right mind. However, it would also be wrong not to return the arms as they do in fact belong to the friend who deposited them. If we accept both these premises, it follows that it is wrong and not wrong to return the arms. Another example is mentioned by Jean-Paul Sartre in Existentialism is a Humanism. Sartre describes a case in which one of his students during World War II “had the choice between going to England to join the Free French Forces or of staying near his mother and helping her to live. He fully realised that this woman lived only for him and that his disappearance – or perhaps his death – would plunge her into despair.”32 According to Sartre, the student had to be loyal to his country. Therefore, it would have been wrong not to join the resistance movement in England. However, the student also had to help his mother to live, meaning that it would have been wrong to travel to England. Hence, all alternatives open to the student were wrong. Generally speaking, a moral conflict is a situation in which an agent has several incompatible but undefeated reasons to perform several acts, but is 31 32

Republic, 1.331c. Sartre (1946/2021: 295).

https://doi.org/10.1017/9781009336772.002 Published online by Cambridge University Press

Introduction

13

unable to perform them all, or has undefeated reasons to perform and not perform one and the same act. That a reason is undefeated means that it is not canceled out when balanced against other reasons. Gradualists, as well as some binary theorists, agree that the examples mentioned by Plato and Sartre can be plausibly understood as situations in which undefeated reasons clash. Cephalus in Plato’s example has an undefeated reason to hand back the arms but another undefeated reason not to do so, and Sartre’s student has an undefeated reason to stay with his mother but another undefeated reason not to do so. By construing moral conflicts as cases in which rightness and wrongness come in degrees, rather than as traditional moral dilemmas in which all options are outright wrong, gradualists can offer a more nuanced analysis of these cases. According to the argument from conflicting reasons, gradualists can analyze moral conflicts in a manner that is less heavyhanded than the analysis offered by traditional binary theorists. Instead of claiming that Plato’s and Sartre’s cases are examples of moral dilemmas, gradualists believe that it is somewhat right but also somewhat wrong to give back the arms in Plato’s example, and somewhat right and somewhat wrong for Sartre’s student to abandon his mother. However, the degree to which these acts are right and wrong need not be equal. Gradualists can compare options across situations and may thus conclude that it would, say, be right to a higher degree for Sartre’s student to join the resistance movement than for Cephalus in Plato’s example to hand back the arms. Such nuanced deontic verdicts cannot be stated if we insist that Plato’s and Sartre’s examples are traditional moral dilemmas in which all options are wrong simpliciter. The gradualist analysis avoids the implausible conclusion that moral conflicts are merely illusory. Some moral theories, such as mainstream consequentialism and Kantianism, are committed to the view that what appears to be a genuine moral conflict can always be resolved in a manner that makes the conflict disappear.33 These theories do this by identifying a set of binary deontic verdicts entailing that at least one act is right and no more than one is obligatory. This guarantees that the alleged dilemmas discussed by Plato and Sartre are merely illusory, which does not square well with what is supposed to be main point of these examples. 33

Kant’s moral theory can be interpreted in more than one way. Some scholars deny that Kant is committed to the binary theory because he is silent about RIGHT and WRONG, he just makes a claim about DUTY. However, note that this does not entail that Kant would be willing to accept gradualist verdicts. Kant makes it very clear that in his view, there is no gray area in ethics.

https://doi.org/10.1017/9781009336772.002 Published online by Cambridge University Press

14

Ethics in the Gray Area

The third argument for the gradualist hypothesis draws on a certain way of thinking about conflicts across different domains of normativity, such as morality and self-interest. Here is an example: From a selfinterested point of view, it would perhaps be best for you to invest the money you currently have in your checking account in a retirement fund, but from a moral point of view it would be better to donate the money to charity. So, what should you do, all things considered, when morality and self-interest clash? The argument from conflicting sources of normativity asserts that conflicts of this type do occur and warrant gradualist verdicts. The argument proceeds in two steps. The first step seeks to show that there exists more than one source of normativity, and that different sources sometimes clash. This is by no means a trivial claim as it seems possible to maintain that all normativity stems from one and the same source. The second step seeks to show that clashes between conflicting sources of normativity warrant gradualist verdicts. This is also a claim that needs to be defended carefully. An influential alternative position, pioneered by Sidgwick, is that the notion of an all-things-considered ought is either meaningless or vacuous. According to Sidgwick, all we can conclude is that sometimes we ought to do something with respect to one source (morality) but something else with respect to another source (self-interest). Therefore, all that can be said is that from a self-interested point of view you ought to invest the money in your checking account in a retirement fund, but from a moral point of view you ought to donate the money to charity. To ask what you ought to do all things considered is a meaningless question that has no answer. My suggestion is to think of different sources of normativity as distinct regions in a shared conceptual space. The term “conceptual space” is a technical term introduced by Peter Gärdenfors.34 By applying this theory to normative issues, we can articulate a nonrigid theory of normativity according to which morality sometimes overrides self-interest and selfinterest sometimes overrides morality. In such cases, it is appropriate to conclude that what the agent ought to do is best captured by gradualist verdicts. To some degree, it is morally right to donate money to charity, but the agent also has a gradualist self-interested obligation to save some money for retirement. All things considered, it is right to some degree to make the donation, at the same time as it is right to some degree to refrain from doing so. 34

Gärdenfors (2000, 2014).

https://doi.org/10.1017/9781009336772.002 Published online by Cambridge University Press

Introduction

15

Preview This book consists of nine further chapters and a conclusion. The meaning-tracks-use argument, the argument from conflicting reasons, and the argument from conflicting sources of normativity introduced in the preceding section are discussed in the first three chapters. I take these arguments to indicate that some version of the gradualist hypothesis is true, but I admit that they leave important questions open. Chapter 4 discusses the pros and cons of the binary theory. One of the six arguments examined in this chapter asserts that gradable notions of RIGHT and WRONG would be semantically incomplete or even incoherent. Another argument holds that there exists a series of analytic connections between the binary notions of RIGHT and WRONG and certain reactive attitudes, such as resentment and forgiveness, which speaks against the idea that RIGHT and WRONG permit of degrees. I respond to these arguments, as well as several others, by showing how gradualists can accommodate the concerns that are thought to support the binary theory at least as well as critics of the gradualist hypothesis. I also introduce an argument for the binary theory that is rebutted in greater detail in another chapter, namely, the idea that choices are ultimately binary, which could be taken to support the binary theory unless we are able to state a decision criterion that fits with the idea that some acts are somewhat right and somewhat wrong. Chapter 5 seeks to determine if the gradualist hypothesis should be understood as an epistemic, semantic, or ontic claim. I defend the ecumenical view that some instances of gradualism are merely epistemic, while others are semantic or ontic. A challenge faced by defenders of the ecumenical view is to explain what ontic moral indeterminacy is. Drawing on influential articles by Elisabeth Barnes and Miriam Schoenfield, I argue that if moral realism is true then we can make sense of ontic moral indeterminacy without postulating the existence of non-bivalent third-category entities.35 Another challenge is to explain why several different analyses are needed. I do this by providing examples of cases that seem to be best understood as instances of epistemic, semantic, and ontic indeterminacy. Chapter 6 discusses the implications of the gradualist hypothesis for normative ethics. Mainstream theories of normative ethics assume that every act is either right or wrong simpliciter, which is incompatible with the gradualist hypothesis. Although nearly any theory could be revised to 35

See Barnes (2010b) and Schoenfield (2016).

https://doi.org/10.1017/9781009336772.002 Published online by Cambridge University Press

16

Ethics in the Gray Area

accommodate gradualist verdicts, the focus of my discussion is on consequentialist theories. The reason for this is largely idiosyncratic. The multidimensional approach to consequentialism defended in Dimensions of Consequentialism squares well with the gradualist hypothesis, and I am keen to show that the moral implications of my preferred normative theory are intuitively appealing. The focus of Chapter 7 is on decision-making. What should a normatively conscientious agent do when making decisions among acts that are neither entirely right nor entirely wrong? I distinguish between an act’s degree of moral rightness and its property of being rationally permissible for a morally conscientious agent. J. Robert G. Williams has proposed a similar distinction between what he calls “moral oughts” and “decision oughts,” and Adam Bales distinguishes between an act’s “moral permissibility” and its “choiceworthiness.”36 If all alternatives available to an agent are entirely right or wrong, then the agent is rationally permitted to perform any of the entirely right acts, but none of the entirely wrong ones. However, it is not clear how this insight should be generalized to options that are somewhat right and somewhat wrong. Williams discusses the idea that it is rationally permissible to perform every act that is not outright morally wrong.37 An alternative view, proposed by Susanna Rinard and discussed in greater detail by Bales, is to maintain that there is no fact of the matter about whether acts that are somewhat right and somewhat wrong are rationally permissible.38 A third alternative is to argue that an act is rationally permissible if it is not more wrong than right, and a fourth option is to claim that agents are rationally obliged to always perform (one of) the option(s) that is morally right to the highest degree, meaning that actions that are morally right to a lower degree are rationally impermissible. Yet another alternative, defended in a different context in Dimensions of Consequentialism, and also by Williams, is that agents making decisions in a nonbinary moral landscape are (only) permitted to randomize their choices.39 According to this proposal, which I believe is the most plausible one, an act that is somewhat right and somewhat wrong is rationally permissible if and only if it is chosen randomly with a probability that reflects its degree of moral rightness. 36 37 38 39

Williams (2017). See also Bales (2018). Williams (2017: 670) expresses this idea in a slightly different way: “A choice to X is decision permissible iff it is not determinately objectively impermissible to X.” Rinard (2015) and Bales (2018). Williams (2014).

https://doi.org/10.1017/9781009336772.002 Published online by Cambridge University Press

Introduction

17

Chapter 8 discusses the relationship between the gradualist hypothesis and the law. What are the legal implications, if any, of the moral conclusion that some acts are somewhat right and somewhat wrong? Should an act that is somewhat right and somewhat wrong remain legal, or should it be prohibited by the law? Several philosophers have discussed indeterminacy and vagueness in the legal domain without taking a stand on foundational issues about legal positivism and its alternatives. In this chapter, I follow their lead. My aim is to articulate and defend a view I call precedentism, which holds that judges can sometimes create new practical oughts by establishing precedents, even though such precedents do not make any vague or indeterminate laws sharp. This is, of course, not a novel proposal. My contribution to the debate is primarily a distinction between “doctrinal oughts” and “practical oughts,” which makes the precedentist’s position clearer. Roughly, precedentists believe that when a legal doctrine is vague or indeterminate, then the letter of the law is indeed vague or indeterminate: Courts do not make vague or indeterminate laws sharp by creating precedents. However, it does not follow that such vagueness or indeterminacy has to be reflected in any practical decision about how the court ought to rule. The practical ought may be determinate, even if the doctrinal ought is not. Chapter 9 explores the idea that the gradualist hypothesis enables its defenders to depolarize controversial moral debates by introducing gradualist positions located between the two endpoints of the deontic spectrum. There is a sense in which this claim about depolarization is trivially true: By merely adding a new, middle ground position to a polarized moral debate, the addition of this third alternative will automatically depolarize the debate in the sense that the variance of the opinions expressed in the group will decrease. However, a more important question, which is the one I focus on in Chapter 9, is whether the introduction of gradualist positions can be expected to lead to depolarization among fully rational peers. My discussion draws on recent work on peer disagreement and Bayesian epistemology, which is applied to a moral setting in a series of computer simulations. I conclude that under many circumstances, especially those in which rational peers are aware of and seriously consider the opinions of people they disagree with, the introduction of the gradualist hypothesis will indeed enable its defenders to depolarize strongly polarized moral debates.

https://doi.org/10.1017/9781009336772.002 Published online by Cambridge University Press

chapter 1

Meaning Tracks Use

Meaning tracks use. If an overwhelming majority of competent language users frequently say that some acts are somewhat right and somewhat wrong, then this indicates that RIGHT and WRONG are gradable concepts. Obviously, for this argument to be convincing, it is not enough to show that a few individuals occasionally use RIGHT and WRONG as gradable concepts. Not every usage of a concept is thoughtful and sincere. What we need to show is that many people persistently and sincerely talk about RIGHT and WRONG as if they permit of degrees.1 The structure of the meaning-tracks-use argument is, thus, as follows: (1) If the vast majority of competent language users frequently and sincerely use RIGHT and WRONG as gradable concepts, then RIGHT and WRONG are gradable concepts. (2) The antecedent of the first premise is true. (3) Therefore, RIGHT and WRONG are gradable concepts. On a strict interpretation of the first premise, the vast majority of competent language users cannot be wrong about their persistent and sincere use of moral concepts. This is, however, not the intended interpretation. The meaning-tracks-use argument merely assumes that the way people use words and phrases is a reliable guide to meaning. This assumption is weaker than Wittgenstein’s claim that meaning is use.2 Regardless of what meaning is, few would question that we can typically figure out the meaning of a concept by studying how it is used. This holds true for moral as well as nonmoral concepts. The epistemic interpretation of the meaning-tracks-use argument is endorsed by modern linguists. For instance, Katrin Erk notes that socalled vector space models of meaning are based on the observation that

1 2

As noted on p. 4, the gradualist hypothesis has a semantic as well as a non-semantic part. This ­chapter is exclusively concerned with the semantic component of the gradualist hypothesis. See Wittgenstein (1953) and the quote in footnote 30 on p. 11.

18

https://doi.org/10.1017/9781009336772.003 Published online by Cambridge University Press

Meaning Tracks Use

19

“we can often guess what a word means from the contexts in which it is used. Thus, we can represent meaning as distribution, as observed contexts.”3 Vector space models have proven to be empirically successful. By observing how words are used in large corpora, by counting their occurrence in different sentences, computational linguists have developed software that accurately predicts whether the meaning of two words is similar or not.4 However, a well-known limitation of the computational approach is that it requires large quantities of empirical data. As Erk puts it, “many phrases do not occur with sufficient frequency in a corpus to be represented through their distributional contexts.”5 This includes phrases relevant for assessing the gradualist hypothesis, for example, “this act is somewhat right and somewhat wrong” and “this act is a bit right and a bit wrong.” To overcome this problem, this chapter uses the methods of experimental philosophy. Data from three surveys are presented. In the largest, which had more than 700 respondents, no more than four percent of ordinary language users persistently used RIGHT and WRONG as binary concepts.6 The statistical analysis also indicates that RIGHT and WRONG are used as gradable concepts to approximately the same extent as color concepts, which suggests that rightness and wrongness come in degrees about as much as colors do. Furthermore, by using multidimensional scaling techniques, it can be shown that RIGHT and WRONG are used as gradable concepts even when no gradable terms (such as “right to some degree” or “somewhat right and somewhat wrong”) appear in the questions or answer options. This reduces the risk of acquiescence bias. The presentation of the empirical evidence for the meaning-tracks-use argument presupposes some knowledge of statistics. Readers who are willing to accept the claim that the vast majority of competent speakers do use RIGHT and WRONG in ways that permit of degrees without scrutinizing the empirical evidence can skip the rest of this chapter. 3 4 5

6

Erk (2012: 635). A popular measure of similarity in linguistic is the cosine of the vector space representation of each word-pair. For details, see Erk (2012: 637) and Turney and Pantel (2010: 160). Erk (2012: 637). According to Google’s word2vec algorithm the five most similar words to “moral” are “ethical” (cosine distance 0.363), “ethics” (0.471), “religious” (0.489), “morality” (0.513), and “philosophical” (0.518). It is not surprising that “ethical” is more similar to “moral” than “religious,” but some readers might find it surprising that “religious” is closer to “moral” than is “philosophical.” This figure is a summary of the overall results based on data from all semiabstract and concrete items included in the largest of the three studies, which included four answer options. The corresponding figure for the second study, in which each respondent evaluated only one semiabstract item and was presented with five answer options was, as expected, higher (twelve percent).

https://doi.org/10.1017/9781009336772.003 Published online by Cambridge University Press

20

Ethics in the Gray Area

Background and Design Questionnaires were distributed to three sets of respondents: students at Texas A&M University taking a class in engineering ethics in Spring 2019 (n = 715), students taking the same class in Spring 2020 (n = 578), and a group of U.S. citizens who voted in the 2016 election (n = 182). Students took the surveys for credit (about 0.5% of the total course grade) while respondents in the third study received between $2.65 and $3.00. In all three studies, respondents were invited to answer up to twelve questions. The order of the questions as well as the answer options was randomized. At the request of the Institutional Review Board at Texas A&M University, no demographic information was collected. Respondents were presented with up to five different types of tasks: abstract, semiabstract, concrete, comparative, and open-ended tasks. All tasks are described in the following sections. By using numerous measures for testing a single hypothesis, the validity of the measurement instrument can be assessed. If the validity is high, we should expect the results to be roughly the same for all types of tasks.

Abstract Tasks Subjects were invited to evaluate abstract statements about ethics, mathematics, colors, scientific evidence, and scientific facts on a seven-point Likert scale ranging from “strongly disagree” (1) to “strongly agree” (7). The following example pertains directly to the gradualist hypothesis: A1. Moral rightness and wrongness come in degrees. The boundary that separates morally right acts from wrong ones is not always sharp. Some acts are somewhat right and somewhat wrong. Study 1: average degree of agreement 5.7 (n = 237, std. dev. 1.4) Study 2: average degree of agreement 5.4 (n = 114, std. dev. 1.3) Study 3: average degree of agreement 5.4 (n = 181, std. dev. 1.5)

The following abstract statements were used as reference points for comparative purposes: A2. In mathematics, truth comes in degrees. The boundary that separates true mathematical statements from false ones is not always sharp. Some mathematical statements are somewhat true and somewhat false. Study 1: average degree of agreement 2.7 (n = 219, std. dev. 1.7) Study 2: average degree of agreement 2.6 (n = 79, std. dev. 1.6) Study 3: average degree of agreement 2.2 (n = 181, std. dev. 1.5)

https://doi.org/10.1017/9781009336772.003 Published online by Cambridge University Press

Meaning Tracks Use

21

Table 1.1 Mann–Whitney U-tests for data in Study 2

A1. Moral rightness and wrongness come in degrees … (n = 114)

A2. In mathematics, truth comes in degrees … (n = 79)

A3. Colors come in degrees … (n = 80)

Mann–Whitney U = 1,111 p-value