Doctrine of Double Effect, The: Philosophers Debate a Controversial Moral Principle [1 ed.] 0268008973, 9780268008970

Citation preview

The Doctrine of Double Effect

The Doctrine of o ble Effect OU e PhilosophersDebate a Controversial Moral Principle Edited by

P. A. WOODWARD

University of Notre Dame Press Notre Dame, Indiana

University of Notre Dame Press Notre Dame, Indiana  All Rights Reserved www.undpress.nd.edu

Copyright ©  by University of Notre Dame Published in the United States of America

This ebook has been made possible in part by a major grant from the National Endowment for the Humanities: Exploring the human endeavor. Any views, findings, conclusions, or recommendations expressed in this book do not necessarily represent those of the National Endowment for the Humanities.

Reprinted in 

Library of Congress Cataloging-in-Publication Data The doctrine of double effect : philosophers debate a controversial moral principle / edited by P. A. Woodward. p. c.m. Includes bibliographical references and index. isbn --- (cloth : alk. paper) isbn : ---- (pbk. : alk. paper) isbn : --- (pbk. : alk. paper) . Double effect (Ethics) I. Woodward, P. A. (Paul A.) BJ.D D  —dc -

∞ This book is printed on acid-free paper.

Contents

Introduction 1 P. A. Woodward PART

I

THE

DOCTRINE

ONE Toward Understanding the Principle of Double Effect Joseph M. Boyle, Jr. PART

Two

II

IN DEFENSE

OF THE

7

DOCTRINE

Actions, Intentions, and Consequences: The Doctrine of Double Effect 23 Warren S. Quinn

THREE Agent-Relative Morality Thomas Nagel

41

FOUR Medalist's Address: Action, Intention, and 'Double Effect' 50 G. E. M. Anscombe FIVE Morality, Action, and Outcome Philippa Foot PART

III

IN OPPOSITION

67

TO THE

DOCTRINE

SIX Foreseen Side Effects versus Intended Consequences Jonathan Bennett s Ev EN The Doctrine of Double Effect: Problems oflnterpretation 119 Nancy Davis

85

vi

EIGHT

NINE

The Problem of Abortion and the Doctrine of the Double Effect 143 Philippa Foot Four Versions of Double Effect Donald B. Marquis PART

TEN

ELEVEN

TW EL

vE

Contents

IV

156

DISCUSSION

Quinn on Double Effect: The Problem of"Closeness" John Martin Fischer, Mark Ravizza, and David Copp The Importance of the Proportionality Condition to the Doctrine of Double Effect: A Response to Fischer, Ravizza, and Copp 211 P. A. Woodward Affirmative Action and the Doctrine of Double Effect William Cooney

TH IR TEEN

The Doctrine of Double Effect and Affirmative Action Jeff Jordan

F o u RTE EN

Affirmative Action Revisited: A Response to Professor Jordan 239 William Cooney PART

V

228

234

APPLICATIONS

FIFTEEN

War and Murder 247 G. E. M. Anscombe

SIXTEEN

Double Effect 261 Michael Walzer

SEVENTEEN

189

Suicide and Self-Sacrifice Robert M. Martin

270

EIGHTEEN

Nuclear Power, Challenging Non-Utilitarian Ethics Stanislaus J. Dundon

NINETEEN

Morphine Use for Terminal Cancer Patients: An Application of the Principle of Double Effect Greg Beabout Index of Names Index of Cases

313 317

298

289

The Doctrine of Double Effect

Introduction

Terrence is a bomber pilot serving his country in what we will assume is a just war. Terrence realizes that his enemy's ability to continue to prosecute the war depends upon, among other things, the willingness of the enemy civilian population to continue to make the sacrifices necessary to the prosecution of the war. Terrence also believes that that willingness is fragile and can be shattered if enough civilians are killed. Terrence, therefore, sets out on a bombing mission not only intent on destroying a munitions factory that is supplying his enemy with the weapons necessary to continue fighting the war, but also intent on killing a large number of civilians. We may be generous to Terrence and assume that it is reasonable to believe, as he does believe, that his successful mission will shorten the war to the degree that fewer lives will be lost should he fly the mission than should he refuse. Samuel is a bomber pilot fighting in the same just war. Samuel also believes that civilian deaths will undermine his enemy's ability to continue the war. Samuel, however, in setting out on the same mission as Terrence, does not set out to kill civilians. Samuel sets out intent only on destroying the munitions factory. Samuel realizes that a large number of civilians will be killed should his mission be successful; but those civilian deaths are not his aim. Similarly, we may assume that Samuel's success will, in the long run, save more lives than it will cost. As philosophers and ethicists we might wonder whether there is a moral difference between what Terrence does and what Samuel does. Terrence, after all, sets out to kill civilians whereas Samuel does not; but Samuel foresees that he will kill as many civilians as Terrence kills. Even so, what Terrence does looks like murder, whereas what Samuel does

2

•

Introduction

looks like a legitimate act of war. (We assume that being an act of war does not, in itself, make what either Terrence or Samuel do immoral.) One way a philosopher may arrive at an answer to this question is to identify some feature that one action has but which the other action lacks. Then the philosopher can look for a moral principle that discriminates in favor of actions that have (or lack) that feature and discriminates against actions that lack (or have) that feature. If such a principle can be identified and rationally defended, then, to the degree that one action exhibits the feature and the other does not, the two actions differ in moral status, assuming that all other morally relevant features are shared equally by the two actions. One principle which has been employed to distinguish actions like Terrence's from actions like Samuel's is the Doctrine of Double Effect (the DDE). The DDE exploits the distinction between intentional production of evil (as exemplified by actions such as Terrence's) and foreseen but unintentional production of evil (as exemplified by actions such as Samuel's). There are many versions of the DDE, but perhaps most dearly it is interpreted as limiting the application of certain prohibitions. According to the DDE, an agent may not run afoul of, for example, the prohibition on killing civilians, if in acting the agent kills civilians unintentionally, even when those killings are foreseen. (I say the agent "may" not run afoul of such a prohibition because, in typical statements of the DDE, there are other conditions that must be satisfied in order for such actions to be licit.) It is important to notice that, if it is a true principle, the DDE is not limited in its application to situations such as the one in which Samuel finds himself. All those who accept a moral theory that includes multiple precepts may find themselves in a situation in which they can avoid violating one precept or they can avoid violating another precept, but they cannot avoid violating some precept. Such situations may be called "moral dilemmas." Samuel is apparently faced with such a moral dilemma. His participation in the bombing mission seems to be prohibited by the prohibition on killing innocent civilians, but his refusal to participate in the mission seems to be prohibited either by the requirement that he obey his orders, or by his obligation to strive to win the war with as few casualties as possible. The DDE is a principle of moral theory that lessens the occurrence of moral dilemmas. It does so by limiting the scope of moral precepts.

Introduction

3

Samuel's action of participating in the bombing mission is not prohibited by the prohibition on killing innocent civilians because, if the DDE is a true principle, that prohibition does not apply to unintentional killings of civilians, if certain other conditions are also satisfied. Thus, the DDE may find application in any situation in which a person encounters conflicting moral precepts. The DDE is not the only moral principle that, if true, will help to resolve moral dilemmas, nor will it resolve all such dilemmas. But, since a moral theory that results in fewer dilemmas is preferable to a theory that results in more dilemmas, a moral theorist may find the DDE attractive. The DDE has a long and rich tradition (for references to the discussion of the history of the DDE, see note 1 in Joseph M. Boyle's essay), and it has given rise to an ongoing debate within the philosophical community. The present volume presents a sample of this debate. The first section consists of just one essay, Boyle's. Although not specifically an endorsement or criticism of the DDE, Boyle's essay is an attempt at understanding the doctrine. Boyle discusses many of the issues that are raised by a careful study of the doctrine: issues, for example, such as those involved in action theory, including the definition and individuation of actions; the distinction between justifying and excusing principles; considerations of the objective rightness or wrongness of behavior as compared with considerations of the moral quality of the agent; and, of course, the formulation of and moral significance of the distinction between what is intended by the agent and what that agent merely foresees will follow from what he or she does. This latter concern raises questions about the nature of the voluntary. The second and third sections of this volume consist of essays in which various philosophers either defend the DDE or criticize it. These defenses and criticisms turn largely on the issues introduced in Boyle's article. Some philosophers in turn have discussed specific defenses or criticisms of the doctrine. The fourth section presents some of this discussion and further illustrates the continuing interest in the DDE. Finally, in the fifth section, entitled "Applications," a number of philosophers show how the doctrine can be employed in particular situations and how it can be misused in certain ways. This collection is not comprehensive. There are other important articles on the DDE and related topics in the philosophical literature,

4

Introduction

many of which are identified in the notes of the selections included here. I hope, however, that these selections offer a thorough introduction to the DDE and the important issues surrounding the DDE. Moreover, I hope that they will inspire the reader to think through these issues for himself or herself. I wish to express my gratitude to all of the authors who have given their permission to present their work here, and to the publishers who granted permission to republish articles that have appeared in print elsewhere. P. A. Woodward

PART

I

The Doctrine

ONE

Toward Understanding the Principle of Double Effect JOSEPH

M.

BOYLE, JR.

The Principle of Double Effect (hereafter PDE) has long been a mainstay of Catholic moral thinking. 1 In recent years, however, the use and discussion of this doctrine have not been limited to Catholics or to theologians. 2 The PDE, or propositions closely related to it, have come up for considerable discussion by English-speaking philosophers. 3 In spite of this discussion, however, the PDE remains something of a mystery. As I hope to show, its purpose, its essential claims, and its presuppositions are not adequately understood. This lack of understanding is due both to the difficulties and ambiguities in traditional formulations of the PDE and to the fact that its central conceptions are either foreign or contrary to much of contemporary ethics and action theory. 4 The purpose of this paper, therefore, is to state plainly the propositions involved in the PDE and at least some of the propositions concerning intention, action, and moral responsibility which must be true if some version of the PDE is to be defensible and morally relevant. I will not seek to defend or criticize these propositions here. My aim is to state them clearly and thus to make possible a more intelligent and more decisive discussion of the PDE.

Reprinted from Ethics90 (July 1980): 527-38, © 1980 by The University of Chicago, by permission of the University of Chicago Press. 7

8

JosephM. Boyle,Jr.

The classic modern formulation of the PDE is presented in J. P. Gury's widely used and often revised manual, Compendium theologiae moralis: "It is licit to posit a cause which is either good or indifferent from which there follows a twofold effect, one good, the other evil, if a proportionately grave reason is present, and if the end of the agent is honorable-that is, if he does not intend the evil effect."5 In a clarification of this statement Gury makes it clear that the PDE contains four conditions, all of which together are required for the type of act in question to be licit: (1) the agent's end must be morally acceptable (honestus), (2) the cause must be good or at least indifferent, (3) the good effect must be immediate, and (4) there must be a grave reason for positing the cause.6 The determination of what constitutes a grave reason is a matter of normative ethics which I will not consider here; moreover it is sufficiently clear that a grave reason is required to bring about an evil state of affairs-one which it would presumably not be licit to bring about except under these conditions. 7 The sense of 'immediate' in the third condition seems to be that the evil effect may not be a means to the good effect. This is suggested by Gury's remark that, if the good came about through the evil effect (mediante pravo effectu), good would be sought through evil, and this can never be moral. 8 Thus this condition allows that the evil effect should follow from the good effect, or that the evil effect and the good effect should follow independently from the cause, but not that the good effect should follow from the evil effect. The second condition-that it must be morally permissible to posit the cause-suggests that the cause which brings about both effects can be morally evaluated independently of either of its effects. Thus we may take the first condition to refer to a human undertaking, the executing of a choice, which can be the subject of moral evaluation independent of the good and evil effects which are brought about. The application of these four conditions to many cases has been widely controverted. But several clear examples show how it is intended to work. The killing of noncombatants in a justified military action is one example used by Gury: the action may be justified-even if it is foreseen that some noncombatants will certainly be killed-if killing them is not a means of achieving one's military objective, since in this case undertaking the action may be presumed to be in itself good, the bringing about of the deaths is not intended, and there is a grave reason

Toward Understandingthe Principleof Double Effect

9

for the action. Gury says that the death of the innocents follows per accidens, and is not intended but only permitted. The justification of killing in self-defense is another frequently used but more controversial application of the PDE. According to Saint Thomas, one intends only one's self-defense, the aggressor's death being outside the agent's intention-an effect of one's defensive act 9 and not a means to one's defense. Discussion of the application of the PDE to cases like this is beyond the scope of this paper. Such a discussion presupposes that the PDE is intelligible. A first step in understanding the PDE is the determination of what it is meant to do. Is it meant to show that certain acts which, except under these conditions, would be morally impermissible, are, if its conditions are met, morally justified?Or is the PDE meant to be a principle of excuse-that is, a principle which lessens the imputability of a bad act?10 The Catholic moralists who developed this doctrine are none too clear on this point. Gury, for example, in stating the PDE, formulates it as a set of conditions for the licitnessor permissibility of a certain action. But Gury in other places and other authors in this tradition write in such a way as to suggest that the PDE is a principle of excuse. They say, in effect, that the unintended consequences of one's acts are not imputable to the agent, or they suggest that the imputability is diminished or somehow indirect. This interpretation is reinforced by the fact that most of these authors deal with the PDE in their treatments of the various kinds of voluntary acts and because of the importance of intention in the PDE. I believe, however, that the PDE should be understood as a principle of justification,and that the tradition, even if it is unclear on this point, is most coherently understood in this way.tz There are two reasons for thinking this. First, the unintended evil effect, the bringing about of which is rendered licit by the PDE, is clearly imputable to the agent: he knowingly and willingly brings it about. The scholastics often say that it is permitted or consented to; one could more clearly say in modern English that it is accepted.But permitting, consenting to, and so on, are volitional acts, or at least volitional dispositions, even if they are not volitional in the paradigmatic sense of intentional actions. t 3 There might be a difference in the mode of responsibility one has for what one intends to bring about from that which one has for what one

10

JosephM. Boyle,Jr.

does not intend but willingly accepts. In both cases there might be degrees of imputability, but there is no necessary difference in degree between the two types of willing. What one intends and what one permits are both voluntarily brought about, and thus both are imputable. If what one permits were not voluntarily brought about, and thus ascribable to the agent, the fourth condition of the PDE would have no use. This condition states a requirement for the permissibility of permitting certain effects which one may not intend to bring about. My second reason is a historical consideration. Aquinas's view of human action and intention forms the background for the distinctions upon which the PDE is based. The moralists who articulate and use the PDE often refer explicitly to his analysis of killing in self-defense, and some of the ancestors of the PDE are the sixteenth- and seventeenthcentury commentaries on Aquinas's theory of action. 14 This is not to say, of course, that modern proponents of the PDE either consistently or clearheadedly develop his action theory. According to Aquinas, an act is, morally speaking, the kind of act it is in virtue of what is intended. 15In other words, Aquinas believes that actions are at least in part defined by the intention with which the agent performs them; intention is an act of volition, and it is necessarily the case that one intend whatever functions as an end or goal in one's actions;16 one must intend not only the more distant and ulterior goals but also the immediate aim one has in undertaking an action. Thus one must intend what Aquinas calls the formal objectof one's act. In more contemporary language, an action is undertaking to bring about a certain state of affairs. This state of affairs is intrinsic or essential to the act and is necessarily intended. 18 If this understanding of action is applied to the conditions of the PDE, it becomes clear that the first condition is invoked to determine the moral kind of the act which is at issue. In the first condition it is required that the agent's end be morally acceptable. The end referred to here might be either the immediate object of the act-the state of affairs the bringing about of which defines the act-or it might be a more remote end. In either case, it is among the things one must intend in acting. 19Moreover, as Gury explicitly states, this condition is meant to exclude the agent's intending the evil effect. If the good effect is intended and the evil effect is not intended, the act will be, morally speaking, a good act. It will be specified by the good effect as a morally good act.

Toward Understanding the Principle of Double Effect

11

The act-defining character of intention is also relevant for understanding the third condition. If the evil effect is brought about as a means to the good effect, then the evil effect must be intended, and the bringing about of the instrumental state of affairs is morally impermissible. The bad effect is intended if it is chosen as a means because it becomes something which the agent is committed to realizing. The bringing about of this instrumental state of affairs is a morally impermissible act because this state of affairs-the bad effect-determines the moral character of the undertaking. Thus, the third condition is implied by the first condition and the definition of a "means." This condition does not assume, therefore, that the causal sequence of the effects is itself morally significant, as is often supposed. 20 In most cases in which one chooses a certain means, it is because of one's beliefs about the causal consequences of what one chooses as a means that one chooses it, but what is morally significant about such a choice is the intention and action involved in executing it.21 The first condition and the third, therefore, are attempts to determine exactly what one's act is in situations where the causal initiative involved in one's undertaking brings about both good and evil states of affairs. Together with the fourth condition, these conditions constitute a basis for the justificationof actions having evil effects:such actions are not themselves evil in kind and there is grave reason for performing them. The foregoing attempt to explain the PDE as a principle of justification by reference to Aquinas's theory of action can also accommodate the second condition of the PDE-namely, that initiating the causal sequence must be morally permissible independently of its good and bad effects. If this "positing of the cause" can be determined to be impermissible independently of the evil effect at issue, or of some other evil end, then one is not justified in initiating the causal sequence. To the extent that this condition suggests that it is possible to determine the morality of initiating the causal sequence without reference to any effect, then it is inconsistent with Aquinas's view of action. But to that extent it is also confused: cause and effect are correlative, and "positing a cause" can be immoral only because it brings about some effect which it is impermissible to bring about. This way of understanding the second condition has the effect of rendering it, strictly speaking, superfluous. If an act is not permissible, then the doing of that act would involve the intention of what is evil, and this is prohibited by the first condition.

12

Joseph M. Boyle, Jr.

By this point my attempt to treat the PDE as a principle of justification may appear to be radically revisionist. The second condition is rendered superfluous, and the third is implied by the first together with the definition of a means. The redundancy of the traditional formulations of the PDE, however, is understandable if one considers the casuistical purpose for which the PDE was meant to be used and the distinctions the scholastics were accustomed to make. Following Aquinas, the scholastics held that the moral character of an act was determined by the end, the object of the act, and the circumstances in which the act was performed. 22 Thus, the first condition could be taken to exclude acts done for an immoral purpose; the second to exclude acts which are intrinsically immoral-that is, whose object it is impermissible to bring about; and the third to exclude immoral means to good ends-an exclusion which is not always covered by the second condition. More concretely, the first condition would exclude, for example, giving someone money in order to get him drunk; the second, making oneself drunk in order to give someone a lesson in the value of sobriety; and the third, giving someone money in order to get him drunk in order to teach him a lesson. In the first case, the "positing of the cause" is permissible and there is no evil means to a good end, but the aim is evil. In the second case, the aim is good but the cause posited-getting oneself drunk-is evil. In the third case, the final end of the action is good, as is the positing of the cause, but an intermediate means is evilnamely, getting the person drunk. The PDE, then, can be understood as a principle of justification. In its briefest form it can be stated in the following way: it is morally permissible to undertake an action when one knows that the undertaking will bring about at least one state of affairs such that, if this state of affairs were intrinsic to the action undertaken, the action would be rendered morally impermissible, if and only if (1) the state of affairs is not intrinsic to the action undertaken-that is, it is not intended-and (2) there is a serious reason for undertaking the action. This formulation, however, gives rise to further questions. In particular, it gives rise to questions about the distinctions made by the PDE and about the moral significance of these distinctions. First, the PDE as explained here forces one to draw in an odd way the distinction between the objective considerations about the rightness

Toward Understanding the Principle of Double Effect

13

or wrongness of behavior and the subjective considerations relevant to evaluating the moral quality of the agent. One might expect that the question of rightness or wrongness is a question about behavior and its consequences, whereas questions about the goodness or badness of the agent would consider such things as the agent's motives and intentions as well as factors related to the imputability of his act. The PDE as explained here, however, requires one to regard the question of rightness or wrongness not simply as a question about behavior but also as a question about actions which contain intentions as an essential part. Second, the PDE requires the moral significance of a distinction which, according to many, cannot even be drawn. Many modern philosophers regard the difference between what is intended and what is foreseen and "permitted," but not intended, as a merely verbal difference.23 Moreover, it is certainly not clear what significance this difference has-supposing it can be drawn-for purposes of the moral evaluation of acts. Even if one admits that acts are defined by the agent's intentions and distinguished from foreseen consequences which are merely "permitted," one might wonder why it is that an agent's acts have a moral significance which is different from-and more decisive than-the moral significance of the foreseen consequences of what he does. It is important to recognize that the PDE does not require that the foreseen consequences of acts be in no way relevant to determining the rightness or wrongness of the agent's concrete behavior; they are relevant, but only in a subsidiary way. Thus, if the action is itself morally permissible, and if there is a serious reason for undertaking it, then it may be done morally no matter what the foreseen consequences may be. The answers to these questions are based on a factor which has been completely overlooked in discussions of the PDE24 -namely, the view of voluntariness and responsibility which the PDE presupposes. An account of this view goes something like the following:25 There are many different types of voluntary acts; in fact, the notion of "voluntariness" is an equivocal one. There is, however, an order in these senses such that one could say with the medievals that voluntariness is an "analogous" notion, or, in more current language, that there are family resemblances between the various senses of"voluntary." In this ordered set of meanings of "voluntary" one is paradigmatic-namely, that sense in which it is said of an act which is the execution of deliberate, free choice. ln this case, the elements which all

14

Joseph AL Boyle, Jr.

voluntary acts have in one way or another are preeminently present; 26 no human act is so clearly or properly self-initiated as is an action which executes a free choice. A free choice is a choice in which all the causal factors other than the agent's choosing are not sufficient to bring about the choice. 27 Moreover, in a deliberate choice one knows what one is doing; one considers and reflects upon the options and their various attractions. By contrast, actions which are voluntary in any other sense of the word involve some diminishing of one or both of these components. Nondeliberate acts-for example, actions done out of passion or fear or under duress-will lack the cognitional component involved in deliberate choices. Similarly, habitual acts, automatic reactions, and the actions of children and of those who are not compos mentis are in various ways not self-initiated as free choices are; they arc self-initiating but the self is not in control. Human behavior which carries out deliberate free choices, therefore, is voluntary in the strongest sense of the term. 28 Thus it is not surprising that such behavior is often regarded as the primary subject of moral evaluation. Such behavior can be understood in the following way: in deliberating one considers various incompatible practical proposalsone's bringing about state of affairs P or one's bringing about state of affairs Q (where P includes non-Q and Q non-P). Choice is one's selection of one's bringing about P rather than Q or Q rather than P. The behavior consequent upon a choice is the undertaking to bring about P or to bring about Q. In other words, the agent acts with the intention of bringing about P or with the intention of bringing about Q. Thus, one who, after considering several practical proposals and freely selecting one of them, executes his choice by undertaking to bring about the state of affairs proposed is performing a paradigmatically voluntary act. This performance is, according to this view of voluntariness, a doing of the type that is either right or wrong. But how is this performance to be understood? Should the performance be understood in a formal way-as simply bringing about the state of affairs which it was undertaken to bring about? Or should this performance be taken as a kind of individual, a concrete event with many causal connections which bring about many states of affairs? It seems to me that the controverted ontological question of whether

Toward Understanding the Principle of Double Effect

15

actions are individuals of some sort or the bringing about of certain abstract states of affairs can be set aside for present purposes. If an action is a concretum, then it will be truly described by a number of propositions. The question relevant here, should such an ontological view be accepted, would be under which of the descriptions the act should be morally evaluated in the first instance. The proponent of the PDE would hold that the performance should be understood formally, since the performance is-insofar as it is the type of voluntary act defined above-the execution of a choice, and the choice is the commitment to bring about a certain definite state of affairs. In other words, while an agent's performance involves a causal initiative that can have many frneseeable consequences, it is not the performance as a willing causal initiative, together with some set of the effects of that initiative, which is primarily voluntary or the primary subject of moral evaluation. 29 A causal initiative-or at least the willful refraining from such an initiative-is part of the notion of a voluntary act in this paradigmatic sense. But it is part of the act only insofar as it executes the choice and not as a concrete event with an indefinite set of effects. This is not to say, of course, that the bringing about of the foreseen effects of one's performances is not voluntary. It is, but not in the way in which the executions of choices-regarded just as such-are voluntary. The foreseen consequences of one's potential performances are no doubt a part of what is considered in the deliberation leading to choices. But they are included in this deliberative process in a unique way. Frequently, they do not appear to be of any practical consequence to the person deliberating, and sometimes they are seen to interfere with the achievement of either the state of affairs one is considering bringing about or some further goal with respect to which this state of affairs is taken to be instrumental. In other words, the foreseen consequences of one's bringing about an intended state of affairs are often considered in deliberating, but not as reasons for the action-rather, they are sometimes conditions in spite of which one acts. It is not for the sake of such conditions that one selects an option; it is not these effects to which one is committed in acting. If this is correct, the agent in acting has a fundamentally different attitude toward what he intends and toward what he foresees and consents to or accepts but does not intend. The agent who acts with the

16

JosephM. Boyle, Jr.

intention of bringing about a certain state of affairs makes that state of affairs his goal; he sees it as worthwhile or as instrumentally valuable and commits himself to bringing it about. Clearly, what is thus regarded as valuable is not all that one foresees will come about by his initiative. For example, pain involved in undergoing surgery is not regarded as valuable by the patient. He would avoid it if he could. The agent's attitude toward such consequences as this is entirely different from his attitude toward what he intends. These consequences need not be seen as good or as desirable; there is no commitment to bring them about; in many cases the agent would avoid them if he could. They are not a part of what one chooses to bring about. This account of voluntary action provides the basis for answering the questions listed above. Given this account of voluntary action, the distinction between factors relevant to determining the rightness of actions and factors relevant to assessing the moral qualities of agents are properly drawn where the PDE requires. The primary object of the moral evaluation of the act is the bringing about of the state of affairs which the agent is committed to realizing. This action includes and is defined by the agent's intention. Moreover, this way of drawing the distinction between factors relevant to determining the imputability of the act and factors relevant to characterizing the act to be evaluated allows for the obvious excusing factors. It allows for questions about whether the agent knows the moral character of his undertaking and questions about whether the agent's choice is free. It does imply, however, that an action will be regarded as if it were a voluntary act even if in fact it is not imputable, because it is only as the execution of a choice that behavior can be characterized as morally significant. Thus, for example, if an omission or piece of habitual behavior is regarded as morally wrong, it is because that omission or piece of behavior would be wrong if it embodied the execution of a deliberate choice. The second question is also resolved by this account of voluntary action; the distinction between what is intended and what is consented to is intelligible in terms of this account. A morally significant act is the execution of a choice. What one chooses is not an indeterminate set of foreseeable results of one's performance but the bringing about of a definite state of affairs regarded as worthwhile or valuable. Thus there is a basis for the distinction between what is intended and what is not intended but consented to, or between actions defined by the state of

Toward Understanding the Principle of Double Effect

17

affairs intended and the consequences of actions defined by what is not intended but consented to. Moreover, this account of voluntary action shows why the distinction between what is intended and what is foreseen and accepted but not intended is taken to be morally relevant. Since it is one's choices and their execution which are primarily voluntary, it is these which are the primary object of moral evaluation. The states of affairs intended in such acts are what the agent, as it were, sets his or her heart on. It is the commitment to these states of affairs which is the basis upon which a person forms his character and makes himself a certain kind of person. Moreover, if moral demands are regarded as unconditional demands upon one's free choice, 30 then it is by choosing and committing oneself to some such states of affairs that one acts morally or immorally. In other words, if moral demands are demands upon one's free choice and if free choice is the adopting of a proposal that one bring about a certain state of affairs regarded as valuable or worthwhile, then what moral normativeness primarily bears upon-as far as behavior goes-is human action in the sense defined here. This account of the view of voluntary action presupposed by the PDE also throws light on the connection between the PDE and the socalled absolutism of traditional Catholic ethics. By "absolutism" I mean the view that there are exceptionless moral proscriptions. First of all, absolutism is not required by the PDE. The PDE does not explain how one is to come to the normative judgments presupposed by the first three conditions and enjoined by the fourth. The first three conditions, if met, are sufficient to characterize the act in question as a good type of act. But they do not explain how acts of that type are judged to be morally good. Likewise, these conditions require that the "evil effect" would be a bad kind of act if any of the first three is not met, and that there must be a grave reason for accepting such an act, but they do not specify what makes an immoral act immoral. This judgment might well be made on the basis of the view that such an act is proscribed by a general exceptionless proscription, but it need not be so based. Any normative theory which allows that there are kinds of acts which are good and bad could be consistent with the PDE and could make use of the PDE. Furthermore, the PDE is not required by all forms of absolutism. An absolutist system such as Kant's, for example, does not make use of

18

JosephM. Boyle, Jr.

the PDE, and it is possible to use principles other than the PDE to resolve cases of moral perplexity. 31 Moreover, an absolutist might limit his exceptionless proscriptions so that cases dealt with by the PDE are either clearly prohibited or clearly allowed by the relevant rule. But there is a vital connection between the PDE and the form of "absolutism" with which it is usually associated. This form of absolutism does not depend on a set of intuited or commanded absolutes, or a set of absolutes based on generalizations from particular cases, but on moral rules which direct one to respect basic human goods or values. Specifically moral precepts mandate that these basic goods be promoted whenever possible and that they not be attacked or acted against. This leads to moral norms-including absolute proscriptions-of a very general sort. For example, human life is taken to be a basic human good. The proscription of killing, therefore, will be quite general. 33 Moreover, these goods are the basis for deliberation and choice. They are pursued-in the morally relevant way-by voluntary acts and especially by acts executing free choices. The demand that we respect all the human goods-and especially that we not act against them-is a demand on our free choice. This normative view does require the PDE, but not simply because it is absolutist. This connection is not a matter of straightforward implication. The normative theory in question does imply a part of the theory of human agency presupposed by the PDE. Thus they both imply the same thing. The necessity of the connection between this normative theory and the PDE is established by the facts that (1) the actual world is such that the realization of most human choices is by way of causal initiatives which bring about many states of affairs other than that state of affairs which the agent intends to realize through his performance, and (2) some of these states of affairs are contrary to one or more basic goods. These facts require that one committed to a normative theory demanding respect for a set of basic goods hold the PDE. Otherwise respecting the goods becomes an impossibility, since any performance can-and many performances do-bring about what is contrary to one or more basic goods. To sum up: the PDE is a coherent doctrine of justification. But it continues to be misunderstood because the theory of agency which it presupposes is ignored. If this view of human agency is false, then the PDE must be abandoned; but if this theory of agency is true, and if

Toward Understanding the Principle of Double Effect

19

the normative theory which makes use of the PDE can be defended, then the PDE is a long way toward vindication.

NOTES

The work for this paper was begun while on a College Teachers in Residence Fellowship of the National Endowment for the Humanities. I have profited from discussing the issues taken up here with R. M. Chisholm and Alan Donagan. I thank Germain Grisez for help on several drafts of this paper. 1. See Joseph Mangan, S.J., "An Historical Analysis of the Principle of Double Effect," TheologicalStudies 10 (1949): 41-61; J. Ghoos, ''L'Acte a double effet: Etude de theologie positive," EphemeridestheologicaeLouvaniensis27 (1951):30-52; and F. J. Connell, in The New CatholicEncyclopedia,s.v. "Principle of Double Effect." 2. E.g., Paul Ramsey, the eminent Methodist theologian, makes extensive use of the PDE; see his War and the Christian Conscience (Durham, N.C.: Duke University Press, 1961). For a nontheologian, see G. E. M. Anscombe, "War and Murder," in War and Morality, ed. R.Wasserstrom (Belmont, Calif.: Wadsworth Publishing Co., 1970), pp. 50-51. [Included here, chapter 15.] 3. See, e.g., Philippa Foot, "Abortion and the Doctrine of Double Effect," in Moral Problems, ed. ). Rachels (New York: Harper & Row, 1971), pp. 28-41; Jonathan Bennett, "Whatever the Consequences," Analysis 26 (1966): 83-102 and the ensuing discussion; Alan Donagan, The Theory of Morality (Chicago: University of Chicago Press, 1977),pp. 122-27, 157-64. 4. See Germain Grisez, "Toward a Consistent Natural Law Ethics of Killing," American Journalof Jurisprudence15(1970): 73-79, for a critique of certain aspects of traditional formulations. 5. ). P. Gury, S.J. (revised by A. Ballerini, S.J.), Compendium theologiae moralis, 2d ed. (Rome and Turin, 1869), p. 7. The translation from the Latin is mine. Mangan, pp. 60-61, provides a translation from the fifth German edition of Gury's entire treatment of the PDE. 6. Gury, p. 8. 7. This condition can, but need not, be understood in a consequentialist way. It can be understood as requiring that relevant obligations other than those bearing on the directness or indirectness of the bringing about of the evil effect be considered. 8. Gury, p. 8. 9. S.T. Il-Il,64,7. 10. See). L. Austin, "A Plea for Excuses," in PhilosophicalPapers,ed.). 0. Urmson and G. J. Warnock, 2d ed. (London: Oxford University Press, 1970), pp. 175-77, for an explanation of the difference between justification and excuse. 11. See, e.g., Arthurus Vermeersch, S.J., Theologiae moralis: Principia, responsa, consilia (Rome: Gregorian University Press, 1922), 1:118.Vermeersch, one of the most important Catholic moralists in the early decades of this century, states the PDE as follows: "Effectus malus qui actionem sequi permittitur, non imputatur, si diversa est efficientia immediata, et permissio ratione proportionata gravi excusatur" (emphasis mine).

20

Joseph M. Boyle, Jr.

12. See William Conway, "The Act of Two Effects," Irish Theological Quarterly 18 (1951):127-29. 13. See R. G. Frey, "Some Aspects to the Doctrine of Double Effect," Canadian Journal of Philosophy 5 (1975):265. 14. See John of St. Thomas, Cursus theologicus, tome 6: De bonitate et malitia actuum humanorum, disputatio 11 (Paris, 1885);and Salmanticenses, Cursus theologicus, tome 7, tractatus 13,disputatio 10, dubium 6 (Paris, 1877). 15. S.T. II-II, 64,7: "Morales autem actus recipiunt speciem secundum id quod intenditur, ... " See also S.T. I-Il,72,1. 16. S.T. I-II,12,1; De Veritate 22,14. For a discussion of these and other relevant texts, see my "Aquinas on Praeter Intentionem," Thomist 42 (1978): 649-65. 17. S.T. I-II,20,4; I-II,72,3, ad 2; I-II,73,1. 18. See G. H. von Wright, Norm and Actions: A Logical Inquiry (New York: Humanities Press, 1963),pp. 39-41; The Varieties of Goodness (New York: Humanities Press, 1963), pp. 39-41, 123-25. 19. Gury, p. 7. 20. E.g., by Frey, pp. 261,280-81. 21. See Grisez, pp. 87-89, for a critique of traditional formulations of the PDE on this point. 22. S.T. I-II,18,2,3,4. 23. See Henry Sidgwick, Methods of Ethics, 7th ed. (New York: Dover Publications, 1966), p. 202; R. M. Chisholm, "The Structure of Intention," Journal of Philosophy 67 (1970): 636; for a response and reference to other literature, see Joseph M. Boyle, Jr., and Thomas D. Sullivan, "The Diffusiveness oflntention Principle: A Counter-Example," Philosophical Studies 31 (1977):357-60. 24. For example, by R.A. Duff, "Absolute Principles and Double Effect," Analysis 36 (1976): 68-80. 25. The following paragraphs are inspired by S.T. I-II,6-21. 26. The notion of"voluntary act" supposed is Aristotle's; see Nichomachean Ethics 3.no9b30-nnb3, and esp. 1nia22-23: "The voluntary would seem to be that of which the moving principle is in the agent himself, he being aware of the particular circumstances of the action." 27. Joseph M. Boyle, Jr., Germain Grisez, and Olaf Tollefsen, FreeChoice:A Self Referential Argument (Notre Dame, Ind.: University of Notre Dame Press, 1976), pp. 11-23. 28. See S.T. I-II,6,1: "Whence when a human being most fully [maxime] knows the end of his act and moves himself, then is his act most fully [maxime] voluntary." 29. See Donagan, pp. 37-52, 112-22,for a contrary view. 30. See Boyle et al., pp. 164-66, for an exposition of this view of moral norms. 31. See Donagan, pp. 1 i9-56. 32. See ibid., pp. 60-65; and Grisez.

PART

II

In Defense of the Doctrine

TWO

Actions, Intentions, and Consequences:The Doctrine of Double Effect WARREN

S.

QUINN

Situations in which good can be secured for some people only if others suffer harm are of great significance to moral theory. 1 Consequentialists typically hold that the right thing to do in such cases is to maximize overall welfare. But nonconsequentialists think that many other factors matter. Some, for example, think that in situations of conflict it is often more acceptable to let a certain harm befall someone than actively to bring the harm about. I believe that this view, which I call the Doctrine of Doing and Allowing, is correct, and I defend it elsewhere. 2 But there is a different and even better known anticonsequentialist principle in the Doctrine of Double Effect (for short, the DDE). 3 According to one of the common readings of this principle, the pursuit of a good tends to be less acceptable where a resulting harm is intended as a means than where it is merely foreseen. 4 It is this controversial idea that I wish to examine here. There are two major problems with the DDE. First, there is a difficulty in formulating it so that it succeeds in discriminating between cases that, intuitively speaking, should be distinguished. In particular, I

Reprinted from Philosophy and Public Affairs 18, no. 4 (1989): 334-51, © 1989 by Princeton University Press, by permission of Princeton University Press.

23

24

Warren S. Quinn

will need to find a formulation that escapes the disturbing objection that under a strict enough interpretation the doctrine fails to rule against many or most of the choices commonly taken to illustrate its negative force. Second, there is a question of rationale. What, apart from its agreeing with our particular intuitions, can be said in favor of the doctrine? Indeed, why should we accept the intuitions that support it? In answer, I shall suggest a rationale with clear Kantian echoes. I Like the Doctrine of Doing and Allowing, the DDE discriminates between two kinds of morally problematic agency. It discriminates against agency in which there is some kind of intending of an objectionable outcome as conducive to the agent's end, and it discriminates in favor of agency that involves only foreseeing, but not that kind of intending, of an objectionable outcome. That is, it favors and disfavors these forms of agency in allowing that, ceterisparibus, the pursuit of a great enough good might justify one but not the other. The doctrine is meant to capture certain kinds of fairly common moral intuitions about pairs of cases which have the same consequential profile-in which agents bring about the same good result at the same cost in lives lost and harm suffered-but in which the character of the intention differs in the indicated way. One such pair of contrasting cases is drawn from modem warfare: In the Case of the Strategic Bomber (SB), a pilot bombs an enemy factory in order to destroy its productive capacity. But in doing this he foresees that he will kill innocent civilians who live nearby. Many of us see this kind of military action as much easier to justify than that in the Case of the Terror Bomber (TB), who deliberately kills innocent civilians in order to demoralize the enemy. Another pair of cases involves medicine: In both there is a shortage of resources for the investigation and proper treatment of a new, life-threatening disease. In the first scenario doctors decide to cope by selectively treating only those who can be cured most easily, leaving the more stubborn cases untreated. Call this the Direction of Resources Case (DR). In the contrasting and intuitively more problematic example, doctors decide on a crash experimental program in which they deliberately leave the stubborn cases untreated in order. to learn more about the nature of the disease. By this

Actions, Intentions, and Consequences

25

strategy they reasonably expect to do as much long-term medical good as they would in DR. Call this the Guinea Pig Case (GP). In neither case do the nontreated know about or consent to the decision against treating them. Another pair of medical examples is found in most discussions of double effect. In the Craniotomy Case (CC) a woman will die unless the head of the fetus she is trying to deliver is crushed. But the fetus may be safely removed if the mother is allowed to die. In the Hysterectomy Case (HC), a pregnant mother's uterus is cancerous and must be removed if she is to be saved. This will, given the limits of available medical technology, kill the fetus. But if no operation is performed the mother will eventually die after giving birth to a healthy infant. Many people see less of a moral difference between these two cases than between the other pairs. This might be for a variety of reasons extraneous to the doctrine: because the fetus is not yet a person and therefore not yet within the moral framework, because the craniotomy is seen as a way of defending the mother against the fetus, because the fetus's position within the mother's body gives her special rights over it, and so on. But the relative weakness of the intuitive contrast here might also signal something important about the doctrine's central distinction. I shall say more about this later. But for the present it will be useful to include this pair of cases under the DDE, if only because it naturally illustrates the objection mentioned earlier. According to that objection, the doctor in CC does not intend, at least not strictly speaking, that the fetus actually die. 5 On the contrary, we would expect the doctor to be glad if, by some miracle, it survived unharmed. It is not death itself, or even harm itself, that is strictly intended, but rather an immediately physical effect on the fetus that will allow its removal. 6 That effect will of course be fatal to the fetus, but it is not intended as fatal. The intentions in CC are therefore really no different from those in HC. It might seem that this kind of point cannot be made about the bombing and nontreatment cases. In GP the doctors seem to need the disease to continue so that they can observe its effects. And in TB the pilot seems to need the deaths of the civilians to lower enemy morale. But Jonathan Bennett suggests a way of extending the objection to the bombing case. 7 The terror bomber does not, he argues, need the civilians actually to be dead. He only needs them to be as good as dead

26

Warren S. Quinn

and to seem dead until the war ends. If by some miracle they "came back to life" after the war was over, he would not object. And something similar might be said about the doctors in GP. While they need the disease to continue its course, they do not need the victims actually to be harmed by it. If by some miracle the victims developed special ways of withstanding the disease so that they remained comfortable and well-functioning despite its progress, the doctors would be glad.8 This line of objection clearly threatens to deprive the doctrine of most of its natural applications. One reply is to say that it surely matters how closethe connection is between that which is, strictly speaking, intended and the resulting foreseen harm. If the connection is close enough, then the doctrine should treat the harm as if it were strictly intended. 9 And, the reply might go on, the connection is close enough in the cases I have used to illustrate the doctrine's negative force. But what does this idea of closeness amount to? H. L. A. Hart suggests a possible answer by way of the example of someone violently striking a glass just in order to hear the sound of the initial impact. In such a case the further outcome, the shattering of the glass, is "so immediately and invariably" connected with the intended impact that the connection seems conceptual rather than contingent. rn The death of the fetus in CC is, arguably, connected with the intended impact on its skull in just this immediate and invariable way. And the deaths, or at least some harms, in TB and GP seem just as closely connected with what is strictly intended in those cases. But what of the contrasting cases? Since hysterectomies are rarely performed on pregnant women, they rarely result in the death of a fetus. So we might say that what is strictly intended in HC (that the uterus be removed) is not, in the relevant sense, closely connected with the fetus's death. And we might hope to find something similar to say in SB and DR. But in taking this way of presenting the contrasts, we would be making everything depend on which strictly intended outcomes of the various choices we fasten upon. This leads to a new problem. For certain things that the doctor in CC strictly intends for the fetus lack an invariably fatal upshot. Indeed, if craniotomies are ever performed on fetuses that are already dead, then a craniotomy is already such a thing. Even more obviously, the doctor in HC might strictly intend something that is invariably fatal to a fetus. Suppose, for example, that hysterectomies performed on patients who are in the early months of pregnancy are distinguished by

Actions, Intentions, and Consequences

27

the use of a special anesthetic that is safer for the patient and, in itself, harmless to the fetus. This peculiarity could hardly make the operation in HC more difficult to justify, but it would imply that the strictly intended medical means were immediately and invariably connected with the death of a fetus. 11 Perhaps similar things can be said about the other cases. A strategic bomber might have as his mission the bombing of automotive factories. This would not make him a terror bomber, for he would still not aim at civilian casualties. But, for obvious reasons, no automobile factories have ever existed completely apart from civilian populations. So the kind of thing the bomber strictly intends immediately and invariably results in some innocent deaths. Two problems have emerged: First, since more than one thing may be strictly intended in a given choice, the pronouncements of the doctrine may depend on how the choice happens to be described. This relativity is embarrassing. We would like the doctrine to speak with one voice in any given case. Second, if we try to get around this problem by saying that the doctrine discriminates against a choice in which anything that is strictly intended is also closely connected with death or harm, the doctrine will make uninviting moral distinctions. As we have seen, it will speak against H C if hysterectomies performed on pregnant patients have some distinguishing surgical feature. Otherwise it will speak in favor. And it will speak against the strategic bomber's attack on an urban factory if he was looking specifically for an automotive plant but not, perhaps, ifhe was looking for a strategically important productive facility. 12 Another approach clearly seems called for. Instead oflooking for a way to identify intrinsically bad effects that are 'close enough' to what is intended, we might look instead for a way to identify choices that are intended under some intrinsically negative description. We might then find a way to show that the actions in TB and CC, but not in SB or HC, are intentional as killings and that the inaction in GP, but not in DR, is intentional as a letting die. Elizabeth Anscombe gives us one such criterion. 13 If we ask a man why he is pushing a mower, he will perhaps say "to cut the grass"; if we ask why he is cutting the grass, he may say "to get things spruced up around here," and so on. The "to ... " answers, or answers that can be understood in terms of them, give further intentions with which the agent acts. If, his choice being described in a certain way, he accepts the 'why' question and replies with a "to ... " answer, then his choice is intentional under that description. But if he rejects the question in a certain

28

Warren S. Quinn

familiar way, his choice is unintentional. If asked why he is cutting the grass he replies, for example, "I don't care about that, I'm just out to annoy the neighbors" or "Can't be helped-it goes with this terrific form of exercise," his cutting the grass is not, as such, intentional. This seems to give the desired result when applied to our cases. If we ask the doctor in CC why he is killing the fetus, he will naturally say "to save the mother." If we ask the pilot in TB why he is killing the civilians, he will say "to help with the war." And if we ask the doctors in GP why they withhold treatment, they will say "to observe the progress of the disease." And it might be thought that if we ask similar questions in the other cases, the 'why' question will be rejected in a way that shows the choices to be unintentional. Thus, if asked why he is killing the fetus, the doctor in HC will avoid a "to ... " answer, saying instead something like "It can't be helped ifl am to save the mother." Actually, this seems not quite right. If the doctors in DR were asked why they weren't treating the group in question, they might naturally reply "to save our resources for more easily treated cases." And this, by Anscombe's criterion, would seem to make the nontreatment intentional. But waiving this difficulty, there is another worry. What if the agents in the problematic cases (TB, GP, and CC) become philosophically sophisticated? Perhaps they will then come to reject the 'why' questions in the manner of their counterparts. The terror bomber, for example, might respond by saying, "The actual deaths can't be helped if I am to create the realistic appearance of death and destruction." By giving such answers, he and the others will be opting for a more demanding criterion of the intentional. All aspects of an action or inaction that do not in the strictest sense contribute to an agent's goal ,vill be trimmed away as unintentional. By this criterion, the action in CC is intentional as a crushing and that in TB is intentional as an apparent killing. But neither is intentional as a killing. And in GP the inaction is intentional as a way of facilitating medical research, but not as a letting die. Now it would be very natural to object that the ordinary, more relaxed criterion of the intentional is the right one, and that the stricter criterion is specious. But how is this to be made out? We might try to introduce a form of essentialism here, claiming that the surgery in CC and the bombing in TB are essentially killings or harmings, while the surgery in HC and the bombing in SB are not. But surely the ground of this essentialism would be the prior conviction that the killings in CC

Actions, Intentions, and Consequences

29

and TB are intentional while those in HC and SB are not. The issue about intentionality seems to be the basic one. And what would we say about the inaction in GP-that it was essentially a failure to prevent harm? But then this would also seem true of the inaction in DR. On the one side we have Anscombe's criterion of the intentional, which pretty well maps our ordinary ways of speaking, while on the other we have a criterion that is structurally similar but stricter. The problem here about intention is reminiscent of a problem about causality that arises in connection with the Doctrine of Doing and Allowing. Certain defenses of that doctrine (which discriminates against active harming and in favor of allowing harm) appeal to a familiar conception of causality according to which active harming causes harm while inactively allowing harm does not. But opponents counter that according to other, philosophically superior conceptions of causality, inaction can be every bit as much a cause of harm. Now I have argued that if DDA is sound theory, it ought to have force on any plausible conception of causality. 14 And I feel much the same here. If the DDE is sound, its force ought to be capturable on any plausible theory of the intentional, even one that would revise ordinary ways of speaking. So, for purposes of argument, I shall grant opponents of the doctrine the greatest latitude in paring back intentional actions to their indisputably intentional cores.

II We must therefore find a different reply to the difficulty with which we started. And I think I see a way. For we have been neglecting one striking respect in which members of our contrasting pairs differ. Take TB and SB. In the former case, but not the latter, the bomber undeniably intends in the strictest sense that the civilians be involved in a certain explosion, which he produces, precisely because their involvement in it serves his goal. He may not, if Bennett is right, intend their deaths. But his purpose requires at least this-that they be violently impacted by the explosion of his bombs. That this undeniably intended effect can be specified in a way that does not strictly entail their deaths is, on the view I am proposing, beside the point. What matters is that the effect serves the agent's end precisely because it is an effect on civilians. The

30

WarrenS. Quinn

case with SB is quite different. The bomber in that case intends an explosion, but not in order that any civilians be affected by it. Of course he is well aware that his bombs will kill many of them, and perhaps he cannot honestly say that this effect will be "unintentional" in any standard sense, or that he "does not mean to" kill them. But he can honestly deny that their involvement in the explosion is anything to his purpose. The same contrast is found in the medical cases. The doctor in CC strictly intends to produce an effect on the fetus so that the mother can be saved by that effect. But the doctor in HC has, as we have seen, no such intention. Even ifhe cannot deny that, in some ordinary sense, he "intends" the fetus's death, he can rightly insist that the effects on the fetus of his surgery are nothing toward his medical purpose. Similarly, the doctors in GP intend, as something toward their further goal, that the disease in the untreated patients work its course. And this could be true even if, wishing to investigate only the effects of the disease within cells, they had no interest in the pain and loss of function it also causes. But in DR nothing that happens to the untreated patients serves the doctors' further goal. 15 The important way in which the cases differ should not be obscured by the following complication. We have seen that a doctor in HC might intend to use the special anesthetic "safest for a pregnant patient." Would it follow from this allusion to the fetus that the doctor does, after all, strictly intend something for it? No. The medical relevance of the patient's pregnancy does not mean that any of the surgical effects on the fetus are medically useful. Something similar holds in SB. Suppose the bomber wants, for moral reasons, to target factories in the least populated district of a certain city. If so, the formulation of his strictly intended means contains an indirect reference to the civilians whom he may kill. But this hardly turns him into a terror bomber. The impact of his bombs on those civilians is still nothing to his military purpose. This clear distinction between the intentional structures of the contrasting cases is the key to a new and better formulation of the doctrine. To put things in the most general way, we should say that it distinguishes between agency in which harm comes to some victims, at least in part, from the agent's deliberately involving them in something in order to further his purpose precisely by way of their being so involved (agency in which they figure as intentional objects)16 and harmful agency

Actions, Intentions, and Consequences

31

in which either nothing is in that way intended for the victims or what is so intended does not contribute to their harm. 17 Let us call the first kind of agency in the production of harm directand the second kind indirect. According to this version of the doctrine, we need, ceterisparibus, a stronger case to justify harmful direct agency than to justify equally harmful indirect agency. 18 Put this way, the doctrine solves the original problem of showing a genuine difference in the intentional structures of our contrasting cases, even under a strict interpretation of what is intended. And it makes no appeal to the problematic notion of "closeness." For direct agency requires neither that harm itself be useful nor that what is useful be causally connected in some especially close way with the harm it helps bring about. 19 There is another, related advantage. With this version of the doctrine, we can sidestep all potentially controversial questions about whether the agents in our various cases kill or harm intentionally. It is enough that we can identify the things they uncontroversially intend as contributing to their goal. One further bit of line-drawing remains. We have not yet defined the difference between the more pronounced moral asymmetry of DR and GP, or SB and TB, and the apparently weaker asymmetry of HC and CC. This difference may partly depend on whether the agent, in his strategy, sees the victim as an advantage or as a difficulty. In CC the doctor wants the fetus removed from the birth canal. Its presence there is the problem. In GP and TB, on the other hand, the availability of potential victims presents an opportunity. By bringing it about that certain things are true of them, the agents positively further their goals. Perhaps it would not be surprising if we regarded fatal or harmful exploitation as more difficult to justify than fatal or harmful elimination. If so, we might say that the doctrine strongly discriminates against direct agency that benefits from the presence of the victim (direct opportunistic agency) and more weakly discriminates against direct agency that aims to remove an obstacle or difficulty that the victim presents (direct eliminativeagency).

III The DDE, of course, has only prima facie moral force. Special rights may allow us to harm someone's interests by way of direct (and even

32

Warren S. Quinn

direct opportunistic) agency. Various rights of competition and the right to punish seem to be examples. Certain other cases may prompt qualifications or special interpretations of the doctrine. Suppose that the doctor in HC needs to alter, harmlessly, the position of the fetus before the womb can be safely removed. Whether the overall surgical procedure would still count as indirect harming seems a matter of interpretation. If we saw the manipulation of the fetus as a partial cause of its later removal, we would presumably count the harming as direct. If we saw the manipulation as a precondition, but not a partial cause, of the removal, we would count the harming as indirect. Another problematic kind of case involves innocent hostages or other persons who physically get in the way of our otherwise legitimate targets or projects. Does our shooting through or running over them involve a direct intention to affect them? I think not. It is to our purpose, in the kind of case I am imagining, that a bullet or car move through a certain space, but it is not to our purpose that it in fact move through or over someone occupying that space. The victims in such cases are of no use to us and do not constitute empirical obstacles (since they will not deflect the missile or vehicle in question). If we act despite their presence, we act exactly as we would if they were not there. If, on the other hand, we needed to aim at someone in order to hit a target, that person would clearly figure as an intentional object. Another tricky case is one in which we could, and would if we had to, accomplish our end by harmful indirect agency; but it is better, perhaps safer for those to be benefited, to pursue the end by harmful agency that is direct. It seems clear why we might wish to make this kind of case an exception. Before we turn to the defense of the doctrine, we should briefly consider the way in which it interacts with the distinction, mentioned in connection with the Doctrine of Doing and Allowing, between what is actively brought about and what is merely allowed to happen. I have claimed that DDE, with the exceptions noted, discriminates against harmful direct agency. But, as we have seen, people may figure as intentional objects not only of a choice to act but also of a choice not to act. DDE therefore cuts across the distinction between harming and allowing harm. Sometimes, as in TB and CC, it discriminates against direct agency in which harm is done. And sometimes, as in GP, it discriminates against direct agency in which harm is allowed.

Actions, Intentions, and Consequences

33

In all of these cases we seem to find an original negative or positive right that, while opposed by other rights, seems to be strengthened by the fact that harm will come via direct agency. 20 Civilians in wartime have negative rights not to be killed. But if their government is waging an unjust war, these rights may conflict with strong rights of selfdefense. A sufficiently developed fetus in utero might also have some negative right not to be killed. But this right may not prevail, either because the fetus is not yet fully one of us or because its mother has strong rights over her body. In TB and CC, the directness of the threatening agency apparently serves to strengthen these negative rights, perhaps giving them a power to stand against moral forces to which they would otherwise give way. Something similar happens in GP. The untreated people have, presumably, some positive right to medical aid. This right might not be binding if doctors could cure more people by directing aid elsewhere. But it stands against any attempt to maximize medical benefit by deliberately letting the people deteriorate. Again, the directness of the intention strengthens the force of the opposing right or claim. It is interesting to consider whether ODE might also come into play where no independent negative or positive right is present. Suppose, in an act of pure supererogation, I am about to aid you but am checked by the realization that your difficulty can be turned either to my advantage or to that of someone I care more for. Does my change of mind, for that reason, violate any of your rights? I am inclined to think not. It might be bad of me to be checked by such a reason, but its appearance cannot create an obligation where none existed before. Rights not to be caught up, to one's disadvantage, in the direct agency of others seem to exist only where some positive or negative right already applies. Their effect always seems to be that of strengthening some other right. The effect of the doctrine is therefore to raise rather than to lower moral barriers. So we should not expect a proponent of DDE to be more tolerant of harmful indirect agency than those who reject the doctrine but share the rest of his moral outlook. We should rather expect him to be less tolerant of harmful direct agency. This point is important. For casual critics of the doctrine sometimes seem to suppose that its defenders must be ready to allow killings or harmings simply on the ground that the agency is indirect. But nothing could be further from the truth. The doctrine in no way lessens the constraining force of any independent moral right or duty.

34

Warren S. Quinn

IV We must now turn to the question of rationale. At first glance, harmful direct agency might seem harder to justify because it requires that the agent welcome something bad for the victim. The terror bomber, for example, must welcome the news that the innocent civilians are blown up, even if he is not glad that they won't be miraculously resurrected after the war. The trouble is that it also seems the strategic bomber must, in some sense, welcome the same news, since if the civilians had been unharmed the factory would not in fact have been destroyed. 21 Of course the news is good for different reasons. It is good news for the terror bomber because it announces the very thing that he intended, while it is good news for the strategic bomber because it announces the thing that he foresaw would be evidence of what he intended. But this difference does little more than register what we already knew-that the terror bomber strictly intended the deaths while the strategic bomber merely foresaw them as necessary costs. So it is hard to see how it could be used to explain the moral difference between direct and indirect agency. Nor is it the case that harms of direct agency need be worse than those of indirect agency. If someone threatened by a terror bomber and someone equally threatened by a strategic bomber both needed rescuing, the former would not seem to have the stronger claim to help. Indeed, there would seem to be no reason to rescue either in preference to someone threatened by purely natural causes. 22 And if we sometimes think that the first rescue must have priority, it seems to be only because we are tempted to regard the violation of a special right against harmful direct agency as a distinctive and additional kind of moral evil. But then it would be circular simply to appeal to the evil in order to explain the existence or force of the right. Perhaps the following rationale is more promising. Someone who unwillingly suffers because of what we intend for him as a way of getting our larger goal seems to fall under our power and control in a distinctive way. And there may be something morally problematic in this special relation-something over and above what is morally objectionable in the simpler relation of bringing about or not preventing harm. If this is right, then harmful direct agency must have two things against

Actions, Intentions, and Consequences

35

it, while equally harmful indirect agency need have only one. This additional negative element can be seen most clearly in the contrast between the doctors' attitudes in GP and DR. In the former, but not the latter, they show a shocking failure of respect for the persons who are harmed; they treat their victims as they would treat laboratory animals. DDE might therefore seem to rest on special duties of respect for persons, duties over and above any duty not to harm or to prevent harm. While this is surely on the right track, we must proceed with caution. For there is also a kind of disrespect in typical cases of wrongful indirect agency. A strategic bomber who ought to have refrained from destroying a rather unimportant target because oflikely civilian casualties has failed to treat his victims with the consideration that they and their interests deserve. So we must look for a kind of disrespect that is peculiar to wrongful direct agency-a kind different from that shown in wrongly giving a victim's interests too little weight. What seems specifically amiss in relations of direct harmful agency is the particular way in which victims enter into an agent's strategic thinking. An indirect agent may be certain that his pursuit of a goal will leave victims in its wake. But this is not because their involvement in what he does or does not do will be useful to his end. The agent of direct harm, on the other hand, has something in mind for his victims-he proposes to involve them in some circumstance that will be useful to him precisely because it involves them. He sees them as material to be strategically shaped or framed by his agency. Someone who harms by direct agency must therefore take up a distinctive attitude toward his victims. He must treat them as if they were then and there for his purposes. But indirect harming is different. Those who simply stand unwillingly to be harmed by a strategy-those who will be incidentally rather than usefully affected-are not viewed strategically at all and therefore not treated as for the agent's purposes rather than their own. They may, it is true, be treated as beings whose harm or death does not much matter-at least not as much as the achievement of the agent's goals. And that presumption is morally questionable. But in a counterpart case of direct agency there is the additional presumption that the victim may be cast in some role that serves the agent's goal. The civilians in TB serve the bomber's goal by becoming casualties, and the infected people in GP serve the doctors' goal by becoming guinea pigs. If things were different, the victims might become these

36

Warren S. Quinn

things only voluntarily. Suppose, for example, the civilians had effective bomb shelters and the sick people medicines of their own. Then the bomber or doctors could succeed only with the cooperation of the victims. The service exacted would then be voluntary. But in cases of indirect agency the victims make no contribution. If the civilians in SB had shelters and if the sick people in DR had medicines, the bomber and the doctors would see no point in their refusing to use them. The DDE rests on the strong moral presumption that those who can be usefully involved in the promotion of a goal only at the cost of something protected by their independent moral rights (such as their life, their bodily integrity, or their freedom) ought, prima facie, to serve the goal only voluntarily. 23 The chief exceptions to this strong presumption are cases in which people have or would have strong moral obligations to give themselves to the service of a goal even at such personal costs-especially cases in which it would be indecent of them to refuse. But surely there is not, or may not be, any such obligation in the cases we have been considering: noncombatants (even those on the wrong side) are not morally obligated to serve the right side by accepting the role of demoralizing civilian casualties, victims of dangerous diseases are not typically obligated to become guinea pigs for the sake of others, and I suppose it is at least open to question whether the fetus in CC, if it could grasp its predicament, would have to accept, for the sake of its mother, the sacrifice of its life. In these cases, but not in their indirect counterparts, the victims are made to play a role in the service of the agent's goal that is not (or may not be) morally required of them. And this aspect of direct agency adds its own negative moral force-a force over and above that provided by the fact of harming or failing to prevent harm. 24 This additional force seems intuitively clearest in direct opportunistic agency, such as TB and GP, where unwilling victims are not only harmed but, in some sense, used. And this must be why the doctrine seems most plausible when it discriminates against opportunistic direct agency. It must also help explain why some of the most perverse forms of opportunistic agency, like torture, can seem absolutely unjustifiable. It is less plausible, on the other hand, to think of the victims of direct eliminative agency as used. This may be why the doctrine seems to discriminate against eliminative agency less forcefully. And it may therefore help explain why some people feel that the direct agency of

Actions, Intentions, and Consequences

37

CC is not much harder to justify than the indirect agency of HC. But something of the questionable character of direct opportunistic agency also seems present in direct eliminative agency. Someone who gets in your way presents a strategic problem-a causal obstacle whose removal will be a service to your goals. And this is quite unlike what we find in harmful indirect agency, where victims can he obstacles only in a moral sense. In discriminating to some extent against both forms of direct agency, the doctrine reflects a Kantian ideal of human community and interaction. 25 Each person is to be treated, so far as possible, as existing only for purposes that he can share. This ideal is given one natural expression in the language of rights. People have a strong prima facie right not to be sacrificed in strategic roles over which they have no say. They have a right not to be pressed, in apparent violation of their prior rights, into the service of other people's purposes. Sometimes these additional rights may be justifiably infringed, especially when the prior right is not terribly important and the harm is limited, but in all cases they add their own burden to the opposing moral argument. The Doctrine of Double Effect thus gives each person some veto power over a certain kind of attempt to make the world a better place at his expense. This would be absurd if the entire point of morality were to maximize overall happiness or welfare. But that is not its entire point. An equally urgent basic task is to define the forms of respect that we owe to one another, and the resulting limits that we may not presume to exceed. The doctrine embodies our sense that certain forms of forced strategic subordination are especially inappropriate among free and equal agents.

NOTES

I am grateful for very helpful suggestions from Rogers Albritton, Philippa Foot, Matthew Hanser, and many others; and for criticism from audiences at New York l.:niversity, the University of California at Irvine, and Princeton University. 1. Harm is meant in a very broad sense that includes the loss of life, rightful property, privacy, and so on. In my examples, the relevant harm will usually be the loss oflife. 2. Warren S. Quinn, "Actions, Intentions, and Consequences: The Doctrine of Doing and Allowing," PhilosophicalReview, July 1989, pp. 287-312.

38

Warren S. Quinn

3. The doctrine, which is usually traced to Thomas A4uinas, Summu Theologiac, II-II, Q. 64, art. 7, is typically put as a set of necessary conditions on morally permissible agency in which a morally questionable bad upshot is foreseen: (a) the intended final end must be good, (b) the intended means to it must he morally acceptable, (c) the foreseen bad upshot must not itself be willed (that is, must not be, in some sense, intended), and (d) the good end must be proportionate to the bad upshot (that is, must he important enough to justify the bad upshot). The principle that follows in the text, which I henceforth treat as if it were itself the doctrine, is really what I find most important and plausible in its first three conditions. I ignore the fourth condition both because it is probably best understood in a way that makes it noncontroversial and because I am concerned here not so much with how choices with a 'second effect' can be justified as with whether, ceterisparilms, the structure of intention makes a justificatory difference. That seems to me the fundamental question. 4. The principle is sometimes put in terms of the difference between a harmful result that is "directly" intended and one that is "indirectly" (or "obliquely") intended. But it also might be put in terms of the difference between a directly and an indirectly intended act of harming. In either variant, the point of calling the merely foreseen result or action "indirectly intended" is to mark a species of linguistic impropriety in an agent's asserting, with a completely straight face, that a clearly foreseen harm or harming is quite unintended. !fl have no desire to wake you hut simply do not care that my fiddling will have that effect, I cannot say that your waking or my waking you is purely unintentional. Whether there is anv natural sense in which they are intentional is a debated point. In the final analysis, I shall sidestep this controversy, concerning myself with a species of intention that an agent clearly does not have toward a merely foreseen result of his agency-namely, the intention that the result occur, or that he bring it about, as a means of achieving his purpose. 5. See Herbert L.A. Hart, "Intention and Punishment," in Punishment and Responsibility (Oxford: Clarendon Press, 1968), p. 123. Hart finds the intentions in CC and HC to be parallel. But he does not argue, and does not seem to think, that a similar point can be made about most other cases that the doctrine might seem to distinguish. Nancy Davis finds more general problems along these lines in "The Doctrine of Double Effect: Problems of Interpretation," PacificPhilosophicalQuarterly 65 (1984): 107-23. [Included here, chapter 7.] 6. If the miracle happened, and after its removal the fetus were quickly restored to its previous healthy condition, we would say that the craniotomy had done no real harm. In the actual case, the harm done to the fetus by the craniotomy consists in the combination of the desired immediate effect on it (which permits its removal) and the further natural effects that flow from that first effect. Since these further effects are not strictly intended, the objection holds that the harm itself is not strictly intended. See Jonathan Bennett, Morality and Consequences,The Tanner l.ectures on Human Values II (Salt Lake City: University of Utah Press, 1981), pp. 110-11. 7. Ibid., p. 111. 8. Perhaps then it would not really be, at least in these people, a disease. But then it might be said that the doctors don't really need it to be a disease in them. It would be good enough if, due to their special powers of compensation, it is for them a harmless condition very much like a disease in others. 9. Philippa Foot perhaps suggests this kind of reply in "The Problem of Abortion and the Doctrine of the Double Effect," in Virtues and Vices and Other Essays(Berkeley

Actions, Intentions, and Consequences

39

and Los Angeles: University of California Press, 1978), pp. 21-22. [Included here, chapter 8.] 10. Hart, "Intention and Punishment," p. 120. 11. Of course this special operation could, however inappropriately, be performed on patients who were not pregnant. And this might lead someone to speculate that the doctrine speaks against a strictly intended and invariably harmful kind of action or omission only if the harm is an empirically necessary consequence. But this cannot be right. Suppose there is some good that will arise immediately upon your being injected with a certain fatal poison. The good does not require that you actually die. But that is what will happen, since the very real and naturally abundant antidote that could save you has not been, and in fact never will be, discovered. In such a case, the doctrine should certainly speak against my poisoning you. But the directly and invariably connected harm would not follow of empirical necessity. 12. If the latter intention sometimes gets fulfilled, for example, by bombing electric power facilities built into remote and isolated dams. 13. G. E. M. Anscombe, Intention, 2d ed. (Oxford: Blackwell, 1963), sec. 25, pp. 41-45. 14. See Quinn, "Actions, Intentions, and Consequences: The Doctrine of Doing and Allowing," pp. 293-94. 15. Not even, I would argue, the fact of their not receiving the treatment. What really furthers the goal is the treatment received by the other, more tractable cases. The nontreatment of the first group contributes, at most, in an odd and secondary sense. This point applies, I think, to a wide range of intentional expressions. Suppose we decide to combat a disease by spending our limited resources on education rather than on inoculation. Education, and not noninoculation, will then be our means of combat; and the way we fight the disease will be by educating, not by not inoculating. 16. I might instead have said "agency in which harm comes to victims ... from the agent's deliberately producing some effect on them in order to further his purpose precisely by way of their being so affected." But there is a certain kind of ingenious case, attributed to David Lewis, that such a formulation might seem to miss. Suppose that another terror bomber wishes to demoralize enemy leaders by bombing a major center of population, and suppose he knows that these leaders will be convinced that the city is destroyed by seeing, from afar, the explosion of his bombs over it. The explosion occurs an instant before the fatal effects below. So in this case the bomber does not, strictly speaking, intend to blow up the civilians, or produce any physical effects on them, as a means to his end. Yet the case seems, morally speaking, to be like TB rather than SB. But notice that while such a strategy does not aim at physically affecting its victims, it does strictly aim at exploding bombs in their vicinity. Whether or not this change in their situation could be counted as an effect on them, as I think it could, the bomber strictly intends to involve them in something (to make his bombs explode over them) in order to further his purpose precisely by way of their being involved. 17. This way of drawing the distinction excludes a pair of cases sometimes used to illustrate double effect: in one we give powerful analgesics to lessen the terrible pain of a dying patient, where we foresee that he will die as a side effect. In the other we relieve his suffering by intentionally killing him with the same or other drugs. In both cases we are to suppose that life is no longer a good and that we act with his explicit or correctly presumed consent. So we cannot see ourselves as infringing, justifiably or unjustifiably, any of his moral rights. For this reason I see these cases as really quite different from the others, in which there is conflict between the moral claims of different people. Indeed,

40

Warren S. Quinn

I think that the doctrine is misapplied in nonconflict cases. I see, for example, no difference between amputating someone's leg to save him and proceeding with some lifesaving treatment that, as a side effect, results in the loss of the limb. And by parity of reasoning it seems to me that if stopping pain is urgent enough from the patient's perspective to make death acceptable as a side effect, it ought to make death acceptable as a means. 18. A terminological point: Something counts as 'harmful direct agency' only insofar as harm comes to the very people who are deliberately affected by the agency. Insofar as harm comes to others, the agency also counts as 'indirectly harmful'. A single act or omission can thus be both directly and indirectly harmful. 19. Nor, of course, does it require that the agent have particularvictims in mind. It is enough, as in the case of a terrorist's car bomb, that he intends something for someone or other. 20. Positive rights are rights to aid while negative rights are rights to noninterference. While borrowed originally from the law, these terms are here used in a moral sense. 21. See Bennett, Moralityand Consequences, pp. 102-3. 22. Samuel Scheffler makes a similar point in The Rejection of Consequentialism (Oxford: Clarendon Press, 1982), p.109. 23. I am deliberately not considering cases where the sacrifice is financial. What to think in such cases partly depends on the sorts of moral rights people really have to keep money or property that is legally or conventionally theirs when others have more pressing material needs. It is quite consistent with everything I say here to deny that the doctrine speaks against liberal schemes of redistributing wealth. 24. Although it is, as we have seen, a kind of negative moral force that is activated only when other rights are present. 25. But there is a way in which the rationale I have provided is not Kantian. For it draws a sharp moral line between adversely affecting someone in the pursuit of an end that he does not share (not treating him as an end in itself) and adversely affecting someone because his being so affected is strategically important to achieving an end that he does not share (very roughly, treating him as a means). Neither the terror nor the strategic bomber treats his victims as ends in themselves, but only the former treats them as something like means. And I have argued that this difference is significantthat morality erects an extra barrier against the strategic posture of harmful direct agency. Kant might disagree, focused as he is on the alleged status of people as ends in themselves. But I have difficulty attaching any sense to that idea except via intuitions that certain forms of treatment are unacceptably disrespectful of rational beings. And the intuition that it is more disrespectful, all other things being equal, to treat someone as if he existed for purposes he does not share than simply not to be constrained by his purposes, seems to me plausible enough to be worth incorporating in a proper idea of what it means for persons to be ends in themselves. On this conception, one aspect of being an end in itself would be to have, ceterisparibus,a stronger right against directly harmful agency than against indirectly harmful agency.

THREE

Agent-Relative Morality THOMAS

NAGEL

Let me turn now to the obscure topic of deontological constraints. These are agent-relative reasons which depend not on the aims or projects of the agent but on the claims of others. Unlike autonomous reasons, they cannot be given up at will. If they exist, they restrict what we may do in the service of either agent-relative or agent-neutral goals. Whatever their explanation, they are conspicuous among the moral appearances. Here is an example to focus your intuitions. You have an auto accident one winter night on a lonely road. The other passengers are badly injured, the car is out of commission, and the road is deserted, so you run along it till you find an isolated house. The house turns out to be occupied by an old woman who is looking after her small grandchild. There is no phone, but there is a car in the garage, and you ask desperately to borrow it and explain the situation. She doesn't believe you. Terrified by your desperation, she runs upstairs and locks herself in the bathroom, leaving you alone with the child. You pound ineffectively on the door and search without success for the car keys. Then it occurs to you that she might be persuaded to tell you where they are if you were to twist the child's arm outside the bathroom door. Should you do it?

Excerpt reprinted from Thomas Nagel's "The Limits of Objectivity," The Tanner Lectures on Human Values (Salt Lake City: University of Utah Press, 1980), 126-35, by permission of the Tanner Lectures on Human Values, University of Utah. Delivered as a Tanner Lecture on Human Values at Brasenose College, Oxford University, 1979. 41

42

Thomas Nagel

It is difficult not to see this as a serious dilemma, even though the child's getting his arm twisted is a minor evil compared with your friends' not getting to a hospital. The dilemma must be due to a special reason against doing such a thing. Otherwise it would be obvious that you should choose the lesser evil, and twist the child's arm. Common moral intuition recognizes several types of deontological reasons-limits on what one may do to people or how one may treat them. There are the special obligations created by promises and agreements; the restrictions against lying; the prohibitions against violating various individual rights, rights not to be killed, injured, imprisoned, threatened, tortured, coerced, robbed; the restrictions against imposing certain sacrifices on someone simply as means to an end; and perhaps the special claim of immediacy, which makes distress at a distance so different from distress in the same room. There may also be a deontological requirement of fairness, of evenhandedness or equality in one's treatment of people. (This is to be distinguished from any agent-neutral value thought to attach to equality in the distribution of benefits, considered as an aspect of the assessment of states of affairs.) In all these cases it appears that the special reasons, if they exist, cannot be explained simply in terms of agent-neutral values, because the particular relation of the agent to the outcome is essential. Deontological constraints may be overriddenby agent-neutral reasons of sufficient strength, but they are not themselves to be understood as the expression of agent-neutral values of any kind. It is clear from the way such reasons work that they cannot be explained by the hypothesis that the violation of a deontological constraint has high negative agentneutral value. Deontological reasons have their full force against your doing something-not just against its happening. For example, if there really are such constraints, the following things seem to be true. It seems that you shouldn't break a promise or tell a lie for the sake of some benefit, even though you would not be required to forego a comparable benefit in order to prevent someone else from breaking a promise or telling a lie. And it seems that you shouldn't twist the arm of a small child to get its grandmother to do something, even if the thing is quite important-important enough so that it would not be reasonable to forego a comparable benefit in order to prevent someone else from twisting a child's arm. And it may be that you shouldn't engagein certain kinds of unfair discriminatory treatment (in an official role, for example) even to produce a good result which it

Agent-Relative Morality

43

would be unreasonable to forego in order to prevent similar unfairness by others. Some may simply deny the plausibility of such moral intuitions. Others may say that their plausibility can be subtly accounted for in terms of agent-neutral values, and that they appear to involve a fundamentally different type of reason for action only if they are inadequately analyzed. As I have said, I don't want to take up these alternative accounts here. They seem to me essentially revisionist, and even if from that point of view they contain a good deal of truth, they do not shed light on the deontological conceptions they are intended to replace. Sometimes, particularly when institutions and general practices are involved in the case, there may be an agent-neutral justification for what looks initially like an agent-relative restriction on action. But I am convinced there are many cases that evoke a different type of moral intuition. Right or wrong, it is this type of view that I want to explore and understand. There is no point in trying to show in advance that the controversy does not exist. One reason for the resistance to deontological constraints is that they are formally puzzling, in a way that the other reasons we have discussed are not. We can understand how autonomous agent-relative reasons might derive from the specific projects and concerns of the agent, and we can understand how agent-neutral reasons might derive from the interests of others, giving each of us reason to take them into account. But how can there be agent-relativereasons to respect the claims of others?How can there be a reason not to twist someone's arm which is not equally a reason to prevent his arm from being twisted by someone else? The agent-relative character of the reason cannot come simply from the character of the interest that is being respected, for that alone would justify only an agent-neutral reason to protect the interest. And the agent-relative reason does not come from an aim or project of the individual agent, for it is not conditional on what the agent wants. Deontological restrictions, if they exist, apply to everyone: they are mandatory and may not be given up like personal ambitions or commitments. There is no doubt that ideas of this kind form an important part of common moral phenomenology. Yet it is tempting to think that the whole thing is a kind of moral illusion resulting either from innate psychological dispositions or from crude but useful moral indoctrination. But this hypothesis faces problems in explaining what the illusion is. It

44

ThomasNagel

may be a good thing if people have a deep inhibition against torturing children even for very strong reasons, and the same might be said of other deontological constraints. But that does not explain why we cannot come to regard it as a mere inhibition which it is good to have. An illusion involves a judgment or a disposition to judge, and not a mere motivational impulse. The phenomenological fact that has to be accounted for is that we seem to apprehend in each individual case an extremely powerful agent-relative reason not to torture a child. This presents itself as the apprehension of a truth, not just as a psychological inhibition. And the claim that such an inhibition is in general very useful does nothing to justify or explain the conviction of a strong reason in every individual case. That conviction is what has to be analyzed and accounted for, and accepted or rejected according to whether the account gives it an adequate justification. I believe that the traditional principle of double effect, despite problems of application, provides a rough guide to the extension and character of deontological constraints, and that even after the volumes that have been written on the subject in recent years, this remains the right point of convergence for efforts to capture our intuitions. 1 The principle says that to violate deontological constraints one must maltreat someone else intentionally. The maltreatment must be something that one does or chooses, either as an end or as a means, rather than something one's actions merely cause or fail to prevent, but that one doesn't aim at. It is also possible to foreseethat one's actions will cause or fail to prevent a harm that one does not intend to bring about or permit. In that case it is not, in the relevant sense, something one does, and does not come under a deontological constraint, though it may still be objectionable for impersonal reasons. (One point worth stressing: the constraints apply to intentionally permitting as well as to intentionally doing harm. Thus in our example, there would be the same kind of objection if with the same end in view you permitted someone else to twist the child's arm. You would have let it happen intentionally, and that would be different from a failure to prevent such an occurrence because you were too engaged in doing something elsewhich was more important.) So far this is just moral phenomenology: it does not remove the paradox. Why should we consider ourselves far more responsible for what we do (or permit) intentionally than for consequences of action

Agent-Relative Morality

45

that we foresee and decide to accept but that do not form part of our aims (intermediate or final)? How can the connection of ends and means conduct responsibility so much more effectively than the connection of foresight and avoidability? It is as if each action produced a special perspective on the world, determined by intention. When I twist the child's arm intentionally I incorporate that evil into what I do: it is my creation and the reasons stemming from it are magnified from my point of view so that they tower over reasons stemming from greater evils that are more 'distant' because they do not fall within the range of intention. That is the picture, but how can it be correct? I believe that this is one of those cases in which the removal of paradox is not a philosophical advance. Deontological reasons are essentially problematic, and the problem is an instance of the collision between subjective and objective points of view. The issue is whether the special, personal perspective of agency has fundamental significance in determining what people have reason to do. The question is whether, because of this perspective, I can have sufficient reason not to do something which, considered from an external standpoint, it would be better if I did. That is, things would be better, what happened would be better, if I twisted the child's arm than if I did not. But I would have done something worse. If considerations of what I may do, and the correlative claims of my victim, can outweigh the substantial impersonal value of what will happen, that can only be because the perspective of the agent has an importance in practical reasoning that resists domination by a conception of the world as a place where good and bad things happen, and have their value without perspective. I have already claimed that the dominance of this agent-neutral conception of value is not complete. It does not swallow up or overwhelm the agent-relative reasons arising from those individual ambitions, commitments, and attachments that are in some sense chosen. But the admission of what I have called autonomous agent-relative reasons does not imply the possibility of deontological reasons. The two are very different. The special paradox of deontological reasons is that although they are agent-relative, they do not express the subjective autonomy of the agent at all. They are demands. The paradox is that this partial, perspectival respect for the interests of others should not give way to an impersonal respect free of perspective. The deontological

46

Thomas Nagel

perspective seems primitive, even superstitious, by comparison: merely a stage on the way to full objectivity. How can what we do in this narrow sense be so important? Let me try to say where the strength of the deontological view lies. We may begin by considering a curious feature of deontological reasons on which I have not yet remarked. Intention appears to magnify the importance of evil aims by comparison with evil side effects in a way that it does not magnify the importance of good aims by comparison with good side effects. We are supposed to avoid using evil means to produce a good end, even though it would be permissible to produce that good end by neutral means with comparably evil side effects. On the other hand, given two routes to a legitimate end, one of which involves good means and neutral side effects, and the other of which involves neutral means and slightly better side effects, there is no reason to choose the first route. Deontological reasons tell us only not to aim at evil;they don't tell us to aim at good, as a means. Why should this be? What is the relation between evil and intention, or aiming, that makes them clash in a special and intense way? The answer emerges if we ask ourselves what is the essence of aiming, what differentiates it from merely producing a result knowingly? The difference is that action intentionally aimed at a goal is guided by that goal. Whether the goal is an end in itself or only a means, action aimed at it must follow it and be prepared to adjust its pursuit if deflected by altered circumstances. Whereas an act that merely producesan effect does not follow it, is not guided by it, even if the effect is foreseen. What does this mean? It means that to aim at evil, even as a means, is to have one's action guided by evil. One must be prepared to adjust it to insure the production of evil: a falling off in the level of the desired evil must be grounds for altering what one does so that the evil is restored and maintained. But the essenceof evil is that it should repel us. If something is evil, our actions should be guided, if they are guided by it at all, toward its elimination rather than toward its maintenance. That is what evil means. So when we aim at evil we are swimming headon against the normative current. Our action is guided by the goal at every point in the direction diametrically opposite to that in which the value of that goal points. To put it another way, if we aim at evil we make what we do in the first instance a positive rather than a negative

Agent-Relative Morality

47

function of evil. At every point, the intentional function is simply the normative function reversed, and from the point of view of the agent, this produces the acute sense of doing something awful. If you twjst the child's arm, in our example, your aim is to produce pain. So when the child cries, "Stop, it hurts!" his objection corresponds in perfect diametrical opposition to your intention. What he is pleading as your reason to stop is precisely your reason to go on. If it didn't hurt, you would twist harder, or try the other arm. You are pushing directly and essentially against the normative force intrinsic to your goal, for it is the production of pain that guides you. It seems to me that this is the essence of deontological constraints. What feels peculiarly wrong about doing evil intentionally even that good may come of it is the headlong striving against value that is internal to one's aim. Some corroboration of this diagnosis may be found by asking what would be the corresponding principle governing the relation between intention and good, as opposed to evil? I have said that there is no deontological requirement to aim at good-only a requirement not to aim at evil. But the analogue of the requirement not to aim at evil would be a requirement not to aim away from good. To aim to prevent something good as a means to a worthy end would have a similar quality of normative reversal, though less acute than that of aiming at evil. And I believe there may be deontological constraints, though not such conspicuous ones, against deliberately preventing something good, in order that good may come of it. (Think for example of someone who resists ameliorating the condition of the poor because he thinks it will reduce their anger and diminish the long-term chance of a social revolution.) I mention the point, but will not pursue it. But all this still leaves unsettled the question of justification. For it will be objected that if one aims at evil as a means only, then one's action is not really being guided by evil but by overall good, which includes a balanceof goods and evils. So when you twist the child's arm, you are guided by the aim of rescuing your injured friends, and the good Cifthat aim dominates the evil of the child's pain. The immediacy of the fact that you must try to produce evil as a subsidiary aim is phenomenologically important, but why should it be morally important? Here I think we have come down to a fundamental clash between perspectives. The question is whether to disregard the resistance encountered by my immediate pursuit of evil, in favor of the overall value of the results of what I do. When I view my act from outside, and think

48

Thomas Nagel

of it as resulting from a choice of the impersonally considered state of the world in which it occurs, this seems rational. In thinking of the matter this way, I abstract my will and its choices from my person, as it were, and even from my actions, and decide directly among states of the world, as if I were taking a multiple-choice test. If the choice is determined by what on balance is impersonally best, then I am guided by good and not by evil. But the self that is so guided is the objective self which regards the world impersonally, as a place containing TN and his actions, among other things. It is detached from the perspective of TN, for it views the world from nowhere within it. It chooses, and then TN, its instrument, or perhaps one could say its agent, carries out the instructions as best he can. He may have to aim at evil, for the impersonally best alternative may involve the production of good ends by evil means. But he is merely following orders. To see the matter in this light is to see both why the appeal of agentneutral, consequentialist ethics is so great and why the contrary force of agent-relative, deontological ethics is so powerful. The detached, objective view takes in everything and provides a standpoint of choice from which all choosers can agree about what should happen. But each of us is not only an objective self but a particular person with a particular perspective; we act in the world with that perspective, and not only from the point of view of a detached will, selecting and rejecting worldstates. So our choices are not merely choices of states of the world, but of actions. From this point of view, the pursuit of evil in twisting the child's arm looms large. The production of pain is the immediate aim, and the fact that from an external perspective you are choosing a balance of good over evil does not cover up the fact that this is the kind of action you are undertaking. This account of the force of deontological reasons applies with special clarity to the constraint against doing harm as a means to your ends. A fuller deontological theory would have to explain the different types of normative grain against which one acts in breaking promises, lying, discriminating unfairly, and denying immediate emergency aid. It would also have to deal with problems about what exactly is being aimed at in cases of action that can be described in several different ways. But I believe that the key to understanding any of these moral intuitions is the distinction between the internal viewpoint of the agent in

Agent-Relative Morality

49

acting and an external, objective viewpoint which the agent can also adopt. Reasons for action look different from the first point of view than from the second. So we are faced with a choice. For the purposes of ethics, should we identify with the detached, impersonal will that chooses world-states, and act on reasons that are determined accordingly? Or is this an evasion of the full truth about who we really are and what we are doing, and an avoidance of the full range of reasons that apply to creatures like us? If both personal and impersonal perspectives are essential to us, then it is no wonder that the reasons for action deriving from them do not fit comfortably together.

NOTE 1.

A good statement of a view of this type is found in Charles Fried's recent book,

Right and Wrong (Cambridge: Harvard University Press, 1978).

FOUR

Medalist's Address:Action, Intention, and 'DoubleEffect' G. E. M. ANSCOMBE

It is customary in the dominant English and related schools of philosophy to restrict the terms "action" or "agency." That is, when the topic is 'philosophy of action'. This is often done by an appeal to intuition about a few examples. If I fall over, you wouldn't usually call that an action on my part; it's not something that I do, it is rather something that happens to me. Donald Davidson has made a more serious attempt than this at explaining a restriction on the term "action," or what he means by "agency." "Intentional action" is an insufficient designation for him: it determines no class of events, because an action which is intentional under one description may not be intentional under another. And anyway there are unintentional actions, which he doesn't want to say are not actions in the restricted sense in which he wants to apply the term. So he suggests that we have an action (in the restricted sense) if what is done (no restriction on the ordinary sense here) is intentional under some description. This allows pouring out coffee when I meant to pour out tea to be an action, being intentional under the description "pouring out liquid from this pot." I fear, however, that it may allow tripping over the edge of the carpet to be an action too, if every part of an intentional progress across the room is intentional under that

Reprinted from Proceedingsof the American CatholicPhilosophicalAssociation56 (1982): 12-25, by permission of the American Catholic Philosophical Association.

50

Medalist's Address

51

description. But Davidson doesn't want to count tripping as an action. If this is right, then his account is wrong because it lets in what he wants to exclude. Furthermore, I don't think it comprises omissions, which are often actions. I am inclined to think that the attempt, brave as it was, was misconceived anyway. There is a goal in view when people want to introduce a restricted sense of "action," but I don't think it can be attained by trying to find a characterization of a sub-class of events. All the same, the point about there being many descriptions of any action, as indeed of any event or any object, is true. It has an importance which we shall see later. What is being aimed at, I think, is the category called "human action" by some scholastics: They made a contrast between 'human action' -actus humanus-and 'act of a human being' -actus hominis. Again, this is usually explained by examples. Idly stroking one's beard, or idly scratching one's head, may be an 'act of a human being' without being a 'human act.' 1 And I expect falling over or tripping up is so too; at any rate I lay it down that I will use the term 'act of a human being' in such a way that those things come under that heading. Also, that 'act of a human being' is a wider notion, which includes 'human action'. The contrast is that human actions are under the command of reason: This does not mean just that reason can intervene to forbid-for that holds of idle actions too. What it does mean beyond that will, I hope, emerge. We might say that human action = voluntary action. But that raises a question of meaning, like what we have just glanced at in the last paragraph. We are speaking of voluntary action not in a merely physiological sense; not in the sense in which idly stroking your beard is a voluntary action. Notice, too, that what is voluntary under one description may be non-voluntary or counter-voluntary under another. We are not, like Davidson, attempting a classification which will divide all events into members and non-members of a class. Nor are we using "voluntary" as Aristotle uses bwumov, generally translated "voluntary." For Aristotle says that beasts and babies have the voluntary, but we would not say so in the sense of "voluntary" that we are trying to introduce. Aristotle too introduces a restricted sense of "action" -praxis, which beasts and babies don't have. But it is a bit too limited for us.

52

G. E. M. Anscombe

It wouldn't include omissions unless calculated, or sudden impulsive actions. Voluntariness does not require that there be any act of will, any formation of intention, any choice of what is voluntary; or even any positive voluntary act. What you were able to do and it was needful you should do, if you omit it in forgetfulness or sloth, falling asleep perhaps and sleeping through the time when you should have been doing ityour omission of it is voluntary. Unless indeed you fell asleep because you were drugged without the slightest consent on your part. Consent may reside in not taking care when you could have, and in the nature of the case as available to your understanding, you needed to. I will now put forward a thesis, which I will later give reason for: All human action is moral action. It is all either good or bad. (It may be both.) This needs a lot of clarification. First, let me point to an implication. It means that "moral" does not stand for an extra ingredient which some human actions have and some do not. The idea of the moral as an aspect that is to be seen in some human actions, or felt by the agent, and which may be lacking, perhaps is lacking if it is not felt by the agent-this idea is rejected by the equation "Human action moral action." Or at least "Human action = moral action of a human being." For I am not concerned with angels, demons, fairies or Martians. I shall hope to shew that this equation is not a mere 'extensional equivalence' -two descriptions which happen to be true of the same things. The descriptions are equivalent in content. All the same, not all human-action descriptions are moral actiondescriptions. E.g. "walking," if a human being other than an infant is said to walk, is normally a human-action description, but it is not a moral action-description. (I say it is normally a human-action description to allow exclusion of sleep-walking (say), which would be the act of a human being, but not a human action.) "Chucking a pebble into the sea" is another such description, or "Picking a flower." Like "walking," these are indifferent human-action descriptions. This fact, that there are indifferent human-action descriptions, might lead someone to think that there are indifferent human actions. But that does not follow at all. It is the particular action that is always good or bad. Note also that not all action-descriptions which can, and in the particular case often do, describe human actions, are such that

Medalist'sAddress

53

whatever acts fall under them are good or bad human actions. For example the description, even when it does describe the action of a human being, still may not be describing a human action. As an instance, take "Killing someone." Suppose I am a parcel-I mean I've been made up into a parcel-and by sheer accident I get set rolling down a hill in such circumstances that I kill someone by knocking him over into the path of a rapid vehicle-that's not a human action on my part. It is, ifl kill someone by an accident in driving my car-I reverse it by mistake, say, and kill a small child standing behind it. As I said: "Walking," though if it is applied to a non-infant human being it is normally a human-action description, is nevertheless not a moral action-description. By a moral action-description I mean one at least suggestingsome specificgoodness or badness about an act that falls under it. "Walking," even when it is used to report a particular human action, does not suggest anything specific. It is not even true that every moral action-description, as a description of a particular human action, entails that that particular action was good, or again, that it was bad. It may only entail that the action was, say, good, unless some aspect makes it bad. Or was bad, unless some excuse or justification either lets it have a certain goodness inasmuch as it is an action which is not wicked, or actually renders it specifically good. This working of excuse may astonish. Someone has killed someone, let us say. We ask: Was it his fault? It turns out that it was not; although the action which as it turned out resulted in death was a human action. So: he has an excuse, he is perhaps exonerated; he is not guilty of homicide. But does that mean his action was good?If all actual human action that occurs is either good or bad, it seems as if his excuserenders his action good! But isn't that too paradoxical? Don't we want something between good and bad in actual particular occurrences of action? Assume his action is not good qua killing, qua resulting in death. And we can suppose it was no particular species of good action; not the act of any specific virtue. But, just as anything that has positive being is good qua being (even if it is something that for some reason ought not to exist) so action is good as being action, i.e., qua belonging to the genus 'action'. (And it would be so, even if there were no specific goodness about it.) Suppose, then, that all the things about it which make one want to say it was bad turn out to be not his fault: suppose they were just accidents for which he had no responsibility, except in the sense of (involuntarily) contributing some of the causation of a bad result. Then,

54

G. E. M. Anscombe

there is not the quicunque defectus to make his action a bad human action. Here I am invoking the principle "Bonum ex integra causa, malum ex quocunque defectu." This might be rendered: "Something is good by reason of being good in every respect, bad from being bad in any." Here respects of good and bad have to belong to the thing under the description under which it is described as good or bad. Thus there might be a good human action-good tout court, not just good in this respect or that-which was by sheer accident an intervenient cause of some evil. Here some might say that that act of a human beinghad something bad about it, and so wasn't good tout court. That is, if I am right in the extensive application I give to the expression "act of a human being." But the human act has not got something wrong with it qua human act from being an involuntary cause of some evil. That respect of badness does not belong to it as a human act. For being the involuntary cause of something does not belong to it as a human act. On the other hand, there might be nothing good about it qua human action, except that it was human action. I am still thinking of the same case, where the evil result is pure accident, not voluntarily caused by the agent. But he is presumably engaged in something or other in his being there or at least in his motion, if that is voluntary; and what he is up to may itself have been for a bad end, and hence bad on its own account. Or it may have been subordinated to a bad end. But here I get into deep water. Is a wicked man, who keeps himself in good trim by physical exercise so as to pursue his life's purposes, acting badly or well in taking that exercise? I cannot answer this question, and to investigate it now would take too long. It will be apparent why I said that an action-a human action-may be both good and bad. It may be good only in a certain respect, and bad in others; then it is not good tout court, but bad. And the respect may be: generically. In this respect all human actions are good, though many are bad tout court. So far we've been speaking in a uniform way about the case of the bad upshot, whether there is excuse or exoneration. Alternatively, we may want to distinguish between these. With exoneration, an agent is in the clear; but if there is less than exoneration, we may say his action was bad as needing excuse and thereforepardon. In such a case we may

Medalist's Address

55

say it was mixed. In that way, it was bad-if he had taken more care, he would have avoided the result; and yet the result was not one that, as we say, 'was to be expected in the nature of the case'. And in this way it was mixed; it had the goodness of being a human action and perhaps of being whatever sort of action he was engaged in intentionally; it lacks the defect of being a specifically wicked action, since it does not get this from the bad accidental result; and yet it has something of badness, that makes it seem to need pardon. For if we do distinguish between exoneration and excuse in this sort of case, this is where the distinction must be made. If it merits complete exoneration, then it has no badness from the bad upshot-e.g. the unfortunate result is 'utterly accidental, he couldn't have avoided it, it was sheer bad luck'. From considering good and bad, we see that the extension of "human action" is wider than that of "intentional human action." That is to say: something may be a human action under a description under which it is not an intentional action. Acts of carelessness, negligence and omission may be of this character. For though they can be intentional, they may not be so, but their not being intentional does not take the character of human action away from them. But a human action does not have something wrong with it qua human act-i.e., so as to be to that extent a bad human action-from being an involuntary cause of some evil. The involuntariness does mean that the badness is not in a respect belonging to it as a human action. By parity of consideration, voluntariness would mean that it was bad in such a respect. To repeat: when I speak of moral action-descriptions, I mean ones which suggest, at least suggest, good or bad-unlike neutral or indifferent action-descriptions, which suggest nothing either way. However, one might object that if just being an action is good, then neutral action-descriptions must after all mention something good. (Even if in the particular case it also fails in goodness because of something bad about it then and there.) So how can I say that there are neutral actiondescriptions? The answer is that the good or bad suggested by moral action-descriptions is specificgoodness or badness. That in the particular case each human action is good or bad is a different point from the generic goodness of any action as being an action: it means that each human action is specifically good or specifically bad. It will be good, if good, for more than just being an action. Firstly, the particular thing that is done may itself be more than generically good. But also

56

G. E. M. Anscombe

there are circumstances and ends to consider. For suppose the particular thing that is done is sufficiently specified by a neutral actiondescription, so that it-the particular thing that is done-might be called "no more than generically good": then, if the agent's purpose or purposes in doing it are specifically good and circumstances import nothing bad about doing that thing for those purposes, the action will be specifically good. A neutral action-description reporting a human act may give 'the particular thing that was done' sufficiently for us to say: "That was all right, unless the purpose or circumstances put a different complexion on it." Or, it may be an inadequate description of the particular thing that was done: he didn't 'merely write his name on a bit of paper', he was signing such-and-such a contract. To say that "human action" and "moral action (sc. of a human being)" are equivalent is to say that all human action in concreto is either good or bad simpliciter. There is no need to insert "morally" and say "morally good or bad." The term "moral" adds no sense to the phrase, because we are talking about human actions, and the 'moral' goodness of an action is nothing but its goodness as a human action. I mean: the goodness with which it is a good action. 'Moral' goodness is: goodness of actions, passions, and habits of action and feeling. The term is however a distinguishing one, restricted to specific goodness, goodness simpliciter, or to what comes closer to that than the generic good of being a human action or passion or disposition. The idea that a human action could be called a "pre-moral evil," or evil in a pre-moral sense, is extremely confused. Examples which are offered to prove it are killing and amputation. Death is an evil, and a killing (of a human by a human) may not be a wicked action. For it may be blamelessly accidental and so, it is suggested, a 'pre-moral evil'. The amount of truth there is in this conclusion consists in this: the description "killing someone" may be the description of an act of a human being (actus hominis) without describing a human act (actus humanus)-as when I was a parcel rolling down the hill. Note that when it is the description of an act of a human being, even though not of a human act, it is still a moral actiondescription. At least this is so if I am right in my application of these terms. For we ask whether it was the person's fault, was there excuse, and so on. If, as in the parcel case, it turns out that the act of knocking the victim over to his death wasn't a human act, that is sufficient to answer those questions: it wasn't his fault, it was in no way voluntary on

Medalist's Address

57

his part. Only in the sense in which "responsibility" means "causality" was he responsible. The fact that "killing," even when applied to the interaction of two human beings, has this somewhat generic character, does not prove that a particular human action of killing someone is not established with a definite character of moral evil once it is clear that it was intentional and also was private-Le., was not an action belonging to the exercise of civil authority. I would prefer to say "a definite character of evil" because of my opinion that the term "moral" there is strictly redundant; but I have to put it in as I am commenting on a conception of 'pre-moral evil' according to which an action of killing someone, fully deliberate, fully intentional, may only be evil in a 'pre-moral' sense. That is, as so far described or proposed it may be so. Now this, if true, would not be true in the way in which the evil was pre-moral in the parcel case: namely, because, while death is an evil, it was not brought about by a human action. It would, rather, be true because of what is excluded by my word "private." Intentional killing in warfare, or by police in fighting violent criminals, or by execution of a death sentence-all these can be murderous, but they are not necessarily so, and that means that "intentional killing of one human being by another" is a moral action-description whose applicability does not eo ipsoprove the wickedness of the deed. The description is more determinate than is the bare word "killing"; yet it is still somewhat indeterminate. But this indeterminateness of description does not signify an indeterminateness in the quality of the human act, given the sanity of the killer and the circumstances which make it 'private'. If anyone thinks otherwise, he must have been misled by bad teaching. Or simply by a bad philosophic tradition, according to which the intentionalness of an action (a) can't be known to anyone but the agent and (b) is a matter of what the agent did it for-intention being often taken to mean purpose, or intention of the end.Now as to (a), "intention can't be known, because it is something private," that is in general absurd. It is often, nay usually quite apparent that someone is doing such-and-such on purpose. It is no objection to this that error on the point is possible. Nor yet, that it is more difficult to know someone's further intentions, or the intentions with which he was doing what he was obviously doing on purpose, or what he was doing it all for. Murder is a complex concept with many disparate elements in it. But you don't have to know what some private person killed his uncle

58

G. E. M. Anscombe

for in order to know he committed murder, so long as he was awake, compos mentis and was doing the killing on purpose. (I say "some private person" to exclude the cases of capital punishment, cops and robbers, and warfare all in a lump.) This means that although the mental element of mens rea is important in identifying a killing as a murder, the question whether someone has committed a murder can usually be answered as soon as the relevant facts are in, and having the facts in doesn't mean that all we have ascertained is the occurrence of a 'pre-moral evil'. Amputation-of a limb, a more pressing example for our immediate enquiry. This is because what someone is doing an amputation for is all-important. If it is needed for the life or health of the victim because of the physical condition or situation of his body, then it is all right as far as purpose goes, otherwise not. And this seems to lend real substance to the idea of a 'pre-moral evil' which is deliberately encompassed by a human action for the sake of a greater good. Someone cut off someone's hand, let us say. That sounds awful-could it be an indeterminate description of an action justified by circumstances: by who he was, and who his victim was? Suppose he was the public executioner and his victim a thief being punished according to law? If that would make it all right, then amputation as a moral problem would be rather like killing. But we don't have amputations as punishment 'in civilised countries'. I don't wish to raise the question whether such punishment is admissible at all: but the fact that we don't have it puts amputation into a special position: an evil inflicted, not as a punishment, but for the sake of a greater good in some other way. The evil is all in the particular thing that is purposely done, but the evil of it and the intentionalness do not determine the quality of the act. That is principally determined by what it was done for. So you have a pre-moral evil, and the goodness or badness of the action of doingit is principally a matter of the intention of the end. This gives us a pattern which, it might be said of, we can surely apply elsewhere: "If deliberately cutting off a hand is so far forth only pre-moral evil, then can't the same be said of killing your uncle, or aborting your baby or ... ? What will give it its moral character (and maybe a lovely one) is what you do it for." "Cutting off a limb" is a moral action-description by my criteria. It is also indeterminate, suggesting evil, raising the question of excuse or justification. Who did it to whom, in what circumstances, and what for?

Medalist's Address

59

When we know the answer to these questions we should be able to tell the quality of the act. The thing that makes it of importance for our discussion is the dominant role of the question "What for?" If this role is described by reference to 'a greater good', then the road is thrown open to a generalisation which threatens to destroy many absolute prohibitions of kinds of act. The important thing about a justified amputation-as we 'in civilised countries' understand it-is that it is for the life, health or capacityto-function of the individual human being who suffers it. The lesser evil for that human being's physical condition or integrity is chosen for the sake of the greater good of that human being's life or condition or capacity. So there really is here a case in which an intrinsic evil is inflicted for the sake of a more valuable good. But the evil and the good concern the bodily condition of the single human individual. He occupies a preeminent position in questions about good and bad action. Once this is clear, it should also be clear that in this case too we have an indeterminate description whose kind of indeterminateness is exploited to argue for a characterisation of many bad actions as so far 'pre-moral evils'. Because in this case the principal determinant is the 'what for', it is suggested that this is quite generally so. Old authorities are invoked, who wrote of all the factors that go to determine the specific goodness or badness of actions. These factors certainly include intention. Against the background of certain modern traditions in philosophy-especially the Cartesian-it is hardly noticed that intention may relate to the intentionalness of the particular act that is done, as well as to the purpose for which it is done. Now there are several kinds of action which, if they are done intentionally, are evidently evil actions, no matter what they are done for. The good end only sanctifies such means as either are good considered in themselves, or would naturally fall merely under a neutral action-description. My phrase "if they are done intentionally" signifies a sufficient condition, not a necessary one. This will become clear if we return to our earlier topic of murder. For here there can be borderline cases arising because murder is not committed only where there was an intention to kill. The arsonist who burns down a house, not caring that there are people there, is as much a murderer if they are burned to death by his action, as if he had aimed to kill them. This action falls squarely within a penumbra

60

G. E. M. Anscombe

surrounding the hard-core part of the concept of murder, which contains only intentional killing. The penumbra is fuzzy at the outer edges-that is, there are borderline cases. But that fact does not mean that an absolute prohibition on murder makes no sense. And similarly for other intrinsically evil deeds. The descriptions might be simple and easy to apply, like "sodomy," or have cases in which the application was somewhat obscure or disputable, like "lying"; but none of that creates a conceptual difficulty about the notion of absolute wrongness-whether great wrongness, or small as in many instances of lying. The fact that there is murder where death foreseeably results from one's action, without the actual intention of killing, naturally leads to a problem. One cannot say that no action may be done which foreseeably or probably leads to some death, or that all such actions are murderous. Why, the very begetting of a child would be murder at that rate-for the child will surely die. Or if that seems too zany an example: you can't build roads and fast vehicles, you can't have various sports and races, you can't have ships voyaging over the seas, without its being predictable that there will be deaths resulting. And much that is done in medicine and surgery is done knowing it involves the risk of death-painkilling drugs which may kill the patient before his disease does, and high-risk surgery. Therefore some distinction needs to be made, or some distinctions. This is a problem for those who hold that deliberately killing is absolutely out in the sort of case we may have to consider. If you don't hold that, it seems you don't get into a bind. For the whole thing begins by saying (a) You must not kill either as an end or as a means to some end. (b) Not all action that leads to someone's, or people's, death, though this is not intended, is murderous and so forbidden. (c) Often such action is murderous, and so is forbidden. It is here that we have the famous 'doctrine of double effect'. "Double effect" is an unfortunate Latinism. What we are talking about is death as a side effect which is brought about as well as the effect being aimed at.

Medalist'sAddress

61

I will call it the "principle of side-effects" that the prohibition on murder does not cover all bringing about of deaths which are not intended. Not that such deaths aren't often murder. But the quite clear and certain prohibition on intentional killing (with the relevant 'public' exceptions) does not catch you when your action brings about an unintended death. So much seems clear. But notice that the principle is modest: it says "where you must not aim at someone's death causing it does not necessarily incur guilt." The principle is unexceptionably illustrated by some examples of dangerous surgery, by some closings of doors to contain fire or water; or by having ships and airlines. In these we are helped by thinking of the deaths as either remote or uncertain. But an unintended death which is a foreseeable consequence of one's action may be neither. Then we are confronted with cases where it strikes people that there is little difference between direct and indirect killing. Imagine a pot-holer 2 stuck with people behind him and water rising to drown them. And imagine two cases: in one, he can be blown up; in the other, a rock can be moved to open another escape route, but it will crush his head. He will be killed by it. Someone who will equally choose either course clearly prefers saving these lives to the avoidance of any intentional killing. Such cases are discussed with a note of absolute necessity about saving lives-a presupposition that is paramount. The example of the stuck pot-holer was invented (without the choice of ways of escape) to illustrate the iniquity of abortion: you wouldn't say you could kill the pot-holer to get out-it was argued. But people did say just that-at least if the posture of the pot-holer was so described that he was going to get drowned too! Yet situations must often have occurred where people were in a tight corner, could have saved themselves by killing one of their number, and that has not occurred to them as an option at all. If that were the attitude of the people in the cave, they wouldn't do either thing. Indeed someone might say of moving the rock and crushing his head: "Isn't that direct killing too?" The point of not calling it direct was that it isn't his being crushed that gives the escape route-and so his being crushed can be something you don't intend.

62

G. E. M. Anscombe

"Direct" and "indirect" are dodgy terms; sometimes they relate to offshoots, as it were, from a given sequence of causes, and sometimes to immediacy or remoteness, and sometimes to what is intended or not. I note with curiosity that McCormick uses the concepts of direct and indirect intention, saying that they are centuries-old tools of moral theology. If that is so, I'd like to know the sources. I myself know of "direct" and "indirect" intention only as terms introduced by Jeremy Bentham. He called it "indirectly intended" if you took a shot at something and hit something else, if you knew that was a possibility. Fr. McCormick's sense seems to be the same. "Indirectly intended" thus means "Unintended, but the possibility was foreseen." While I am speaking of Fr. McCormick, I will also remark on his curious use of "intention in a psychological sense." 3 With respect, there is no other relevant sense. Returning to the pot-holer, I note that the 'doctrine of double effect' is supposed to say that the people could move the rock, though they must not, not on any account, blow up the pot-haler. And this is what is found intolerably artificial and unnatural. But we may ask: Is that because of what is allowed, or what is forbidden? What I've called "the principle of side effects" is related to an absolute prohibition on seeking someone's death, either as end or as means. I've spoken of that. Now, if the objection is to what is here allowed, note that the principle of side effects does not state or imply that they can move the rock. It does not say when you may foreseeably cause death. And why should we have a package deal here, as in what is called the "doctrine of double effect"? Now-perhaps because of this teaching-there might be people among those in the cave who, even seeing the undesired consequence, would move the rock, though they would refuse to blow the man up because that would mean choosing his death as means of escape. This is not a meaningless stance. They'd be shewing themselves as people who will reject any policy making the death of innocent people a means or an end. But, to repeat: you cannot deduce the permissibility of moving the rock from the principle of side effects. It determines nothing about that, except perhaps that it is not excluded on the score of being intentional killing. We have to say "perhaps" because of the possible closeness of the result. As I have described the case, we are only given that moving

Medalist's Address

63

the rock will crush the pot-holer's head. This might be a result so immediate that the action could not be called "taking the risk that that would happen." If so, then it is at best a dubious business to say "We don't intend that result." At most, it is no part of the aim, in securing the opening, that the man's head should be crushed. At this point the Doctrine of Double Effect helps itself to an absurd device, of choosing a description under which the action is intentional, and giving the action under that description as the intentional act. "I am moving what blocks that egress," or "I am removing a rock which is in the way." The suggestion is that that is all I am doing as a means to my end. This is as if one could say "I am merely moving a knife through such-and-such a region of space" regardless of the fact that that space is manifestly occupied by a human neck, or by a rope supporting a climber. "Nonsense," we want to say, "doing that is doing this, and so closely that you can't pretend only the first gives you a description under which the act is intentional." For an act does not merely have many descriptions, under some of which it is indeed not intentional: it has several under which it is intentional. So you cannot choose just one of these, and claim to have excluded others by that. Nor can you simply bring it about that you intend this and not that by an inner act of 'directing your intention'. Circumstances, and the immediate facts about the means you are choosing to your ends, dictate what descriptions of your intention you must admit. Nota bene that here "intention" relates to the intentionalness of the action you are performing as means. Suppose for example that you want to train people in habits of supporting the Church with money. If you exact money from them as a condition of baptism you cannot say you are not making them pay for it. All this is relevant to our pot-holer only where the crushing of his head is an immediate effect of moving the rock. Here a ground for saying you can intend to move the rock and not intend to crush the head is that you might not know that in moving the rock you would crush his head. That is true, we may suppose; which does differentiate this case from the simoniacal one. But if you do know, then where the crushing is immediate you cannot pretend not to intend it if you are willing to move the rock. Let us now consider the case where the result is not so immediate-the rock you are moving has to take a path after your immediate moving of it, and in the path that it will take it will crush his head. Here

64

G. E. M. Anscombe

there is indeed room for saying that you did not intend that result, even though you could foresee it. And that is the sort of case we have to consider. The Doctrine of Double Effect is supposed to allow you to move the rock, if the balance of good over bad results is favourable. The Principle of Side Effects says no more than that moving the rock is not exeluded by the prohibition on intentional killing. For, as I have explained it, that principle is not a package deal and it does not say what circumstances or needs excuse unintended causing of death. Some principle or principles are needed, and if we adopt that one principle, of the balance of good over evil in the expected upshot, then it becomes obscure why we could not do this where the causation of death was perfectly intentional. And that seems to be the principal ground on which some thinkers throw the whole package out of the window, and talk about a deliberate killing, for example, as sofar a 'pre-moral evil'. They may help themselves by the confused considerations I discussed earlier, but the nerve of the rejection of former doctrine is here. I formerly understood the doctrine of double effect, not as a package deal, but rather as what I have called the "principle of side effects," and I thought I was only doing a work of clarification in formulating it and remarking that it does not tell us what to allow and what to forbid when we have left the area of intentional killing.4 I have come to realise that the Doctrine of Double Effect comprises several things, merely includingthis 'principle of side effects', and that we should split it up. Having accepted the principle of side we need some further principle or principles on which to judge the unintended causing of death. There is one which both seems obvious and covers a good many cases. The intrinsic certainty of the death of the victim, or its great likelihood from the nature of the case, would exclude moving the rock. Here is a reasonable principle. Surgery would be thought murderous, even though it was not done in order to kill, but, say, to get an organ for someone else, if the death of the subject were expected as a near consequence, pretty certain from the nature of the operation. It will be apparent that this principle tells you rather what you can't do than what you can. Also, it is particularly devised for the causing of death; causing of other harm is not covered by it. I think that these are not faults. I will end by protesting at the ascription of the Doctrine of Double Effect to Aquinas: The phrase "duplex effectus" occurs in his discussion

Medalist's Address

65

of killing in self-defence. (Summa Theologica Ila Ilae, Q.64 art. 7) "Ex acu ergo alicuius se defendentis duplex effectus sequi potest: unus quidem, conservatio propriae vitae; alius autem, occisio invadentis." His doctrine is severe: he holds that one may not aim at the death of one's assailant, but one is guiltless if it occurs unintended as a result of one's use of means proportionate to the end of repelling him. One must moderate the force one uses, to suit the end; if one uses more than is required (si utatur maiori violentia quam opporteat) one's act is rendered illicit. Those who wished to see in this text the package Doctrine of Double Effect claimed that in speaking of proportionate means St. Thomas was introducing their doctrine of a proportion of good over evil in the upshot. I am not concerned to discuss St. Thomas' view on self-defence, only to note the false interpretation. If we want to know St. Thomas' general opinion on responsibility for evil consequences of actions, this is not the place to look, but rather at what he says about the relation of an eventus sequens to the goodness or badness of an action (Ia Ilae, Q.20 art. 5): Si est praecogitatus, manifestum est quod addit ad bonitatem vel malitiam. Cum enim aliquis cogitans quod ex opere suo multa mala possunt sequi nee propter hoc dimittit: ex hoc apparet voluntas eius esse magis inordinata. Si autem eventus sequens non sit praecogitatus, tune distinguendum est. Quia si per se sequitur ex tali actu, et ut in pluribus, secundum hoc eventus sequens addit ad bonitatem vel malitiam actus: manifestum est enim meliorem actum esse ex suo genere, ex quo possunt plura bona sequi; et peiorem ex quo nata sunt plura mala sequi. Si vero per accidens, et ut in paucioribus, tune eventus sequens non addit ad bonitatem vel ad malitiam actus: non enim datur de re aliqua secundum illud quod est per accidens, sed solum secundum quod est per se.5

NOTES 1. I use the words "act" and "action" interchangeably without intending any difference of sense.

66

G. E. M. Anscombe

A cave explorer. 3. My references are to his Ambiguity in Moral Choice. 4. See Chapter III, section of Euthanasia and Clinical Practice, The Report of a Working Party, published by The Linacre Centre, London, 1982. 5. Translation: If it (the consequent event) is pre-conceived, it manifestly adds to the goodness or badness of the action. For when someone considers that much that is bad can follow from what he does, and does not give it up on that account, this shows that his will is tlte more inordinate. But if the consequent event is not pre-conceived, then it is necessary to distinguish. For if it follows from that kind of action per se and in most cases, then the consequent event does accordingly add to the goodness or badness of the action; for it is clear that that action is better in kind, from which more goods can follow, and worse, from which more evils are liable to follow. But if it is per accidens, and in ratlter few cases, then the consequent event does not add to the goodness or to the badness of an action: for there isn't judgment on any matter according to what is per accidens,but only what is per se. 2.

FIVE

Morality, Action, and Outcome PHILIPPA

FooT

No decision is more important for practical ethics than that by which we come to embrace or reject utilitarianism. For although nonutilitarian principles are apparently deeply embedded in our ordinary morality, theoretical justification often seems hard to find; and some common intuitions are in danger of being disregarded on theoretical grounds. I want to consider two of these intuitions, and to defend them. The first is that there is a morally relevant distinction between what we do and what we allow to happen, and the second that there is a similarly relevant distinction between what we aim at and what we foresee as the result of what we do. I believe it is rather generally thought that the moral relevance of these distinctions is impossible to maintain. I shall, however, deny this, arguing that both differences are defensibly as well as widely recognised in the moral judgments we ordinarily make. Let us consider first of all the distinction between 'doing' and 'allowing', the moral judgments which seem to depend upon it, and the nature of the distinction itself. It is implied, it seems, in many decisions about what it is right to do, e.g. in cases dealt with in medical ethics. So, for instance, if some medical resource is in short supply and it would be possible to deploy it either to save a number of patients or to save one, then the policy would be to save as many as possible. It does not follow,

Reprinted from Morality and Objectivity: A Tribute to J.L. Mackie, edited by Ted Honderich (London and Boston: Routledge and Kegan Paul, 1985), 23-38, by permission of International Thomson Publishing Services Ltd.

68

PhilippaFoot

however, that any and every decision could rightly be taken which resulted in the same net saving of lives. For although it might be called a regrettable moral necessity that a smaller group should be left to die while a larger group was saved, the same could not be said in the case that the few were to be killed. It is not respectably believed that medical experimentation is justifiable so long as the benefit to some outweighs the cost to others; nor do those doctors whose patients will die without organ transplants think that they should secretly kill one healthy person if his body could save several. And while it might be right to leave one injured person to die by the road if hurrying to the rescue of several, it would be outrageous to drive over a recumbent person in order to reach them in time. It seems, therefore, that the 'negative duty' of noninterference is stronger than the 'positive duty' of assistance. 1 The same distinction between 'doing' and 'allowing' appears again when we think of what someone is obligated to sacrifice, on the one hand to avoid being the agent of some mischief to another and on the other to prevent or remedy the same evil when it is in no way due to him. We are inclined to think that no one is required to bring financial ruin on himself in order to give funds to Oxfam, but to export poisonous food even to stave off ruin is dearly another thing. So much is implied in our ordinary moral judgments; but it is not easy to see just what principle is at work Is the operative difference the difference between act and omission, between what someone does and what he refrains from doing? Obviously this cannot be it. For then it would be possible to change the moral character of certain trains of events by such simple expedients as building respirators which needed to be turned on each day. The difference between act and omission is in fact irrelevant to any moral issue except in so far as it corresponds to the distinction between allowing something to happen and being the agent to whom the happening can be ascribed. The difference we should have in mind is roughly this: in the case of allowing, a train of events must already have started or be on the horizon; an agent who could stop or prevent it does not do so, and therefore allows it to go on. In the other case it is he who initiates a sequence, as for instance by shooting someone who dies of the shotgun wound, or by pushing into the water someone who then drowns. 2 No one who has worked on these questions in vexed areas such as medical ethics can think that the distinction between allowing something to happen and being its agent

Morality, Action, and Outcome

69

is simple or clearcut, if only because what goes on in the world is not neatly divided into differentiated sequences. 3 Nevertheless it is from the moral point of view a different thing to start up a new train of events in someone else's life and to refuse to intervene to stop one. This may strike us again if we think about objections to paternalistic action. For while it is not normally permissible to seize an object from a man because he will harm himself with it, it does not follow that there is an obligation to help him to keep such a thing if it is falling out of his reach. 4 It seems, then, that the moral character of an action is on occasion affected by the position of the agent in the causal nexus: by the fact that he is on the one hand the initiating agent of a sequence or happening, or by contrast merely one who does not intervene. But whether someone's role in a train of events is permissive or initiatory is something that makes no difference to its outcome. The good or harm for which the action is a necessary or sufficient, or necessary and sufficient, condition may be the same in either case. And therefore utilitarians, who place the whole moral significance of an action in its production of good or harm, must treat the difference between initiating and allowing as having no independent influence on morality. So much, for the moment, about the first of the principles mentioned in the opening paragraph of the present piece as being inconsistent with utilitarianism. Let us now consider the second; namely the one that allows a moral distinction between what the agent 'directly intends', that is what he aims at either as means or end, and what he 'indirectly intends' in foreseeing it as a consequence of his action. The moral relevance of this distinction has often been challenged. 5 I think, however, that it must be allowed. To be sure it often makes no difference to the injustice of an action whether an injury which it causes is something the agent aims at or is something he foresees but has not made the object of his will. A merchant who sold food he knew to be poisonous in order to make money would be morally no better than an unemployed grave digger who deliberately killed to get trade. Nevertheless there are circumstances in which it is morally permissible to bring something about without aiming at it although it would not be morally permissible to aim at it; even though the balance of benefit and harm in the consequences remained the same. That this is so is proved, I think, by some facts about the permissibility of allowing an evil to come on some for the sake of saving others. For sometimes this is a regrettable moral

70

Philippa Poot

necessity, as in our previous examples having to do with scarce medical resources and with the person lying injured by the roadside. But it does not follow that it would be morally unobjectionable deliberately to leave someone unattended because his death would allow us to save others. We said earlier that it would be objectionable to kill even for such a good purpose, and now we must add that it would also be wrong to serve that same purpose by deliberately allowing someone to die. 6 There is, moreover, a very interesting and significant extension to the principle that bans direct intention in certain cases where indirect intention would be permissible. For it seems that what matters morally is not only how someone acts, which is what we have so far been considering, but also how his will is disposed even when this cannot affect the course of events. To see this one has only to think of a case in which someone is a spectator rather than a participant in a complex of circumstances where several lives might be saved through one man's death. For then one sees that while it is wrong to aim at the death of one in order to save others, and wrong of course to go along with the doing of this action by somebody else, it is also morally objectionable to be glad, as a spectator, if it is done; or indeed if such a thing comes to pass without human agency. No doubt some will deny this conclusion about the mere spectator's wishes; but simpler examples from more familiar surroundings might make them think again. For would it not be objectionable if, in a hospital where patients were dying for lack of a transplant, the death of someone in the hospital with several transplantable organs was automatically treated as good news? It follows from what has been said in the last two paragraphs that the recognition of a moral distinction between allowing and initiating evil is not able by itself to take care of all our non-utilitarian moral intuitions. But would the other principle, the one telling us that indirect intention may be permissible where direct intention is not, be enough on its own to do all the work that needs to be done? It is perhaps tempting to think so, but would nevertheless be wrong. For although many cases of the initiation of harm are also cases of direct intention the example of the wicked merchants shows that it is not always so. Moreover when the evil is merely foreseen and so not directly intended the morality of the action of which it is an effect may vary according to whether the agent is or is not the originator of that effect. So, although we may have to allow an individual to die because we must use the scarce medical

Morality, Action, and Outcome

71

resources for others it would nevertheless be impermissible to manufacture a curative substance for the same number at the cost of sending fumes to poison one. It seems, therefore, that we need both the principles discussed in preceding paragraphs, and may of course need others as well, in order to bring moral theory into line with our ordinary moral thoughts. There is, I think, no way of amending utilitarianism which will cure the discrepancies we have noticed. They will remain so long as the moral character of an action, motive, or any other subject of moral judgment is taken wholly to depend on its causal relation to the general welfare: whether welfare is conceived as pleasure, happiness, or preferential choice; whether maximum welfare is taken as an average or a sum total; and whether it is supposed to be distributed more or less equally. It is another question, however, whether the trouble is endemic towelfare utilitarianism in particular or whether it rather belongs to consequentialist theories as a whole. Might the situation be remedied if some other form of consequentialism replaced utilitarianism? \iVhat if, for instance, the violation of rights should be recognised as one element in the goodness or badness of states of affairs; the general idea being that the moral character of an action should depend, as in utilitarian philosophies, on their outcome-on the goodness or badness of states of affairs in which they play a causal, or a constitutive, role-but that the judgment of these states of affairs should no longer be directed solely to welfare? By this amendment it would become possible to count any violation of rights as itself contributing an element of badness to the total outcome of an action, and in principle in this 'goal rights system' such badness could be supposed to outweigh any good the action might produce. Such a provision would take care of many of our recalcitrant examples because they do in fact have to do with the violation of rights. 7 This interesting attempt to remain within the boundaries of outcome morality while throwing off the shackles of welfare utilitarianism has the great advantage of giving rights a more basic role than they can have within the latter theory. The theory fails, however, to accord with common morality in the verdict it implies in a special range of cases having to do with the causal interaction of one agent ·with another. Suppose, for instance, that you threaten (reliably) to kill or torture several people if I do not kill or torture one. Then it may be that by doing one of these actions I can lessen the total number done; and it would follow

72

Philippa Foot

in such a theory that whatever weight of badness is given to this kind of action it could be my duty to do it. 8 We can therefore see that no amendment of utilitarianism which retains its consequentialism can get away from the following implication: that there is nothing so bad that it cannot be done to prevent others from doing more things of the same kind. Consequentialist systems have this implication because in basing moral judgment solely on the evaluation of states of affairs they allow of no distinction between what an agent does himself and what he allows others to do. Each contributes its quota of goodness or badness to the total, along with other elements such as happiness or pain. So within a consequentialist system even of the goal rights kind a violation of rights such as murdering or torturing will be justifiable if the alternative is more murdering or torturing; and it follows that any powerful and ruthless person can make it right, can even make it our duty, to do these things, by threatening to do more of them himself. Some people may think that it would indeed be our duty to give in to him if enough were at stake; but if they are consistent consequentialists they must believe the more obviously bizarre proposition that any net balance of 'goodness' over 'badness' in the total outcome would do. So far as I know the only way to meet this objection from common morality while staying within anything that could possibly be called a consequentialist system is to introduce 'agent-relativity' into the evaluation of states of affairs. By this expedient each agent would be enabled to value his mm performance of an action differently from its performance by others; so that each could think it a worse total outcome when he killed or tortured than when someone else did, other elements of the situation always remaining the same. Amartya Sen has suggested this as a viable account of the matter. 9 But it seems to me that it is implausible. For it is not true that one thinks that a given act is worse when done by oneself than when done by another, unless of course there is some relevant difference between us, as when only one of us is a doctor or a parent or a friend. It is after all generally agreed that an action done by one individual is not, except in such cases, morally different from the same action as done by someone else. I do not refuse to kill or torture to prevent others from killing or torturing because I think that killing or torturing is, in the ordinary sense, worse when I do it than when they do. So what is it that one is supposed to think about one's own action as opposed to that of others? Sen speaks of 'evaluating' differently the

Morality, Action, and Outcome

73

states of affairs of which each is a part, suggesting as an illustration that if Othello and another gave a moral evaluation of the state of affairs in which Othello had murdered Desdemona the murderer himself would think it morally worse. 10 And he also uses the idea of an agent-relative moral judgment to explain, e.g., why someone with moral objections to a certain kind of work might refuse to do it himself though he felt no obligation to influence others similarly placed. The agent is supposed to think that it is worse if he takes the job than if someone else does, even when their situations do not relevantly differ Y It is to be morally worse from his point of view. But it is hard to know what this means. Sen suggests a comparison with an aesthetic proposition to the effect that an object is beautiful (he must surely mean 'looks beautiful') from a certain position. 12 But the comparison will not go through. Anything which is, visually, beautiful is so because it looks beautiful. This implies that it looks beautiful from some distances and from some angles but not that it must look beautiful from all; and the question therefore arises of whether it looks beautiful (or of how beautiful it looks) from some particular place. But it would be impossible even to start to give a similar explanation of moral badness from an agent's point of view because there is nothing that stands to being morally right or wrong as looking beautiful from certain positions stands to being beautiful. I believe therefore that Sen has failed to explain a concept which is central to his attempt to reconcile consequentialism with our moral intuitions. It seems, then, to go back to the main line of the argument, that it is impossible to believe in utilitarianism or any other form of consequentialism and at the same time to hold on to more or less well entrenched moral opinions. And this naturally raises the question: why believe in those theories? Why do they seem to be plausible? Why in particular does utilitarianism seem to be so in spite of the difficulties it creates for practical morality? Different people no doubt have their own reasons for believing in utilitarianism, and some think that they have proof of its truth. But many more appear to embrace the theory because they are convinced that utilitarian morality is the only rational morality, and we should ask about the cause of this conviction. It lies, I think, first of all in the fact that utilitarians insist that moral goodness must somehow be connected with what is good to and for human beings. But it also lies in the utilitarian, or basically consequentialist, idea that it must always be right for an agent to bring about the best state of

74

Philippa Foot

affairsthat is within his reach. And it is this last thought that makes the theory discordant with our ordinary moral intuitions. For the question raised is why any action should not be done if the doing of it will produce the best state of affairs or total outcome; and it is exactly this which tends to make us think that it must be morally permissible, or even obligatory, to do to individuals anything that 'needs to be done for the general good'. How can it be right, we ask ourselves, to choose to produce a state of affairs less good or worse than another that is equally within our reach? If this state of affairs is the best state of affairs then we ought to produce it. To be sure we are not always able to do what would bring about what is in other contexts referred to as the best state of affairs, but that is because there may be restraints on our actions coming e.g. from law, morality, or etiquette. When it is a question of bringing about the 'morally best state of affairs' there can be no such restrictions because only moral restraints could operate against morality, and only an irrational moral rule could stand against the production of the morally best state of affairs. It is such thoughts that tend to make utilitarianism seem uniquely rational and therefore irresistible. Those who hold out against it usually argue that it is not always right to produce, or try to produce, the best state of affairs; but naturally find it hard to see how indeed it can be right to prefer a worse over a better state of affairs. What does not occur to them is that with the posing of the question, which they thought to be on neutral territory, they are already deep within utilitarianism itself. This is what I shall now try to show. Let me begin this part of the discussion by making some observations about the peculiar and problematic character of judgments about good states of affairs when construed as they are apt to be construed by both sides in this debate. For of what form are they supposed to be? Clearly 'state of affairs' is not a description like 'game' or 'holiday' or 'committee meeting' giving the possibility of what Peter Geach has called an attributive judgment of goodness of the ordinary 'good F', 'good G' form. 13 But nor, again, does the description 'a good state of affairs' refer to a good like pleasure, or health, or affection, or understanding. So where do propositions about good states of affairs belong in the language? And in particular how do they get the connection which they are supposed to have with human choice? In the case of the two other types of judgments the relationship, although complex, is

Morality, Action, and Outcome

75

usually in principle fairly clear. Good Fs and good Gs may in certain cases (such as good roots and good claws) have no particular connection with the human will; but where a connection does exist it is not magical, nor merely linguistic, but depends upon the way in which those in some standard position are interested in choosing that type of person, role, or thing. And for the case of a good the conceptual connection with choice or reasons for choice though hard to explain, and still very much of a philosophical problem, obviously exists. For the reason for calling pleasure and health and affection and understanding goods is clearly something to do with the fact that they are goods to or for people; and thus give these people a reason to seek them out. How is it then with good states of affairs? We often speak of them in everyday life-such locutions are not the invention of philosophers-and nothing then seems clearer than the fact that we have reason to try to produce the states of affairs we call good and avoid the ones we call bad. But is not the explanation of this that in these pieces of everyday, non-philosophical, usage we use expressions such as 'a good state of affairs' and 'a good thing' as a way of saying how things fit in with our interests and with things we are interested in? And of course this makes these particular usages quite unsuitable as a pattern for the way in which the expressions might function in moral contexts, where a more impersonal reference would be required. 14 I do not know why these impersonaluses of 'good state of affairs' and likewise 'good thing' and 'It is good that ... ' are supposed to be unproblematic. It seems to me that there is every reason for being suspicious of them as they are talked about in philosophy, where it is supposed that judgments of this form, though having no original connection with anything that a particular agent sees as a good, or ,vith any object of his interest or desire, are nevertheless held to compel, persuade, or at least give him reason for choice. Let us consider, then, what use there can be in moral contexts for expressions such as 'a good thing' and 'a good state of affairs'. There seems to be no doubt that kindly people agreeing with each other in their attitudes will take for granted that certain occurrences-such as natural disasters for instance-produce 'a bad state of affairs'; and among those who accept morality a certain modicum of kindness will be assumed. It will also be taken for granted that the avoidance of suffering is among the ends which can determine what it is right to do.

76

Philippa Foot

This must be so given any morality in which benevolence is recognised as a virtue; and a further end is implied by the acceptance of a morality in which the virtue of justice is held to include fighting against injustice as well as performing just acts. We find these ends, and perhaps others, deeply embedded in our morality, and for this reason there is a special way of talking about states of affairs as good and bad, particularly when a moral issue is actually being discussed. In such contexts, unlike those of which we spoke earlier, there is implicit reference to ends belonging to moral persons as moral persons, rather than to idiosyncratic or personal aims. It is not, therefore, to be denied that there is a special way of talking about good and bad states of affairs in what we might broadly call moral contexts. But this in itself does not mean that we are back with the old puzzle about how it can be right not to do anything and everything which will be for most people's good. For it is one thing to think that morality dictates an end such as the happiness of others, and even directs us, on occasions, towards doing as much good as we can, and quite another to think that any action is morally permissible if directed to this end. For what we have said so far does not give any reason why the pursuit of the end should not be restricted by moral rules forbidding e.g. certain kinds of interference with individuals even for the sake of the general good. The fact that benevolence is a virtue giving persons who possess the virtue certain ends does indeed determine a special use of expressions such as 'a good state of affairs' in moral contexts. (And as already mentioned the virtue of justice can also play its part.) But there is nothing in the virtue of benevolence licensing us to say that it is 'a good thing' when a benefit comes to many through injustice to a few. So benevolence gives us no reason to say, for instance, that it would be a 'good state of affairs' or 'good total outcome' if the sacrificing of a few experimental subjects allowed us to get cancer under control. The operation of benevolence is circumscribed by justice, and even the end which the virtue prescribes is qualified, in that we are not told to be glad that good should come to some when it comes through evil to others. The fact that benevolence is a virtue, and a virtue which dictates attachment to the good of others, does not, then, give morality a universal end or goal; and the same is true of other parts of morality which have to do with what must be aimed at or desired. Thus the moral requirement to fight injustice does not imply that one must, or may, fight

Morality, Action, and Outcome

77

it by any means. And if there is a separate requirement having to do with the promotion of good action in others (which perhaps there is), this too can be circumscribed. So if, at this point, we return to consider the morality of doing some apparently bad action such as torturing, or killing innocent people, to stop someone else doing more acts of the same kind, and think that it must be right to do it, we can see where we have gone wrong. For whether we think of the moral objection to such actions in terms of suffering inflicted, injustice, or immorality, we have no reason to suppose that our sole moral duty is to minimise it. We are not even mandated to minimise immorality. For moral injunctions can be like orders which say not 'See to it that there is less shouting' (which might be obeyed by shouting) but rather 'don't shout'. There is thus nothing puzzling about the fact that many moral directives give rules for an agent's own conduct and tell him nothing about what he is to do about influencing the actions of others. In so far as one is required to see that other people behave well these requirements come from other parts of morality and have no priority over moral rules. The argument of the last few paragraphs has suggested that the use of the notion of a good state of affairs is very limited in moral contexts. This restriction may, however, be disputed, on the ground that any reasonablerecognition of rules in morality must allow that they may sometimes be broken, and must therefore deploy the notion of better and worse states of affairs in deciding when that may be. This has recently been argued by Amartya Sen who insists that any theory which allows deontological restraints to be overridden in certain circumstances must collapse into a form of consequentialism, because if it is determined that a constraint may be overriden in specified circumstances (as e.g. a right violated when it will save lives) then the badness of the violation is being weighed against that of non-violation. For example, it can be specified that if the badness of the state of affairs resulting from obeying the constraint exceeds some 'threshhold', then the constraint may be overridden. Such a thresholdbased 'constraint' system must rest ultimately on consequential analysis, comparing one set of consequences (badness resulting from obeying the constraint) with another (badness of violating the constraint itself, given by the threshold), and its distinguishing feature will be the particular form of the consequence-evaluation function. 15

78

Philippa Foot

This is an interesting argument, but I think it is mistaken. To meet Sen on his own ground I shall assume that moral rules can indeed be broken on occasions, which is certainly true of some even if not of all. So why do we say that in these circumstances this rule but not that one could be broken, or that this rule could be broken in these circumstances but not in those?Sen supposes that we make the decision by estimating the goodness and badness in the total outcomes of the alternative choices. But this seems to me to be either vacuous or wrong. If it merely says that our decision may be put in terms of 'better' and 'worse' ('It is better to do without a cure for cancer than to perform this experiment') it is vacuous. But if it says something about the way in which the decision is arrived at it is dubious. For what entitles Sen to assume that this is how we judge rightness, i.e. by comparing total outcomes; when what is, or should be, at issue is the concept of a good total outcome as intended here? It seems, then, that we can accept the idea that we talk in special ways in moral contexts about good and bad states of affairs without implying anything that could make the existence of non-utilitarian principles problematic. Some virtues do indeed give us aims, but nothing from within morality suggests the kind of good state of affairs which it would seem always to be our duty to promote. And why indeed should there be any such thing? It will be helpful here to consider a parallel between the case of morality and the case of etiquette. Why is it, we may ask, that we do not have good and bad states of affairs from the point of view of etiquette? The notion seems incomprehensible. But why is it so? It is so because a good state of affairs from the point of view of etiquette would be one which from the point of view of etiquette it must be right to aim at or produce. But in fact there can be no state of affairs which stands in this position because even if there are aims prescribed by etiquette (as e.g. avoiding causing social embarrassment), etiquette is also a matter of following rules, and the rules circumscribe the manner in which the aims may be followed. So what of morality? A morally good state of affairs would, it seems, a state of affairs which morally right or good action must aim at or produce. But since what is morally good or right is what a good moral system calls good or right there can be a state of affairs which stands in this position only if a good moral system contains no radically non-utilitarian rules. So if a good moral system does

Morality, Action, and Outcome

79

indeed contain rules of the kind we have been discussing then there cannot be a good state of affairs in the sense which consequentialism requires. There is, as we saw, a limited use for the concept within morality; but a virtue such as benevolence does not give an end to which all moral action must aspire. So far, then, the argument has tended to resist the encroachment of any form of consequentialism on the 'mixed' aim-and-rule morality that we actually seem to have. But nothing has yet been said of the rationale of this partly deontological system. If we do in fact recognise virtues such as justice which consist mainly in adherence to rules of conduct, as well as those like benevolence which we might call virtues of attachment, why do we do so? Will utilitarianism not re-establish itself when it comes to defending the morality we have? This is an important line of thought: it is responsible for the faith of many utilitarians, and the answer to it goes to the heart of the debate. So let us see why it is that a non-utilitarian morality might be thought to need a utilitarian defence. The idea is, I think that a morality can be shown to be a good morality only by being shown to do better than its rivals; that is to work better or produce a better state of affairs. And this generates the old problem about individual actions which do not, because of exceptional circumstances, have the usual effect. For then violation (perhaps secret violation) of the code could produce a better state of affairs than obedience, and it will seem irrational to insist on adherence nevertheless. The crucial assumption in this argument is the one that links the goodness of a moral system to a prior goodness in states of affairs. For states of affairs are things that can be brought about or influenced by actions, and this is why we have only to suppose strange causal circumstances to get a violating action producing 'the best state of affairs'. The question we should ask is, therefore, why this criterion of goodness should be assumed for a moral system. Perhaps some think that there just are good states of affairs and that this is a fact which can hardly be ignored in moral judgment, but I hope that the argument in the early part of this paper has helped to make this seem more problematic. Others believe that the moralpoint of view relates to the unique purpose of maximising welfare, and try to prove it, but it is significant that the general line of argument outlined above seems irresistible to many who are not actually satisfied by any such proof, and therefore have no

Bo

Philippa Foot

warrant such as Hare and Harsanyi believe themselves to have for operating with the idea of states of affairs which are good from the moral point of view. t(, But is it then so obvious that the criterion of goodness in a moral system must lie in its relation to a state of affairs that is judged to be good? Perhaps it will seem obvious so long as no alternative comes to mind. But of course there are alternatives which make a clean break with the idea of morally criteria! 'good states of affairs'. To sec that this is true we have only to look at the way in which Rawls, for instance, sees the distinction between a good moral system and a bad one. For in his moral philosophy any good moral system is necessarily a fair one, and a fair system one which operates on principles which would be chosen by rational agents in the original position. Thus a good moral system will be one that has a certain characteristic which is not a causal property, and there need be no place in this theory for an original judgment about good states of affairs. 17 The same is true of Scanlon's more thoroughgoing contractarianism, under which a moral rule's justification depends on its acceptability (a rather complex kind of acceptability) by, or on behalf of, every individual whatever his position in the world. 18 It is not necessary to commit oneself to either system to see in outline how a justification might be produced for the two principles of non-utilitarian morality discussed earlier in the present paper. Suppose for instance that we simply start from the rough idea that a good moral system must be one that could use a demand for reciprocity to urge conformity on any individual; and that the necessity that it produce benefit to him should be linked to this condition if to no other. 19 Then it would be intelligible that the more a morality rendered benefits from which each and every person stood to gain, the more acceptable, and so far forth the better, the system would be. And of course there are such benefits from the existence of a morality which refuses to sanction the automatic sacrifice of the one for the good of many because it secures to each individual a kind of moral space, a space which others are not allowed to invade. Nor is it impossible to see the rationale of the principle that one man should not want evil, serious evil, to come on another even to spare more people the same loss; it seems to define a kind of solidarity between human beings, as if there is some sense in which no one is to come out against one of his fellow men. 20 In both cases the good of the rule is a good that comes from having the system. But the

Morality, Action, and Outcome

81

justification is not, as with rules that limit the direct pursuit of the general good in rule utilitarian systems, that those who accept them will be most likely actually to bring about the best result. For if this were it the supposition of exceptional circumstances would give cases in which the agent would be obliged to break the rule. What I am suggesting is that the concept of'the best state of affairs' should disappear from moral theory, though not, as explained earlier, from all talk in moral contexts. If this would allow some of the insights of rule utilitarians to be validated, so much the better. But the starting point would be so different that the results would be unlikely to be the same. It has been suggested here that one criterion for a good moral system is that it should be possible to demand reciprocity from every individual because of the good the system renders to him. But I am sure that this is not the only condition for a good moral system. It has also, for instance, to be such that anyone can conform to it and still live well in the ordinary, non-moral, sense. 21 This condition may well be what limits the demands of altruistic action, and a whole new non-utilitarian enquiry should open up here.

NOTES

I want to thank Warren Quinn and David Sachs who have helped me by criticising a draft of this essay. 1. I have discussed this idea more fully in 'The problem of abortion and the doctrine of the double effect', Oxford Review, no. 5, 1967. Reprinted in Philippa Foot, Virtues and Vices (Blackwell, 1978). [Included here, chapter 8.] 2. In some cases the agent will not start a sequence but rather keep it going when it would otherwise have come to an end. Then he initiates a new stage of the sequence rather than the sequence itself. 3. One interesting complication is that diverting a harmful sequence from one victim to another seems not to be viewed in the same way as starting one up. A pilot whose plane is going to crash should steer from a more to a less inhabited area, and the principle apparently holds even in case it should be certain that someone will die. It does not follow, however, that one could, as it were, start a flood to stop a fire. 4. Cp. Foot, 'Killing, letting die, and euthanasia: a reply to Holly Smith Goldman', Analysis, vol. 41, no. 4 (June 1981). 5. See, e.g., H. L.A. Hart, 'Intention and Punishment', Oxford Review, no. 4, Hilary 1967. Reprinted in Hart, Punishment and Responsibility (Oxford, 1968). 6. In 'The problem of abortion and the doctrine of the double effect' I argued (wrongly as I now think) that the distinction between direct and indirect intention was

82

Philippa Foot

irrelevant to moral judgment. Not surprisingly I was then in difficulties about the wrongness of deliberately allowing a beggar to die in order to use his body for medical research. 7. Amartya Sen discusses goal rights systems in his excellent article 'Rights and agency', Philosophy and Public Affairs, vol. 11, no. 1. It would also, I suppose, be possible to recognise other non-welfarist values within a theory of this general type, saying for instance that there is more badness in the direct than in the indirect intention of harm. 8. Samuel Scheffler has an interesting discussion of this kind of case in The Rejection of Consequentialism (Oxford, 1982), especially chapter IV. 9. Op. cit., sections VI and VII. 10. Ib., pp. 29-30. 11. lb., pp. 24-6 and 32. 12. lb., pp. 35-6. 13. Peter Geach, 'Good and evil', Analysis, vol. 17, no. 2 (December 1956). 14. I have discussed this topic at more length in 'Utilitarianism and the virtues'; Presidential address to the American Philosophical Association, Pacific Division, 1983. Proceedings and Addresses of the A.P.A, vol. 57, no. 2 (November 1983). 15. Op. cit., pp. 6-7, note 8. 16. See R. M. Hare, Moral Thinking (Oxford, 1981), and John C. Harsanyi, 'Morality and the theory of rational behaviour', Social Research, Winter 1977, vol. 44, no. 4. Reprinted in Amartya Sen and Bernard Williams (eds), Utilitarinanism and Beyond (Cambridge, 1982). Hare's position has been criticised, to my mind effectively, by Thomas Nagel in The London Review of Books, 1-15 July 1982. 17. John Rawls, A Theory ofJustice (Harvard, 1971). 18. T. M. Scanlon, 'Contractualism and utilitarianism', in Sen and Williams, op. cit., pp. 103-28. I find this a bold and very suggestive piece. 19. Cp. Scanlon, op. cit., pp. 118-19. 20. Perhaps it is this idea that is partly responsible for the peculiar outrage that we feel about torture. The principle must obviously be qualified to take care of the case of guilty men. Yet witnesses to judicial executions have reported a primitive reaction of horror at the fact that no one went to help the man in the electric chair, as if it was impossible really to believe that the assembled company wanted him to die. 21. Nietzsche's suggestion that some men's lives are necessarily deformed by obedience to morality is not, I think, one that we could see as leaving morality intact.

PART

III

In Opposition to the Doctrine

SIX

ForeseenSide Effectsversus Intended Consequences JONATHAN

INTENTIONS

BENNETT

AND

FIRST-ORDER

MORALITY

Some philosophers have thought that morality forbids some conduct for reasons having to do with the concept of intention. In many possible crises, performing the atrocity will involve not merely bringing about something terrible but intending to bring it about as a means to averting a calamity; and this 'intending as a means' is thought lo make a big moral difference. There are sharp differences of opinion on this issue; properly to grapple with it, we need a good understanding of what it means to say 'Agent intends to bring it about that P'. I mean what he intends in or by