Intention, Character, and Double Effect

Intention, Character, and Double Effect

LAWRENCE MASEK

INTENTION, CHARACTER,

and

DOUBLE EFFECT

University of Notre Dame Press

Notre Dame, Indiana

University of Notre Dame Press

Notre Dame, Indiana 46556

undpress.nd.edu

Copyright © 2018 by University of Notre Dame

All Rights Reserved

Published in the United States of America

Library of Congress Cataloging-in-Publication Data

Names: Masek, Lawrence J., 1976– author.

Title: Intention, character, and double effect / Lawrence J. Masek.

Description: Notre Dame : University of Notre Dame Press, 2018. | Includes bibliographical references and index. |

Identifiers: LCCN 2018043825 (print) | LCCN 2018044731 (ebook) | ISBN 9780268104719 (pdf) | ISBN 9780268104726 (epub) | ISBN 9780268104696 (hardback : alk. paper) | ISBN 0268104697 (hardback : alk. paper)

Subjects: LCSH: Double effect (Ethics)

Classification: LCC BJ1500.D68 (ebook) | LCC BJ1500.D68 M37 2018 (print) | DDC 170/.42—dc3.

LC record available at https://lccn.loc.gov/2018043825

∞ This paper meets the requirements of ANSI/NISO Z39.48-1992

(Permanence of Paper).

This e-Book was converted from the original source file by a third-party vendor. Readers who notice any formatting, textual, or readability issues are encouraged to contact the publisher at [email protected]

For Maria and Mark

Contents

Preface

Acknowledgments

Introduction

One The Rational Basis of the Principle of Double Effect

Two An Agent-Based Definition of Intended Effects

Three The Strongest Objection to the Principle of Double Effect

Four Trolley Cases and an Objection from Neuroscience and Moral Psychology

Five Hard Cases in Medicine and War

Appendix: Case Summaries

Notes

Bibliography

Index

Preface

My interest in the principle of double effect (PDE)—which I define as the principle that the difference between intended effects and foreseen side effects is relevant for judging actions—began when I was a graduate student and took an ethics course that covered PDE. The principle interested me because of its applications to fictitious cases of runaway trolleys, flooding caves, and organ transplants, as well as to real cases of war, euthanasia, and pregnancy. PDE seemed intuitively plausible, but I did not think about the basis of these intuitions.

My thinking about PDE, and about moral philosophy in general, changed a few years later when I read Alasdair MacIntyre’s After Virtue. MacIntyre’s book says little about specific questions in applied ethics and nothing explicit about PDE, but he persuaded me that philosophers should not settle for arguments that rely on mere intuitions. Instead of using intuitions to defend a moral theory against its rivals, a proponent of the theory must show that the theory can identify, explain, and overcome limitations of its rivals.¹ For example, someone cannot refute Kant merely by citing the case of lying to a murderer at the door, and someone cannot refute utilitarianism merely by citing the case of framing and executing an innocent scapegoat to prevent deadly riots. I concluded—and I still believe—that mere intuitive appeals cannot settle the debate about PDE. The same cases can elicit different intuitions from different people. Critics cannot design a counterexample that refutes PDE once and for all, and proponents of PDE cannot hope to refine the principle so that it fits everyone’s intuitions about every case. To resolve the debate, a proponent of PDE must find the different theories of morality that support these conflicting intuitions and then show which theory is best.

MacIntyre also persuaded me that moral rules make sense only as directions for people to use to fulfill their potential as human beings. This teleological view of moral rules may be unfashionable in contemporary moral philosophy, but MacIntyre argues that philosophers who reject this view cannot explain why people should follow moral rules, given that they have strong inclinations to disobey them.² At the time, I did not see MacIntyre’s defense of this view of morality as relevant for debates about PDE.

When I applied for jobs a few years later, an interviewer noted that I had published a paper about PDE.³ He asked why PDE first developed in the Catholic tradition and why it is still more widely accepted among Catholics than non-Catholics. I did not have a good answer at the time. I said that the Catholic tradition includes exceptionless moral rules, such as the prohibition of murder, and that these rules would create inescapable moral dilemmas without PDE. I was unsatisfied with my answer, so my interviewer’s question stayed in the back of my mind. I still covered PDE when I taught courses about ethics, but my scholarly work focused on other issues.

A few years later, I read a critique of my paper about PDE.⁴ I welcomed the chance to restate and improve my argument, which I had come to see as flawed in many ways, so I wrote a reply.⁵ Writing this reply made me think more seriously about PDE, including my interviewer’s question about PDE and the Catholic tradition. The close relation between PDE and the Catholic tradition did not seem like a historical accident, but I did not believe that PDE depended entirely on theological premises or appeals to authority. The intuitions that support PDE still seemed clear to me. Further, some philosophers, such as Thomas Nagel and Philippa Foot, accept PDE despite being atheists.⁶ (Foot even describes herself as a card-carrying atheist.⁷) I wanted to find the basic reasons that people disagree about the relevance of an agent’s intention. Assuming that one side of the debate is simply stubborn, confused, or blinded by religious beliefs would have cut off the analysis too soon, before I had any real insight into the disagreement.

One cause of my confusion was that many contemporary philosophers divide moral theories into consequentialism, which says that the right action is the one that causes the best consequences, and deontology, which says that people should follow duties or obligations even when violating them would cause better consequences than following them. PDE does contradict consequentialism, but calling it a deontological principle is misleading, because the disagreements about PDE do not line up with the disagreements between deontologists and consequentialists. Prominent deontologists such as Alan Donagan, T. M. Scanlon, and Judith Jarvis Thomson reject both consequentialism and PDE. My thinking about PDE began to focus on finding the basic disagreement that separates proponents and critics of PDE.

This disagreement came into clearer focus after I studied Philippa Foot’s critique of consequentialism.⁸ She tries to explain why consequentialism is plausible and why it is ultimately mistaken. Her answer is that consequentialists correctly see benevolence as one of the virtues but that they treat benevolence as the only virtue, ignoring justice and other virtues. In other words, she argues that consequentialists confuse part of morality (i.e., causing happiness for other people) with the whole of morality. I came to believe that contemporary deontologists who reject PDE make a mistake similar to the one that Foot sees in consequentialism. While consequentialists reduce acting morally to causing happiness for other people, many contemporary deontologists reduce acting morally to treating other people properly. Consequentialists reduce all the virtues to benevolence, and contemporary deontologists reduce all the virtues to justice. Deontologists do not accept consequentialism because they deny that people may mistreat an individual to cause more happiness. They agree with consequentialists, however, that whether an action is right or wrong depends only on its effects outside the agent. According to this deontological view of morality, the difference between intention and foresight might be relevant for judging the agent’s character, but the difference is irrelevant for judging whether the action itself is right or wrong. Victims are no worse off when their killers intend death than when their killers merely foresee death.

An alternative to both consequentialism and deontology is the view of morality that I accepted after studying After Virtue. According to this view, an action can be wrong because of the way it forms the agent’s character, even if the action does not have any other harmful effects. This view explains why the difference between intention and foresight is relevant for judging both actions and agents. I first sketched this account of the basic disagreement about PDE in a paper published in 2010.⁹ In this book I develop my argument and explain how many disagreements about PDE arise from disagreements about the nature of moral rules.

After I was convinced that PDE makes sense only if the agent’s character is relevant for judging actions, I began to see that confusion results when philosophers set aside the agent and focus on causal relations outside the agent. For example, many proponents of PDE replace the distinction between intention and foresight with a distinction between direct and indirect actions. Critics of PDE then have an easy time arguing that there is no plausible definition of direct and indirect.

My defense of PDE appeals to the agent both to explain why intentions are relevant for judging actions and to define what agents intend in specific cases. An agent-based version of PDE resolves some controversial issues in debates about PDE, including debates about medical procedures that save a pregnant woman while killing her unborn child. I argue that someone who accepts the view of morality that supports PDE can identify and resolve some problems better than consequentialists and many contemporary deontologists. I use some cases as illustrations, but I do not rely only on intuitions about these cases to defend PDE or to criticize its rivals.

I now have a better answer to the question about why PDE developed in the Catholic tradition. As I explain later, people do not need to be Catholic to accept PDE, but many philosophers and theologians in the Catholic tradition see developing a good character as an important part of living morally. Ultimately, PDE does not depend on appeals to authority or to intuitions about cases. Instead, it depends on a view of morality that treats a person’s character as relevant for judging actions, a view that I defend throughout this book.

Acknowledgments

I begin by thanking Michael Dougherty for his comments on early drafts of every chapter of this book. For comments on drafts of chapters and papers that developed into chapters, I thank Brian Besong, Thomas Cavanaugh, Walter Kokernot, Matthew Ponesse, Michael Storck, and Christopher Tollefsen. I thank the late Joseph Boyle and an anonymous referee at The Philosophical Quarterly for their comments on a paper in which I first presented a justification of PDE and a definition of intended effects. In the introduction, chapter 1, and chapter 2 I develop the arguments in that paper. My defense of PDE does not depend on the version of natural law theory defended by Boyle and Tollefsen, but their writings, along with those of Germain Grisez and John Finnis, have shaped my views about PDE. Their influence is evident throughout this book.

I further thank participants in a 2013 conference at Franciscan University, where I presented an earlier version of the argument in chapter 1. I thank John Zeis for comments on my article in American Catholic Philosophical Quarterly, which presents the genealogy of the trolley problem and the minimalist version of PDE that I develop in chapter 4. I thank Marc Maitre for his help with the figures and illustrations in this book. A sabbatical from Ohio Dominican University enabled me to write chapters 1, 4, and 5.

I am grateful to Matthew Dowd, Stephen Little, and Wendy McMillen at the University of Notre Dame Press, as well as the manuscript reviewers, for their guidance and comments. I also am grateful to Marilyn Martin for her meticulous copyediting of the manuscript.

Finally, I thank my parents, who provided me with the education that makes my work possible, and my wife and children, who supported me throughout this project. As will become clear in my defense of PDE, being a husband and father has shaped my thinking about moral philosophy. I trust that my family will forgive me for all of the unflattering examples about children, parents, and spouses. These examples are, of course, fictitious.

Introduction

According to the principle of double effect, the difference between intention and foresight is relevant for determining whether an action is right or wrong. To illustrate this difference, consider two cases of bomber pilots who kill innocent civilians:

• One pilot targets a hospital so that the deaths of the hospital’s patients will weaken the enemy’s morale.

• Another pilot targets a weapons factory, despite foreseeing that the explosions will kill some patients at a nearby hospital.

Both pilots kill innocent people, but the first pilot intends their deaths as a means of weakening the enemy’s morale, while the second pilot knowingly causes or foresees their deaths as a side effect of destroying the weapons factory. According to PDE, the agent’s intention is relevant for judging actions in two ways. First, it is relevant for judgments about how to act (e.g., I will not drop the bombs because I would intend to kill innocent people or The pilot should not drop the bombs because the pilot would intend to kill innocent people). Second, the agent’s intention is relevant for evaluating actions that have occurred (e.g., The action was wrong because the pilot intended to kill innocent people).

A writer can say little about PDE without stepping into a briar patch of thorny issues. My definition and illustration of PDE already raise some disputed questions. Is PDE one principle or a set of principles? What if a bomber pilot needs only for innocent people to look dead, not to be dead? What about empirical studies in neuroscience and moral psychology that seem to debunk PDE? What if a bomber pilot targets a large man who is stuck in the mouth of a cave where five people who need organ transplants are trying to escape from a runaway trolley? The last question is a parody, but not much of a stretch from cases that philosophers actually use.¹ A writer can discuss only one issue at a time, so I will set aside these questions and many other objections and alleged counterexamples for now, or else my defense of PDE never will begin. For example, I consider the famous trolley problem—which is the challenge of explaining common intuitions about cases in which agents save several lives from runaway trolleys at the expense of one—only after I develop my version of PDE. My delay in considering the trolley problem is intended, not merely foreseen. When I discuss the trolley problem in chapter 4, I explain why confusion results when philosophers assume that PDE is supposed to explain common intuitions about trolley and transplant cases.

To defend PDE, some philosophers rely on intuitions about cases, but I use cases only to illustrate PDE. My main theses are these: (1) the difference between intended effects and foreseen side effects is relevant for judging actions because what people intend forms their character differently from what they knowingly cause, and (2) to define intended effects, someone must consider the action from the agent’s perspective, not the perspective of an observer or victim of the action.

Like the bombing cases presented above, many examples of PDE also involve causing death and serious bodily harm:

• Torture is wrong, but causing pain as a foreseen side effect can be right. For example, interrogators may not electrocute a prisoner of war to extract information about a hidden weapons depot. Pilots may, however, bomb power lines to a weapons depot even if they foresee that the bombing will knock down power lines and electrocute an enemy soldier.

• Intending to kill a patient is wrong, but rationing medical care can be right. For example, physicians may not withhold drugs from one patient so that the patient’s death will make organs available for five life-saving transplants. Physicians may, however, ration a scarce drug so that they can save five patients instead of only one. ²

• Suicide is wrong, but causing one’s own death as a side effect can be right. For example, a mother may not kill herself so that her son can collect on a life insurance policy in order to afford medicine that he needs to survive. The mother may, however, run into a burning building to save her son even if she foresees that she will die from smoke inhalation a few hours later.

Some critics challenge common moral intuitions about these cases, and other critics try to explain these intuitions without PDE. For now, I use the cases only to illustrate the distinction between intended effects and foreseen side effects and to show why many people believe that this distinction is relevant for judging actions.

That these cases are interesting is good enough for me, but some critics with more refined tastes urge other philosophers to spend their time on more important topics. I respond to these highbrow critics when I discuss the trolley problem in chapter 4. For now, I note that the link between these cases and real choices is indirect, like the link between an athlete’s weight training and a game. Basketball players lift weights to build muscles, not to be ready in case they need to lie down and lift other players off their chests. The point of contrived cases is to analyze moral rules and definitions of terms, not to be ready when a trolley comes speeding down a hill.

Dramatic cases of murder and torture are not the only illustrations of PDE. The distinction between intention and foresight is also relevant in ordinary cases:

• Intending to embarrass an opponent is wrong, or least an example of bad sportsmanship, but knowingly causing an opponent’s embarrassment is not. For example, coaches may not run up the score so that an inferior rival will feel embarrassed. Coaches may, however, tell their team to play hard even if they know that doing so will cause a lopsided score that will embarrass their opponent.

• Lying is wrong, but knowingly causing a false belief as a side effect can be right. For example, I may not tell my boss that I was stuck in traffic when I really overslept. I may, however, apologize for being late and omit the reason why even if I foresee that my boss will assume that I was stuck in traffic.

• Plagiarizing is wrong, but knowingly causing a misattribution of ideas can be right. For example, I may not use Plato’s allegory of the cave without citing Plato so that my students will attribute the allegory to me and overestimate my brilliance. I may, however, use the allegory of the cave in a conference presentation without citing Plato so that I do not insult my audience’s intelligence, even if I foresee that some students in the audience will attribute the allegory to me and overestimate my brilliance.

• Inciting envy is wrong, but knowingly causing envy can be right. For example, a child may not choose a toy in order to incite a sibling’s envy. A child may, however, choose a toy to have fun despite foreseeing that a sibling will feel envy.

PDE seems to explain why the two actions in each pair elicit different moral intuitions even though they have similar effects. In each pair, the first agent intends a bad effect (i.e., embarrassment, a false belief, a misattribution of ideas, or envy) as a means, while the second agent knowingly causes or foresees the same effect without intending it.

Why would a person act wrongly by intending an effect but not by knowingly causing the same effect? In chapter 1 I answer this question by explaining the rational basis of PDE. Agents can act wrongly by corrupting their own character, not only by mistreating other people. Agents have a closer relation to what they intend than to what they cause as a side effect, so intending an effect can corrupt the agent’s character even if knowingly causing the same effect does not. For example, intending to embarrass an opponent can block someone from respecting opponents, even if causing embarrassment as a side effect does not. After presenting this agent-based justification of PDE, I argue that proponents of PDE cannot overcome objections if they turn away from the agent’s character and appeal to intuitions, exceptionless moral rules, or phenomenological descriptions of intending evil.

How can people draw the line between intended effects and foreseen side effects? In chapter 2 I answer this question by considering actions from the agent’s perspective. I argue that an agent intends an effect if and only if it is part of a true and complete answer to the question What is the agent’s goal, and what are all the steps in the agent’s plan to achieve that goal? Instead of defining intended effects from the agent’s perspective, some proponents of PDE set aside the agent’s perspective and define intended effects by analyzing causal relations outside the agent. For example, some proponents of PDE claim that agents intend all the effects that closely or directly follow from their actions. These definitions inevitably entail false conclusions because they classify some effects as intended even though the agent regards them as irrelevant or even when the agent works against them—as when a sniper uses a flash suppressor and muzzle to reduce the flash and noise that follow a gunshot more closely or directly than the bullet’s hitting the sniper’s target.³ More importantly, these definitions undermine the basis of PDE. When proponents of PDE set aside the agent’s perspective to define intended effects, they cut the link between intended effects and the agent’s character, leaving them unable to explain why intentions are relevant in the first place. Unlike definitions that refer to closeness or directness, my agent-based definition of intended effects fits the agent-based justification of PDE that I present in chapter 1. One objection to PDE in general is that it is merely a tool for rationalizing moral judgments, because people classify a bad effect as an intended effect when they disapprove of the action and as a side effect when they approve of the action. To support this objection, some critics cite empirical studies in which subjects’ moral judgments seem to influence their judgments about what people intend. These studies have flaws, but even if moral judgments cloud some people’s judgments about intentions, it does not follow that all people’s judgments about intentions merely rationalize their moral judgments.

Do people accept PDE because they confuse judgments of actions and judgments of agents? This challenge strikes at the heart of PDE because it seems to explain both PDE’s appeal and its failure. According to the objection, PDE is appealing because intentions are relevant for some moral judgments, but PDE is false because intentions are relevant only for judging agents, not for judging actions. In chapter 3 I argue that this objection arises from an unnecessarily narrow view of morality. The objection seems plausible because the agent’s intention is irrelevant for judgments about whether the action treats someone unjustly, but the objection is mistaken because treating someone unjustly is not the only way to act wrongly. The objection also seems plausible because many philosophers, including proponents of PDE, speak about actions as if they existed apart from intentions, as a hammer sits on a shelf apart from a user’s intention. Anyone who treats a bent nail as evidence of a bad hammer confuses a judgment of the hammer with a judgment of the user, so treating actions like tools on a shelf exposes PDE to the objection about confusing different types of moral judgments.

What does my version of PDE say about the famous trolley cases? Very little, but I argue in chapter 4 that these cases are widely misunderstood. Many critics assume that intuitions about trolley cases are the only basis of PDE, but PDE predates trolley cases by centuries, and the trolley cases come from critics of PDE, who use them to analyze other moral principles. When philosophers expect PDE to explain intuitions about all the trolley cases, they end up defining intended effects too broadly. Like the proverbial carpenter who has only a hammer and sees every problem as a nail, philosophers who expect PDE to solve the trolley problem end up seeing too many actions as examples of intending evil. When critics assume that PDE depends on intuitions about trolley cases, they end up rejecting PDE too quickly. Some of these critics claim that recent studies in moral psychology and neuroscience debunk PDE by showing that it is the product of unreliable emotional processes rather than more reliable cognitive processes. I challenge these critics’ assumptions, especially the assumption that PDE depends on intuitions about trolley cases and other bizarre scenarios.

What does my version of PDE say about the famous case of obstetric craniotomies and other hard cases in war and medicine? Caesarean sections have replaced craniotomies as remedies for obstructed labor, but the principles used to analyze the craniotomy case are still relevant. According to a traditional view of PDE, performing an obstetric craniotomy is wrong because the surgeon directly kills the fetus as a means of saving the mother. The traditional view has been