Death, Dying and the Ending of Life 9780754621744

Addressing key issues arising from the nature of death, 'Death, Dying and the Ending of Life' examines importa

1,601 60 26MB

English Pages 500 [1015] Year 2007

Report DMCA / Copyright

DOWNLOAD PDF FILE

Table of contents :
Death, Dying and the Ending of Life
Volume Set
Volume I
Half Title
Series Page
Title
Copyright
Contents
Acknowledgements
Series Preface
Introduction: Death and Consent
PART I DEATH
The Nature of Death
1 'On Defining a "Natural Death'", Hastings Center Report, 7, pp. 32-37.
2 'Why Is Death Bad?', Philosophical Studies, 50, pp. 213-21.
3 'Some Puzzles About the Evil of Death', Philosophical Review, 100, pp. 205-27.
Brain Death and the Uses of the Dead
4 'Brain Death and Personal Identity', Philosophy and Public Affairs, 9, pp. 105-33.
5 'Brain Death: A Durable Consensus?', Bioethics, 7, pp. 239-46.
6 'The Dead Donor Rule: Should We Stretch It, Bend It, or Abandon It?', Kennedy Institute of Ethics Journal, 3, pp. 263-78.
7 'Some Must Die', Zygon, 38, pp. 705–24.
PART II DECISION-MAKING AT THE END OF LIFE
Competent Patients
8 'Medical Paternalism', Philosophy and Public Affairs, 7, pp. 370-90.
9 'Arrogance', New England Journal of Medicine, 303, pp. 1507-11.
10 'Depression, Competence, and the Right to Refuse Lifesaving Medical Treatment', American Journal of Psychiatry, 151, pp. 971-78.
Advance Directives
11 'Do-Not-Resuscitate Orders: No Longer Secret, But Still a Problem', Hastings Center Report, 17, pp. 24-33.
12 'Advance Directives and the Personal Identity Problem's Philosophy and Public Affairs, 17, pp. 277-302.
13 'Why I Don't Have a Living Will', Hastings Center Report, 19, pp. 101-104.
Incompetent Patients
14 'Deciding for Others', Milbank Quarterly, 64, Supplement 2, pp. 17-94.
15 'The Severely Demented, Minimally Functional Patient: An Ethical Analysis', Journal of the American Geriatric Society, 88, pp. 938-44.
16 'Terminating Life-Sustaining Treatment of the Demented', Hastings Center Report, 25, pp. 25-31.
17 'Quality of Life and Non-Treatment Decisions for Incompetent Patients: A Critique of the Orthodox Approach', Law, Medicine and Health Care, 17, pp. 234–44.
18 'Continued Treatment of the Fatally Ill for the Benefit of Others', Journal of the American Geriatrics Society, 36, pp. 63-67.
19 'The Problem of Proxies with Interests of Their Own: Toward a Better Theory of Proxy Decisions', Journal of Clinical Ethics, 4, pp. 20-27.
20 'Courts, Gender and "The Right to Die"', Law, Medicine and Health Care, 18, pp. 85-95.
Children
21 'Moral and Ethical Dilemmas in the Special-Care Nursery', New England Journal of Medicine, 289, pp. 890-94.
22 'Involuntary Euthanasia of Defective Newborns: A Legal Analysis', Stanford Law Review, 27, pp. 213-69.
23 'Toward an Ethic of Ambiguity', Hastings Center Report, 14, pp. 25-33.
Futility
24 'Judging Medical Futility: An Ethical Analysis of Medical Power and Responsibility', Cambridge Quarterly of Healthcare Ethics, 5, pp. 23-35.
25 'Is the Treatment Beneficial, Experimental, or Futile?', Cambridge Quarterly of Healthcare Ethics, 5, pp. 248-56.
26 'Informed Demand for "Non-Beneficial" Medical Treatment', New England Journal of Medicine, 325, pp. 512-15.
27 'The Significance of a Wish', Hastings Center Report, 91, pp. 27-29.
Name Index
Volume II
Half Title
Series Page
Title
Copyright
Contents
Acknowledgements
Series Preface
Introduction: Justice and Hastening Death
PART III HASTENING DEATH
Killing vs Letting Die
28 'Active and Passive Euthanasia', New England Journal of Medicine, 292, pp. 78–80.
29 'The Intentional Termination of Life', Ethics in Science and Medicine, 6, pp. 59–64.
30 'The Ambiguity of Clinical Intentions', New England Journal of Medicine, 329, pp. 1039–40.
31 'Killing, Letting Die, and the Trolley Problem', Monist, 59, pp. 204–17.
32 'Taking and Saving Lives', Columbia Law Review, 93, pp. 1063–1156.
Physician-Assisted Suicide and Euthanasia
33 'Euthanasia', Philosophy and Public Affairs, 6, pp. 85–112.
34 'Some Non-Religious Views Against Proposed "Mercy-Killing" Legislation', Minnesota Law Review, 42, pp. 969–1042.
35 'Voluntary Active Euthanasia', Hastings Center Report, 22, pp. 10–22.
36 'It's Over, Debbie', Journal of the American Medical Association, 259, p. 272.
37 'Death and Dignity: A Case of Individualized Decision Making', New England Journal of Medicine, 324, pp. 691–94.
38 'Euthanasia: The Way We Do It, The Way They Do It', Journal of Pain and Symptom Management, 6, pp. 298–305.
39 'Doctors Must Not Kill', Journal of Clinical Ethics, 3, pp. 95–102.
40 'When Self-Determination Runs Amok', Hastings Center Report, 22, pp. 52–55.
41 'Assisted Death – A Compassionate Response to a Medical Failure', New England Journal of Medicine, 327, pp. 1384–88.
42 'A Right of Self-Termination?', Ethics, 109, pp. 606–28.
43 'The Supreme Court and Physician-Assisted Suicide: Rejecting Assisted Suicide But Embracing Euthanasia', New England Journal of Medicine, 337, pp. 1236–39.
44 'Physician-Assisted Suicide: Two Moral Arguments', Ethics, 109, pp. 497–518.
PART IV DISTRIBUTIVE JUSTICE AND DECISIONS ABOUT LIFE AND DEATH
Justice, Health and Death
45 President's Commission for the Study of Ethical Problems in Medicine and Biomedical and Behavioral Research (1983), An Ethical Framework for Access to Health Care, in Securing Access to Health Care, Washington, DC: US Government Printing Office, pp. 11–47.
46 'For and Against Equal Access to Health Care', Milbank Quarterly, 59, pp. 542–60.
47 'Class, Health, and Justice', Milbank Quarterly, 76, pp. 449–67.
48 'The Prostitute, the Playboy, and the Poet: Rationing Schemes for Organ Transplantation', American Journal of Public Health, 75, pp. 187–89.
49 'The Survival Lottery', in John Harris (ed.), Violence and Responsibility, London: Routledge and Kegan Paul, pp. 66–84, 84a.
Age Rationing
50 'Aging and the Ends of Medicine', Annals of the New York Academy of Sciences, 530, pp. 125–32.
51 'Am I My Parents' Keeper?', Midwest Studies in Philosophy, 7, pp. 517–40.
Is There a Duty to Die?
52 'Age Rationing and the Just Distribution of Health Care: Is There a Duty to Die?', Ethics, 97, pp. 317–40.
53 'Is There a Duty to Die?', Hastings Center Report, 27, pp. 34–42.
54 '"For Now Have I My Death": The "Duty to Die" versus the Duty to Help the Ill Stay Alive', Midwest Studies in Philosophy, 24, pp. 172–85.
Name Index
Recommend Papers

Death, Dying and the Ending of Life
 9780754621744

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

Death, Dying and the Ending of Life

The International Library of Medicine, Ethics and Law Series Editor: Michael D. Freeman

Death, Dying and the Ending of Life, Vols I and II Margaret P Battin. Leslie P Francis and Bruce M Landesman Abortion Belinda Bennett Health, Rights and Globalisation Belinda Bennett Health and Human Rights Rebecca Cook and Charles Ngwena Children, Medicine and the Law Michael Freeman Ethics and Medical Decision-Making Michael Freeman The Ethics of Public Health Michael Freeman Mental Illness, Medicine and Law Martin Lyon Levine The Elderly Martin Lyon Levine

Genetics and Gene Therapy Sheila A.M McLean Rights and Resources Frances H. Miller Organ and Tissue Transplantation David Price AIDS: Society, Ethics and Law Udo Schuklenk Women, Medicine, Ethics and the Law Susan Sherwin and Barbara Parish Legal and Ethical Issues in Human Reproduction Bonnie Steinbock Medical Practice and Malpractice Harvey Teff Human Experimentation and Research George F. Tomossy and David N. Weisstub Medicine and Industry George F. Tomossy

Death, Dying and the Ending of Life

Volume I

Edited by

Margaret P. Battin, Leslie P. Francis and Bruce M. Landesman University of Utah, USA

ASH GATE

©Margaret P. Battin, Leslie P. Francis and Bruce M. Landesman 2007. For copyright of individual articles please refer to the Acknowledgements. All rights reserved. No part of this publication may be reproduced, stored in a retrieval system or transmitted in any form or by any means, electronic, mechanical, photocopying, recording or otherwise without the prior permission of the publisher. Wherever possible, these reprints are made from a copy of the original printing, but these can themselves be of very variable quality. Whilst the publisher has made every effort to ensure the quality of the reprint, some variability may inevitably remain. Published by Ashgate Publishing Limited Gower House Croft Road Aldershot Hampshire GUll 3HR England

Ashgate Publishing Company Suite 420 I 0 I Cherry Street Burlington, VT 05401-4405 USA

Ashgate website: http://www.ashgate.com British Library Cataloguing in Publication Data Death, dying and the ending of life.- (The international library of medicine, ethics and law) I. Terminal care -Moral and ethical aspects 2. Euthanasia I. Battin, M. Pabst II. Francis, Leslie, 1946 TIT. Landesman, Bruce M. 179.7 Library of Congress Cataloging-in- Publication Data Death dying and the ending of life I edited by Margaret P. Battin, Leslie P. Francis, Bruce M. Landesman p. em.- (The international library of medicine, ethics and law) Includes index. I. Euthanasia- Moral and ethical aspects. 2. Assisted suicide- moral and ethical aspects. 3. Right to die- moral and ethical aspects. 4. Advance directives (Medical care) 5. Death. I. Battin, M. Pabst. II. Francis, Leslie, 1946- TIT. Landesman, Bruce M. IV. Series. [DNLM: I. Euthanasia- ethics. 2. Decision Making. 3. Right to Die- ethics. 4. Terminal Care- ethics. WB 60 D285 2007] R726.D4413 2007 179.7-dc22 ISBN 978-0--7546-2174-4

Printed in Great Britain by TJ International Ltd, Padstow, Cornwall

2007009794

Contents Acknowledgements Series Preface Introduction: Death and Consent PART I

IX XI

DEATH

The Nature of Death 1 Daniel Callahan ( 1977), 'On Defining a "Natural Death"', Hastings Center Report, 7, pp. 32-37. 2 Anthony L. Brueckner and John Martin Fischer (1986), 'Why Is Death Bad?', Philosophical Studies, 50, pp. 213-21. 3 Fred Feldman (1991), 'Some Puzzles About the Evil of Death', Philosophical Review, 100, pp. 205-27. Brain Death and the Uses of the Dead 4 Michael B. Green and Daniel Wikler (1980), 'Brain Death and Personal Identity', Philosophy and Public Affairs, 9, pp. 105-33. 5 Daniel Wilder (1993), 'Brain Death: A Durable Consensus?', Bioethics, 7, pp. 239-46. 6 Robert M. Arnold and Stuart J. Youngner (1993), 'The Dead Donor Rule: Should We Stretch It, Bend It, or Abandon It?', Kennedy Institute ofEthics Journal, 3, pp. 263-78. 7 Stuart J. Youngner (2003), 'Some Must Die', Zygon, 38, pp. 705-24. PART II

VII

5 11 21

47 77 85 101

DECISION-MAKING AT THE END OF LIFE

Competent Patients 8 Allen Buchanan (1978), 'Medical Paternalism', Philosophy and Public Affairs, 7, pp. 370-90. 9 Franz J. Ingelfinger (1980), 'Arrogance', New England Journal of Medicine, 303, pp. 1507-11. 10 Mark D. Sullivan and Stuart J. Youngner (1994), 'Depression, Competence, and the Right to Refuse Lifesaving Medical Treatment', American Journal of Psychiatry, 151, pp. 971-78. Advance Directives 11 Stuart J. Youngner (1987), 'Do-Not-Resuscitate Orders: No Longer Secret, But Still a Problem', Hastings Center Report, 17, pp. 24-33.

125 147 153

163

vi

Death, Dying and the Ending of Life, I

12 Allen Buchanan (1988), 'Advance Directives and the Personal Identity Problem', Philosophy and Public Affairs, 17, pp. 277-302. 13 Joanne Lynn (1991), 'Why I Don't Have a Living Will', Hastings Center Report, 19, pp. 101-104. Incompetent Patients 14 Allen Buchanan and Dan W. Brock (1986), 'Deciding for Others', Milbank Quarterly, 64, Supplement 2, pp. 17-94. 15 John D. Arras (1988), 'The Severely Demented, Minimally Functional Patient: An Ethical Analysis', Journal of the American Geriatric Society, 88, pp. 93 8-44. 16 Daniel Callahan (1995), 'Terminating Life-Sustaining Treatment ofthe Demented', Hastings Center Report, 25, pp. 25-31. 17 Rebecca S. Dresser and John A. Robertson (1989), 'Quality of Life and NonTreatment Decisions for Incompetent Patients: A Critique of the Orthodox Approach', Law, Medicine and Health Care, 17, pp. 234-44. 18 Mark Yarborough (1988), 'Continued Treatment of the Fatally Ill for the Benefit of Others', Journal of the American Geriatrics Society, 36, pp. 63-67. 19 John Hardwig (1992), 'The Problem of Proxies with Interests of Their Own: Toward a Better Theory of Proxy Decisions', Journal of Clinical Ethics, 4, pp.20-27. 20 Steven H. Miles and Allison August (1990), 'Courts, Gender and "The Right to Die'", Law, Medicine and Health Care, 18, pp. 85-95. Children 21 Raymond S. Duff and A.G .M. Campbell (1973), 'Moral and Ethical Dilemmas in the Special-Care Nursery', New England Journal ofMedicine, 289, pp. 890-94. 22 John A. Robertson (1974), 'Involuntary Euthanasia of Defective Newborns: A Legal Analysis', Stanford Law Review, 27, pp. 213-69. 23 John D. Arras (1984), 'Toward an Ethic of Ambiguity', Hastings Center Report, 14, pp. 25-33. Futility 24 Nancy S. Jecker and Lawrence J. Schneiderman (1995), 'Judging Medical Futility: An Ethical Analysis of Medical Power and Responsibility', Cambridge Quarterly ofHealthcare Ethics, 5, pp. 23-35. 25 Lawrence J. Schneiderman and Nancy S. Jecker (1996), 'Is the Treatment Beneficial, Experimental, or Futile?', Cambridge Quarterly ofHealthcare Ethics, 5, pp. 248-56. 26 Steven H. Miles ( 1991 ), 'Informed Demand for "Non-Beneficial" Medical Treatment', New England Journal ofMedicine, 325, pp. 512-15. 27 Felicia Ackerman (1991), 'The Significance of a Wish', Hastings Center Report, 91, pp. 27-29.

Name Index

173 199

205 283 291 299 311 317 325

339 345 403

415 429 439 443 447

Acknowledgements The editor and publishers wish to thank the following for permission to use copyright material. American Psychiatric Publishing, Inc. for the essay: Mark D. Sullivan and Stuart J. Youngner (1994), 'Depression, Competence, and the Right to Refuse Lifesaving Medical Treatment', American Journal of Psychiatry, 151, pp. 971-78. Copyright © 1994 American Psychiatric Association. American Society of Law, Medicine and Ethics for the essays: Rebecca S. Dresser and John A. Robertson ( 1989), 'Quality of Life and Non-Treatment Decisions for Incompetent Patients: A Critique of the Orthodox Approach', Law, Medicine and Health Care, 17, pp. 234--44; Steven H. Miles and Allison August ( 1990), 'Courts, Gender and "The Right to Die"', Law, Medicine and Health Care, 18, pp. 85-95. Blackwell Publishing for the essays: Michael B. Green and Daniel Wilder (1980), 'Brain Death and Personal Identity', Philosophy and Public Affairs, 9, pp. l 05-33; Allen Buchanan (1978), 'Medical Paternalism', Philosophy and Public Affairs, 7, pp. 370-90; Allen Buchanan (1988), 'Advance Directives and the Personal Identity Problem', Philosophy and Public Affairs, 17, pp. 277-302; Daniel Wilder (1993), 'Brain Death: A Durable Consensus?', Bioethics, 7, pp. 239--46; John D. Arras (1988), 'The Severely Demented, Minimally Functional Patient: An Ethical Analysis', Journal of the American Geriatric Society, 88, pp. 93 8--44. Copyright © 1988 American Geriatrics Society; Mark Yarborough (1988), 'Continued Treatment of the Fatally Ill for the Benefit of Others', Journal ofthe American Geriatrics Society, 36, pp. 63--67. Copyright © 1988 American Geriatrics Society; Allen Buchanan and Dan W. Brock (1986), 'Deciding for Others', Milbank Quarterly, 64, Supplement 2, pp. 17-94. Copyright © 1986 Millbank Memorial Fund. Cambridge University Press for the essays: Nancy S. Jecker and Lawrence J. Schneiderman ( 1995), 'Judging Medical Futility: An Ethical Analysis of Medical Power and Responsibility', Cambridge Quarterly of Healthcare Ethics, 5, pp. 23-35. Copyright © 1995 Cambridge University Press; Lawrence J. Schneiderman and Nancy S. Jecker (1996), 'Is the Treatment Beneficial, Experimental, or Futile?', Cambridge Quarterly of Healthcare Ethics, 5, pp. 24856. Copyright© 1996 Cambridge University Press. Copyright Clearance Center for the essay: John A. Robertson (1974), 'Involuntary Euthanasia of Defective Newborns: A Legal Analysis', Stanford Law Review, 27, pp. 213-69. Hastings Center for the essays: Daniel Callahan ( 1977), 'On Defining a "Natural Death"', Hastings Center Report, 7, pp. 32-37; Stuart J. Youngner (1987), 'Do-Not-Resuscitate Orders: No Longer Secret, But Still a Problem', Hastings Center Report, 17, pp. 24-33; Joanne Lynn

viii

Death, Dying and the Ending of Life, I

(1991), 'Why I Don't Have a Living Will', Hastings Center Report, 19, pp. 101-104; Daniel Callahan (1995), 'Terminating Life-Sustaining Treatment of the Demented', Hastings Center Report, 25, pp. 25-31; John D. Arras (1984), 'Toward an Ethic of Ambiguity', Hastings Center Report, 14, pp. 25-33; Felicia Ackerman (1991), 'The Significance of a Wish', Hastings Center Report, 91, pp. 27-29. Johns Hopkins University Press for the essay: Robert M. Arnold and Stuart J. Youngner (1993), 'The Dead Donor Rule: Should We Stretch It, Bend It, or Abandon It?', Kennedy Institute of Ethics Journal, 3, pp. 263-78. Copyright© 1993 Johns Hopkins University Press. Journal of Clinical Ethics for the essay: John Hardwig ( 1992), 'The Problem of Proxies with Interests of Their Own: Toward a Better Theory of Proxy Decisions', Journal of Clinical Ethics, 4, pp. 20-27. Massachusetts Medical Society for the essays: Steven H. Miles ( 1991 ), 'Informed Demand for "Non-Beneficial" Medical Treatment', New England Journal of Medicine, 325, pp. 512-15. Copyright© 1991 Massachusetts Medical Society; Franz J. Ingelfinger (1980), 'Arrogance', New England Journal of Medicine, 303, pp. 1507-11. Copyright © 1980 Massachusetts Medical Society; Raymond S. Duff and A.G.M. Campbell (1973), 'Moral and Ethical Dilemmas in the Special-Care Nursery', New England Journal ofMedicine, 289, pp. 890-94. Copyright © 1973 Massachusetts Medical Society Philosophical Review for the essay: Fred Feldman (1991 ), 'Some Puzzles About the Evil of Death', Philosophical Review, 100, pp. 205-27. Copyright© 1991 Cornell University. Springer for the essay: Anthony L. Brueckner and John Martin Fischer (1986), 'Why Is Death Bad?', Philosophical Studies, 50, pp. 213-21. Copyright© 1986 D. Reidel Publishing Company. University of Wisconsin Press for the essay: Stuart J. Youngner (2003), 'Some Must Die', Zygon, 38, pp. 705-24. Copyright © 2003 University of Wisconsin Press and Board of Zygon. Every effort has been made to trace all the copyright holders, but if any have been inadvertently overlooked the publishers will be pleased to make the necessary arrangement at the first opportunity.

Series Preface Few academic disciplines have developed with such pace in recent years as bioethics. And because the subject crosses so many disciplines, important writing is to be found in a range of books and journals, access to the whole of which is likely to elude all but the most committed of scholars. The International Library of Medicine, Ethics and Law is designed to assist the scholarly endeavour by providing in accessible volumes a compendium of basic materials drawn from the most significant periodical literature. Each volume contains essays of central theoretical importance in its subject area, and each throws light on important bioethical questions in the world today. The series as a whole, makes available an extensive range of valuable material (the standard 'classics' and the not-so-standard) and should prove of inestimable value to those involved in the research, teaching and study of medicine, ethics and law. The fifteen volumes together- each with introductions and bibliographies- are a library in themselves- an indispensable resource in a world in which even the best-stocked library is unlikely to cover the range of materials contained within these volumes. It remains for me to thank the editors who have pursued their task with commitment, insight and enthusiasm, to thank also the hard-working staff at Ash gate- theirs is a mammoth enterprise -and to thank my secretary, Anita Garfoot for the enormous assistance she has given me in bringing the series from idea to reality. MICHAEL FREEMAN Series Editor F acuity of Laws University College London

Introduction Death One of the initial issues in the field of bioethics was death itself. What is it - physically, metaphysically and morally? For what purpose is a declaration of death required: cessation of medical treatment, burial, grief, organ transplantation, inheritance or a decision to charge the offence of assault or homicide? Is death an evil, the consummate evil, a blessing, or of little or no significance to the person whose death it is? Some of bioethics' debate about the nature and definition of death was spurred by new possibilities of organ transplantation arising in the twentieth century, but equally important, from the very beginning of religious and philosophical discussion, has been the issue whether death is a harm and thus whether the preservation of life has value in its own right- regardless of the quality of the life or the choices ofthe person whose life it is. The Greek philosopher Epicurus and the Roman philosopher Lucretius argued two millennia ago that death is not bad for the person who dies. Death is non-existence. Persons who do not exist cannot feel pain, even the pain of death, since they do not exist at all. These philosophers were hedonists: that is, they believed that pleasure is the only good and pain is the only bad. Since ancient times, philosophers have sought to answer the Epicurean challenge by providing an account of why death is bad. Historical and contemporary challenges to the Epicureans include rejecting hedonism and rejecting the metaphysical view that the person ceases to exist at death. Several recent philosophers have tried to answer the Epicureans on their own terms, without rejecting either hedonism or the metaphysical assumption that death is the end of the person who dies. Part I of this volume collects classic essays about the nature of death, about defining death, and about resulting issues concerning organ donation. These include some of the most influential pieces from the philosophical literature about the meaning of death itself, a topic that is always not far from the surface in debates in bioethics that range as far as cloning and life-extending technologies.

The Nature ofDeath In one of the seminal essays about what deaths are bad, Daniel Callahan (Chapter 1) develops a definition of a 'natural death'. Callahan begins with four images of dying: an anticipated death accompanied by graceful goodbyes, a young person struck down suddenly, a painless death of an infant, and a death of a very elderly person after a long period of physical and mental decline. He uses these images to construct a concept of a 'natural' death, a death that occurs when one's life work has been accomplished, when one's moral obligations to others are completed, when one's death is not seen as an offence to sensibility or a reason for rage by others, and when one's process of dying is not marked by degrading or unbearable pain. Callahan's contrast is between a 'natural' death and a death that seems 'wrong', or even

xii

Death, Dying and the Ending of Life, I

metaphysically 'evil'. Callahan admits ambivalence about whether any death is an 'evil' in the metaphysical sense, but contends that we intuitively differentiate between premature deaths and deaths that come at the end of a full life. He draws public policy consequences from this distinction: most importantly that there is no right to a natural death - it should instead be regarded as an ideal- and that society should not pursue immortality or a significant extension ofthe lifespan. Instead, efforts should be devoted to supporting people over a full life and not abandoning those near life's end to abuse or pain. In Chapter 2 Anthony Brueckner and John Fischer take up the question left open by Callahan- the Epicurean challenge, why death is a bad thing for the person whose death it is. There are obvious ways in which death could be bad: the process of dying could be horribly painful, or death might open the door to an afterlife of Hell. But these accounts attribute the badness of death to the unpleasant experiences that accompany it; Brueckner and Fischer explore whether death itself is bad for the person whose death it is, without accompanying bad experiences. Thomas Nagel's answer is that death is a deprivation of the good things of life. But this answer must confront the Epicurean challenge that it appears to assume an irrational asymmetry between prenatal non-existence and death after life. After all, if we regard death as bad because it is an absence of the good things of life, why should we not also regard prenatal non-existence as bad because it too is an absence of the good things of life, just at an earlier time? Derek Parfit argued that the asymmetry reflects a rational bias towards the future: we fear the prospect of future pain in a way we do not fear the knowledge that we experienced pain in the past. This reply, Brueckner and Fischer contend, trades on the anticipation of bad experiences to come; it thus does not show that the asymmetry is rational without the accompaniment of bad experiences. Another account of the asymmetry is that the deprivation of goods a person now possesses is bad for him in a way that the absence of the goods is not. Brueckner and Fischer counter that this account, too, capitalizes on the unpleasantness of experiences; what makes deprivation worse is the disappointment of not having what one once had. An alternative explanation of the asymmetry, Brueckner and Fischer contend, is that we care about the possibility of goods to be experienced in the future in a way that we do not care about goods that have already been experienced. It is thus rational for us to want more of life in the future instead of more of life in the past, and that is why we regard death as bad. Fred Feldman, in Chapter 3, also takes up the Epicurean challenge to explain why death is a misfortune for the person who dies. The Epicureans pose a series of puzzles about the badness of death. If the dead do not exist, how can death be a misfortune for them? If death deprives someone of goods- goods anticipated (as in the view of Brueckner and Fischer), goods that would have been enjoyed or goods of some other sort- how can that loss be bad unless it leaves the person worse off than he would have been had he not died? But this comparison would appear to be incoherent if the dead do not exist- there is no person who is worse off, to be compared with a person who is better off. And when is the person made worse off by death: before death, when she has not died -or after death, when she does not exist? To answer these puzzles, Feldman constructs a possible worlds axiology. Suppose there are two possible worlds, each with a given amount of pleasures and pains. Then the world in which there is more pleasure is better than the world in which there is less pleasure. Or suppose there are two possible worlds, in each of which a given person experiences different welfare levels, based on the pains and pleasures she experiences in that world. Then, the world in which she experiences the higher welfare level is better for her than the world in which she

Death, Dying and the Ending of Life, J

xiii

experiences the lower welfare level. States of affairs can be compared for persons in terms of the value of possible worlds for them. For example, suppose we wanted to compare the state of affairs for a person in which she undergoes chemotherapy for breast cancer with the state of affairs for a person in which she refuses the chemotherapy; we would then compare the welfare level for her in the nearest possible world to the actual world in which she undergoes the therapy, with the welfare level for her in the nearest possible world to the actual world in which she does not. If she experiences a higher welfare level in the first world, it would be better for her to undergo the chemotherapy. In the same way, the possible world in which someone dies at a given time can be compared to the possible world in which his death does not occur at that time. Feldman provides an explanation of how states of affairs can be comparatively better or worse for a person- not of how something can be intrinsically bad for a person (like the pain of a tumour) or cause something that is intrinsically bad for a person (like the tumour causing the pain). Feldman attributes the Epicurean puzzles to confusion between what is intrinsically bad for someone (feeling pain) and the relative goodness or badness of states of affairs for that person. His account explains how the state of affairs in which a person dies is bad for him- but not how death is intrinsically bad for the person who dies in the same way that being in pain is intrinsically bad. It also explains when death is bad for the person: because the comparison between possible worlds holds at all times, if the world in which a person dies young is worse for her than a world in which she does not, death is bad for a person 'eternally'. To at least this extent, the asymmetry of our emotional responses to time lost by a later birth and time lost to an earlier death is irrational. Brain Death and the Uses of the Dead The vastly increased potential for the transplantation of human organs opened up the question of standards for determining death. If organs continue to be perfused by circulation and respiration, prospects for transplant success are significantly increased. But a patient whose circulation and respiration continue, even with mechanical support, has not died by traditional cardio-respiratory criteria. And such a patient looks and feels alive: chest rising and falling, breath condensing on a mirror, hands and face warm to the touch. So an alternative proposal for determining death was developed. If a person has suffered irreversible cessation of all brain function, including the function of the brain stem, she is dead, just as if she had already suffered the cessation of cardio-respiratory function that would surely follow without medical support. Many writers have considered the plausibility of thus regarding 'whole brain death' as an alternative to the declaration of death when the patient's heart has ceased to beat and her lungs no longer draw air. Some writers argue that 'whole brain death' should not be viewed as death as long as bodily functions continue, mechanically supported or not. Others argue that 'higher' brain death- complete loss of all cognitive function- is the concept that really should be identified with the death of a person. Daniel Wikler has been a principal proponent of the latter view, arguing that the concept of whole brain death is conceptually confused and inherently unstable. In Chapter 4 Michael Green and Daniel Wilder contend that the standard reasons offered for regarding whole brain death as death- medical equivalence and moral equivalence - are misguided. Instead, they argue for an ontological account of brain death: that it represents the

xiv

Death, Dying and the Ending of Life, I

end of the life of the person. The principal medical-equivalence argument is that since the cessation of cardio-respiratory functions is likely to follow very quickly upon the cessation of the function of the entire brain, in the absence of mechanical intervention, the two should be regarded as functionally the same. However, we do not typically conclude that a person has died because other vital functions- such as kidney function- require mechanical support to continue. The cessation of other bodily functions - for example, neural connections to the heart or muscular capacity for breath - might portend cardio-respiratory cessation, as does brain death - but that does not mean they should be equated with that death. The moral argument is that it is morally right to discontinue care of a brain-dead patient. But this puts the cart before the horse; the moral rightness of discontinuing care is a conclusion reached because the patient has died (although, of course, these are not the only grounds on which the cessation of care might be defended). If, for example, we believe that life should always be preserved, then the question remains as to what is 'life' and what is 'death', and whether brain death is death has not been settled. Neither medical nor moral arguments succeed in capturing the ontological point that the body of a brain-dead patient no longer remains the body of the person it was, or so Green and Wikler contend. Their view is that the existence of a person presupposes the possibility of psychological properties (the further view, that we identify a person as a given individual say, Jones- by whether the individual has psychological properties related by causal processes to Jones, is not necessary for this point). An individual who has lost all brain function lacks the capacity for psychological properties and thus cannot be a person at all. Thus the person ceases to exist at the time the capacity for having psychological properties ends. Green and Wikler defend this thesis about personhood through complex thought experiments about when we would regard someone as the same person if memories were lost, brains were transplanted and the like. This account leaves open, however, whether it is morally or legally appropriate to continue to maintain the vital functions of the brain-dead. It also entails that complete cessation of all higher brain function ('neocortical' or 'higher brain' death) amounts to the death of the person. Green and Wikler believe that there is no obligation to continue to treat patients who have lost all capacity to benefit from care in the sense that they cannot experience the benefits of care. However, there is a good reason to draw a legal line between those who have lost function of the entire brain and other patients: because people regard the loss of all brain function as equivalent to death, defining death as death of the entire brain, including the brain stem, does not risk moving down a causal slippery slope to more widespread acceptance of euthanasia. Some I 3 years later, Daniel Wikler (Chapter 5) returned to the issue of whether drawing a line at declaring death at complete and irreversible cessation of all functions of the entire brain, including the brain stem, courts a slippery slope towards euthanasia. Recognizing the immense benefits that have come from the organs made available for transplantation once brain death has been accepted, Wikler nonetheless worries about the risks of drawing boundaries on such utilitarian grounds. He renews his attack on the intellectual coherence of medical and moral arguments for the standard of whole brain death. Why should whole brain death- cessation of function of the entire brain, including the brain stem- be regarded as death? The standard answer is that with such cessation of functioning, cardio-respiratory failure will ineluctably follow, without mechanical intervention. But this connection will occur for other patients too -patients with amyotrophic lateral sclerosis (Lou Gehrig's disease), for example, will cease

Death, Dying and the Ending of Life, J

XV

to breathe without the mechanical intervention of a respirator. Conceptual coherence requires that we draw the line at higher brain death, but this line may be subject to risks of a slippery slope because it is less intuitively obvious to the public. So, Wikler contends, we should continue to permit transplantation when patients meet whole brain death criteria, but we should be wary of two concerns. The first is that the instability of the whole brain definition may actually impede organ donation; family members of patients who meet the criteria may find it difficult to regard them as 'dead'- warm-bodied as they are. The second is that we may eventually be able to diagnose higher brain death with complete certainty and we then may need to reassess our insistence on the standard of whole brain death. The literature, as well as medical practice, is replete with confusion about the description of patients who are brain dead, in permanent or persistent vegetative states, or comatose. Even medical professionals have difficulty in referring to patients who meet criteria for whole brain death as deceased. Family members understandably are troubled when a patient who is still breathing, however mechanically, is described as 'dead'. Descriptions in the media of patients in permanent vegetative states - such as the American patient Terri Schiavo - often confuse whole brain death, the permanent vegetative state (in which the patient has lost so much higher brain tissue that there is no possibility of higher brain function or its return), the persistent vegetative state (in which loss of higher brain function has been diagnosed but its irrevocability has not yet been determined) and coma (in which a patient may suffer from varying degrees of unconsciousness). Taking the instability of the definition of brain death still further, Robert Arnold and Stuart Youngner, in Chapter 6, observe that the 'dead donor' rule requires that patients be declared dead before organs may be harvested for transplant. Coupled with a prohibition against active euthanasia and the requirement of consent to donation from patient or family, the dead donor rule seems an important safeguard against the abuse of patients who are near death. It is also useful reassurance to patients and their families that hospitals do not kill for the sake of obtaining organs. Recent protocols for organ donation from 'non-heartbeating' donors appear to put pressure on this framework, however. Once a decision has been made to withdraw life-sustaining treatment from a patient who has been sedated for that purpose, these protocols would permit organ transplantation to occur after the patient's heart has stopped for a determinate period of time (usually two to six minutes), the patient has been declared dead and the mechanical support has been restarted to allow organ perfusion. Arnold and Youngner explore whether developments such as the non-heartbeating donor protocol suggest that we should abandon the dead donor rule. These debates are ongoing; non-heartbeating donor protocols have become common, but medical practice and organizations such as the American Medical Association continue to oppose vital organ donation from patients such as anencephalic infants who have not yet been declared 'dead'. In his later essay 'Some Must Die' (Chapter 7) Stuart Youngner turns to the human side of organ transplantation. Youngner is a psychiatrist, and he explores the meaning of organ donation as it is presented in the experiences of donors and recipients, in the media and in literature and film. Rationally, we think of organ donation as a beneficial process, making what good can be made out of tragedy. Yet the fear that organs will be stolen surreptitiously is the stuff of horror movies. Non-heartbeating donor protocols may exacerbate this dissonance and, Youngner contends, it may take some time for society to adjust.

xvi

Death, Dying and the Ending of Life, I

Decision-Making at the End of Life

In the United States at least, bioethics came into being as part of the civil rights movement. The rights of patients to control their own bodies- to resist unwanted treatment, to exercise reproductive liberty and, later on, perhaps even to control the time and manner of their dying- were core to the field. Within a very few years, medical practice in the United States shifted from paternalistic refusal to tell patients the truth to insistence on infonned consent to treatment. Protection of the autonomy of research subjects to choose whether to enter studies or to withdraw from them, to know that they were being asked to enter into a research study and to be infonned about risks and benefits was at the core of international and national regulatory regimes governing research with human subjects. And, in law, the standard for a patient to recover damages for hanns resulting from a failure of informed consent shifted from what a reasonable physician would disclose to what a reasonable patient would want to know. Competent Patients

Behind this insistence on individual rights to control their health care was the rejection of paternalism: the idea that it is permissible to coerce people to do what would be good for them, even if they don't see things that way. Two forms of paternalism are generally recognized: 'soft' paternalism licenses interference when the individual's decision-making processes are flawed- when the individual, if he had the appropriate knowledge, time or calm, would agree with what he is being required to do. 'Hard' paternalism licenses interference on the basis that it would be for the individual's good, even if she has considered the matter and concluded otherwise without such decision-making flaws. Allen Buchanan's 'Medical Paternalism' (Chapter 8) is perhaps the classic essay dissecting and rejecting the arguments for medical paternalism. Writing when paternalism was still the dominant paradigm in medical practice, Buchanan aimed to undermine the paternalist's own argument, rather than to advance a view about rights that supporters of paternalism might reject. He examines three arguments. First, defenders of paternalism contend that telling patients the truth will harm them and thus violate the duty of health-care providers not to harm their patients. Buchanan points out that this argument is invalid; it requires a further assumption to the effect that telling the patient the truth will do more harm than withholding the information will do. But there is no reliable evidence that supports this generalization about patients, much less about family members who are decision-makers for incompetent patients such as children. To make such comparative judgements in particular cases, physicians would need to know intimate details of their patients' lives. A second argument is that the physician-patient relationship is implicitly contractual, representing an agreement that the physician is to protect the patient. Buchanan finds no reason to think that contracts with such terms either are or should be implicit in the physician-patient relationship. A final argument for paternalism is that patients cannot understand the truth about their conditions. This argument, too, represents an unwarranted generalization. Even worse, it may create a self-fulfilling prophesy, if physicians fail to explain matters to their patients from the belief that their patients will not understand. Buchanan's essay laid out the problems with paternalism. As the debates about paternalism have continued, discussions have become increasingly nuanced about both the importance of

Death, Dying and the Ending of Life, J

xvii

autonomy and the strains upon it in medical contexts where patients are vulnerable and afraid. Franz lngelfinger's 'Arrogance' (Chapter 9) is one of the most powerful descriptions of the situation of a patient faced with a terminal illness. lngelfinger had been editor of the New England Journal of Medicine and had specialized in the area treating his disease- oesophageal cancer. Yet, facing an illness about which he was especially knowledgeable, he did not want to be informed about options and enabled to choose. Instead, he wanted advice, from a 'doctor', someone he could trust. Paternalism can be arrogant, but it can also be arrogant to leave a patient to his own devices in deciding what to do. Depression has also been thought to interfere with the exercise of autonomy. While it is surely over simplistic to assume that a refusal of care by a seriously ill patient is rooted in the clinical depression that may accompany serious illness, nonetheless it is surely also simplistic to assume that because someone is seriously ill, decisions to refuse treatment should be honoured unquestioningly as realistic assessments of devastating circumstances. In Chapter I 0 Mark Sullivan and Stuart Youngner explore the apparent disconnection between the assumption in psychiatric practice that decisions to refuse treatment warrant an exploration of the patient's competence, and the assumption in end of life decision-making that treatment refusals should be honoured. They argue that depression should be appropriately evaluated in patients who are seriously medically ill. Such patients should be offered treatment, but if they are competent their wishes to die should be honoured. Advance Directives

One of the issues addressed earliest in the bioethics literature concerning end-of-life decisionmaking was 'do not resuscitate' orders ('DNR orders' - now sometimes called 'do not attempt resuscitation' or 'DNAR' orders, to avoid the implication that resuscitation is likely to succeed). When patients were unlikely to survive a cardiac or respiratory arrest, it had become common practice in some hospitals to write DNR orders on their charts. But these orders were written in ways that were not shared with patients and their families - in order to avoid difficult discussions or perhaps concerns about liability. The apparent clandestine nature of these orders created something of a scandal about hospitals' failure to treat their patients honestly and respect their choices. When these practices came into the public eye, hospitals developed open policies about DNR orders. Often, however, these policies did not reach beyond DNR orders; New York law, for example, to this day authorizes DNR orders but not advance orders for the withholding or withdrawal of other forms of life-sustaining treatment for gravely ill patients. In Chapter 11 Stuart Youngner explores the need to have open policies and careful decisions when decisions are made to withdraw or withhold other forms of therapy. Decisions to withdraw or withhold some forms of care, moreover, should not be taken to suggest that the patient is being abandoned- but only that decisions have been made to shift the emphasis from certain forms of aggressive intervention to other forms of more palliative care. DNR orders are medical orders written for the care of a given patient in particular circumstances. They represent a medical decision made in the context of a given episode of illness or institutionalization. They are not, therefore, what has become generally known as an 'advance directive' -although they may be written as a response to wishes written out in an advance directive. Two kinds of advance directive have become familiar- the 'living

xviii

Death, Dying and the Ending of Life, I

will', in which a patient writes out guidance for the care he or she would or would not want to receive in the case of a terminal illness (or, in some jurisdictions, a persistent coma or an illness that cannot be remedied, such as a stroke), and the 'special power of attorney', through which the patient selects a surrogate decision-maker. Several commentators have objected to living wills in particular as providing imperfect guidance for future circumstances. A quite radical metaphysical claim is that if a person is so changed as to have become incompetent, she cannot be thought of as 'the same' person and thus her earlier directives should not be taken to be authoritative for her present circumstances. What gives this claim purchase is the thought that someone- say, with dementia- may be quite happy in her present circumstances, and thus should be treated, despite earlier pronouncements that she 'would not ever want to live that way'. If there is a loss of continuity of identity at the metaphysical level, however, we literally have one person expressing a view about the life of another. On the other hand, it seems that the earlier person has at least some claim to write out the last narrative chapter of her life, no matter how she may appear to have changed. Allen Buchanan tackles this set of questions about the authority of living wills in Chapter 12. (Commentators seem not to have directed the lack-of-continuity objection against special power of attorney designations, perhaps because it seems less likely that there would be changes on this score; after all, wills governing the inheritance of property are not thought to be undermined by dementia.) Buchanan argues that the loss of psychological capacity would have to be very radical indeed for us to be willing to say that there had been sufficient discontinuity for us to judge that there had been a genuine change of person. In such circumstances of limited mental function, however, the image of a person with dementia enjoying life would be inapposite. Joanne Lynn has long experience as a hospice physician who specializes in the care of the elderly. In 'Why I Don't Have a Living Will' (Chapter 13) she raises a set of concerns about these documents that rests not on personal identity but on an individual's ability to know what she will want in future circumstances. It may be hard for patients to anticipate what they will want in the circumstances of illness generally, or in the exact circumstances of their particular illness. Modalities of care change, in their efficacy, intrusiveness and even expense. A trusted surrogate can adjust to these changes, but a living will cannot. So, Lynn urges, designating a special power of attorney is a much better way of dealing with changes in medical condition than a directive that may become frozen in time. Incompetent Patients When patients cannot speak for themselves, and have not left an advance directive, how should decisions be made for them? In such circumstances, someone will need to serve as a surrogate decision-maker - that is, as someone stepping in for the patient to make the decisions the patient would otherwise have made for him or herself. The bioethics literature has developed two basic approaches to how surrogates should be guided in making these decisions. The 'substituted judgement' approach requires surrogates to attempt to reconstruct the decision the patient would have made, given the patient's known values and preferences. The 'bestinterests' standard requires the surrogate to choose what would be best for the patient in terms of the patient's health-related and other interests. The primary difficulty for the substituted judgement standard is that there may be insufficient information about the patient's values

Death, Dying and the Ending of Life, J

xix

and preferences, at a minimum for the circumstances at hand but more generally as welL The primary challenge for the best-interests standard is developing an account of interests. Take even freedom from pain: while this is surely an interest for most, others would argue that suffering is ennobling. One response of the law has been to insist on a heightened evidentiary standard, requiring clear and convincing evidence (rather than a preponderance of evidence) of preferences or interests when decisions to withhold or withdraw life-sustaining treatment are being made. The selection of a surrogate is a difficult question. The patient herself may designate a surrogate by means of an advance directive- the special power of attorney. Court-appointed guardianship is also a possibility. More commonly, surrogate decision-makers are family members or friends who have knowledge of the person and who can be expected to care about her. In many jurisdictions, a prioritized list of surrogates is authorized to speak for the patient. The standard order of priority is the patient's selected power of attorney, the court-appointed guardian, spouse, parents, adult children; lists also sometimes extend to other relatives or friends. People involved in providing health care are prohibited from serving as their patients' surrogates, out of concern for conflicts of interest. Some of the most difficult ethical questions emerge when family members disagree about what the patient wanted or what would be in the patient's best interests, as family members did in the highly publicized Florida case of Terri Schiavo. In 'Deciding for Others' (Chapter 14) Allen Buchanan and Dan Brock consider the threshold question of what competence is. They argue that, in order to be considered competent, the patient must have a basic understanding of her condition and the available alternatives, of her values and of how her values might be implemented into a choice. A determination of competence is decision-relative, moreover. When the choice is momentous and irrevocable, such as a decision to withdraw life-sustaining treatment, the standard for competence should be set higher, Buchanan and Brock contend. But they also emphasize the importance of not confusing the patient's competence with whether or not the patient is making the choice preferred by the health-care team. Patients may reasonably decide to forego medical treatment that in the judgements of physicians would be medically helpful -and they may also reasonably decide to undergo 'long-shot' or 'last-ditch' therapies that have only minimal chances of success. What matters is that they understand what they are doing and relate this understanding to their own values, not that they do what medical practice would recommend. The process of surrogate decision-making is a complex one. In Chapter 15 John Arras presents a detailed account of the challenges facing a surrogate. The first step in the exercise of substituted judgement is to try to determine the patient's subjective wishes as far as they can be known. These can be drawn from information about attitudes and behaviour, as these features of the person are known to family members and friends. They can also be drawn from present reactions- for example, withdrawal from aversive stimuli. But the knowledge thereby gained can be radically insufficient to yield conclusions about what a patient would want done in the exact present circumstances of his life. The alternative is to attempt to determine the patient's interests in terms of what would be needed to further the best result for the patient. Here, the difficulties begin with the understanding of what would be a good result; Arras suggests using a method of 'moral triangulation' in trying to understand the best interests of a severely demented, minimally functional patient. The method suggests starting with clearer

XX

Death, Dying and the Ending of Life, I

cases on either side of the patient, and then trying to extrapolate from these cases to the best interests of the minimally functional patient. On the one side, patients in permanent vegetative states do not have subjective interests, Arras maintains, because they do not have conscious experiences. On the other side, patients who are more functional do have interests in the continued enjoyment of experiences that give them pleasure and avoidance of pain, Arras contends. Patients who are minimally functional may have limited capacities for pleasure- but limited capacities for pain as well. These patients are in a 'grey area', where it is appropriate to let trusted surrogates make decisions either to withdraw or to continue care. In Chapter 16 Daniel Callahan asks the specific question ofwhether dementia is ethically relevant to decisions to discontinue care. He rejects the view defended by Ronald Dworkin that 'critical interests' that guided the patient's life before dementia should continue to guide care thereafter. Instead, Callahan maintains, decisions should be based on the current patient's experiential interests. There is much to be known about the 'selfuood' and experiences of persons with dementia; these patients should not be treated any differently than any other cognitively disabled people. In particular, it is important to resist economic pressures in these cases to the extent that we would in other cases of cognitive loss or life-limiting illness. Dementia of the Alzheimer's type is a progressive and ultimately fatal illness, and Callahan postulates three criteria that are relevant to decisions to terminate treatment in these cases: no one should have to live longer in dementia than he would have in the pre-technological era; the more advanced the damage ofthe dementia, the more legitimate it is to overturn biases in favour of treatment; and the obligation to prevent a painful and degrading death is as strong as the obligation to preserve life. Decision-making strategies that focus on the patient's prior preferences or current interests eschew concern for 'quality of life', except insofar as it may be relevant to these other standards. Indeed, the concerns about 'quality of life' judgements run deep: that decisionmakers may disguise their own ethical or preferential views as 'quality of life' judgements; or, even worse, that a judgement that a life is of poorer quality for the person who lives it might be confused with the judgement that the person's life is itself less valuable. Rebecca Dresser and John Robertson (Chapter I 7) reject what they call the 'orthodox' substituted judgement standard. This standard made decisions seem deceptively easy, they contend, by fictionalizing a present account of the patient's own choices and relying on it instead of on an appreciation of the patient's current interests. In determining those interests, Dresser and Robertson contend, we must confront directly the question of the patient's quality of life. But we should do so openly and honestly, not in a disguised fashion. The right question to ask, they say, is: what is the value of the patient's life to him now, as he is living it? Both the substituted judgement and the best-interests standard are patient-centred. To be sure, the patient may select his surrogate from the belief that the surrogate has interests of her own that the patient would want to see taken into account. But if substituted judgement and best interests are the standards that must guide the surrogate, the surrogate may take her own interests into account only as they derive from the interests of the patient. In Chapter 18 Mark Yarborough argues that it may be morally permissible to continue medical treatment of the patient to benefit others. The classic example of this is continued treatment to allow family members to come to terms with the patient's inevitable death. Although ethically justified out of respect for the patient and family in some cases, nevertheless such continued treatment

Death, Dying and the Ending of Life, J

xxi

does not comport with good medical practice when it represents futile care or care that will not achieve goals of healing the patient. It thus represents a misuse of medical resources. John Hardwig, in Chapter I 9, advances the other side of this dilemma. Suppose that a surrogate (designated or otherwise) has interests in the termination of a patient's treatment. There are, of course, the unsavoury examples of greedy relatives eager for inheritances. But there are also examples of sacrificial daughters (yes- usually women) foregoing their own opportunities for happiness while caring for a relative with a lengthy terminal course. Should these more sympathetic surrogates be permitted- morally or legally- to act on these interests of their own? Or, should the law declare them to have the same kind of conflict of interest as the unpleasantly avaricious, and appoint a guardian for the patient to try to ensure a neutral proxy? In an essay that has proved highly controversial, Hardwig argues that in such cases the surrogate should be permitted to balance interests, not giving undue consideration to her own interests but also not giving undue consideration to the patient's interests. Hardwig suggests that Norman Daniels' concept of the normal opportunity range over an ordinary lifespan can be used as guidance about what might constitute 'undue' consideration. A decision would give undue consideration to the patient's interest if it recommended continuing treatment beyond the normal opportunity range for that patient, in a context that would require the sacrifice of the normal opportunity range for a care-giving family member. The images called to mind in Hardwig's essay- of dutiful, sacrificial daughters- remind us that our attitudes towards end-of-life care-giving may be gendered. In an influential essay Steven Miles and Allison August (Chapter 20) present data that indicate that our attitudes towards terminally ill patients may be gendered as well. Of the 22 reported court decisions about withdrawal of life-sustaining treatment that were available at the time Miles and August report, there was an almost perfect correlation between the willingness of courts to construct the patient as an autonomous chooser and the patient's sex. Males were autonomous choosers with clear preferences; women were incapable of choosing and in need of protection. Although Miles and August wrote over I 5 years ago, and the intervening years have seen increased attention to concerns about the roles of race and gender in the care patients actually receive, we know of no effort to follow up on their original data. The essay must remain as a caution to scholars, practitioners and patients thinking about end-of-life decisions to be sure that inappropriate factors -whether sex or race, disability or income status - do not pass unnoticed into influencing care. Children

The standard of substituted judgement can be brought into play for decisions about adults who have lived lives in which they developed and acted on preferences and plans. But it is unavailable for children whose lives are as yet unknown. As the frequency of premature deliveries increases, and as medical interventions can increasingly prolong the lives of extremely premature neonates, the problem of decision-making in such cases is apparent. In a famous- and infamous- essay Raymond Duff and A. G .M. Campbell (Chapter 21) determined the extent to which deaths in the Yale-New Haven Hospital's newborn intensive care unit resulted from decisions to withdraw or withhold care. Of299 NICU deaths between l January 1970 and 30 June 1972, 86 per cent (256) resulted from underlying pathologies despite continued aggressive care. Other infants died after prolonged treatment as a result of decisions

xxii

Death, Dying and the Ending of Life, I

on intervention that had not resulted in satisfactory outcomes; the infants remained ventilatordependent, for example. There were also examples of decisions not to intervene surgically: with an intestinal condition (duodenal atresia) of a child with Down's syndrome and with a myomenigocele. Duff and Campbell view these decisions as difficult but ethically appropriate in the face of uncertainty about what would be best for the child. Outcries about similar non-treatment decisions, especially of infants with Down's syndrome, led to the 'Baby Doe' regulations in the United States, which require states receiving federal child abuse prevention funds to have programmes in place that ensure that infants receive continuing care unless it would be futile or the infant is permanently comatose, or under circumstances in which it would be both virtually futile and inhumane. These regulations remain controversial. In Chapter 22 John Robertson addresses the legal status of the non-treatment decisions described by Duff and Campbell. Robertson's essay was published in 1975, and its title bears the terminology of the period; 'involuntary euthanasia' means withdrawal or withholding of care on the view that it is in the interests of the patient to do so, but without the patient's consent. Although a formal analysis of the law suggests that parents may be prosecuted for child neglect if they withhold needed medical care, or even for homicide if the withholding is the cause of the child's death, the fact that no one had ever been prosecuted in such cases (at the time of Robertson's writing) suggests a discontinuity between law and practice. That discontinuity may represent a conscious policy decision; more likely, in Robertson's view, it reflects the fact that many medical decisions are made outside of public view. Robertson regards widespread practices of involuntary euthanasia as deeply troubling and advocates the development of legal strategies for ensuring that such decisions meet acceptable guidelines. his view is that the guidelines should be shaped by the best interests of the child and that parents' choices should prevail where interests are unclear. The Baby Doe regulations would appear not to meet these guidelines, as they require treatment in all but the direst cases. John Arras, in 'Toward an Ethic of Ambiguity' (Chapter 23), criticizes the regulations on just this basis. Arras contends that there are cases in which the prolongation of life, although medically possible, is not in the interests of the infant. Death is not always the worst outcome on a best-interests standard. Yet the interests of 'normal' adults should not set the template for determining the interests of an impaired newborn. The primary consideration should be the child's well-being, seen from the experiences of the child. We should openly recognize the ambiguities we face in decisions about whether to continue treatment that sustains life in the face of the infant's limited experiences and any continuing pain of treatment. Both decisions mandating treatment -like the Baby Doe regulations- and decisions allowing widespread discretion - such as the practices in the Yale-New Haven nursery- nearly always err in failing to recognize these ambiguities, Arras contends.

Futility Physicians as professionals are not required to render futile care; even the Baby Doe regulations recognize this. But the definition of futility has been problematic. The essays in this section are the classic pieces on how futility has been understood and what its significance might be for medical decision-making. A medical procedure can be said to be futile when it is powerless to achieve a desired result. It does not work. Suppose a patient, in danger of death, wishes treatment that physicians deem futile. Should the patient be entitled to have that treatment?

Death, Dying and the Ending of Life, J

xxiii

The issue of futility is complicated because physicians must act on judgements of probabilities, not certainties. There is often anecdotal evidence that a treatment might work, even though studies do not support it It is theoretically possible that a given treatment might cure a disease, even though it has never done so before. Nancy Jecker and Lawrence Schneiderman, writing together in a pair of essays addressing issues of futility (Chapters 24 and 25), call the principle that patients should be entitled to treatment in such a case 'the criterion of potential benefit'. This criterion holds that if there is any conceivable possibility of benefit, the patient should be entitled to the treatment This view is often justified on grounds of patient autonomy. Patients should make these decisions. Judgements of futility, it is said, are value judgements. Doctors are experts on medical, but not ethical, matters. People have different values about when continued life is worthwhile and what risks are worth taking. Therefore, respect for patient choice means that patients should make these decisions, not doctors. For a doctor to decide that a treatment a patient wants is futile and to deny it is a misuse of power. Jecker and Schneiderman reject the criterion of patient benefit They argue that patient autonomy is not an absolute value, but can conflict with other values, such as professional and personal integrity, avoiding harm to the patient and justice in the allocation of health care. More importantly, they argue that the idea of medicine as a value-free, technical discipline is mistaken. Medicine is shot through with ethical values, as evidenced in the presentation of alternatives, the ordering of tests, the prescribing of medications and so on. There are ethical standards of responsibility inherent in the practice of medicine. Physicians collectively need to have the power to develop and act on such standards. Only then will they be able to promote patient well-being and earn patient trust In the first of these essays, then, Jecker and Schneiderman concede that judgements of futility have an ethical dimension. But it is part of the professional responsibility of physicians to offer only appropriate care, and futile care is not appropriate. Thus they reject the 'criterion of potential benefit'. The decision of when care is futile, however, should not be made by individual physicians at the bedside; rather, it should be the result of a consensus among physicians based on our best scientific understanding. In the second essay, Schneiderman and Jecker are concerned with a similar problem, the tendency of patients to demand, and physicians to use, unproven treatments. They argue that it is important to distinguish between the following: i) beneficial treatments: those which have been shown to be effective; ii) experimental treatments: those which show promise but have not yet been proved effective or tolerable; and iii) futile treatments: those which provide no benefit and sometimes make things worse. Schneiderman and Jecker argue that physicians, regulatory agencies, like the FDA, and patients are too quick to make use of treatments that are still experimental with no proven benefit Most experimental drugs fail; only I in 5000 ultimately gains approval for use. The public is under the illusion that 'new is likely to be better', and physicians and the government have done a poor job convincing people otherwise. The use of experimental drugs, except as part of controlled experiments, does little good and often great harm. Schneiderman and Jecker cite three reasons why physicians fail to make the needed distinctions: the 'optimistic but erroneous belief' that new is better, the power of pharmaceutical companies to promote new drugs, and the fear of malpractice. They also note that the use of unproven drugs interferes with the ability of researchers to carry on with controlled

xxiv

Death, Dying and the Ending of Life, I

experiments that will lead to genuine knowledge about drugs and prove more beneficial in the long run. In the bioethics literature there have been many high-profile cases of requests for both treatment and treatment refusals. One of the earliest and most discussed cases of requests for ongoing treatment was that of Helga Wanglie, an 85-year-old woman who had lapsed into a permanent vegetative state. In Chapter 26 Stephen Miles discusses the case as one in which treatment is not futile but is deemed non-beneficial. Little was known of Helga Wanglie's preferences for treatment or non-treatment in such a condition. After several months, the physicians at her Minnesota hospital suggested to her family that she be removed from the respirator and be allowed to die. Her husband and children refused. They argued that physicians should not play God, that she would not be better off dead and that a miracle could occur. Attempts to get the Wanglie family to change their minds failed. Attempts to find other health-care facilities more willing to care for Wanglie also failed. She remained on a respirator for another year. Her insurance company paid the considerable costs of her treatment. Miles, who was also the hospital's ethics consultant, concedes that the treatment was not futile because it prolonged Wanglie's life. He argues, however, that it was non-beneficial because it could not heal her, palliate her suffering or enable her to experience the benefit of life. On the basis of its not being beneficial, he argues further that it is inappropriate care. In general, patient requests for care do not obligate physicians to provide that care when it violates professional standards- for example, requests for medicines known to be ineffective. Miles notes that patients in a persistent vegetative state typically stay on a respirator no longer than six months. He uses this as a reason to think that Wanglie's care beyond that period was inappropriate, as was the fact that other health-care facilities would not accept her. This shows a consensus that such non-beneficial care violates professional standards. He also notes that although her insurance company paid for her care, its doing so still imposes unfair costs on others in her insurance pool, since if such care must be paid for, more beneficial care must be foregone or premiums raised. Miles argues that there needs to be professional procedures and standards to deal with such cases. There was, in his view, a 'social consensus that intensive care for a woman as "overmastered" by disease as this woman is inappropriate' (p. 44 I). Felicia Ackerman disagrees with much of Miles's analysis. In Chapter 27 she contends that the ongoing use of the respirator in Wanglie's case was medically appropriate in the sense that it achieved its medical end - maintaining her respiratory function and thus her life. Ackerman criticizes hasty judgements about the quality of life of others, particularly judgements made by the able-bodied about the lives of people with disabilities. She also points out that when patients' values are unknown (as they may have been in the Wanglie case), keeping someone alive who would want to die and allowing someone to die who would want to live are not parallel wrongs- only in the former case does treatment against values affect conscious experience. Finally, Ackerman objects to the argument based on insurance: the Wanglie family's insurance specifically and publicly provided for the care she was given and, Ackerman contends, it would be a violation of the expectation of those who had purchased the insurance to deny the promised care.

Part I Death

The Nature of Death

[1] On defining a 'natural death'

E

images of death stay fresh in my mind. The first is that of an old friend who died at the age of sixty-one of cancer of the liver. He knew at least two months beforehand he was going to die, as did his family and friends. During that time he received a constant stream of visitors. They knew and he knew they were taking their leave. He was a model of gaiety and hospitality; it is hard to imagine how any dying person could have put people more at ease. When his time came, his family and many of those friends were at his bedside in his home, not in a hospital. Without any significant pain, he slipped into a coma some six hours before he died. Until then he was fully lucid. I arrived ten minutes after his death. I wept, as did everyone else there, but I could not feel that his death was "untimely"; one could only hope to die with equal grace, encircled by one's friends and family. That kind of image invites sentimentality. It was almost too perfect a death, and those of us prone to take a dark view of the world tend to be skeptical of such perfection, even when we witness it with our own eyes. But I did witness· it, and the image is fresh and powerful. Naturally, his death was inevitably easier for me to bear-a friend-than for his family, those who lived with him day-in and day-out; their sense of loss was greater than mine. Yet, however great their sadness, I had no sense that they felt his death was wrong. I might add that he left his wife healthy and financially secure, his children grown and self-sufficient, and his business affairs in order. My second image is that of the funeral of the twenty-year-old son of another friend, a bright, attractive, promising boy who died with total unexpectedness of a coronary caused (it was learned during the autopsy) by an undetected congenital heart de-. feet. It was ah..unbearable funeral. I have never been able to think of anything good to say for that death. My third image is that of the death of my own son, Thomas, at the age of six weeks in 1961. The cause of death, also unexpected, was that catch-all phrase "crib-death." Alive and well one hour, dead the next. While I certainly was devastated at the time, and even now miss Thomas, I find it hard to know what to think about his death. It seems in some sense wrong that he died, but then again, his attachment to others in the world was tenuous, he felt (so far as we know) no pain in his dying, and he had had no chance to develop hopes or goals. It

DANIEL CALLAHAN was hardly. a routine event in our lives that he died. But I do not know what kind of event irwas in his life. My fourth image is that of the death of a very elderly patient in a chronic care facility, a death preceded by a long period of physical and mental decline, with the last few years of life marked by an almost total isolation, certainly from others but seemingly from self as well. In this particular case, the isolation did not come from staff insensitivity. The time had long passed when it was possible to do more than care for the body of the patient. I mention these images for several reasons. I believe it impossible for anyone to consider the concept of "natural death" totally free of one's personal images of death, particularly those witnessed at firsthand. Moreover, given the varied forms death can take, we are each likely to have rather different images. I have witnessed at least one death which seemed to me (intuitively at any rate) to be a "natural death," at least one that seemed wrong and "unnatural," at least one that does not easily fit into either class, and one that seems to have come much too late. That there can be so many images of death points to a problem in any attempt to define, or to characterize, the notion of a "natural death." Any socially significant definition will and must be affectively colored, evoke different private images, and take its point of departure from certain powerful cognitive-emotional exemplars. A Definition of Natural Death

My goal is the immodest one of proposing a definition of "natural death," which will be at once rationally persuasive, emotionally satisfying, socially advantageous, and politically attractive. It will, of necessity, have to be a stipulative definition, since, so far as I know, there exists no commonly accepted meaning; it is a popular rather than a technical phrase. Why should one want to define and gain cultural support for the concept of "natural death"? First, because it is becoming socially, economically, and politically imperative that there be such a concept. Without one we will not, for instance, be able to DANIEL CALLAHAN is director of the Hastings Center. This paper was originally prepared for a project of the Research Group on Death and Dying on "Alternative Life-Extending Technologies," supported in part by a grant from the Ittleson Foundation.

6

Death, Dying and the Ending of Life, I

specify some rational limits to the aspirations of medicine, some limits to the kind and extent of medical research and care individuals can legitimately ask of society, and some limits (quite apart from economics) to the medical care that ought to be invested to keep people alive. Second, a concept of natural death is psychologically imperative in order that we may know, as individuals, what it is right and reasonable to hope for; the alternative is both spiritual and psychological chaos. Third, a concept of natural death is a moral necessity, in order that we may gain an idea of what we owe, as human beings, to those who are going to die. I mean by "death" more the event itself than the process of dying. I would like to make a total separation between the event and the process but never found that to be altogether possible. My definition of "natural death" is this: the individual event of death at that point in a life span when (a) one's life work has been accomplished; (b) one's moral obligations to those for whom one has had responsibility have been discharged; (c) one's death will not seem to others an offense to sense or sensibility, or tempt others to despair and rage at human existence; and finally (d) one's process of dying is not marked by unbearable and degrading pain. Each part of this definition requires some explanation. By the accomplishment of one's life work I mean primarily one's vocational or professional work. Some obvious problems, however, present themselves. If the paradigm instance of a life's work is that of pursuing and attaining a particular goal, some goals are obviously more easily achieved than others. It will be comparatively easier to become president of a bank than to discover the cure for cancer, and easier to gain leadership of a national political party than to achieve perfect world justice. Some life goals, that is, focus on limited and finite projects; others have as their goals what may well, and probably do, transcend the grasp of any one individual, whether because of their inherent difficulty or because they are dependent upon the joint effort of many individuals, no one of whom can fully control what all the others will do. One might, by my criteria, say that a person who died before achieving the long-sought bank presidency died (in his terms) prematurely. Yet if, for lack of talent or luck, a person well prior to death had lost any chance of achieving the goal-and no further time would allow a regaining ·of the opportunity-it would not then be possible to say the death was premature; more life would not have helped. If there is disappointment in such a life, its reason has nothing to do with death. In the instance of someone whose life goal was the perfectly just society, and the goal perhaps totally out of reach regardless of how much time was invested in it, then death prior to its achievement could not really be said to be premature or untimely either. There would be· no way of knowing if more time, a longer life, would. have made any difference anyway. How long would a person have to

Institute of Society, Ethics and the Life Sciences

live if his ambition was to see a permanent eradication of war? The model of a life work as the seeking of some specific goal is by no means the only possible one. Much more co=on perhaps is a vocational aim of simply doing a certain kind of work, where the dominant motive is the satisfaction of the process itself. A person becomes a shoemaker, on the one hand, or a physician, on the other, because he would prefer to spend his time doing the kinds of activities that go with those roles. In that sense, such people have achieved their life's goal when they first undertake the activities. If it is their ambition to do such work forever, then I suppose one could call their deaths, whether after a year of the work or two hundred years, premature. In that event, nothing but immortality would suffice; a much longer life span would only stay the day of reckoning, not do away with the problem. The second element of my definition also requires examination. When I speak of having discharged one's moral obligations, I have in mind primarily family obligations, particularly to one's children. Obligations to children are very special and inescapable. The death of a parent at a time when children are still wholly dependent upon them is easily and rightly seen as particularly sad and wrong. Yet if the children are grown, and have achieved a selfsupporting and self-directing status, it is then fair to say that the parents no longer have special obligations toward them; they have done their work. We may be sad when an elderly person with grown children dies; but we are rarely sad (save in very special cases) because of any feeling that the children still need the parents as parents. Of course there are other kinds of obligations as well. Wives and husbands, by virtue of their marriage, take on mutual obligations. An ideal of many, if not most, marriages is that the spouses will so prepare themselves that the death of one will not mean the ruination or impoverishment of the other. A premature death can then be seen as one in which, for lack of time, a spouse has not been able to provide the basis for the (at least) material security of the survivor. But if that has been accomplished, or one lives in a society which does not allow the death of a spouse to undercut all future social security of the survivor, then it easily becomes possible to imagine situations in which it can be said that spouses have discharged their obligations to each other. This ·is hardly to deny, of course, that the survivor may severely and permanently be grieved by the death of the other. But that possibility does not seem to me sufficient to call :the death wrong or untimely. There is still another set of obligations. What about the situation where a person has voluntarily undertaken the obligation to bear the burdens of others or to work in behalf of their improved welfare? A premature death would, in that context, be one in which death occurred before the chosen obligations could be discharged. There are a number of difficulties in

33

Death, Dying and the Ending of Life, I evaluating that kind of situation, however. First, though a person may undertake voluntary obligations, it is not evident that an inability to discharge them can be seen as an evil. Surely there is loss, on the part of those who would have benefited as well as on the part of he whose sense of obligation was thwarted. Yet since the obligation was voluntary, it could not easily be claimed that its beneficiaries had any right to it; the death of their benefactor cannot be counted as an injustice. Second, some voluntary obligations can be seen as finite and time-limited, others as open-ended and infinite. It is one thing to nake responsibility for the welfare of an individual orphan and quite another to take responsibility for the welfare of all orphans. More time may help in the first instance, but (assuming one can talk sensibly at all about voluntarily undertaking an obligation toward the welfare of all orphans) no amount of time is likely to enable achievement of the second goal, if for no other reason than that more children destined to become orphans will continually be born. The third aspect of my definition, that a given death will not seem to others an offense to sense or sensibility, or elicit feelings of rage and despair at human existence, brings us to some fundamental problems about death itself. From all I can gather, the death of an elderly person who has lived a rich and full life is not, in any society, counted as an evil, as if symptomatic of a deranged and cruel universe. It may well be of course that mankind has simply been led to rationalize that kind of death; since it cannot be helped, it might as well be accepted. There is no way to determine if that is the case. Nor is there any way to determine if, as many (though not all) elderly testify, their proclaimed readiness to die is rationalization either. One can only point out the brute fact that neither in the eyes of others nor in their ·own is death seen as an evil. But the main reason normally given in both cases is that they have lived a life, have done what they can, and thus are not the victim of the malevolence either of the forces of divinity or of nature. This seems especially true if they have, in their own eyes and that of others, completed their life's work and discharged their moral obligations to others. It is not as if life has ceased to have a purpose because of that, but instead that its main purposes have been achieved-both self-realization and assistance to others in their self-realization. That point may be underscored by thinking of the death of an elderly person not at the time of death or immediately thereafter, but in very distant retrospect. I can speak only from limited experience here, but I have never heard anyone remember with bitterness, or sharp regret, the death of an elderly person who lived a full life, but whose death occurred twenty, thirty, or forty years ago. I can recall being sad at the death of my eighty-year-old paternal grandmother some thirty years ago. But I do not now feel sad that she is dead, nor do I know of any of her own children, my father, uncles, and aunts (who were of course much closer to her than I was) who are still

34

7

sad that she is no longer alive" That many nf them are now dead, and their lingering sadness, if any,. is buried along with them, only adds to the difficulty of imagining her death as an evil-for surely if an event was an evil, as distinguished from an occasion for sadness, it should be possible to look back and identify just what makes it an evil even now. Finally, I have added a special stipulation about natural· death-that it not be marked by unbearable and degrading pain. Pain, whether physical or psychological, can destroy personality, sense· of self, and the ability to relate to others; it separates a person not only from himself but from others as well. Without further attempting to argue the case, I will assert that it is simply impossible that a death amidst pain could be seen either by the victim or those around him as in any sense acceptable. Is Death an Evil? In explaining my stipulative definition of "natural death," I have provided a variety of arguments in support of the cogency of its various parts. For the most part, they stem from the observation that hardly anyone seems to consider the death of an elderly person, after a full life, as an evil; and, moreover, that there seem to be a number of intuitive considerations to support this common feeling. That kind of death is my paradigm of a natural death. The obvious premise behind my definition, assuming all the criteria are met, is that such a death is not an evil. To make that case, which can be done only in part here, it is necessary to specify what I mean by evil. Death could not, in my paradigm case, be called a moral evil in the narrow sense of the term, that is, in the sense that rights have been violated, a social injustice done, or clear obligations defaulted. No person or persons, that is, have done wrong to the one who dies. I believe that there is a broader sense of evil, however, one which places the emphasis upon the absurdity and irrationality of life itself, where nature or fate of some other implacable and uncontroll~ble power victimizes hapless human beings, forcing them to be born and to die in a world not of their choosing. One might call that "metaphysical" as distinguished from "moral" evil. I admit to an enormous ambivalence about whether human life as such, and the inevitable death that goes with it, is a metaphysical evil. Paul Ramsey, 1 Robert Morison," and Leon Kass3 have, most recently, wrestled with that question, to which Robert M. Veatch• added some additional points. I will not try to directly join that argument, but will only highlight what seem to me some important confusions that it revealed. One confusion was introduced by Paul Ramsey, who tried ro argue "the contradiction death poses to the unique worth of an individual human life."5 But Ramsey does not really defend that proposition at all. To do so, he would have to show that, to be worthy, an individual human life would have to be one in which death has no part. That he does not show, or even try to show. Instead, he contents himself with

Hastings Center Report, June 1977

8

Death, Dying and the Ending of Life, I

arguing that death is a threat to human life, an enemy of life. But that is a very different matter from demonstrating that both the threat and the reality of death remove from human life some of the value and worthiness of that life. One can well acknowledge that "there is grief over death which no human agency can alleviate,''" without thereby accepting the notion that, because people do •and must die, an absence of worth is thus displayed. Kass very effectively points all that out, but he then goes on to make the further case that death itself cannot be judged an ultimate evil: ". . . death is natural and necessary and inextricably tied to life. To live is to be mortal; death is the necessary price of life ... When we 'buy' life we 'buy' dea~h." 7 The general thrust of Kass's argument is, in the end, that since death is, it cannot be considered an ultimate evil. But he does not really ask the question whether this is a good situation for humans to be born into, i.e., metaphysically good. His use of the metaphor "buy" betrays that omission. One does not "buy" life, as if it were some voluntary transaction into which one knowingly and freely chases to enter; one has no choice about being born and dying. If ever there was a forced sale, this is one. Indeed, life is a "bitter-sweet bargain,''8 but one in which the ultimate transaction is not of our choosing. What is finally missing in Kass's account is the centrality of human individuality and subjectivity in human life. Kass appears to consider a dwelling on that part of human nature as, somehow or other, harmful to the more important quest for some larger truth about human life. 9 I will here only point out one very basic problem with that tack. Let us agree for the sake of argument that death is the price of life, and life and death inextricably and necessarily related. We then "see nature as it is."10 But if that is our goal, what else can we discern about nature "as it is,'' and particularly human nature? It seems obvious that we will at least see, as a hard fact, that people want to live and not to die; that drive seems to also be an inextricable and inescapable part of life, as much a part as the fact that we die. I did not "buy" that drive nor was I given an opportunity to bargain about its strength. Thus even if we are wise enough to accept the necessity of death, we will have an internal war on our hands-between that hard-won wisdom and our own unchosen psychic makeup (seemingly a generic and not 'individuated, idiosyncratic part of human nature) that makes us want to live. The paradox (or, if you will, the sheer perversity) of this situation can best be seen by rephrasing a passage of Kass. "Death,'' he says, "is necessary not just in the sense of 'unavoidable,' which it is also for the individual, but in the sense of 'indispensable.' " 11 Consider this rephrasing: "A drive for life is necessary not just in the sense of 'unavoidable,' which it is for the individual, but in the sense of 'indispensable:' " I believe that to be a true sentence, and if Kass's sentence is true as well-then it is no wonder that we have so

Institute of Society, Ethics and the Life Sciences

much trouble figuring out the meaning of life and death. In expressing his worries about the social wisdom of installing a concept of natural death, Veatch is, as one must be, driven back to the question of whether death is an evil. He considers two options: the rationalist and the social-eschatological case for life. The first rests on the premise that, as an individual, I do not want to be dead, that is, to be. nonexistent; I can have an "anticipatory regret of my nonexistence."12 But Veatch finally rejects this line of argument, on the ground that a desire for continued life does not prove that continued life is therefore desirable. He rests his own convictions on the "socialeschatological" case for life, by which he means the persistent Western eschatological vision of perpetuation. A central part of that vision is the conquest of the evil of death-"life shall prevail ·and death shall be no more." 13 Veatch's initial case would ·be stronger if he had distinguished between religious and secular eschatological visions. While a great part of the Western religious tradition, at least the Christian, has seen the overcoming of death as a legitimate part of human aspiration, this is by no means the case with the major secular eschatologies. Neither Marxism, nor other variant socialist eschatologies, appear to hanker for the perpetuation of individual lives. It is the perpetual life of the just community, a community of people who live and die, that is sought. In the end, Veatch does not confront directly the proposition that "life is indeed a good and death is an evil. " 14 In fact, by even initially posing the problem that way, as if a belief in the goodness of life necessarily entails the judgment that death must then be an evil, he gets off on the wrong foot. Though he tries in his discussion of death as a relative rather than an absolute eviJ15 to rectify that narrow posing of the question, he never, I think, solves his own problem; if anything, he makes it worse. He contends, for one thing, "that death is fundamentally incompatible with the ideal human community."16 But nowhere, other than invoking the very general Western eschatological vision, does he say just what he takes the ideal human community to be. I too would not want to be forced to specify just what that would be. But it is not at all clear to me just what, even as an inchoate vision, such a community would be like or what would be particularly good about it. Selfevidently, it would be ·good in the sense that no one would have to fear death; but beyond that, I am hard-pressed to think of any other benefits. The very notion of a "human community" is bound up with what we presently know of human communities: life and death, hope and fear, parents and children, the succession of generations, being young and growing old. The removal of death would entail a totally different kind of human community; and not being able to imagine what that would be like, I cannot judge whether it would be desirable. More than that, how could anyone presume to know enough about such a community even to give its advantages the

35

Death, Dying and the Ending of Life, I benefit of the doubt? An equally serious difficulty arises in Veatch's attempt to hold that "immortality is a desirable goal (but) some deaths may be preferred to a painful and dehumanizing struggle to that goal." 17 Veatch asserts that these two values are not incompatible, but nowhere actually tells us why. The most he says is that "to hold death as an evil and still feel that individual deaths are acceptable may be a tragic view of death. . . ." 18 But that hardly deals with the problem. What is meant by a "tragic view of death"? That life is absurd., that things just do not come out right? If nothing else, there are some logical puzzles in his position. If life is inherently good, then he cannot say that some people are better off dead. If not all life is good, then it cannot be concluded that life as such is good; it all depends upon what kind of life one is talking about. If death is an evil, then it is not less of an evil simply because some people find life unbearable or degrading. What is finally missing in Veatch's position is some account of what makes life good and what makes death evil. Lacking that, it is impossible to find any coherent criteria of why life as a permanent state is better than death as a permanent state. The Implications for Pnblic Policy

Defending my definition of natural death (and the paradigm of death it draws upon) as a basis for public policy involves considering the implications (a) for those, not yet dead, whose lives have reached the stage where, if they died, their deaths would be considered "natural," and (b) for those whose lives are threatened with premature or untimely deaths. However, an important qualification is in order. A public policy based on a concept of natural death is not, as I use the term, a wholly formal policy. The most absurd manifestation of such a policy might be, for instance, a federal or state definition of natural death, followed ·by a rigorous attempt to conform law, regulation, and practice to the definition. Enamored as I am of my own definition, it would strike me as pure Orwellianism for a legislature or a committee to formally deliberate about whether a given person did or did not meet the criteria, and then to decide whether to extend further medical or social assistance. The most that ought to be hoped for is a broad social consensus, dipping deeply into what Philip Rieffl9 has called the cultural unconscious, which will tend to push society in the direction of recognizing rough limits to any desire to conquer death or greatly extend the human life span. That would leave considerable room for dissent and variation as well as considerable leeway for dealing with particular cases. The most important social implication of my definition is that it does not lend itself to the language of rights. As I have defined it, there could not be a "right to die a natural death." Given the subjective nature of life goals, no society could guarantee their realization. The most that could be guaranteed

36

9

would be the provision of those material necessities that are a minimal condition for the realization of any human goals-food, clothing, shelter, education, and decent health care. Nor could a society, at least in the foreseeable future, guarantee that a person would live long enough to discharge all family obligations; quite apart from diseases that can kill people early in life, accidents, suicide and other misadventures are hardly likely to be banished. Neither could a society guarantee that there would be no deaths of the kind that lead people to fear their own death, or wonder about the rationality and benignness of the universe. The only aspect of a natural death that society might conceivably be able to guarantee, though who knows when, is a death totally free of pain and degradation; that is, perhaps, a fulfill able technological and cultural possibility. If a natural death could not, given its combination of both subjective and objective elements, ever be a legal or social right, could it at least be seen as a moral right? No, for if it is impossible to imagine ways in which societies could fulfill such a right, it is no less impossible to imagine ways in which individuals, in their lives together, could ever have a duty to insure the natural death of another. The same contingencies that make it impossible for society to have meaningful duties with respect to the natural death of its citizens, make it no less impossible for individuals. How could it ever be said, for instance, that I have a duty to see to it that my neighbor achieves his life goals, whatever they might be? If, then, the language of rights, duties, and obligation cannot appropriately be used in the case of natunrl death, the concept itself then stands revealed as a quasi-utopian ideal, one which, while it admits of frequent empirical realization, will remain as an ideal. It would .:ease to be such only when all, or most, people would have a good chance of dying such a death. And that possibility still seems far distant. But it is, nonetheless, a meaningful ideal, precisely because my paradigm case of the death of an elderly person has, more than once, been historically realized. To note that a natural death is an ideal, and to note that it cannot be a right, is to defuse concern that the political acceptance of such a concept would open the way for abuse. Veatch has expressed this concern, observing how easy it would be for a society to tum a right to natural death, if such was affirmed, into duty to die a natural death. But that move could only come--which is not to say it would notby virtue of an almost total lack of reflection on the nature of a natural death. More pointedly, there would already have to ·be a potential for abuse in a society for it to seize upon natural death as a pretext to rush people into the grave. Are there not, however, even in my ideal, the seeds for abuse? An implication of my definition might be that those who have lived a full life, and whose death would be called natural, should it occur at some point in that realized life, would be less entitled to expensive and socially marginal high technology medi-

a

Hastings Center Report, June 1977

10

Death, Dying and the Ending of Life, 1

cal support, than those who have not reached that stage. I do not think that would be an unfair or an unreasonable outcome, at least as a broad result. But how far should that be pushed? Not very. Without trying to fully develop the case here, it strikes me as reasonable for a society. not to invest large amounts of money in expensive life-extending therapies for those who have already lived a full life, particularly when, as is now the case in this country, there are many others in the society who need that money even more -if they are to have their chance to live out their lives. At the same time, in order to achieve the ideal of a death not marked by unbearable pain or degradation, supportive care of the elderly ought to be fully financed--care designed not necessarily to extend their lives, but to make those lives as comfortable and carefree as possible. That would of course cost . money, money which could be spent otherwise. But · it is doubtful that people .could tolerate life in a society, however fine the other social conditions, where they would have to look forward to abandonment in old age. That would be a fearful cloud under which to live out a life. The point about a "natural death," as I have defined it, is not that it would condemn the elderly to neglect and mistreatment; on the contrary, it could not even be conceived as an ideal unless the conditions of old age were tolerable. The forces of nature, finitude, and mortality will be quite enough for the elderly to cope with; at a minimum, society should not make matters worse. Another social implication df my definition is that society would eschew any quest for immortality or for a significant extension of the human life span. If it is possible for people to die a natural death, as it is, and possible to hold out the ideal that most people could die such a death, there seems no reason at all to seek immortality. The ideal of a natural death presupposes that one has lived a full life, that completion of that kind of life is not an evil. Only those will be left dissatisfied who would say that, should all the conditions I have indicated for a natural death be met, such a death would still be an evil. But the only plausible reason I would see for such a dissatisfaction would have to rest on a 'broader complaint about the nature of reality-the problem of metaphysical evil, which sees us born into a world not of our choosing, with the ·best to be hoped for being a natural death. But that moves one, then, into the question of religion. While I do not accept it, the Christian belief in the resurrection of the body seems to me to make more sense than a quest for earthly immortality-for the former at any rate presupposes a transcendent reality which it does not pretend to understand. If the quest for earthly immortality makes no particular sense, the same will be true of any significant attempt to extend the average life span. No case has

Institute of Society, Ethics and the Life Sciences :

yet been presented as to why an average life of onehundred or one-hundred-fifty years would be significantly better than one of seventy-five years. More of the same is not, by itself, a very good argument. One might as well contend that the plays of Shakespeare, or the sonatas of Beethoven, are flawed because they all come to an end. If a full and complete life can be lived in seventy-five years, what is the gain of another few decades? If it could be shown that the ideal of a natural death could be more easily attained with a longer life span, that would be a persuasive argument. But I do not believe that is at present the major obstacle to a natural death. A gradual improvement in medical care, in the eradication of the most common diseases, in general living conditions, appears to bring a gradual increase in the average life span. The biblical three score and ten is now well exceeded in Sweden, where the life expectancy for women is seventy-nine years. The main problem for medicine now is to find ways to deal with those remaining diseases and conditions that stand in the way of the living of a full life. A by-product of that effort will be a longer life for most people. If, along the way, medicine can find better ways of coping with those conditions that make ·being old so often a burden, well and good; research on aging is a valid and potentially fruitful area. The possibility of a natural death will then be enhanced. But until some good reasons have been presented why a longer life per se is good, as distinguished from a long life where the evils of life have been as far as possible minimized, there is no public policy case to be made for the investment of so much as one cent in efforts to extend life for its own sake. REFERENCES

lPaul Ramsey, 'The Indignity of 'Death With Dignity,'" in Peter Steinfels and Robert M. Veatch, ed., Death Inside Out (New York: Harper & Row, 1975), pp. 81-96. 2Robert Morison, "The Dignity of the Inevitable and Necessary,'' in Death Inside Out, pp. 97-100. 3Leon Kass, "Averting One's Eyes, or Facing the Music?On Dignity and Death,'' in Death Inside Out, pp. 104-14. •Robert M. Veatch, "Natural Death and Public Policy," in his Death, Dying, and the Biological Revolution (New Haven: Yale University Press, 1976), pp. 277-305. 5Ramsey, p. 81. •Ibid., p. 96. 7Kass, p. 109. Sibid., p. 113. 9Ibid., p. 103. lOibid., p. 110. llibid., p. 113. 12Veatch, p. 294. lSibid., p. 296. H[bid., p. 292. '"Ibid., p. 302. 16Ibid. 17Ibid. lSibid., p. 303. 19Philip Rieff, The Triumph of the Therapeutic (New York: Harper & Row, 1968).

37

[2] WHY IS DEATH BAD? ANTHONY L. BRUECKNER AND JOHN MARTIN FISCHER

(Received 14 October, 1985)

I. WHY IS DEATH BAD?

It seems that, whereas a person's death needn't be a bad thing for him, it can be. In some circumstances, death isn't a "bad thing" or an "evil" for a person. For instance, if a person has a terminal and very painful disease, he might rationally regard his own death as a good thing for him, or at least, he may regard it as something whose prospective occurrence shouldn't be regretted. But the attitude of a "normal" and healthy human being - adult or child toward the prospect of his death is different; it is not unreasonable in certain cases to regard one's own death as a bad thing for oneself. 1 If this is so, then the question arises as to why death is bad, in those cases in which it is bad. If one believes in an afterlife, one could explain how death (conceived of roughly as the cessation of bodily functioning) can be bad insofar as it can involve eternal torment - an indefinitely long sequence of (highly) unpleasant experiences. Of course, on this sort of account, death needn't be bad, even for a normal and healthy human being, since he may experience eternal bliss in the afterlife. If there is an afterlife, and for some it includes unpleasant experiences, then this would explain how death can be a bad thing, but it is controversial whether there is an afterlife. Since it is quite possible to deny the controversial assumption that there is an afterlife and yet regard death as a bad thing, it would be desirable to produce an explanation of death's badness which doesn't presuppose that there are experiences after death. Many have thought that such an explanation can be given. If death can be a bad thing for a person, though not in virtue of including unpleasant experiences of that person, then death is a bad thing for a person in a way that is different from the way in which, say, pain is a bad thing for a person. That is, some things which are bad (or evil) for a person (such as pain) are "experienced as bad by the person", whereas other things which are bad for a person (such as death) are not (ever) experienced as bad by the per-

Death, Dying and the Ending of Life, I

12

214

A. L. BRUECKNER AND J. M. FISCHER

son. 2 Death, then, is assimilated to such bads as betrayal by a friend behind one's back, which, though never experienced as bad (one never finds out and suffers no bad consequences), are nevertheless bad for a person. 3 Let's suppose that some things which are never experienced as bad by a person are nevertheless bad for the person. Death could then be an experiential blank and still be a bad thing for an individual. And one plausible explanation of why this is so is that death (though an experiential blank) is a deprivation of the good things of life. That is, when life is, on balance, good, then death is bad insofar as it robs one of this good: if one had died later than one actually did, then one would have had more of the good things in life. This is the sort of explanation of death's badness which is adopted by Thomas Nagel. 4 But a problem emerges. We intuitively think that it is appropriate to have asymmetric attitudes toward prenatal nonexistence and death. We think that it is reasonable to regard death as a bad thing in a way in which prenatal nonexistence is not. If death involves bad experiences in an afterlife, then this asymmetry could be explained. But we are assuming here that death's badness is not experienced as bad by the individual who dies. If this is so, how can we explain the intuitive asymmetry between prenatal and posthumous nonexistence? Both periods are, after all, experiential blanks. And it seems that prenatal nonexistence constitutes a deprivation in a sense analogous to that in which death is a deprivation: if a person had been born earlier than he actually was born, then he would have had more of the good things in life. (When it is supposed that one is born earlier here, we hold fixed the date of one's death. Similarly, when it is supposed above that one dies later, we hold fixed the date of one's birth.) Being born at the time at which one was born (rather than earlier) is a deprivation in the same sense as dying at the time when one dies (rather than later). Both Epicurus and Lucretius argued that our ordinary asymmetric attitudes are irrational and since we don't regret prenatal nonexistence, we ought not regard death as a bad thing. If death is a bad insofar as it is a deprivation, the challenge posed by Epicurus and Lucretius is pressing: why should we treat prenatal and posthumous nonexistence asymmetrically? One way to respond to the challenge (and thus defend the Nagelian explanation of death's badness) is to say that, whereas one could (logically) have lived longer, it is logically impossible that one should have been born much earlier. Further, the claim is that it is irrational (or impossible) to regret that a proposition which is necessarily false isn't true. 5 This response is un-

Death, Dying and the Ending of Life, I WHY IS DEATH BAD?

13

215

satisfying. It is not clear that it is logically impossible that an individual should have been born substantially earlier than he actually was. It is not at all clear, for instance, that Socrates - the very same Socrates - couldn't (logically) have come into being ten years earlier than he in fact did. Why exactly should (roughly) the actual time of one's birth be an essential property of a person? Given that the essentiality of the actual time of birth is a controversial metaphysical claim, it is unsatisfying to use it as part of an explanation of the intuitive asymmetry. 6 The explanation will not be acceptable to anyone who denies the assumption. 7 If it is at least logically possible that one should have been born much earlier (and no reason has been offered to rule this out), then we still need to develop a response to the challenge raised by Epicurus and Lucretius (insofar as we cling to the explanation of death's badness in terms of deprivation). Recently, Derek Parfit has suggested another response. 8 His position could be put as follows. We have a (not irrational) bias toward the future to the extent that there are cases where we are indifferent toward (or care substantially less about) our own past suffering but not indifferent toward our own future suffering. Since there are such cases, and the attitudes therein seem rational, the general principle that it is always rational to have symmetric attitudes toward (comparable) past and future bads is false, and so it might be true that it isn't irrational to have asymmetric attitudes toward our own past and future nonexistence (where such periods of nonexistence are taken to be bads). Thus, death could be considered a bad thing for us, and yet we needn't assume symmetric attitudes toward death and prenatal nonexistence. Consider Parfit's example: I am in some hospital, to have some kind of surgery. This kind of surgery is completely safe, and always successful. Since I know this, I have no fears about the effects. The surgery may be brief, or it may instead take a long time. Because I have to co-operate with the surgeon, I cannot have anaesthetics. I have had this surgery once before, and I can remember how painful it is. Under a new policy, because the operation is so painful, patients are now afterwards made to forget it. Some drug removes their memories of the last few hours. I have just woken up. I cannot remember going to sleep. I ask my nurse if it has been decided when my operation is to be, and how long it must take. She says that she knows the facts about both me and another patient, but that she cannot remember which facts apply to whom. She can tell me only that the following is true. I may be the patient who had his operation yesterday. In that case, my operation was the longest ever performed, lasting ten hours. I may instead be the patient who is to have a short operation later today. It is either true that I did suffer for ten hours, or true that I shall suffer for one hour. I ask the nurse to f"md out which is true. While she is away, it is clear to me which I prefer to be true. If I learn that the first is true, I shall be greatly relieved. 11

Death, Dying and the Ending of Life, 1

14

216

A. L. BRUECKNER AND J. M. FISCHER

Parfit's claim is that it seems to be a deep-seated feature of us that we regard our own past and future sufferings asymmetrically. He doesn't explicitly defend the rationality of this sort of asymmetry, but he has pointed to a class of examples involvihg bads other than death in which it doesn't appear obviously unreasonable to hold asymmetric attitudes. 10 Let us grant, for the sake of argument, that Parfit is correct about his example. The problem is that it cannot be extended to the case of death. The reason is that Parfit's case involves a bad for a person which is experienced as bad by the person. One's own pain is perhaps paradigmatic of such bads. But death is not a bad of this kind; indeed, the entire problem of justifying our intuitive asymmetric attitudes arises precisely because death is a bad for a person which is not experienced as bad by the person. Further, it seems that it is plausible to suppose that Parfit's conclusion will only apply to cases involving bads experienced as bad by the person. Cases which are structurally similar to Parfit's except involving bads not experienced as bad by the person yield symmetric attitudes. Suppose, for instance, that you know that either some friends of yours have betrayed you behind your back nine times in the past or some friend will betray you behind your back once in the future. Here, it seems that you should prefer the one betrayal in the future (given that the betrayals are comparable, etc.). It also appears that, given a choice between being mocked once behind your back in the past and being similarly treated once in the future, you should be indifferent. (Of course, we assume here that you know that you can have no effect on the future events). 11 These cases suggest that Parfit's point only applies to the class of bads experienced as bad by the person, and not to the class of bads (like death) which are not experienced as bad by the person. Note that there are two different kinds of cases within the class of things which a particular person might reasonably regret (or wish wouldn't happen or take to be bad), but which he himself doesn't experience as bad. One kind contains things which no person experiences as bad (such as death). Another kind contains things which are experienced as bad by another person (such as another's pain). If it is reasonable to take temporally symmetric attitudes toward regrettable things which we don't experience as bad and which no one experiences as bad, then it shouldn't be 'surprising that we take temporally symmetric attitudes toward regrettable things which are experienced as bad by others. And Parfit has produced just such an example:

Death, Dying and the Ending of Life, I WHY IS DEATH BAD?

15

217

I am an exile from some country, where I have left my widowed mother. Though I am deeply concerned about her, I very seldom get news. I have known for some time that she is fatally ill, and cannot live long. I am now told something new. My mother's illness has become very painful, in a way that drugs cannot relieve. For the next few months, before she dies, she faces a terrible ordeal. That she will soon die I already knew. But I am deeply distressed to learn of the suffering that she must endure. A day later I am told that I had been partly misinformed. The facts were right, but not the timing. My mother did have many months of suffering, but she is now dead. 12

Parfit claims, about this example, that the new piece of information - that my mother's suffering is in the past - should not have a crucial impact on my attitude. Concerning the suffering of others it is rational to have temporally symmetric attitudes. This is precisely what one should expect in the light of the foregoing discussion of the appropriateness of temporally symmetric attitudes toward certain bads not experienced as bad by the person - those not experienced by anyone. The difference between our symmetric attitudes toward another's past and future suffering and our asymmetric attitudes toward our own past and future suffering is a special case of the difference between our attitudes toward bads not experienced by us and bads experienced by us. If this is correct, it is appropriate to have temporally symmetric attitudes toward the class of regrettable things experienced by others, even if it is appropriate to have temporally asymmetric attitudes toward the class of regrettable things experienced by us. 13 Thus Parfit's own example highlights the inadequacy of the present response to the challenge posed by Epicurus and Lucretius, viz. the response suggested by Parfit's examples of temporally asymmetric attitudes toward experienced bads. It might seem appealing to suggest that what makes death a bad thing for a person is that it is the deprivation of good things already had by the person. On this account, the asymmetry between our attitudes toward prenatal and posthumous nonexistence is due to the fact that the time before our birth cannot be conceived as a deprivation of good things we have already had, whereas the time after our death clearly can be so conceived. But why exactly should we care especially about the lack of good things we already have had, in comparison with the lack of good things which we could have had, had we been born earlier? The plausibility of the suggestion may come from a psychological truth which says that, in general, if a person has experienced a good thing and then been deprived of it, he tends to lament its absence (to "miss it") in a way in which a person who has never experienced the good doesn't. If a person has regularly drunk fine wines with dinner, he regrets the lack of a fme

Death, Dying and the Ending of Life, 1

16

218

A. L. BRUECKNER AND J. M. FISCHER

wine at tonight's dinner more than someone who has never had a fme wine with dinner. But why would one regret the absence of something good to which one has grown accustomed? Presumably, because one tends to be frustrated by the lack of such goods - their absence causes unpleasant experiences. When a person accustomed to fine wines must do without, he is likely to have unpleasant experiences caused by the (partially involuntary) comparison of his present quite ordinary wine with his past delightful wines. In general, it is true that, when one is accustomed to a good thing, its absence causes unpleasant experiences and is therefore expecially regrettable. But clearly this principle is not applicable to death, since death deprives a person of goods without causing any experiences at all (according to our supposition). The psychological principle may apply to bads which are experienced as bad by a person (or which cause unpleasant experiences had by the person), but it doesn't apply to death, since it is not such a bad. So this explanation of our asymmetric attitudes suffers from the same problem as the above strategy. Suppose, on the other hand, that we do not appeal to the psychological principle and instead conceive of death as a bad which is not experienced. Then insofar· as it is held that in regretting the prospect of death we regret the future deprivation of goods we have already had, it would be equally reasonable to regret the prenatal deprivation of such goods, goods which, we now know, could have graced our life had it begun earlier. If death is taken to be a bad thing for a person, and it is appropriate to take symmetric attitudes toward past and future bads that are not experienced as bad by the person, then either we ought radically to revise our attitudes toward prenatal nonexistence, or we haven't explained why death is a bad thing for a person. In "Annie Hall", Woody Allen says, "We have two complaints about life. First, life is terrible. And second, life is too short." If life is terrible, it is - in the typical case- because of bad experiences. But if life is too short, why? II. WHY DEATH IS BAD

Imagine that you are in some hospital to test a drug. The drug induces intense pleasure for an hour followed by amnesia. You awaken and ask the nurse about your situation. She says that either you tried the drug yesterday (and had an hour of pleasure) or you will try the drug tomorrow (and will have an

17

Death, Dying and the Ending of Life, I WHY IS DEATH BAD?

219

hour of pleasure). While she checks on your status, it is clear that you prefer to have the pleasure tomorrow. There is a temporal asymmetry in our attitudes to "experienced goods" which is parallel to the asymmetry in our attitudes to experienced bads: we are indifferent to past pleasures and look forward to future pleasures. Perhaps it is this temporal asymmetry in our attitudes toward certain goods, and not the asymmetry in our attitudes toward bads, which explains our asymmetric attitudes toward prenatal and posthumous nonexistence. Death is a bad insofar as it is a deprivation of the good things in life (some of which, let us suppose, are "experienced as good" by the individual). If death occurs in the future, then it is a deprivation of something to which we look forward and about which we care -future experienced goods. But prenatal nonexistence is a deprivation of past experienced goods, goods to which we are indifferent. Death deprives us of something we care about, whereas prenatal nonexistence deprives us of something to which we are indifferent. Thus we can defend Nagel's account of the badness of death by explaining the asymmetry in our attitudes toward prenatal and posthumous nonexistence. This explanation makes use of a principle clearly related to (but different from) Parfit's principle concerning the asymmetry in our attitudes toward past and future experienced bads. If we have asymmetric attitudes toward past and future experienced goods, then death is a bad thing in a way in which prenatal. nonexistence is not. 14 Let us end with a fanciful example which illustrates the present point. It is now 1985 and you will live eighty years in any case. Suppose you are given the following choice. Either you were born in 1915 and will die in 1995, or you were born in 1925 and will die in 2005. In each case, we will suppose, your life contains the same amount of pleasure and pain, distributed evenly through time. It is quite clear that you would prefer the second option- you want your good experiences in the future. Note that the periods before 1915 and after 2005 involve "experiential blanks" in any case. However, on the first option there is an "extra" blank between 1995 and 2005, and on the second option this extra blank is placed between 1915 and 1925. If one focuses simply on this experiential blank of ten years and asks whether it would be better to have the blank in the past or the future, it seems that one shouldn't care. That is, as argued above, it is rational for a person to have temporally symmetric attitudes toward bads not experienced by him. Thus, our preference for the second option - living more in the future - cannot be

Death, Dying and the Ending of Life, 1

18

220

A. L. BRUECKNER AND J. M. FISCHER

explained directly by an alleged asymmetry in our attitudes toward experiential blanks. Rather, it is crucial that the placement of the "extra" experiential blank of ten years determines the temporal distribution of experienced goods, since we do have temporally asymmetric attitudes toward experienced goods. Nagel is correct to assimilate death to a bad such as betrayal by a friend behind one's back - both bads do not involve unpleasant experiences. But the two sorts of bads are interestingly different. If death occurs later than it actually does, we will have a stream of good experiences in the future. The alternative to death is good experiences, whereas (in the typical case, at least) the alternative to a future betrayal behind one's back is not good experiences. Thus prenatal and posthumous nonexistence deprive us of things to which we have temporally asymmetric attitudes, whereas past and future betrayals do not. Death's badness is similar to the badness of betrayal behind one's back, but different in a way which explains why death is rationally regarded as worse than prenatal nonexistence. 1 5 NOTES 1 This does not imply that it is rational to preoccupy oneself with one's own death or to focus one's attention upon it constantly, etc. 2 Something is "experienced as bad by a person" roughly speaking insofar as that thing causes unpleasant experiential episodes in the person (and perhaps, the person believes that the thing is causing such experiences). 3 Thomas Nagel discusses such bads in: "Death", reprinted in Thomas Nagel, Mortal Questions (Cambridge: Cambridge University Press, 1979), pp. 1-10. Also, Robert Nozick discusses similar examples in: "On the Randian Argument", in Jeffrey Paul (ed.), Reading Nozick (Totowa, N.J.: Rowman and Littlefield, 1981), pp. 218-222. 4 Nagel, Ibid. 5 Ibid., pp. 7-8. 6 Even if one - controversially - held that generation from such and such gametes is an essential property of an individual, this would not commit one to the further essentialist claim in the text. 7 Nagel himself is unsatisfied with this response. (Nagel, Ibid. fn. 3, pp. 8-9). He points out that "it is too sophisticated to explain the simple difference between our attitudes toward prenatal and posthumous nonexistence." (Ibid.) To explain his doubts, he presents an example (attributed to Robert Nozick) in which it is granted that it is logically possible that an individual be born years before he is actually born (by prematurely "hatching" the spore from which one develops), and yet it seems that even here the intuitive asymmetry is justified. Thus, the logical impossibility of being born earlier cannot explain the asymmetry in our attitudes. 8 Derek Parfit, Reasons and Persons (Oxford: Oxford University Press, 1984), pp. 165185, esp. p. 175. 9 Ibid., 165-166. 10 Nagel seems to have been aware of some version of Parfit's claim. Given his worries about the view that it is logically impossible that one should have been born much earlier

Death, Dying and the Ending of Life, I WHY IS DEATH BAD1

19

221

than one actually was, Nagel admits that "Lucretius' argument still awaits an answer". He continues (Ibid., fn. 3, p. 9): "I suspect that it might require a general treatment of the difference between past and future in our attitudes toward our own lives. Our attitudes toward past and future pain are very different, for example. Derek Parfit's unpublished writings on this tropic have revealed its difficulty to me." 11 So a symmetric attitude towards past and future betrayals involves preference for one betrayal over several comparable ones regardless of when they occur and indifference between two comparable betrayals regardless of when they occur. 12 Ibid., p. 181. 13 Parfit (Ibid., p. 182), says: "My own examples reveal a surprising asymmetry in our concern about our own and other people's pasts. I would not be distressed at all if I was reminded that I myself once had to endure several months of suffering. But I would be greatly distressed if I learnt that, before she died, my mother had to endure such an ordeal." This asymmetry is not the same as the asymmetry between my attitudes toward my own past and my own future, yet the two asymmetries are connected as follows. The first asymmetry consists in my indifference to my own past suffering paired with my concern for another's past suffering. Given my concern for my own future suffering, it follows that I have asymmetric attitudes toward my own past suffering and my own future suffering. Given my concern for another's future suffering, it follows that I have symmetric attitudes toward another's past suffering and another's future suffering. Thus the contrast between temporally asymmetric attitudes regarding my own suffering and temporally symmetric attitudes regarding another's suffering stems from the 'surprising' asymmetry Parfit notes in the above-quoted passage. But the contrast in question, which arises from the 'surprising' asymmetry, is precisely what one should expect given the discussion in the text: the contrast matches up with the contrast between bads which one experiences and bads which one does not. 14 Though Parfit focuses upon examples involving temporally asymmetric attitudes towards pain, he speaks of our "bias toward the future" with respect to experienced goods such as pleasure as well. So he would endorse the principle about temporally asymmetric attitudes toward experienced goods, which grounds the foregoing explanation of the asymmetry in our attitudes toward prenatal and posthumous nonexistence. Though this explanation is consistent with Parfit's remarks in the passages surrounding his discussion of Epicurus on death, that discussion itself does not indicate that he had the explanation in mind: "Epicurus's argument fails for a different reason: we are biased towards the future. Because we have this bias, the bare knowledge that we once suffered may not now disturb us. But our equanimity does not show that our past suffering was not bad. The same could be true of our past non-existence. Epicurus's argument therefore has force only for those people who lack the bias towards the future, and do not regret their past non-existence. There are no such people. So the .argument has force for no one." (Ibid., p. 175) In any case, it is crucial to see that only the principle about temporally asymmetric attitudes toward experienced goods such as pleasure will afford an explanation of why death is bad. The principle about experienced bads which is suggested by Parfit's examples, it has been argued, will not generate such an explanation. 15 We would like to thank Phillip Bricker for helping us to arrive at the foregoing explanation of why death is bad.

Yale University, P. 0. Box 3650 Yale Station, New Haven, CT 06520, U.S.A.

[3] Some Puzzles About the Evil of Death Fred Feldman l. THE PUZZLES

D

eath is nothing to Epicureans. They do not fear or hate death. They do not view death as a misfortune for the deceased. They think death is no worse for the deceased than is not yet being born for the as yet unborn. They say that ordinary people, who look forward to their deaths with dismay, are in this irrational. Why do they hold these odd views? In his central argument for these conclusions, Epicurus says: So death, the most terrifying of ills, is nothing to us, since so long as we exist, death is not with us; but when death comes, then we do not exist. It does not then concern either the living or the dead, since for the former it is not, and the latter are no more. 1

The argument seems to turn on what has been called "The Existence Condition"-nothing bad can happen to a person at a time unless he exists at that time. 2 If we agree that the dead don't exist, we seem driven to the conclusion that nothing bad can happen to us once we are dead. It is just a small step then to the conclusion that death itself is not bad for those who die. Although some may find reassurance in this ancient bit of reasoning, most of us cannot help but view it as sophistry. Except in cases in which continued life would be unbearable, death is taken to be a misfortune for the one who dies. We cry at funerals; we 1Epicurus, "Letter to Menoeceus," trans. C. Bailey, The Stoic and Epicurean Philosophers, edited and with an introduction by Whitney J. Oates

(New York, N.Y.: The Modern Library, 1940), pp. 30-31. Lucretius presents essentially the same argument. See On the Nature of Things, trans. H. A.J. Munro, and The Stoic and Epicurean Philosophers, p. 131. 2jeff McMahan, "The Evil of Death," Ethics 99 (1988), pp. 32-61, at p. 33. He calls it "The Existence Requirement."

22

Death, Dying and the Ending of Life, I FRED FELDMAN

grieve for the deceased. Especially when a young person dies, we feel that she has suffered a great misfortune. And it apparently seems to most of us that our attitude is perfectly rational. So we have our first puzzle: how can being dead be a misfortune for a person, if she doesn't exist during the time when it takes place? According to the most popular anti-Epicurean view, death is bad for a person primarily because it deprives him of certain goodsthe goods he would have enjoyed if he had not died. 3 This socalled "Deprivation Approach" thus seems to require that we make a certain comparison-a comparison between (a) how well off a person would be if he were to go on living and (b) how well off he would be if he were to die. The claim is that when death is bad for a person, it is bad for him because he will be worse off dead than he would have been if he had lived. The second puzzle arises because it appears that any such comparison is incoherent. It seems to be, after all, a comparison between (a) the benefits and harms that would come to a person if he were to live and (b) those that would come to him if he were to die. However, if he doesn't exist after his death, he cannot enjoy or suffer any benefits or harms after death. So there apparently is no second term for the comparison. Thus, the Deprivation Approach seems in a covert way to violate the Existence Condition, too. 4 Suppose we find some coherent way to formulate the view that a person's death is a misfortune for him because it deprives him of goods. Then we face another Epicurean question: when is it a misfortune for him? It seems wrong to say that it is a misfortune for 31 am by no means the first to defend this sort of answer. Similar views are defended (or at least discussed with some enthusiasm) by a number of philosophers. See, for example, Jeff McMahan, "The Evil of Death"; Thomas Nagel, "Death," Noiis 4 (1970), pp. 73-80, revised and reprinted in Moral Problems, ed.James Rachels (New York, N.Y.: Harper and Row, 1975), pp. 401-409; Roy Perrett, Death and Immortality (Dordrecht, The Netherlands: Martinus Nijhof Publishers, 1987); L. S. Sumner, "A Matter of Life and Death," Noiis 10 (1976), pp. 145-171; Douglas Walton, On Defining Death (Montreal, Quebec: MeGill-Queen's University Press, 1979); and Bernard Williams, "The Makropulos Case: Reflections on the Tedium of Immortality," in B. Williams, Problems of the Self (New York, N.Y.: Cambridge University Press, 1973). 4 For a vigorous defense of the claim that the standard view involves an illegitimate comparison, see Harry Silverstein, "The Evil of Death," The journal of Philosophy 77 (1980), pp. 401-424.

206

Death, Dying and the Ending of Life, I PUZZLES ABOUT EVIL OF DEATH

him while he is still alive-for at such times he is not yet dead and death has not yet deprived him of anything. It seems equally wrong to say that it is a misfortune for him after he is dead-for at such times he does not exist. How can he suffer misfortunes then? Another problem confronts the anti-Epicurean. If we can find a coherent way to say that early death is bad for us because it deprives us of certain goods, then we probably will have found a coherent way to say that late birth also deprives us of certain goods -the goods we would have enjoyed if only we had been born earlier. Yet virtually nobody laments his late birth, or thinks it a misfortune that he wasn't born years or decades earlier. Lucretius presented a forceful statement of this puzzle. He said: Think too how the bygone antiquity of everlasting time before our birth was nothing to us. Nature therefore holds this up to us as a mirror of the time yet to come after our death. Is there aught in this that looks appalling, aught that wears an aspect of gloom? Is it not more untroubled than any sleep? 5

So another puzzle that must be confronted is this: if early death is bad for us because it deprives us of the goods we would have enjoyed if we had died later, then why isn't late birth just as bad for us? After all, it seems to deprive us of the goods we would have enjoyed if we had been born earlier. There are other puzzles about the evil of death. Some of these will be addressed as we go along. But these are the main questions I mean to discuss here. II. METAPHYSICAL AND AXIOLOGICAL ASSUMPTIONS

Before I propose my answers to these questions, I should mention some of my metaphysical and axiological assumptions. First among these, perhaps, is the assumption that there are possible worlds. I am inclined to think that a possible world is a huge proposition fully describing some total way the world might have been, including all facts about the past, present and future. Nothing I say here depends on this particular view about possible worlds. So long as it countenances an appropriate number of ap5 0n

the Nature of Things, p. 134.

207

23

24

Death, Dying and the Ending of Life, I FREDFEWMAN

propriately detailed possible worlds, any other coherent view will do as well. I write as if a given individual may exist at several different possible worlds. This may seem controversial, but I think it is really not. Suppose Myron is an actual person. Suppose he actually smokes. I may ask you to consider some possible world in which Myron does not smoke. This may seem to commit me to the view that there are other worlds relevantly like our (concrete) world, and that in addition to being here in our (concrete) world, the actual concrete Myron (or perhaps a counterpart) is also located at these other places. That, it seems to me, would be strange. In fact, however, I hold no such view. When I ask you to consider some world in which Myron does not smoke, I am just asking you to consider a huge proposition that fully describes some total way the world might have been, and which entails the proposition that Myron exists but does not smoke. Since it is more convenient to do so, I write in a "realistic" way about other possible worldsas if they were giant, concrete planets far from Earth, but populated by many earthlings. I assume that it makes sense to speak of the degree of similarity between possible worlds. Indeed, it seems to me that there are many similarity relations among possible worlds. Later I will have more to say about the details of the similarity relations that are most important for present purposes. However, if we have some particular similarity relation in mind, then it will make sense to speak of some world as being "most similar" in that way to a given world. Sometimes instead of speaking of similarity I speak of "nearness." It is just another way of expressing the same idea. 6 Now let us briefly turn to axiology. Possible worlds can be evaluated in various ways. One sort of evaluation is "objective" and "non-relational." Suppose that the very simplest form of hedonism is true. According to this view, pleasure is intrinsically good and pain is intrinsically bad. Nothing else has any (basic) intrinsic value. Let's suppose that there is a way to measure the amount of pleasure contained in an episode of pleasure; let's suppose simi-

6 The locus classicus of many of these ideas is David Lewis, Counterfactuals (Cambridge, Mass.: Harvard University Press, 1973).

208

Death, Dying and the Ending of Life, I PUZZLES ABOUT EVIL OF DEATH

larly that there is a way to measure the amount of pain contained in an episode of pain. Suppose further that the pleasure-measure and the pain-measure are commensurate, so that it makes sense to subtract amounts of pain from amounts of pleasure. 7 We can then say that the intrinsic value of a possible world is determined as follows: consider how much pleasure is experienced throughout the history of that world; consider how much pain is experienced throughout the history of that world; subtract the latter value from the former; the result is the hedonic value of the world. The simplest form of hedonism says that the intrinsic value of a world is equal to the hedonic value of that world. 8 Another way to evaluate worlds is equally "objective," but is "person-relative." That is, instead of asking how good a world is, we ask how good it is for a certain person. When I speak of how good a world is for a certain person, I mean to indicate the portion of that world's goods and evils that the individual in question enjoys and suffers at that world. Suppose again that the simplest form of hedonism is true. Then the value of a world, w, for a person, s, is determined in this way: consider how much pleasure s enjoys throughout his lifetime at w; consider how much pain s suffers throughout his lifetime at w; subtract the value of the latter from the value of the former. The result is the value of w for s, or V(s,w). I assume that these values can be expressed with numbers in such a way that higher numbers indicate greater value for the person; zero indicates neutrality for the person; negative numbers indicate badness for the person. Since V(s,w) is a measure of how well s fares at w, I sometimes refer to this as s's "welfare level" at w. 9 7 I attempted to present a clear formulation of this view about axiology in my Doing the Best We Can: An Essay in Informal Deontic Logic (Dordrecht, The Netherlands: D. Reidel, 1986). See especially Section 2.2. 8 I doubt that many moral philosophers would endorse anything like this simplest form of hedonism. Indeed, I wouldn't endorse it either. My point here is primarily to indicate something about the structure of an axiological view-it should yield an ordering of worlds in terms of value. In an attempt to make this conception most obvious, I have assumed that there is a value function taking us from worlds to numbers. This structural approach is consistent with a wide variety of substantive axiological theories. 9 lt should be obvious that in interesting cases, no one could possibly

209

25

26

Death, Dying and the Ending of Life, I FREDFEWMAN

There is a question concerning a person's welfare level at worlds at which he does not exist. The proposed account leaves this value undetermined. Although it plays no role in my argument, I stipulate that if s fails to exist at w, then V(s,w) = 0. This thesis is suggested by the proposed account of relativized value, since if a certain person does not exist at a world then he enjoys no pleasure there and suffers no pain there. In fact, I do not think that a person's real welfare level is determined in the simple-minded hedonistic way I have sketched. I am inclined to think that several other factors may contribute to determining how good a world is for a person. Among other things, I suspect that the amounts of knowledge and freedom that a person enjoys, as well as the extent to which he is forced to suffer injustice are also important. However, I prefer to proceed here on the pretense that hedonism is true. I have several reasons. First and foremost, there is the historical reason. I am engaged in a debate with Epicurus about the evil of death. Epicurus was a hedonist. Some commentators have suggested that in order to answer Epicurus, we must reject his axiology-that his view about the evil of death is inextricably tied to his hedonism. I think this is a mistake. I want to show that, even if we accept the Epicurean axiology, we can still reject the Epicurean conclusion about the evil of death. A second reason for assuming hedonism is strategic. The central intrinsic value-bearing properties associated with hedonism are ones that a person can have at a time only if he is alive and conscious then. I want to show how death can be an evil for the deceased even if this hedonistic axiology is assumed. Thus, I take myself to be trying to show that death may be an evil for a person even according to an axiology maximally hostile to this notion. If I succeed, it will be pretty easy to see how to extend the solution in the direction of more plausible axiologies. It should be clear, then, that certain sorts of solution are ruled out by my axiological assumptions. I will not be able to say (as Thomas Nagel and others have suggested) 10 that death is bad in calculate the value of a world for a person. On the other hand, we could have reason to believe that worlds of a certain specified sort would be uniformly worse for someone than worlds of some other specified sort. 10Nagel discusses this idea in his now classic paper, "Death," cited above

210

Death, Dying and the Ending of Life, I PUZZLES ABOUT EVIL OF DEATH

something like the way in which being the subject of nasty rumors is bad. Clearly enough, one can be the subject of nasty rumors even after one has died. If we think this is bad for a person, then we will want to say that one's welfare level at a world can be adversely affected by things that happen after one ceases to exist at that world. Another sort of example involves the failure of one's life projects. One's life projects may come unraveled after one has died. If we think this is bad for a person, then we can cite another way in which one's welfare level at a world may be reduced by things that occur after one's death. These claims about welfare levels are controversial, and strike me as being implausible. I would rather stick to a much more hard-nosed axiology-an axiology according to which one's welfare level at a world is determined entirely by things that happen during one's life there. Thus (for purposes of illustration) I have adopted a form of simple hedonism. According to this view, if a person never learns of nasty rumors, and never suffers from them, then they don't affect his welfare level. If a person never learns that his life project has come to naught, and never suffers from this frustration, then it doesn't affect his welfare level. Only pains and pleasures can affect a person's welfare level at a world-and these he must experience during his life. A final advantage of the hedonistic axiology is its simplicity. If we assume that the fundamental bearers of intrinsic value are experiences of pleasure and experiences of pain, and we assume that these are in principle subject to unproblematic quantification, then the determination of a person's welfare level at a possible world becomes quite straightforwardly a matter of simple arithmetic. To finds's welfare level at w,just subtract the amount of pains suffers at w from the amount of pleasures enjoys at w. Although the axiology is admittedly quite crude, its simplicity makes it especially useful for illustrative purposes. I assume that any statement to the effect that something is good (or bad) for a person can be paraphrased by a statement to the effect that some state of affairs is good (or bad) for the person. Fur-

in fn. 3. A similar approach to the evil of death is suggested by George Pitcher in "The Misfortunes of the Dead," American Philosophical Quarterly 21 (1984), pp. 183-188. 211

27

28

Death, Dying and the Ending of Life, I FREDFEWMAN

thermore, I assume here that a state of affairs (such as the state of affairs of Myron smoking) is just a proposition (in this case, the proposition that Myron smokes). Thus, for present purposes, it makes no difference whether we say that a certain state of affairs obtains, or whether we say that a certain proposition is true. In any case, instead of saying that smoking (apparently an activity) would be bad for Myron, we can say instead that that Myron smokes (a state of affairs) is bad for Myron. Instead of saying that a bowl of hot soup (apparently a physical object) would be good for me, we can say that what would be good for me is that I have a bowl of hot soup, and thus again represent the thing that is good for me as a state of affairs. I prefer to write in this way, since it induces a sort of conceptual tidiness and uniformity. I am also going to assume that when a person dies, he goes out of existence. In fact, I think this assumption is extremely implausible. No one would dream of saying that when a tree dies, it goes out of existence. Why should we treat people otherwise? My own view is that a person is just a living human body. In typical cases, when the body dies, it continues to exist as a corpse. So the thing that formerly was a person still exists, although it is no longer alive (and perhaps no longer a person). Of course, I recognize that some people go out of existence at the moment of death-for example, those located at Ground Zero at the moment of a nuclear blast. For present purposes, I will assume that everyone does. Once again, I do this in part for historical reasons-Epicurus seems to have accepted this view about death and nonexistenceand in part for strategic reasons. I want to show how death can be bad for the deceased even on the assumptions (a) that things that directly affect a person's welfare level can happen to that person only at times when he exists, and (b) that death marks the end of existence for the deceased.u Ill. THINGS THAT ARE BAD FOR PEOPLE

The central question here is how a person's death can be bad for him. The claim that someone's death is bad for him is an instance llSome commentators suppose that we stop existing when we die, but we don't stop "being." They also suppose that appealing to the existence/ being distinction helps solve the problem about the evil of death. For an

212

Death, Dying and the Ending of Life, I PUZZLES ABOUT EVIL OF DEATH

of a more general sort of claim- the claim that some state of affairs is bad for some person. It would be surprising if it were to turn out that we need two independent accounts of what's meant by statements to the effect that something is bad for someone: one account of the meaning of such a statement when the relevant object is the person's death, and another account of the meaning of such a statement when the relevant object is something other than the person's death. Surely the statement about death ought to be nothing more than an interesting instance of the general sort of statement. So let's consider the more general question first, and then focus more narrowly on the specific case concerning death. What do we mean when we say that something would be bad for someone? It seems to me that when we say that something would be bad for someone, we might mean either of two main things. One possibility is that we mean that the thing would be intrinsically bad for the person. So if someone says that a state of affairs, p, is intrinsically bad for a person, s, he presumably means that p is intrinsically bad, and s is the subject or "recipient" of p. Given our assumed hedonistic axiology, the only things that could be intrinsically bad for someone would be his own pains. Thus, Dolores suffering pain of intensity 10 from tl to t3 would be intrinsically bad for Dolores. On the other hand, when we say that something would be bad for someone, we might mean that it would be "all things considered bad" for him. At least in some instances, this seems to mean that he would be all things considered worse off if it were to occur than he would be if it were not to occur. In this case, the thing itself might be intrinsically neutral. The relevant consideration would be the extent to which it would lead to or prevent or otherwise be connected with things that are intrinsically bad for the person. Consider an example. Suppose we are interested in the question whether moving to Bolivia would be bad for Dolores. Intuitively, this question seems to be equivalent to the question whether Dolores would be worse off if she were to move to Bolivia than she would be if she were to refrain from moving to Bolivia. example of this approach, see Palle Yourgrau's "The Dead," The Journal of Philosophy 84 (1987), pp. 84-101. In this paper, I have made no such distinction. 213

29

30

Death, Dying and the Ending of Life, I FREDFEWMAN

Letting "b" indicate the state of affairs Dolores moves to Bolivia, we can say this: b would be all things considered bad for Dolores if and only if she would be worse off if b obtained than she would be if b didn't obtain. Now, if we employ the standard account of the meaning of subjunctive conditionals, together with the assumptions about values of worlds for individuals, we can rewrite this as follows: b would be all things considered bad for Dolores if and only if the value for Dolores of the nearest possible b-world is less than the value for her of the nearest possible -b-world. 12 Correspondingly, to say that a state of affairs would be all things considered good for a person is to say that she would be better off if it were to obtain than she would be if it were to fail to obtain. More exactly, it is to say that her welfare level at the nearest possible world where it obtains is higher than her welfare level at the nearest possible world where it does not obtain. If we make use of the abbreviations introduced above, we can restate these claims as follows: Dl: p would be good for s if and only if (3:w) (3:w') (w is the nearest p-world & w' is the nearest -p-world & V(s,w) > V(s,w')) 12 1 am suppressing consideration of certain complexities. One that should be addressed concerns cases in which there is no unique nearest world in which a certain state of affairs occurs-several worlds are tied for this distinction. What shall we say then? Suppose that at the real world Dolores does not move to Bolivia. Then the real world is the nearest world in which she does not move to Bolivia. Suppose that among worlds in which she does move to Bolivia, there are two that are equally near and most near the real world. Then I want to say this: if each of these worlds is worse for Dolores than the real world, then moving to Bolivia would be bad for her; if each is better for her than the real world, then moving to Bolivia would be good for her; if one is better and the other is worse, then it's not the case that moving to Bolivia would be good for her, and it's not the case that moving to Bolivia would be bad for her; moving to Bolivia might be good for her and might be bad for her. If all the nearest b-worlds have the same value for Dolores, then we can use this value when we compute the value of b for Dolores. On the other hand, if the nearest b-worlds differ in value for Dolores, then the computations become more problematic. One possibility would be to make use of the average value for Dolores of these nearest b-worlds. Another possibility would be to say that the value of b for her might be the result of subtracting the value for Dolores of the real world from the value for her

214

Death, Dying and the Ending of Life, I PUZZLES ABOUT EVIL OF DEATH

D2: p would be bad for s if and only if ('lw) ('lw') (w is the nearest p-world & w' is the nearest -p-world & V(s,w) < V(s,w')) If we make use of our assumption that worlds have numerical values for individuals, then we can say precisely how bad or how good something would be for someone. Suppose that if Dolores were to move to Bolivia the rest of her life would be a nightmare. Considering all the pleasures and pains she would ever experience (including the ones she has already experienced), her life would be worth + 100 points. Thus, the value for Dolores of the nearest world in which she moves to Bolivia is + 100. Suppose on the other hand that the value for her of the nearest world in which she does not move to Bolivia is + 1000. Then she would be 900 units worse off if she were to move to Bolivia. That tells us precisely how bad it would be for her to move to Bolivia. The value for her of moving to Bolivia is - 900. So the general principle says that to find the value for a person of a state of affairs, subtract the value for him of the nearest world where it does not obtain from the value for him of the nearest world where it does obtain. Precisely the same thing happens in the case of a state of affairs that would be good for a person. Suppose it would be good for Dolores to move to Boston. To find out how good it would be for her, consider the value for her of the nearest world in which she does move to Boston. Suppose it is + 1100. Consider the value for her of the nearest world in which she does not move to Boston. Suppose it is + 1000. Subtract the value for her of the latter from the value for her of the former. The result ( + 100) is the value for Dolores of moving to Boston. In its most general form, then, the principle may be formulated as a principle about the overall value (good, bad, or neutral) of states of affairs for persons. The overall value of a state of affairs for a person is the result of subtracting the value for him of the

of one of them, and it might be the result of subtracting the value for her of the real world from the value for her of another. In such a case, we would have to say that there is no number, n, such that the value of b for Dolores = n. In what follows, I shall write as if there is always a unique nearest world. My main points are not affected by this simplifying assumption.

215

31

32

Death, Dying and the Ending of Life, I FRED FELDMAN

nearest world where it does not occur from the value for him of the nearest world where it does occur. In other words: D3: The value for s of p = n if and only if (3:w) (3:w') (w is the nearest p-world & w' is the nearest -p-world & V(s,w) minus V(s,w') = n). IV.

THE EVIL OF DEATH

The application of these ideas to the case of one's own death is straightforward. Suppose we are wondering whether it would be bad for a certain person, s, to die at a certain time, t. Then we must ask about the value for s of the possible world that would exist if s were to die at t; and we must compare that value to the value for s of the possible world that would exist if s were not to die at t. If the death-world is worse for s than the non-death-world, then s's death at t would be bad for s; otherwise, not. Let's consider a typical example to see how this works. Suppose I am thinking of taking an airplane trip to Europe. Suppose I'm worried about accidents, hijackings, sabotage, etc. I think I might die en route. I think this would be bad for me. D3 directs us to consider the nearest possible world in which I do die en route to Europe on this trip, and to consider my welfare level at that world. I see no reason to suppose that interesting parts of my past are any different at that world from what they are at the actual world. So I assume that all my past pleasures and pains would be unaffected. The main difference (from my perspective) is that in that world I suffer some terminal pain and then a premature death, and never live to enjoy my retirement. Let's suppose that that world is worth + 500 to me-+ 500 is the result of subtracting the pain I there suffer from the pleasure I there enjoy. Next D3 directs us to consider the nearest world in which I do not die en route to Europe on this trip. The relevant feature of this world is that I do not die a painful and premature death in an airplane accident. Suppose I there do live to enjoy many happy years of retirement. Let's suppose my welfare level at that world is + llOO. D3 implies that my death on this trip would have a value of - 600 for me. It would be a terrible misfortune. Two points deserve mention here. One is the fact that D3 is a proposal concerning how good or bad a state of affairs is for a 216

Death, Dying and the Ending of Life, I PUZZLES ABOUT EVIL OF DEATH

person, and not a proposal concerning the extent to which a state of affairs benefits or harms a person. I am inclined to suspect that the concepts of benefit and harm are in certain important ways different from the concepts of being good for and being bad for a person. One such respect might be this: it might be that it is impossible for a person to be harmed or benefitted by things that happen at times when he no longer exists. It is nevertheless still possible that something bad or something good for a person might occur at a time when the person no longer exists. D3 is not intended to have any direct implications concerning harm and benefit. It is intended to be restricted to the concepts of being good for a person and being bad for a person. The second point is that nothing I have said here implies that death is always bad for the one who dies. Suppose a person is suffering from a painful terminal disease. Suppose he is considering suicide, and is inclined to think that death might be a blessing. He might be right. If his welfare level at the nearest world where he thus commits suicide is higher than his welfare level at the nearest world where he doesn't commit suicide, then committing suicide would be good for this person. 13 My point in formulating D3 is simply to show how it is possible for a person's death to be bad for him, not that everyone's death must be so. Perhaps we can now see where Epicurus went wrong in his argument for the conclusion that one's death cannot be bad for him. Perhaps Epicurus was thinking that the only states of affairs that are bad for a person are the ones that are intrinsically bad for him. Since (given our axiological assumptions, which are intended to be relevantly like his) death is not intrinsically bad for anyone, it would follow that death is never bad for the one who dies. But even the most fervent hedonist should acknowledge a distinction between things that are intrinsically bad for a person (which he will 13 1 think these remarks provide the basis for a reply to one sort of argument concerning the alleged irrationality of suicide. Some have said that suicide is always irrational since it is impossible to calculate the value of death for the deceased. See, for example, John Donnelly's "Suicide and Rationality," in Language, Metaphysics and Death, ed. John Donnelly (New York, N.Y.: Fordham University Press, 1978); and Philip Devine's The Ethics of Homicide (Ithaca, N.Y.: Cornell University Press, 1978), esp. p. 25. If what I have said here is right, the calculations are in principle possible, and some suicides are perfectly rational.

217

33

34

Death, Dying and the Ending of Life, I FRED FELDMAN

take to be pains) and things that are bad for the person in other ways. D3 is designed to calculate an important sort of non-intrinsic value. It tells us the degree of overall badness for a person of a state of affairs. Even though my death on my imagined European trip would not be intrinsically bad for me, D3 tells us that it would be overall bad for me. Another possibility is that Epicurus was thinking that if a state of affairs would be bad for a person, then it must at least cause something intrinsically bad for him. Since (given our axiological and metaphysical assumptions) nothing intrinsically bad can happen to me after my death, my death cannot cause anything intrinsically bad for me. Thus, Epicurus might have concluded that my death cannot even be extrinsically bad for me. However, D3 does not calculate extrinsic value by focusing exclusively on intrinsic goods and evils that would befall the person as a result of the state of affairs. Rather, it calculates the value of a state of affairs for a person by considering what would happen (whether as consequence or not) if the state of affairs were to occur, as compared to what would happen (whether as consequence or not) if it were to fail to occur. Thus, according to D3, my death would be bad for me not because it would cause me to suffer pain, and not because it would itself be intrinsically bad for me. Rather, it would be bad for me because it would deprive me of 600 units of pleasure that I would have had if it had not happened when it did. More precisely, it would be bad for me because my welfare level at the nearest world where it occurs is 600 points lower than my welfare level at the nearest world where it does not occur. V. SOME PROPOSED ANSWERS

At the outset, I mentioned some questions about the evil of death. These were prompted by the Epicurean challenge. I will now attempt to answer those questions. The first question was the question how, given that he doesn't exist after he dies, being dead can be a misfortune for a person. The simple answer is this: a state of affairs can be bad for a person whether it occurs before he exists, while he exists, or after he exists. The only requirement is that his welfare level at the nearest world where it occurs is lower than his welfare level at the nearest world where it does not occur. It may be interesting to consider an 218

Death, Dying and the Ending of Life, I PUZZLES ABOUT EVIL OF DEATH

example in which something bad for a person occurs before the person exists. Suppose my father lost his job shortly before I was conceived. Suppose that as a result of the loss of his job, my parents had to move to another town, and that I was therefore raised in a bad neighborhood and had to attend worse schools. I would have been happier if he had not lost his job when he did. In this case, the fact that my father lost his job was bad for me-even though I didn't exist when it occurred. It was bad for me because the value for me of the nearest world where he didn't lose his job is greater than the value for me of the actual world (which, on the assumption, is the nearest world where he did lose his job). The same may be true of cases involving things that will happen after I cease to exist (although, of course, such cases will illustrate deprivation of happiness, rather than causation of unhappiness). It should be clear, then, that the plausibility of the Existence Condition derives from a confusion. Given our hedonistic axiology, it would be correct to say that nothing intrinsically bad can happen to a person at a time unless he exists at that time. You cannot suffer pains at a time unless you then exist. However, even on the same axiology, the overall value version of the thesis is not true. That is, it would not be correct to say that nothing overall bad for a person can happen at a time unless he exists at that time. Perhaps some Epicureans have been induced to accept the Existence Condition because they fail to notice this distinction. 14 The second puzzle concerns an allegedly illegitimate comparison. It may seem that I am maintaining that when a person's death is bad for him, it is bad for him because he's worse off being dead than he would have been if he had stayed alive. Yet this suggests 14 In "How to be Dead and Not Care: A Defense of Epicurus," American Philosophical Quarterly 23 (1986), Stephen Rosenbaum proposes an interpretation of the Epicurean argument. He suggests that one crucial premise is "A state of affairs is bad for a person P only if P can experience it at some time" (p. 218). I would say that the premise is ambiguous. If taken to mean that a state of affairs is intrinsically bad for a person only if he can experience it, then (assuming hedonism or any other "experiencebased" axiology) the premise may be true-but it is not relevant to the claim that death is bad for the one who dies, since it is most reasonable to take this as the claim that death is extrinsically bad for the one who dies. If the claim is understood in this more plausible way as the claim that a state of affairs can be extrinsically bad for a person only if he can experience it, then, as I have attempted to show, the premise is false.

219

35

36

Death, Dying and the Ending of Life, I FRED FELDMAN

that there is some degree of "bad-offness" that he endures while dead. However, since he doesn't exist while he is dead, he can have no degrees of "bad-offness" then. The question, then, is this: doesn't my answer presuppose an illegitimate comparison? My answer presupposes no such comparison. I am not proposing that we compare a person's welfare level during life to his welfare level during death. I have assumed that one's welfare level at a world is determined entirely by pleasures and pains that one experiences during one's life at that world. Thus, the comparison is a comparison between one's welfare level (calculated by appeal to what happens to one during his life) at one possible world with his welfare level (also calculated by appeal to what happens to him during his life) at another possible world. I have provisionally agreed that nothing intrinsically good or bad can happen to a person at times when he does not exist. In effect, then, my proposal presupposes what Silverstein calls a "life-life comparison." 15 To see how this works, consider again the example concerning my imagined death en route to Europe. My proposal requires us to compare the values for me of two livesthe life I would lead if I were to die on the plane trip and the life I would lead if I were not to die on the plane trip. Since (according to our assumptions) the shorter life is less good for me, my death on that trip would be correspondingly bad for me. The third puzzle was a puzzle about dates. I have claimed that a person's death may be bad for her because it deprives her of the pleasures she would have enjoyed if she had lived. One may be puzzled about just when this misfortune occurs. The problem is that we may not want to say that her death is bad for her during her life, for she isn't yet dead. Equally, we may not want to say that it is bad for her after her death, for she doesn't exist then. In order to understand my answer to this question, we must look more closely into the question. Suppose a certain girl died in her youth. We are not concerned here about any puzzle about the date of her death. We may suppose we know that. Thus, in one sense, we know precisely when the misfortune occurred. Nor are we concerned about the dates of any pains she suffered as a result of that death. We assume that there are none. The present question is,

15Silverstein,

"The Evil of Death," p. 405.

220

Death, Dying and the Ending of Life, I PUZZLES ABOUT EVIL OF DEATH

rather, a question about when her death is a misfortune for her. If Lindsay is the girl, and d is the state of affairs of Lindsay dying on December 7, 1987, then the question is this: "Precisely when is d bad for Lindsay?" I have proposed an account of the evil of death. According to that account, when we say that d is bad for Lindsay, we mean that the value for her of the nearest world where d occurs is lower than the value for her of the nearest in which d does not occur. So our question comes to this: "Precisely when is it the case that the value for Lindsay of the nearest world in which d occurs is lower than the value for her of the nearest world in which d does not occur?" It seems clear to me that the answer to this question must be "eternally." For when we say that her death is bad for her, we are really expressing a complex fact about the relative values of two possible worlds. If these worlds stand in a certain value relation, then (given that they stand in this relation at any time) they stand in that relation not only when Lindsay exists, but at times when she doesn't. If there were a God, and it had been thinking about which world to create, it would have seen prior to creation that d would be bad for Lindsay. In other words, it would have seen that the value for Lindsay of the relevant d-world is significantly lower than the value for Lindsay of the relevant -d-world. And it would have seen this even though Lindsay did not yet exist at that precreation moment. A final puzzle concerns the fact that we feel that early death is a greater misfortune for the prematurely deceased than is "late birth" for the late born, even though each may deprive us of as much happiness as the other. Suppose Claudette was born in 1950 and will die somewhat prematurely in 2000 as a result of an accident. We may want to say that her premature death will be a misfortune for her. Consider the nearest possible world (call it "w3") in which she does not die prematurely. Suppose that at w3 she lives happily until2035. Since she has 35 extra years of happiness in w3, her welfare level there is higher than her welfare level in the actual world. D3 yields the result that her premature death is bad for her. But now consider the claim that Claudette suffered an equal misfortune in not having been born in 1915. This fact seems to deprive her of 35 happy years too-the years from 1915 to 1950 when she was in fact born. Yet we feel uncomfortable with the idea that her late 221

37

38

Death, Dying and the Ending of Life, I FRED FELDMAN

birth is as great a misfortune for Claudette as her premature death. Why is this? Consider the state of affairs of Claudette being born in 1915. Call it "b." In the actual world b is false. Consider the nearest world where b is true. 16 (In other words, consider what would have happened if Claudette had been born 35 years earlier.) Call this world "w4." I see no reason to suppose that Claudette lives any longer in w4 than she does here in the actual world. Any such change in lifespan strikes me as being superfluous. I am inclined to suppose that Claudette's welfare level in w4 is slightly lower than her welfare level in the actual world-after all, in w4 she probably endures hard times during the Great Depression, and maybe even catches measles, whooping cough and other diseases that were rampant in those days. If she has just fifty years to live, she's better off living them in the second half of the twentieth century, rather than thirty-five years earlier. Thus, given my intuitive sense of how to calculate what would have happened if Claudette had been born earlier, it follows that early death is worse for Claudette than late birth. Her late birth deprived her of very little value; her early death would deprive her of a lot. The proposed reply to Lucretius' challenge is thus based on an asymmetry between past and future. When I am asked to consider what would happen if Claudette were to die later, I hold her birthdate constant. It has already occurred, and I tend to think that unnecessary differences in past history are big differences between

16 In "Death" (p. 67), Thomas Nagel claims that late birth does not deprive anyone of anything, since no one could have been born much earlier than she was in fact born. This provides the basis for a quick answer to Lucretius. Derek Parfit makes a similar claim in Reasons and Persons (Oxford, England: Oxford University Press, 1984), p. 351. The argument might be based on the essentiality of origins. However, with the development of techniques for the cryopreservation of sperm and eggs, the view seems false. Even if we grant the controversial claim that each person has her origins essentially, we have to acknowledge that once the relevant sperm and egg have been frozen, it is in principle possible for her to be conceived at any time in the next thousand years or so. I grant, of course, that the issue of the essentiality of origins deserves independent discussion. I simply assume that it makes sense to speak of what would have happened if Claudette had been born earlier. This makes it possible to look more deeply into the puzzle suggested by Lucretius.

222

Death, Dying and the Ending of Life, I PUZZLES ABOUT EVIL OF DEATH

worlds. Thus, it is more natural for me to suppose that if she were to die later, it would be because she lives longer. On the other hand, when I am asked to consider what would have happened if she had been born earlier, I do not hold her deathdate constant. Instead, I hold her lifespan constant, and adjust the deathdate so as to accommodate itself to the earlier birthdate. Someone might claim that I have made an unfair comparison. They might want to insist on holding lifespans constant. They might say that Claudette would be better off living longer if the extra time were tacked on to the end of her life. They might say that Claudette would not be any better off if the extra time were tacked on to the beginning of her life. (That is, if she were born in 1915 instead of 1950 but lived until2000 anyway.) The question is vexing, since it is hard to discern Claudette's welfare levels in the appropriate worlds. My own inclination is to say that if she lives 85 happy years in each world, then her welfare level at the one is equal to her welfare level at the other. In this case, I can't see why anyone would think it would be better for her to have the 35 years tacked on at the end of her life rather than at the beginning. When the comparison is fair, D3 generates what seem to me to be the correct results. And the results are that the deprivation of 35 happy years of life is a bad thing, whether these years would have occurred before the date at which Claudette was in fact born, or after the date on which she in fact died. There are, after all, two ways in which we can rectify the apparently irrational emotional asymmetry. On the one hand, we can follow Lucretius and cease viewing early death as a bad thing for Claudette. On the other hand, we can at least try to start viewing late birth as a bad thing. My suggestion is that in the present case, the latter course would be preferable. I think it must be granted that our emotional reactions toward pleasures lost by early death are quite different from our emotional reactions toward similar pleasures lost by late birth. If my proposal is right, this emotional asymmetry is irrational. To see this, consider a variant of the case involving Claudette. Suppose (to make the case very "clean") that Claudette never experienced any pleasures or pains, but that if she had died later, she would have enjoyed one especially great pleasure ("the Late Pleasure") in her old age. Suppose similarly that if she had been born earlier, she 223

39

40

Death, Dying and the Ending of Life, I FRED FELDMAN

would instead have enjoyed an equally great pleasure ("the Early Pleasure"). In either case, her life would have contained exactly one pleasure. Given natural assumptions, my proposal yields the result that Claudette's late birth was just as bad for her as was her early death. Yet I suppose that at times near the end of her life, Claudette and her friends would have been more upset about her impending early death than they would have been about her late birth. Perhaps this emotional asymmetry is to be explained by the fact that we tend to think that the past is fixed, whereas the future is still open. Thus, we may feel that there's no point in lamenting the fact that Claudette missed the Early Pleasure. On the other hand, we may feel that there was a "real chance" that she might have enjoyed the Late Pleasure. Her loss of that seems a greater misfortune. Another possibility is that we have what Derek Parfit has called "a bias toward the future." Once they are past, we become indifferent toward our pleasures and pains; while they are still in the future, we care deeply about them. 17 If hedonism is true, this sort of asymmetry is wholly irrational. Nevertheless, it might be a deep-seated feature of human psychology. I want to emphasize the fact that my central proposal here concerns a value-theoretic question, not a question in psychology. I mean to be discussing the question about the relative evil of early death and late birth. I have not attempted to answer the psychological question about the differences in the ways in which we react to early death and late birth. If my proposal is right, then (to a large extent) our emotional reactions may be irrational. VI. AN OBJECTION AND A REPLY In "Death and the Value of Life," Jeff McMahan considers and rejects an account of the evil of death very much like the one I mean to defend. 18 He cites a number of difficulties for any such 17 Derek Parfit, Reasons and Persons. An interesting proposal based on some Parfittian ideas can be found in "Why is Death Bad?" by Anthony Brueckner and John Martin Fischer, Philosophical Studies 50 (1986), pp. 213-221. 18 McMahan (in "The Evil of Death," cited above in n. 3) discusses what he calls "the revised possible goods account." This is relevantly like my

224

Death, Dying and the Ending of Life, I

PUZZLES ABOUT EVIL OF DEATH

view. One concerns a young cavalry officer who is shot and killed in the charge of the Light Brigade. According to the story, the officer was shot by someone named "Ivan." McMahan stipulates that if the officer had not been killed by I van's bullet, he would have died just a few seconds later by a bullet fired by Boris. McMahan says that" ... our answer to the question of what would have happened had the officer not died when and how he did will be that he would have lived for a few seconds, and then he would have been killed. This leads to the unacceptable conclusion that his actual death was hardly a misfortune at all." 19 McMahan goes on to offer various revisions of the original proposal, but these seem to me to be changes for the worse (and I

explain why below). It seems to me that D3 generates appropriate results. It is important to distinguish several different things that happen in this example. Let us call the gallant officer "Herbert," and let us suppose the time of his death was 3:30p.m., October 25, I854-or "t." Here are some states of affairs that we should distinguish: PI: P2: P3: P4: P5:

Herbert dies at exactly t. Herbert dies near Balaclava. Herbert dies in the charge of the Light Brigade. Herbert dies as a result of being shot by Ivan. Herbert dies in his youth.

It should be clear that we have five different states of affairs here. In fact, each is logically independent of each of the others. Furthermore, it should come as no surprise if some of these are worse for Herbert than others. Given the details of the story, it turns out that PI and P4 are not very bad for Herbert. Neither of these deprived Herbert of much happiness, since if he hadn't been killed at t by I van, he would have been killed seconds later by Boris. It's hard to see why this calls for any alteration of D3. These states of affairs seem to me not to be very bad for Herbert. The real tragedy here is not that he died exactly at t, or that he died as proposal. He claims that it runs into the "problem of specifying the antecedent" (p. 43). 19 lbid., p. 46.

225

41

42

Death, Dying and the Ending of Life, I FRED FELDMAN

a result of being shot by Ivan; the real tragedy is that he died so ·young. Thus, P5 should be the focus of our attention. We must consider the nearest possible world in which P5 does not occur. Let's call it "w5." What sort of life does Herbert live there? Perhaps in w5 Herbert is one of the few survivors of the charge; perhaps he is wounded, but recovers and goes on to live a long and happy life. Of course, I don't know precisely what happens to Herbert in w5-but it is reasonable to suppose that in w5 Herbert's welfare level is significantly higher than it is here in the actual world. After all, in w5 Herbert does not die in his youth, but is otherwise as much as possible like he is here in the actual world. In any case, according to D3, the badness of P5 for Herbert is equal to the difference in value for Herbert between w5 and the actual world. This might be a significant difference. He might have led a long and happy life if he had not died in his youth. I mentioned earlier that I think that McMahan's view is less plausible than D3. On McMahan's proposal, we are asked to consider what happens in a world far more distant than w5. McMahan asks us to consider the nearest world in which the whole causal sequence leading up to Herbert's death fails to occur. As McMahan remarks, in the example cited, this may mean considering a world in which the Crimean War does not occur. 20 This strikes me as being implausible. To see how it could go wrong, suppose that Herbert loved excitement. If there had been no Crimean War, he would have sought excitement elsewhere. He would have taken up mountain climbing, and would have been killed in 1853. Given these assumptions, McMahan's proposal yields the surprising result that being killed in the Crimean War was good for Herbert. It seems to me to make much more sense to consider a nearer world -a world in which the Crimean War occurs, Herbert participates, but does not die a premature death. w5 is supposed to be such a world, and Herbert is better off in w5 than he is in the real world. VII.

CONCLUSION

I have attempted to formulate a coherent answer to the ancient challenge set by Epicurus. I have claimed that there is nothing 20 " . . . we must presumably imagine that the Crimean War did not occur, in which case the threat from Boris would not have occurred either" (ibid., p. 47).

226

Death, Dying and the Ending of Life, I PUZZLES ABOUT EVIL OF DEATH

paradoxical about the idea that death may be bad for the one who dies. My answer is a version of the traditional view that death is bad (when it is bad) primarily because it deprives the deceased of goods-the goods he would have enjoyed if he had lived. I have attempted to provide my answer within a predominantly Epicurean framework. I have assumed that hedonism is true, and I have assumed that when a person dies, he goes out of existence. I have attempted to show that even if we grant these assumptions, we can still maintain that death can be evil for the deceased. I have furthermore attempted to show that if we formulate our account properly, we can provide satisfactory answers to some puzzling questions: "How can death be bad for the deceased if he doesn't exist when it occurs?" "When is death bad for the deceased?" "Is there an illegitimate comparison between the welfare of the nonexistent and the welfare of the existent?" "Why is death worse than prenatal nonexistence?" Along the way, I have also discussed the merits of some other proposed solutions to the puzzles. 21 University of Massachusetts at Amherst

21 Many friends provided much-needed criticism and support, for which I am thankful. I am especially grateful to Gary Matthews for his encouragement on this and related projects. Earl Conee, John Fischer, Ed Gettier, Ned Markosian, Neil Schaefer, Harry Silverstein, and the editors of The Philosophical Review made helpful suggestions. Earlier versions were subjected to useful criticism at Montclair State College and Drew University.

227

43

Brain Death and the Uses of the Dead

[4] Brain Death and Personal Identity MICHAEL B. GREEN & DANIEL WIKLER

The legal and medical definition of death has recently changed in many states from cessation of heart and lung function to so-called brain death. Patients who have suffered irreversible loss of brain function but continue to breathe would have been accounted alive under previous medical practice and legal statute. They are now pronounced dead. Though the changes are sanctioned by leading medical and legal authorities, they have proceeded in a climate of some confusion, a symptom of which was a recent ruling by a Florida judge: "This lady is dead and has been dead and she is being kept alive artificially."' In part this confusion is merely the result of misunderstanding on the part of judges and the public of what those authorities proposing redefinition have in mind, but it also mirrors the conceptual disarray in the brain-death literature. 2 Though a large number of physicians, jurists, and philosophers now hold that brain death is death, there is little agreement about the justification for the redefinition or about the nature of the task of "redefinition" itself. New York Times, 5 December 1976. Though the literature on brain death is large, it- has been remarkably free of argument. The medical literature, especially, gives the reader the impression that no argument is needed. (This sentiment is discussed in part I and II below.) The arguments which we seek to refute are our reconstruction of what we believe to be the assumptions of the leading writers on the subject. Our references to that literature will be accordingly sparse. Two appropriately argumentative papers, with which we take special care to note parts of agreement and contrast, appeared in Philosophy & Public Affairs: Lawrence Becker's "Human Being: The Boundaries of the Concept," vol. 4, no. 4 (Summer 1975): 334-359, and David Lamb's "Diagnosing Death," vol. 7, no. 2 (Winter 1978): 144-153· I.

2.

Death, Dying and the Ending of Life, I

48 106

Philosophy & Public Affairs

The plincipal arguments for classifying brain-dead patients as dead can be sorted into two groups. Those of the first type, which we 'will call the biological arguments, hold that redefinition is required by new developments in biomedical science. The second sort of argument proposes the redefinition as a solution to a moral problem, that of indefinitely and pointlessly maintaining the irreversibly comatose, and hence justified on moral grounds. Each kind of argument has an initial persuasiveness, but this is lost when the arguments are· set out and examined in detail. We will argue, in Sections I and II, respectively, that neither of these kinds of arguments supports the thesis that brain-dead patients are dead. In so arguing, we undennine the principal theoretical sources of support for the new definition of death. Our ultimate intention is, however, to support the brain-death definition. To justify it, we provide, in Section III, what we regard as the first satisfactory rationale for regarding brain death as death. Our argument is ontological rather than biological or moral, having to do with the conditions of existence of persons. 3 We sketch what we believe to be the best theory of personal identity and draw a corollary on brain death which supports the view that persons cease to exist at that moment. Our conclusion, then, is that brain-dead patients are indeed dead, though not for the reasons that they are now thought of as dead. Whether the brain dead should be considered dead under the law is, we argue in Section IV, another issue entirely; we think they should, but provide an argument for the legal redefinition which is wholly independent of our philosophical claim. I.

THE BIOLOGICAL ARGUMENTS

Brain death-the irreversible cessation of brain function-involves two catastrophic changes in functioning. One is coma, the permanent loss of consciousness and awareness of the world. The other is the loss of 3· The notion that brain death marks the "death of the person" is common in writing on brain death, but usually receives little explicit explication or support, deriving instead either from an intuitive essentialism (see Robert Veatch, Death, Dying and the Biological Revolution: Our Last Quest for llesponsibility, New Haven: Yale University Press, 1976) or moral considerations (see below, Section II). We distinguish these approaches below, see fn. 28.

Death, Dying and the Ending of Life, I 107

Brain Death and Persanal Identity

the brain's ability to regulate certain autonomic body processes, such as respiration, which contribute to maintenance of internal homeostasis. These losses involve cessation of functioning of different parts of the brain (here for convenience to be called "upper" and '1ower," respectively-the details do not matter) and each can occur without the other.• The early, influential Harvard Report used the title "A Definition of Irreversible Coma," but consisted of instructions on diagnosing cessation of both upper and lower brain functioning, and left the reader unclear as to which event was meant to mark the patient's death. • The subsequent medical literature is not unanimous on this point, but the established view now seems to be that brain death is to be understood as cessation of all brain functioning. 6 This point is widely misunderstood. In the celebrated case of Karen Quinlan, for example, none of the parties to the dispute over termination of care advanced the claim that she was dead or brain dead, even though she was thought to be in a "persistent vegetative state," that is, shorn of mental capacities. 7 It was thus unfortunate Lhat the case was publicized as a test case for the new definition. Ms. Quinlan was (and is, as of this writing) alive according to the dominant medical brain death conception, since her lower brain continued to regulate her breathing and other life processes. Death is marked, on this view, by death of the whole brain. Permanent loss of consciousness has no bearing on the matter if the lower brain continues to do its work. Why have most medical authorities thought that loss of this capacity, as occurs in whole brain death, should be counted as death? As we shall 4· We will use the terms "upper" and "lower" to designate the parts of the brain which sponsor cognitive and regulative functions, respectively. These are not terms of the physiologist's art; it is possible that this neat division of brain parts is false to the facts and that some sections of the brain are involved in both kinds of activity, but we do not see how the present discussion would be thereby unde1·mined. 5· ]oumal of the AmeTican Medical Association 205 ( 1968): 337-340. 6. The physiological facts. are elegantly summarized in Peter MeL. Black, "Brain Death," New England ]ouTnal of Medicine 299 ( 17 August 1978): 338344; and (24 August 1978): 393-401. 7· New Jersey Supreme Court: In the Matter of Karen Quinlan, an alleged incompetent. 79 NJ 10, 355 A2nd 647 ( 1976 ). Our analysis of the matter will vindicate common intuition that Karen Quinlan was already dead wbile wbat was once her body continued to live.

49

Death, Dying and the Ending of Life, I

50

108

Philosophy & Public Affairs

argue in Section II, most of the support for redefinition derives from moral considerations. Nevertheless, two arguments have appeared which attempt to uphold the redefinition on medical grounds. The first claims that despite appearances, this is nothing more than an application of the traditional definition of death, one which has heretofore been almost universally accepted. The second sees whole brain death as a departure from tradition, but one justified on scientific grounds. We shall present and counter these in turn. Brain Death as the Traditional Standard

Has brain death been our traditional definition of death? According to Blach's Law Dictionary, death is "the ceasing to exist; defined ... as a total stoppage of the circulation of the blood, and a cessation of the animal and vital functions consequent thereon, such as respiration, pulsation, etc:: Since a brain-dead patient, with machine assistance, can exhibit all of these vital signs, it would seem that a pronouncement of death on such a patient would be a clear deviation from the old standard. Yet some of those arguing for the redefinition of death deny this. 8 Their argument first makes a distinction between the state of death and the signs or clinical indicators by which the presence of that state is detected. Thus, a switch from a set of cardiovascular indicators to the set of brain-death indicators need not mean that one tests for different states before declaring brain death.• Indeed, according to their argument, the traditional heart-lung test has been a test for brain death all along. The claim that the traditional heart-lung test has always been used 8. This version of the argument is implicit In much of the brain-death Jltecature, but it has not been spelled out In the form given here. Arguments resembling (inexactly) the one given here may be found in Clarence Crafoord, "Cerebral Death and the Transplant Era," Diseases of the Chest 55, no. 2 (February 1969): 141-145; and Robert Schwager, "Medicine and Irreversible Coma," in Tom Beauchamp and Seymour Perlin, eds., Ethical Issues in Death and Dying (Englewood Cliffs, 1977 ) . g. For certain practical reasons, some statutes have written brain death into the legal definition of death alongside, rather than in place of,. the traditional heart-lung definition. See Alexander M. Capron and Leon R. Kass, "A Statutory Definition of the Standards for Determining Human Death: An Appraisal and a Proposal," University of Pennsylvania Law Review 121, no. 1 (November 1972): 87-118.

Death, Dying and the Ending of Life, I 109

Brain Death and Personal IdentitlJ

to detect brain death is to be secured by the following sub-argument. The presence of heartbeat and breathing in the healthy individual is a sign, not only of heart and lung function, but also of a certain underlying capacity, in particular the capacity for spontaneous respiraLion and heartbeat. Call this Capacity S. The state of death is assumed, on this account, to be the loss of this capacity. The presence of heartbeat and breathing is a good indicator of the presence of S under ordinary circumstances. 'Vhen heartbeat and breathing are occurring as the result of machine maintenance of life-functions, however, the test yields a false positive. The advent of these artificial aids has made development of a more accurate test mandatory. And (the argument continues) this is what the direct tests for brain function are. Capacity S is lodged in the lower brain; use of the EEG and similar indicators of brain death give us a direct reading on the presence of S regardless of clinical context. Hence brain death represents only a technological refinement of the traditional indicators of death, and no change in the definition of death at all. The troubles with this argument are twofold. First, Capacity S-the capacity for spontaneous heartbeat and respiration-is not lodged in the lower brain. If it were, a functional lower brain would guarantee the presence of S. But it does not. If "spontaneous" means unassisted by machine, S is a capacity of the body as a whole. Hence brain death is not the same as loss of Capacity S. Second, loss of Capacity S was not the stale tested for by the traditional tests for death. Persons requiring pacemakers because of heart injury, or respirators due to spinal injury, also lack this capacity. But they are surely not to be counted as dead for that reason by the concept of death embodied in the tradilional definition. Machine dependence in ge1wral has no bearing on one's status as alive or dead.'" But Capacity S involves nothing more than general machine independence: the ability to breathe and to circulate blood without artificial support. But since the absence of S has never been the state of death, it hardly matters that brain death is an unfailing indicator of it. Hence it is no argument for adoption of a brain-death definition tlu~t the new clinical indicators of brain death are simply refinements of the traditional cardiovascular indicators for the state of death. 10.

A point well argued in Becker, "Human Being." Seep. 355·

51

Death, Dying and the Ending of Life, I

52

IIO

Philosophy & Public Affairs

Brain Death as a New Standard

What is lodged in the lower brain is not Capacity S but neural capacity for spontaneous respiration and heartbeat. The second argument seeks to provide justification for attaching special importance to this source of machine independence. Two grounds have, in fact, been independently advanced. The first is that brain death is, as a matter of medical fact, soon followed by death of the living system as a whole. The second sees brain death itself as the death of that system!' Neither ground is sufficient. The first is easily dispatched. As Becker observed, "though the loss of one vital function (say loss of the capacity to eliminate wastes) may inevitably bring about death, it does not constitute death by itsel£."'2 Brain death portends bodily death, but does not constitute it. The interval during which a brain-dead patient can be maintained by artificial life-supports is at present quite limited, a fact which Lamb cites in support of a brain-death definition of death; but surely it could be extended, perhaps indefmilely. 13 It is difficult to see why the brevity of the interval should have any bearing on the definition of death. There are a host of medical conditions which, given the current power of medicine, also inevitably lead to death of the system as a whole, just as renal failure did only a few years ago. 111ere was and is no temptation to regard the onset of these conditions as the occasion of death, and, failing further argument, this judgment extends naturally to the brain ... I 1. As Black notes in "Brain Death," these two grounds have given rise to two other kinds of validation studies for brain-death indicators. Those interested in the role of brain death in causing so-called somatic death (death of the rest of the body) have been concerned to demonstrate that somatic death is inevitable once the appropriate indications of brain death appear. Those emphasizing the central place of the brain in the body system are concerned to learn just when the brain disintegrates; what happens to the rest of the body, and when, is treated as immaterial. 12. Becker, "Human Being,'' p. 353· 13. See Lamb, "Diagnosing Death," p. 174; Black, "Brain Death." 14. There is a series of classic eighteenth-century experiments by LeGallois on rabbit fetuses and pups which makes this point forcibly. LaGallois noted that if these organisms were decapitated above the pneumogastric nerve, the body of the organism remained alive for a period of time equal to that required for suffocation to occur in an organism at that developmental level. For

Death, Dying and the Ending of Life, I

III

Brain Death and

Personal Identity

The other ground for attaching special importance to death of the (lower) brain, its alleged central role in the functioning body system, purports to offer a way of distinguishing brain death from other nusfortunes which may eventually be fatal. The contention here is that the brain is more than merely one vital organ among others; in some important sense it is the body system. Lamb states: Since the brain stem, not the heart, is recognized as the specific area which regulates all vital processes, it follows that after brain stem death the heart and other organs can never function again naturally .... The recognition that the brain stem, not the heart, is the central vital agency suggests a recent "paradigm shift" within the medical profession. 15 Lamb then proceeds to document the shift to a definition of death based on brain death-even in patients with hearts still beating-on the part of several eminent medical specialists. Lamb's argument deserves careful attention, not least because the almost universal acceptance of the death of the lower brain as a definition of death in the medical community might seem to invoke the authority of expert opinion. We shall question this authority. The argument involves inter alia two steps, both of which we find questionable. First, that a paradigm shift has occurred among scientists; and second, that this gives the rest of us reason to identify death with brain death. The problem with the first step lies in what is promised by the term "paradigm shift." If this is more than mere phrase-mongering, we will want evidence that there has been some sort of theoretical crisis in medical science which preceded the alleged shift. \Ve must see medical science in theoretical crisis, analogous to other critical junctures in very young rabbits, this was a period of several hours! Thus, the proximal cause of death in each case was asphyxiation, not the decapitation which produced it. The moral is obvious. If there is nothing peculiar in maintaining that these decapitated bodies, though waning, were still alive, there is nothing peculiar about maintaining the same position vis-a-vis brain-dead organisms so long as we are constrained to consider biological comiderations alone. For an account of the LeGallols experiments, see Solomon Diamond, Tile Roots of Psychology (New York: Basic Books, 1974), pp. 41-44. 15. Lamb, "Diagnosing Death," pp. 146-147·

53

Death, Dying and the Ending of Life, I

54 112

Philosophy & Public Affairs

the history of science, replete with rival comprehensive theories which appear to be incommensurable. But there are none here. As Lamb himself states, the essential scientific facts were all quite well known before anyone gave any thought to changing the definition of death: There was, and still is, no doubt conceming the causal relationship between cerebral death and a cessation of the heartbeat. However, it was then (i.e., before the new definition became popular) maintained that death came only with the latter.'" Scientists command autonomy in setting some definitions, whether in the course of paradigm shifts or in ordinary science. If, as has happened in the history of science, a chemist wants to change the definition of "acid" to suit theoretical needs, it is up to him and his peers to decide the issue. The revised definition of acid is correct if it augments predictive and explanatory power, and this is a matter in which only the scientist is expert; the rest of us accept the redefinition on faith. The change in the definition of death, however, does not seem to be the sort of response to scientific needs that commands our allegiance. Many of the specialists supporting the change are not scientists at all in the narrow sense of being theory-builders, as they would have to be if we are to be bound to follow their example. They are, rather, medical practitioners. And even those scientists who support this change might not be doing so in their role as theory-builders, since they may simply be responding to the practitioner's statement of need for a definition which facilitates the acquisition of fresh organs and of more room in the intensive care units. To accept Lamb's argument we would need to see new scientific laws formulated or other evidence of the new definition's theoretical, as opposed to moral or practical, utility. Without such evidence, recent wholesale change of opinion within the medical community tells us nothing more than that many physicians have chosen to pronounce brain-dead patients dead. The reader may feel that even if Lamb's appeal to authority fails to establish the centrality of the brain in the body's system of lifefunctions, that role is intuitively clear. Certainly the brain's work is 16. Ibid., p. 147.

Death, Dying and the Ending of Life, I

II3

Brain Death and Personal Identitlj

different from that of any other vital organ. It is the organizer, the integrator; the other organs form the work-force regulated by its commands. And, as Becker points out, the body's life is surely a matter of systemic functioning: the continued interaction of a hierarchy of biological and chemical subsystems. 11 What we need is a criterion for determining death of the system; and what better candidate than loss of the command center which maintains systemic integration? A more careful assessment of the lower brain's role, however, does not support the conclusion that brain death constitutes the cessation of systemic functioning. The fact that the lower brain is the element in the system which keeps other elements acting as a system does not make its continued functioning essential. It is still one among many organs, and, like other organs, could conceivably be replaced by an artificial aid which performed its function. The respirators and other life-supports which maintain body functioning after lower brain death collectively constitute a sort of artificial lower brain, and development of a more perfect mechanical substitute is merely a technological problem. When the lower brain's job is performed by these substitutes, the body's life-system continues to function as a system. The nonessential character of brain death may be brought out by some mechanical analogies: the heating system in a home can continue to function even after its thermostat fails, so long as the furnace is turned on and off manually (or by a substitute machine); an airplane continues to fly even after the autopilot fails if human pilots are able to take control. The source of control is not important; what matters is \;Vhether the job is done. The artificial life-supports now in use perform the brain's work rather poorly, as shown by the rapidity with which death of the body usually follows brain death; but not so poorly that the artificially maintained system is no system at allY 17. Becker, "Human Being," p. 353· 18. In denying the relevance of lower brain death in determining death of the

body, we follow Becker; we have been concerned here to defend him against Lamb. Were it not for our views on the identity of the brain-dead patient, we would also endorse his positive view that "a human organism is dead when, for whatever reason, the system of those reciprocally dependent processes which assimilate oxygen, metabolize food, eliminate wastes, and keep the organism in relative homeostasis are arrested in a way which the organism itself cannot reverse" ("Human Being," p. 353 ). Except for the final clause concerning spon-

55

56

Death, Dying and the Ending of Life, I

Philosophy & Public Affairs

Thus permanent cessation of lower brain functioning, which is the event known as brain death in the medical literature on the subject, does not mark the moment of death. Indeed, it does not even beat; on the question of whether the patient is alive or dead, for any of the reasons found in the medical literature. Just how these arguments miss their mark so badly will be discussed later in this essay. II.

THE MoRAL ARGUMENTS

It is morally right to discontinue care of a brain-dead patient. Does this show that brain-dead patients are dead? The answer depends on what one takes the activity of defining death to be. One often cited paper states that when we speak of human death . . . we are making a statement with policy implications. We are saying that it is now appropriate to behave towards the individual in a different way .... It is appropriate to begin burial ritual and for the deceased's friends and family to begin the mouming process .... If the individual is a holder of public office, say, the President of the United States, the Vice President would assume the office of the President. Thus, human death is a social and moral concept quite beyond the biological. ... The only reason the definition of death receives any attention at all in the realm of public policy is that the term summarizes and legitimates what might be called "death behavior," a radically different set of social relationships and actions.•• This passage suggests that to redefine death is to call for new patterns of "death behavior." If declaring a patient dead upon final cessation of breathing amounted to calling for burial and mourning at that time, taneous reversal, the fundamental idea is sound: bodily death consists of functional disintegration. This definition is appropriately vague; it cannot be used to specify an instant of bodily death in most cases, but it serves to rule out a host of illegitimate candidates. · 19. Robert M. Veatch, "The Whole-Brain Oriented Concept of Death: An Out-Moded Philosophical Formulation," ] ournal of Thanatology 3 (I 975): IJ-JO. Our continuation of the argument is not meant to represent Veatch's view since our development of this argument is not entirely consonant with his. We endorse Veatch's conclusions and our arguments are complementary to his (though see below, fn. 28 ). See his Death, Dying ancl the Biological Revolution, chaps. 1-2.

Death, Dying and the Ending of Life, I

Brain Death and Persanal IdentitiJ

then to pronounce death upon brain death would be to prescribe these practices at the earlier time. Presumably, then, on this view, the way to decid,e when death occurs is to determine when "death behavior" should begin. "Death behavior" at the bedside might include mourning, turning off life-support equipment, and removing vital organs for transplantation. Ordinarily there will be little point to delay these acts past the time of brain death, at least not for the patient's sake. Roland Puccetti writes: If someone suggested to me that my body might survive death of the neocortex for several months or years, provided it were fed and

cleaned properly, etc., that would have no greater appeal to me than preservation of my appendix in a bottle of formaldehyde. For in the sense in which life has value for human beings, I would have been dead all that time. 20

Taken to mean that life would be no more valuable to the patient than death, there can be no nonreligious objection to this assessment. The brain-dead patient has no capacity for happiness, has no interests, and arguably has no rights. Further attempts to maintain the patient's functional integrity would do him no good and are not morally required of us. The question to be answered, however, is whether the moral proposition that maintenance of the brain dead preserves· nothing of value and may be ceased when convenient shows that the brain dead are dead. Many of the medical articles on brain death seem to suggest that it does. There is little real argument of any kind for regarding the brain dead as dead, but the authors regularly mention the pointlessness of maintaining those in irreversible coma and the difficulties for the transplant surgeon caused by waiting for breathing to stop. These concerns clearly motivate the redefinition-again, no new scientific data appeared to support the shift-and, in the absence of other argument, seem to appear to the medical authors to justify the redefinition as well. 21 This view of redefinition is explicitly stated by one philosopher: 20. Roland Pucetti, "The Conquest of Death," Tire ./Houist 59 (1976): 252.

If the chief medical impetus for recognizing brain death as death was a moral one, why would the medical criteria for brain death focus on death of 21.

57

58

Death, Dying and the Ending of Life, I

II6

Philosophy & Public Affairs

The only way of choosing (between competing definitions of death) is to decide whether or not we attach any value to the preservation of someone irreversibly comatose. Do we value 'life' even if unconscious, or do we value life only as a vehicle for consciousness? 22 This account of the task of defining death, however, has unacceptable, even absurd, consequences. If our society came to value sports so much that the cripple's sedentary existence was thought to have no value, we would hardly find it congenial to reclassify the lame as dead. In any case, the value of life often drops well before brain death; years before, in some cases of senility. Indeed, many of us would prefer death over continued existence at the level of some profoundly retarded. institutionalized human beings. For all that, there is no temptation to say that these human beings are dead. When they do die, the change is obvious. A reclassification of healthy, profoundly retarded humans as dead would amount to ontological gerrymandering. Thus the account of defining death which is assumed by the moral arguments is a faulty one, and a source of its error is the way in which the speech act of pronunciation of death is construed. It is certainly true that a person who pronounces a patient dead usually acts differthe whole brain rather than on death of the upper brain alone? The patient whose upper brain is lost has, after all, a life just as low in quality as that of the patient whose whole brain is gone. Medical articles are mostly silent on this issue. An exception is The Lancet's editorial on the occasion of a statement on brain death by the Conference of Royal Colleges and Faculties of the United Kingdom. Their remark that "How long life support should be provided for such survivors of brain damage is a harrowing question, but vegetative patients are not brain dead and the questions of ethics and of the use of resources are different" ( 13 November 1976, p. 1065) reaffirms the focus on lower rather than higher, brain death, but does not explain the reasons for this focus. The tests for whole brain death were more reliable than those for upper brain death alone, but recent articles disavow this as the reason. Our conversations with medical specialists give us the impression that physicians, so used to attempting to restore the body to functional integrity, cannot bring themselves to pronounce death upon a breathing body which maintains itself with only nutritional and custodial care. Our diagnosis, then, is that moral concern to let the patients lapse, constrained by tradition, caused the physicians to adopt an otherwise unmotivated attitude of respect for the lower brain, identifying its continued functioning with life itself. 22. Jonathan Glover, Causing Death and Saving Lives (Middlesex, 1977 ), p. 45-

Death, Dying and the Ending of Life, I

117

Bmin Death and Personal Identity

ently and expects others to do likewise. But it does not follow that to pronounce a patient dead is to state an intention to act differently and to call upon others to follow suit. The conclusion does not follow, even if the sole motives of the person pronouncing death are the statement of intent and the prescription to others. The argument conflates three aspects of speech acts: what is said, the effects of saying it, and the motives for saying it. Two speakers can presumably agree on whether a patient has died even if they endorse different sorts of mourning practices; indeed, even if one of them thinks the practices should begin before death, or only after a decent inten,al. And, of course, the speaker's intent in pronouncing death may be just to inform his listeners about the condition of the deceased, with no intention of prescribing any practice. The reason that pronouncing death seems to be prescribing termination of care, and the reason that citing the moral advantages of medical abandonment seems to be a way of arguing for a redefinition of death as brain death, is that certain moral premises arc simply assumed without question or argument: one is that there is no point to giving medical care to the dead and the other is that life should always be preserved. If these positions are abandoned, the pressure to change the definition of death is immediately relieved. We have only to realize that the moment of pulling the plug need not be the moment of death to see that defining death is a different job from deciding when it is best to remove the life-support systems. The heart-lung definition of death did not, and could not, itself, have required pointless maintenance of the brain dead. That severe prescription emerges only when we add the premise that the living must not be abandoned. What the moral arguments show, then, is not tl1at the brain dead are dead but that the brain dead need not be cared for. The moral argument addresses a moral issue which is, unfortunately, confused by many with the task of defining death. Our argument in Section III supports the redefinition, not by demonstrating that the brain dead have a worthless existence, but by showing that they have no existence at all. III.

AN ONTOLOGICAL ARGUMENT

To state that an ailing patient, Jones, is still alive, is in fact to make two claims; the second of which is usually taken for granted. One is

59

Death, Dying and the Ending of Life, I

60

118

Philosophy & Public Affairs

that the patient is alive. The other is that the patient is (remains) ]ones. 23 It is natural to assume that the living patient, who entered the hospital as Jones, must still be Jones (who else could it be?).u But we will show that this is mistaken. If we do establish that the patient, even if alive, is not Jones, and if no one else is Jones, then we will have established that Jones does not exist. And this, of course, establishes that Jones is dead. ]ones' death thus occurs either at the time that the patient dies, if the patient has remained Jones; or at the time the patient ceased to be Jones, whichever comes first. If, as we contend, the patient ceased to be Jones at the time of brain death, then Jones' brain death is Jones' death. 25 Thus, if the loss of capacity for mental activity which occurs at brain death constitutes death, it is not for moral reasons, nor for biological reasons, but for ontological reasons. 26 We need, then, to show that the patient ceases to be Jones when brain death strips the body of its psychological traits. We accomplish this by drawing a corollary about brain death from what we believe 23. The term "patient" is used neutrally to designate the entity in the hospital bed. For a discussion of the relation of existence and identlly, see Roderick Chisholm, "Coming into Being and Passing Away," in Stuart F. Spieker and H. Tristram Englehard~. eds., Philosophical Medical Ethics: Its Nature and Significance (Dordrecht, Holland, 1977), especially p. 171. 24. That brain-dead patients can nevertheless be alive is argued in Section I, above. · 25. It is worth emphasizing that we are not speaking of two distinct events, loss of identity and death, which occur in the life of every individual. Once Jones is no longer the living entity we call "the patient," Jones is gone and dead. That his body lives or dies is immaterial. As Uecker himself notes ("Human Being," p. 354), the continued life of organs and other entities which were once a part of a living person does not constitute continued life of that person. What matters is whether these tissues are that person. Indeed, Jones would be dead, even if his living body were to become another person-assuming, as we shall show, that such talk makes sense. This shows that the issue is not whether the patient is a person after brain death. It is whether the patient is that person, Jones. 26. Gareth Matthews offers an ontological argument for brain death in his "Life and Death as the Arrival and Departure of the Psyche," American Philosophical Quarterly 16, no. 2 (April 1979): ISI-157· Matthews' argument, which proceeds from a nco-Aristotelian notion of psyche distinct from current conceptions of "person," identifies death of an organism with the loss of the capacity for self-regulation in the way distinctive of its kind. We find his proposal unsuccessful; a comprehensive examination of Matthews' position would not fall within the scope of this article.

Death, Dying and the Ending of Life, I

II9

Brain Death and Personal Identity

to be the correct theory of personal identity. 27 In what follows, we sketch the theory briefly, emphasizing the role which the theory attributes to the brain function. Then we state the corollary and use it to show that the criteria of personal identity thus derived do not permit identity to survive the kinds of changes which brain death involves. The result justifies a definition of death as brain death. 28 27. TI1e theory is essentially the view developed by John Perry in a series of articles; see his contributions to his edited volume Persoualldeutity (Be1·keley and Los Angeles: Unlve1·s1ty of California Press, 1975) and to Amelie Rorty, ed., The Identities of Persons (Berkeley and Los Angeles: University of California Press, 1976); and his review of Bernard Williams' Problems of the Self in Journal of PhilosO}Jhy 73, no. 13 (15 july 1976): 416-428. Bernard Gert's "Personal Identity and the Body," Dialogue, September 1971, pp. 458-478, from which we also learned much, is in important respects similar. A similar them·y is embodied in jewish Jaw, according to Rabbi Azriel Rosenfeld, who has noted its importance for the brain-death controversy; see his "The Heart, the Head, and the Halakhah," New Yorh State Journal of .Medicine (15 October 1970): 2615·2619. 28. The notion that the brain-dead patient has "ceased to exist as a person" (see fn. 3, above) has both a moral and an ontological interpretation. The moral claim is simply that the patient's life now lacks the features that make life more valuable for people than death. Our ontological claim is that the person who entered the hospital, he whose body is now brain dead, no longer exists (though his body or some of its parts may both exist and live). Robert Veatch, in his Death, Dying and the Biological Revolution holds that "to define the death of a human being, we must recognize the characteristics that are essential to humanness ... , We use the term death to mean the loss of what is essentially significant to an entity-in the case of man, the loss of humanness" (p. 26). "To ask what is essentially significant to a human being is a philosophical questiona question of ethical and other values. Many elements make human beings unique-their opposing thumbs, their possession of rational souls ... and so on. Any concept of death will depend directly upon how one evaluates these qualities" ( pp. 29-30 ). This seems to be a hybrid: moral considerations dictating our ontological conclusion. Veatch conducts his search for what is "essentially human" and "essential to man's nature" by contrasting various "traditions" ("the empiricalist philosophical tradition seems to be represented in the emphasis on consciousness .... At least in the \Vestern tradition, man is seen as an essentially social animal," p. 41 ), by drawing upon his own and others' intuitions on man's essence, and by citing perceived moral dangers of adopting the views he opposes. It would be hru·d to argue for or against this view except by citing one's own intuitions on "man's nature," if any. We avoid the evident difficulties of this approach by constructing a step-by-step argument using the intuitive "data" on personal identity drawn on by Locke and his successors. Though our argument, like Veatch's, appeals in the end to intuitions, it appeals to conceptual intuitions, not moral ones, on which there is considerable consensus, and which are used to test a concept (personal identity) which is considerably better defined

61

Death, Dying and the Ending of Life, I

62 120

Philosophy & Public Affairs

Accounts of Personal Identity It will seem to some-particularly to philosophers familiar with the literature on personal identity-that there exists a very short argument proving that brain-dead patients are not identical to the persons who once were associated with their bodies. Bernard Gcrt stales:

... if a body does not have any psychological features, then it is not a person, and hence the question of personal identity cannot even arise. 2 " This argument, however, must be recognized as an enthymeme, a full statement of which involves the acknowledgment of two quite substantial metaphysical assumptions. The first of these is an account of what is essential for an individual to belong to a certain kind; in the present case, the claim is that an entity is a person only if it has psychological properties. The second assumption is that it is essential for the continued existence of an individual that it remain a member of the kind to which it belongs. Otherwise, although having the capacity for psychological slates might be essential to being a person, it could simply be the case that being a person, like being a musician, was a property that a given individual could acquire or lose in the course of a lifetime without ceasing to be the same individual. Each of these assumptions is notoriously controversial. There is nothing approaching a definitive account of the essence of personhood. And apparent counterexamples to the second assumption abound in nature: a bit of radioactive uranium, for example, may decay and become lead without ceasing to exist.•• Indeed, on some accounts of what a person is, even persons come into the world as non-persons and, retaining their identity, enter that category upon suitable psychological development. Thus at least one boundary of that kind is penneable. The approach we take here permits us to remain agnostic on the than Veatch's "human existence." Our individual-essentialism is, relative to Veatch's (and others') kind-essentialism, better methodology if not better metaphysics. 29. Gert, "Personal Identity and the Body," pp. 475-476. 30. Berent En~. "Natural Kinds and Linnaean Essentialism," Canadian Journal of Philosophy 5, no. 1 (I September 1975): 83-102. See also Marjorie Price, "Identity through Time," Journal of Philosophy 74, no. 4 (April 1977): 201-217.

Death, Dying and the Ending of Life, I 121

Brain Death and Personal Identity

issue of kind-essentialism and on whether membership in a kind is essential to retention of identity. We offer instead an argument which establishes a claim about the essential properties of a given individual. The claim is that the continued possession of certain psychological properties by means of a certain causal process is an essential requirement for any given entity to be identical with the individual who is Jones. Thus, we can afford to remain uncommitted on whether persons are essentially beings with psychological properties and on whether Jones is essentially a person. \Ve demonstrate instead that Jones, whateve1· kind of entity he is, is essentially an entity with psychological properties. Thus, when brain death strips the patient's body of all its psychological traits, Jones ceases to exist. What is required to show that Jones is an individual who essentially possesses certain psychological capacities is an adequate account of personal identity. The literature on this topic generally assumes that certain individuals are identifiable as persons and then gives an account of the criteria according to which a person existing at one time is to be identified with one existing at a later lime. For convenience of exposition, we will continue to talk as though the individuals we pick out are persons, but this will not affect the force of our argument. We need only claim that in normal cases we are able to identify the individual Jones (whether or not he is a person) in order to show that this individual's identity is essentially connected with the possession of certain psychological properties. Two sorts of personal identity criteria have been proposed. One concems continuity and connectedness of personality, memory, and other mental phenomena; the other stresses spatia-temporal continuity of the physical body. On the "mentalist" view, two "person-stages" are stages of the same person, just in case the latter is a continuation of the earlier personality and can remember what the earlier one has done. 81 The body-continuity view makes no such requirement; having the same body is sufficient and necessary. The mentalist's basic argument is the apparent conceivability of 31. A person-stage is a person in a given time interval. A person simpliciter is a series of person-stages. The problem of personal identity, thus conceived, is to state the criteria determining which person-stages are stages of the same entity.

63

Death, Dying and the Ending of Life, I

64 122

Philosophy & Public Affairs

body-switching. We can, the mentalist asserts, readily imagine persons coming to inhabit other person's bodies, or switching bodies between them. It certainly seems possible to imagine oneself waking up in another's body. We might have a hard time convincing others who we were, but we feel that we should know it in an instant. The fact that our body might be unfamiliar would be beside the point from the subjective point of view. When we narrate these examples as we usually do, it is plain that we are using a mentalist criterion of personal identity. To the extent that the descriptions of events in these examples are intuitively correct, argues the mentalist, we have established the soundness of the mentalist criterion. The body-continuity theorists, however, have an effective counter to this argument. What is disputed is not whether these examples can be imagined, but how they ought to be interpreted. Let us suppose that a person did wake up feeling that he was Jones, but had a new body, or that two persons seemed able to remember events which had happened in the lives of the persons who had heretofore had the other's body. This does not show that a transfer of bodies has occurred. The subjective phenomenon might be delusive. Many persons have thought they were Napoleon or some other person; they may even have known enough about the other's life and personality to mimic the other person convincingly. If Jones and Smith woke up, each claiming to be the other, this folie a deux would simply be the situation which the mentalist mistakenly calls a body-switch. The difficulty for the mentalist is that an additional criterion of personal identity must be found, one which would establish that the memories are not delusional. But the surest criterion for genuine memory is that the remembered event be one which happened in the rememberer's own history. The circularity is apparent. Can the mentalist show that the putative identity changes in the body-transfer examples are not merely identity-preserving changes of psychological qualities? Not unless the examples are narrated in greater detail; everything that happens in the example mentioned above is consistent with a body-theorist's interpretation. But the mentalist can elaborate the story in a way that rules out this possibility and thus demonstrates the conceptual possibility of a person being serially identical with distinct bodies.

Death, Dying and the Ending of Life, I 123

Brain Death and Personal Identity

This consists of making clear why the narrated changes occur. The body theorist's counter-construal is silent on this point, but one's imagination can supply a range of possible causes. Jones might have been subjected to brainwashing by someone who wanted him to feel and act like Smith. Or Jones might have endured a psychotic episode in which the character of his friend, Smith, provided the organizing motif for his delusions. Or the delusions might coincidentally have formed a coherent personality and imagined life-history which duplicated that of Smith, a total stranger. Other explanations could be dreamed up, but we need not extend this list. What is important for our purposes is the fact that the bodytheorist's counter-construal appears plausible only because the alleged body-transfer narrative permits us to assume the applicability of one or another of these explanations. But suppose a body-switch narrative could be told in sufficient detail to preclude the typical body-theorist interpretations. If all these possible explanations of the qualitative changes were ruled out, we would have no reason to suppose that mere qualitative change had occurred. Then, unless the body-theorist is prepared to offer, in addition to a reconstrual of identity change as qualitative change, some new and powerful explanation for the latter, his argument will fail to convince. We would look for some other account. The likely alternative, of course, would be that of the mentalist: these changes occurred because the persons switched bodies. This account, too, can be convincing only upon the assumption ~hat an adequate explanation of events is offered, that is, an explanation of how (what we want to call) Jones' assumption of Smith's body occurred. And the explanation or explanations which would be acceptable here will characteristically be different from those which would have been accepted as accounting for personality changes. In fact, the explanations can be enumerated. If one believes in a Cartesian soul, the body-transfer must involve some sort of incorporeal unhitching from Jones' body and subsequent rehitching onto Smith's. The mentalist, however, is not tied to Cartesian dualism. Even if he grants that all mental events are physical events, there is a way that an event in Smith's life could come to be remembered by a person with Jones' body. Since memory is (presumably) stored in the brain, and since recall is (presumably) an activity of the brain, the switching of bodies

65

66

Death, Dying and the Ending of Life, I 124

Philosophy & Public Affairs

between Smith and Jones could be brought about by transplanting Smith's brain into Jones' body and vice versa. 32 To put the point in general terms: body-transfer will have occurred if and only if the sort of explanation of the continuity of a person's psychological properties is the same as that which explains this continuity in ordinary life histories (that is, those involving no body transfer). Personal identity presupposes a characteristic causal Lie between person-stages (what the cause is, exactly, depends on what, empirically, causes psychological continuity). Indeed, this causal tie is the criterion of personal identity. 33 Persons can switch bodies, then, if the causal connections between the psychological events in the respective bodies are the same as those which nonnally obtain among the psychological events which occur in the history of a nonnal human body. \Ve know these causes to operate primarily within the neural system; specifically the brain. Thus it is that brain-transplantation would also constitute transfer of a person from one body to another. Personal identity, then, seems to be congment with lmlill identity. This proves to be a happy result in con32. That a materialist view of mind and person can coexist with our view that persons cease to exist before their bodies do is denied by Saul Kripke ("Naming and Necessity," in G. Harman and D. Davidson, eds., Semantics of Natural Languages, Dordrecht, 1972, p. 334 and fn. 73). Kripke, after attacking a crude materialist view, admits that a sophisticated materialist cum mentalist would identify the person with a body whose brain enjoyed the proper physical organization. Kripke claims that even this latter approach falls prey to modal difllculties, but he does not attempt to show this, and we have not been persuaded of any incompatability of materialism with our position by the objections he does offer (for a fuller treatment of Kripke's views; see Fred Feldman, "Kripke on the" Identity Theory," Journal of Philosophy 71, no. 24 (October 1974): 666 If. The outcome of this debate is not relevant to our present purposes, however. Although we have spoken of memory as though it is a brain process, thus suggesting support of a mind-brain identity thesis, all we require is that mental life be nomologically dependent on brain-functioning and that the capacity for consciousness should follow the undamaged brain wherever it might go. This we take modem neuroscience to have established beyond controversy. 33· Body-continuity theorists aim their fire at examples which, like Locke's, leave the causes unspecified. The brain-transplantation variants escape their attention (a point made by Perry in his review of B. Williams, Problems of the Self, and by Gert, "Personal Identity and the Body") and the various arguments used against the Lockean cases score no points against them.

Death, Dying and the Ending of Life, I 125

Brain Death and Personal Identity

sideration of several of the familiar cases. Consider, for example, a decapitated person whose head and body are each sustained by hightechnology medicine. 34 If the head can communicate with us, and shows psychological connectedness and continuity with the erstwhile person, there can be little doubt that it is that person and the body is not. But the equation of personal and brain identity needs to be qualified. Suppose that during a brain-transplantation experiment, the brain is inadvertently subjected to a process which removes all memory traces from it; indeed, "unwires" the brain so completely that the owner's entire complement of mental traits and capacities is permanently erased.•• We feel no urge to regard the individual resulting from placing the unwired brain in a "new" body as the person previously associated with the brain. Brain identity alone is insufficient for personal identity. The reason for this follows immediately from the account given above: the ordinary causal processes which link events in a personal history involve more than spalio-tcmporal 34· Or, substitute brain for head, assuming that the brain can be hooked up to sensory devices and provided with some means of self-expression. An early treatise on medical jurisprudence, Manual of llledical jurisprudence and State Medicine by M. Ray (London, 1836, 2d ed. ), p. 499, stated that "individuals who are apparently destroyed in a sudden manner, by certain wounds, diseases or even.decapitation, are not really dead, but are only in conditions incompatible with the persistence of life." On our own view, the body may well be alive-so may the head-but only one is the person who was once whole. Becker, "Human Being," p. 358, says: "Bizarre questions may be raised, of course. Is a human brain separated from its body and kept 'functioning' a human being? ... I admit to being at a loss for a reply to such cases, let alone an answer." Again, Becker has addressed the ·wrong question: we want to know not (only) if the brain is a human being, but whicl1 human being, where the "human being" means "person." Becker would presumahl)• find our question \Ill· answerable as well; we simply regard the case as more easily conceivable than Becker professes to do. On the conceivability of isolated brains having psychological states, see Bernard Gert, "Can a Brain have a Pain," Philosophy and Phenomenological Research 27, no. 3 (!\'larch 1967): 432-436, and references cited therein. For a case in which such a state may already have existed in an isolated brain, see fn. 19 in Pucetti, "The Conquest of Death." 35· This sortof imagery is, of course, fatuous. We can foresee no reason why a variant which is truer to the facts should not provide the same conclusion, however. Perry refers to this process as a "brain zap" in his review of B. Williams, Problems of the Self.

67

68

Death, Dying and the Ending of Life, I 126

Philosophy & Public Affairs

continuity of brain tissue, They also require continuity of certain brain processes, carried out through microstructural and microfunctional registrations in the brain tissue. Two body-stages which fail to be linked by continuity of these processes will fail to be stages of the same person, even if identity of the brain is preserved."'' Identity a11d Brain Death

Our argument that personal identity does not survive brain death requires but one further observation. \Ve have throughout this discussion spoken of body transfers, cases in which a person's identity "leaves" one body and "arrives" at another. \Ve have undertaken to state the conditions under which the entire transfer occurs. We have a fortiori stated the conditions under which each component of the transfer occurs; in particular, under which the ceasing to be associated with one body occurs. When a brain and its ongoing ordinary physiological processes are removed from one body to another, personal identity follows. But it is not necessary lhat the functioning brain reach its destination for identity to have vacated the donor body. If the brain fails to find a new home and is destroyed, the person's history comes to an end; the continued life functioning of the person's former body is as irrelevant to the issue as it would be had the brain been lodged in the intended new body and the resulting person continued to live. \\'e submit that, in quite ordinary clinical circumstances, a braindead body has similarly been stripped of the identity of the formerly associated person; and this holds true even if that body continues to live. It is not necessary that the brain actually be removed for personal identity to quit the body. The reason that brain removal cancels personal identity in the donor body, after all, is that the resulting 36. Contrast this case from Gert's "Personal Identity and the Body": Jones is hypnotized and made to think, feel, and otherwise resemble Smith in all mental respects. Then the brain is transplanted into Smith's body. The resulting individual is Jones, even though it is psychologically similar to Smith and has Smith's body. The reason is that, in this case, the causal relations between body-stages (hypnosis as well as transplantation) are personal-identity preserving. (In fact, this case involves considerable philosophical license: if hypnosis or other psychological procedures could produce the vast changes alleged here, our concept of personal identity would be considerably different from what it is.)

Death, Dying and the Ending of Life, I 127

Brain Death and Personal Identity

body cannot thereafter have the kind of causal relation to earlier person-body stages which is required for the personal identity to hold. Brain death has the same result. Indeed, it is unimaginable that the identity question could be much affected by the physical removal of a permanently malfunctioning (and possibly liquified) brain from a living brain-dead body. Brain death and brain removal have much the same result; the dead brain serves only to add bulk to the body if left intact. If, as has been established above, removal of the conscious, functioning brain leaves us with a body not identical to the person formerly associated with it, surely removal of a dead brain leaves just the same thing; and no more remains when the brain dies in place. Personal identity can, in theory, survive an operation in which the brain is removed from the body; it will lodge either in the brain (if suitably maintained) or in a new host body. If the brain dies, whether in its customary body or elsewhere, so does the person whose brain it is. The continuation of life in the body is immaterial in deciding this question. 37 That body is no more Jones than would be the continuation of the life of each of his cells in Petri dishes, or of each of his organs in the proper solution, after an ordinary death; it is merely better integrated tissue and organ functioning. We have, then, an argument for regarding a person's brain death as that person's death tout court which avoids both covert moral prescriptions and spurious appeals to the authority of biomedical science. The death of persons, unlike that of bodies, regularly consists in their ceasing to exist. We have now only to relate our ontological argument to the braindeath controversy. We have argued, following other personal identity theorists, that a given person ceases to exist with the destruction of whatever processes there are which nom1ally underlie that person's psychological continuity and connectedness. We know these processes are essentially neurological, so that irreversible cessation of upperbrain functioning constitutes the death of that person. Whole-brain death is also death for persons, but only because whole-brain death is partly comprised of upper-brain death. Tests for either will be tests for death. 37· The same would be true if a new person's brain, carrying that person's brain structure and processes, were transplanted into the body. The result would be a living person, but not the person, now dead, whose body it formerly was.

69

70

Death, Dying and the Ending of Life, I 128

Philosophy & Public Affairs

Of course, our view does not imply that a person dies with his last moment of consciousness. ·what matters is the preservation of the substrate, not the psychological states which it produces. Hence a person who suffers brain death during sleep dies at the time of brain death, not the time of onset of sleep. Similarly, a person in a persistent coma might be alive if enough of the brain remained structurally and functionally intact. Further, it does not follow from our argument that all humans lacking the substrate of consciousness are dead. Anencephalic infants are lacking at birth the cortical material necessary for the development of cognitive functioning and, arguably, consciousness. Still, due to possession of a functioning brain stem, they may have spontaneous breathing and heartbeat, and a good suck. Accounts which simply identified life with upper brain function would have to classify these infants as dead, which they obviously are not. We, on the other hand, need only point out that the identity criteria for the anencephalic, never-to-be conscious infant do not involve causal substrates for higher level psychological continuity. The conditions for life and existence will be those of human bodies rather than those of persons. Finally, we emphasize that our argument, requiring no moral premises, is not sufficient to yield moral conclusions. We show that the brain dead are dead, but it does not follow that brain death is the appropriate moment to turn off ventilators or to remove organs. These acts. and other "death behavior" might be in order at some earlier time, as circumstances dictate. The relationship between defining and pronouncing death, and discontinuing care will receive further discussion in our section on the legal definition of death. IV.

THE LEGAL DEFINITION OF DEATH

We have argued for the definition of death as brain death, and hence have no major quanel with the cunent trend toward that definition in statutory and case law. 38 But neither, in our opinion, should those who 38. Two unresolved problems of detail pose problems for those writing brain-death statutes. One is how to specify medical criteria of brain death in a way that assures a minimum of false positives. The second is the harmonious accommodation of the quite different criteria which will be used by physicians and coroners called upon to pronounce death in settings as diverse as intensive

Death, Dying and the Ending of Life, I 129

Brain Death and Personal Identity

reject our arguments and who cleave to the traditional heart-lung definition. The reasons we will give for this are moral ones. Our position calls for some explaining, especially in view of our insistence (in Section II) that moral considerations should be resolutely ignored in formulating a definition of death. To Becker, allowing moral considerations to dete1mine a definition of death is a cowardly evasion: Rigging the definition of death ... while tempting, is an avoidance of the real issue. The real issue is whether, and if so when, it is moral to give up trying to prolong the patient's life ... It seems best to face this problem directly-by defining when it is permissible to give up life-saving efforts-and not to evade the problem by introducing an ad hoc definition of death. 30 But Becker offers no argument to support this last claim. Perhaps he felt he could assume agreement on the importance of being true to the facts. If so, his confidence is misplaced. However imperative truth is in theoretical work, the cardinal virtues of our social institutions are justice and utility. These will, we feel, attach to a policy that does evade the problem (assuming, for the sake of argument, that it is a problem) worrying Becker. Becker feels, as do we, that discontinuation of medical care of brain-dead patients is morally acceptable, even mandatory. By Becker's definition of death, such action amounts to giving up life-saving efforts. If this sort of medical action were to be endorsed by statute or other governmental sanction, it might take the form of releasing physicians from the legal obligation to preserve the life of the irreversibly comatose. 40 The standard argument against this sort of policy is that it threatens a slide down a slippery slope: today we endorse the withdrawal of care from the comatose; perhaps tomorrow from the senile, the moderately retarded, the nonproductive. The likelihood of such a slide is probably impossible to measure with any confidence, care units, accident sites, and cornfields. Neither of these has significant theoretical iuterest. 39· Becker, "Human Being," pp. 356-357. 40. Since doctors are l'arely, if ever, prosecuted for decisions to cease medical support of the permanently comatose, such a statute might not be needed.

71

72

Death, Dying and the Ending of Life, I

Philosophy & Public Affairs

but it surely is not negligible. Some persons now being retained in institutions for the mentally retarded, after all, have a cognitive life not much different from Karen Quinlan's. And the historical precedent (admittedly under vastly different sociopolitical conditions) has already been set." Whatever danger exists of a dtift toward unjustified euthanasia would be significantly lessened if the statute licensing termination of care of brain-dead patients were one which classified them as dead rather than alive; one which, given a heart-lung definition of death, evaded the euthanasia issue. For the public, whether justifiably or not, seems willing to regard brain-dead patients as dead. A brain-death statute thus has the virtue of leaving official public policy on euthanasia, presently quite restrictive, apparently unchanged. This legal step would portend a threat to the senile and retarded only if the public came to think of them as dead, an eventuality which is most unlikely. The danger of the slippery slope is, then, blocked by setting a precedent which would be nearly invulnerable to distortion. Since almost all parties to the euthanasia dispute agree that the best (minimum) policy would be one which authorized withdrawal of care pf the permanently comatose but no one else, licensing "letting die" by braindeath statute would be an ideal step. Are there any serious moral objections to such an evasion? The fact that the policy would (in the view of those endorsing the heart-lung definition of death) embody a falsehood may count against it. But public policy should be granted some autonomy in these matters. The question of whether (to take an historical example) tomatoes should be classified as a vegetable or a fruit for tax purposes ought to be decided by political and economic considerations, not by biological ones." It is true that an evasion in brain-death legislation would serve to dampen rather than stimulate public debate on these matters. But given the dangers alleged to be inherent in such debate, this may count against an "honest" statute. 43 41. Leo Alexander, "Medical Science Under Dictatorship," New England Journal of Medicine 241, no. 2 (14 July 1949): 39-47. 42. Though once the rules are set, those adversely affected may try to use a biological argument to reverse the ruling and to advance their· interests. See Charles M. Rick, "The Tomato," Scientific American 239 (2 August 1978): 76-86. 43· There are two positions which are immune to our argument on public

Death, Dying and the Ending of Life, I 131

Brain Death and Personal Identity

Those who (mistakenly, in our view) do not agree that brain death is death, then, may still have reason to support brain-death legislation authorizing withdrawal of life supporls from brain-dead patients. Such support will be automatically coming from those convinced, as we are, that brain dead patients are dead (though, as we argue here, such a conclusion is not always sufficient to dictate public policy). The philosophical arguments determine the truth of the definition but not the morality of propounding it. V.

DEATH AS BRAIN DEATH

We have argued that the death of a person's brain is that person's death. Since we believe that the argument we provide constitutes the only grounds for accepting the theoretical definition of death as brain death, it is important to reemphasize the differences between our argument and those which have been most widely accepted. We have distinguished a number of biological arguments for defining death as brain death. They can all be faulted for their focus on the brain stem since (by our argument) it is loss of upper brain function which marks the person's death. Death of the brain stem is no more constitutive of death simpliciter than death of the kidneys or of other vital organs. The arguments we have criticized in Section I rest either on unwarranted identification of the brain stem with the policy. We shall state these without any discussion, though both raise important issues deserving greater consideration. First, those who both disagree with our conclusions and believe that it would be a good thing to have a debate leading to a policy of limited euthanasia would not want to take the muted course we advocate. Second, there are those who will be concerned about an important difference between the "tomato" case nut! the topic of this paper. The justillcntion for the evasive public policy in the present case rests on the assumption that those making policy have made a correct moral judgment on a very important topic and that this decision justifies the evasive legislation. There is no comparable moral decision in the tomato case, just prudential considerations, so that no "evasion" arises, merely pragmatic redefinition. Thus, the tactic advocated here, even if it is supported in this case by a correct moral judgment, is implicitly authorizing lawmakers to employ evasive or conceptually mistaken legislation when, in their best judgment, moral considerations lead them to conclude it would be justified. The path advised here may be an instance of a kind of policy making which, even if justified in this individual case taken in isolation, should probably not be encouraged.

73

74

Death, Dying and the Ending of Life, I 132

Philosophy & Public Affairs

body's biosystem as a whole or on erroneous appeals to the authority of medico-legal tradition or current medical opinion. On our view, the central question about death-which state of a person's body constitutes death-is answered in large part by means of a metaphysical argument. TI1e only purely "biological" question is how this state can be clinically identified, and it is on this issue-not the· former-that we must defer to medical expertise. TI1e moral argument proceeds from the intuition that the existence of the brain-dead patient is without any value to itself or others. Those using this argument construe the definition of death as a prescription of certain actions. The argument is successful in supporting the conclusion that care should be withdrawn from the brain dead. It can even be used to support a statute calling these patients dead. But it in no way shows that the brain dead are dead; this issue is never engaged. Our argument, by contrast, is free of moral premises. The notion of "person" enters in, not because of any moral view concerning what sorts of entities have rights, but because the most likely account of personal identity serves to show that after brain death the person who entered the hospital has literally ceased to exist. Our claim that the person has died, of course, follows immediately from this. The account of personal identity uses as "data" determinations of the identities of persons and bodies in certain circumstances, but involves no testing of moral intuitions. And the moral issues concerning the patient's care are left open. There is one more point our argument clarifies. It comes as no surprise that the continued capacity for a mental life is what makes our life valuable. Without it, life would not be worth living. But the connection between our having a capacity for a mental life and that life being of value to us is even tighter than many have supposed. For the position we hold here is not simply that a mindless future life would be of no value to us, but that such a life, whatever else might be true of it, could not be ours. Finally, we have shown that whatever one's stance on the theoretical issues mentioned above, there may be sound moral reasons to support the current trend in public policy which extends the definition of death to include the brain dead. For those who think that brain-dead patients are still alive, this is a moral evasion. But if we wish to keep our official

Death, Dying and the Ending of Life, I

133

Brain Death and Personal Identity

policy on euthanasia strict to avoid any slippery-slope difficulties, it is an ideal evasion. It is extremely implausible that the public, which believes that the brain dead are dead, could ever come to believe that Lhe profoundly retarded or the grossly senile are dead; even though the mental life of some of these may not be much greater than that of brain-dead patients. This would prevent an extension and abuse of the statute which might follow upon (what those not persuaded by our previous arguments would consider) a more straightforward account assessing the "quality o[ life," one which would involve prescriptions of the form "killing (or 'letting die') is impermissible except when ...." And we have noted that public policy is entitled to such autonomy. Brain dead patients are in fact dead but that is not the reason to have their bodies die. We have tried, then, to deal with the delicate topics of brain death, death, and personal identity, keeping in mind that the spheres of morality, ontology, science, and public policy, while related, are quite distinct. Our solution to the problem, we believe, gives each sphere its due without arrogating the domain of the others. Daniel Wikler gratefully acknowledges suggestions from those who commented on an ancestor of this essay read at the University of Wisconsin, Georgetown University, and Oberlin College, and from Norman Fost, David Mayo, and Elliott Sober, and thanks the joseph P. Kennedy, Jr. Foundation for financial support. The present work is a descendant of a union of that paper and a paper by Michael B. Green, who expresses his appreciation for the support of the Minnesota Center for Philosophy of Science and of its Director, Grover Maxwell. Dr. Green also thanks William Winslade, Co·Director of the UCLA Program in Medicine, Law & Human Values, for providing the opportunity to investigate the topic of this essay.

75

[5] BRAIN DEATH: A DURABLE CONSENSUS? DANIEL WIKLER

I. INTRODUCTION A remarkable consensus, bordering on unanimity, exists among the world's medical and legal experts on the definition of death. It says that death, in man, consists of the irreversible loss of functioning in the brain as a whole. According to this definition, even a patient with a beating heart may be pronounced dead; and, as a result, all medical care may be withdrawn, or vital organs removed for transplantation. This view has been endorsed by leading medical, legal, and governmental authorities in all but a few countries around the world. Though a few holdouts remain, the definition of death as death of the brain as a whole is widely regarded as a settled matter. It is a point of pride to bioethicists, who are vulnerable to the charge that they offer arguments but not solutions; for here is a once-controversial issue settled, seemingly once and for all, with hardly any dissent. The array of kinds of expertise and sources of authority now marshalled in support of the whole-brain definition of death constitute a powerful endorsement of the concept for the average citizen. Bound up in the arcane terminology and concepts of modern neurology, perhaps the least familiar of medical specializations, the notion of brain death is intellectually intimidating. Patients and their families, and also their governmental representatives, must accommodate themselves to this innovation of medical science, perhaps adjusting their native conceptions oflife and death to fit (what they have been assured are) the modem realities. Indeed, those nations which have so far resisted the global movement toward the whole brain death criterion risk being seen as retrograde, held back perhaps by widespread misinformation, anti-scientific attitudes, or archaic folkways. Is it even conceivable that this global consensus could, in time, be regarded as a very temporary and makeshift expedient, a momentary substitute for a resolution of some profoundly difficult issues which for a time, perhaps a brief time, fit with both the tech·

78

Death, Dying and the Ending of Life, I

240 DANIEL WIKLER

nical capacities and the legal needs of those who endorsed it? And that in the long run it could linger as a footnote, or perhaps a chapter heading, in the long history of man's conceptions of life and death? This suggestion is so far from conventional wisdom today that one who espouses it risks being regarded as a crank. Nevertheless, I believe that the argument in its favor, while not conclusive, is much stronger than the argument against it (and in favor of the prevailing consensus). I will state the argument briefly, with particular reference to the landmark report' in 1982 in Washington of the President's Commission for the Study of Ethical Problems in Medicine, and will situate the argument in the context of trends in contemporary bioethics. I do not expect to win over, in this one pass, those who have been convinced of the validity of the conventional view. I do hope, however, to re-open the issue; in particular, to provide reasons to regard the issue as far from settled. II. THE CENTRAL ARGUMENT FOR THE WHOLE-BRAIN DEFINITION There can be little doubt - at least, I will not voice any - that acceptance of the definition of death as death of the brain as a whole has on balance been beneficial. As expected, it has provided a measure of social endorsement and legal protection to physicians who reallocate expensive medical resources to those who can benefit and to surgeons who take donated organs from those whose use has ended so as to provide the gift of life to recipients. There have been some ill effects as well, and I refer to these below, but they have not been of comparable importance. Even though these beneficial effects arguably include the savings of thousands oflives, however, they do not in themselves constitute an argument in support of the thesis that a patient is dead whose brain as a whole has suffered irreversible cessation of functioning. They show merely that good things happen if we choose to operate with that definition of death. Unless one is an extreme pragmatist, the utility of such a belief does not demonstrate, let alone constitute, its validity or truth. The most thoughtful legal and medical authorities on brain death have recognized the inadequacy, indeed, the danger, of appearing to draw the boundary between the living and the dead on such utilitarian grounds. Otherwise the doctors might be suspected of play1 President's Commission for the Study of Ethical Problems in Medicine and Biomedical and Behavioral Research, Dtjining Death: J.fedical, Legal and Ethical Issues in the Determination of Death, Washington: Government Printing Office, July 1981.

Death, Dying and the Ending of Life, I

BRAIN DEATH: A DURABLE CONSENSUS? 241 ing with words, and with the feelings of the patients' loved ones, in order to assure a supply of fresh organs without risking a homicide charge. Both the President's Commission, in the United States, and those such as the neurologist Christopher Pallis2 who have contributed to the British consensus, have expressly taken on the task of arguing that the whole-brain definition stands on its own, i.e., that it can be argued for directly as intellectually sound, without asking for indulgence on the ground that accepting it will save lives. These arguments differ on many points, but in my view the central argument is the same, and a successful attack on it will remove the chief intellectual foundation of the whole-brain definition of death. For Pallis, death is, in the most general terms, the irreversible cessation of functioning of the organism as a whole; and the President's Commission report for the most part agrees. Fitting this general concept to the particulars of whole-brain death is a challenge, however. Maintained in the modern intensive care unit, the whole-brain dead patient's body functions almost normally in a host of respects, including circulation of the blood, growth, digestion, and even gestation. In some cases there is no "irreversible cessation" of these functions for many days, unless one deliberately terminates life supports. This objection is overcome as follows. The key phrase is "organism as a whole". It directs us to ignore, when seeking evidence of continued life, functioning in particular cells or organs which falls short of constituting organismic function. Life in the patient, that is, or even in the patient's body, is more than the sum of the workings of some or even many of its parts - which after all might separately survive the patient by many years if successfully transplanted into the bodies of various donors. But in the well-maintained brain-dead patient, the organs are all in the patient's body, and they do seem to be working together: the heart pumps the blood, which carries oxygen to the cells, where metabolic processes continue; wastes are carried off and excreted; the system perseveres, at least for a while. Why, then does this not constitute function in the organism as a whole? The precise answer found in the literature on brain death varies .. Pallis holds that organismic functioning in man requires two capacities: that for cognition, and that for breathing. The brain-dead patient lacks them both, he says; consciousness has been terminated, and the delivery of oxygen to the blood is not the result of any capacity of the patient's body, but of the action of the artificial ventilator. 2 Pallis, C., "Brainstem Death", in R. Braakman, ed., Handbook of Clinical Nturology 13(57), 1990, pp.441-496.

79

Death, Dying and the Ending of Life, I

80

242 DANIEL WIKLER The President's Commission report offers not one but two distinct accounts; which, however, it declares mutually consistent3 • The "integrated functioning" view holds that the system of organs whose functions constitute life has a center in the triangle of heart, lungs, and brain; when the brain dies, and lung function is taken over by the ventilator, other organs may function, but in this state their workings are not truly integrated. The "primary organ" view, on the other hand, regards the brain, and in particular the brain stem, as the one organ which integrates the functioning of the others. Continued activity in the other organs, even if seemingly in concert with one another, is ipso facto unintegrated if it takes direction from the paraphernalia of the intensive care unit rather than from the intact brain. The "integrated functioning" and "primary organ" accounts, according to the Commission, differ in emphasis but not in their implications for the definition of death. These constitute, in my view, the best arguments in support of defining death as irreversible cessation of functioning in the brain as a whole. Nevertheless, I will argue that they commit the rather elementary error of confusing necessary and sufficient conditions. The error lurks deep in these arguments and they cannot be repaired. Pallis's argument, because it is so lucid, serves as a good first target. We should begin by sharpening it further. Though Pallis holds that loss of the capacity for consciousness is one of the two hallmarks of death of the human organism, this claim actually plays a negligible role in his account. His definition of death as death of the brain as a whole (which, incidentally, he further refines as death ofthe brain stem) serves to divide patients who may be permanently unconscious but who retain brain stem function from those lacking both capacities. Thus the line between life and death is crossed when one, not two, of these capacities is lost, and that is the capacity for the socalled vegetative functions, primarily lodged in the brain stem. Why attach life-and-death significance to this capacity? For Pallis and others it is essentially equivalent to the capacity to breathe; and it not merely a turn of phrase when we speak of "the breath of life." But here is the confusion of necessary and sufficient conditions. Most patients with intact brain stems can breathe, and they can do it on their own. But the brain dead patient also "breathes", in the sense that oxygen is delivered to the blood. This patient does not do this on his own; it is not spontaneous. But it occurs. Brain stem capacity is not necessary for this to happen, provided one is situated in an intensive care unit. Similar remarks apply to the two accounts given by the President's 3

President's Commission, ibid. pp.32f.

Death, Dying and the Ending of Life, I

BRAIN DEATH: A DURABLE CONSENSUS? 243

Commission. The brain-dead patient's organs function in an inte· grated way, even without the intervention of the important triangle of heart, lung, and brain, if the patient is artificially maintained. The "primary organ, account draws the same reply: the intensive care unit constitutes a demonstration that one needn't have one's primary organ intact in order to display integrated functioning of most of the major organ systems. The Achilles heel in each of these accounts is the same. Whether or not the word appears as such in their official formulations, they all insist that whatever coordinated functioning we see in the braindead patient's body counts as "unintegrated" unless it is in some sense commanded by the brain stem. In essence, these accounts focus on one particular function, breathing, as constituting integrated func· tioning - life itself - and even this is dismissed as inconsequential if it is artificially assisted. Thus they, along with most of the standard accounts, insist on speaking of "spontaneous" respiration. The problem is that any number of patients, be they sufferers of polio, ALS, or other maladies, are incapable of spontaneous respiration and yet are indisputably alive. Of course, most of them are also conscious. Occasionally, however, these respirator-dependent patients lapse into seemingly permanent unconsciousness, i.e., into persistent vegetative state (PVS) (at present, of course, this cannot be determined with certainty). At that moment they become patients who lack both the capacity for consciousness and the capacity for spontaneous respiration, which in turn is a stand-in for integrated functioning. According to the concepts underlying the current consensus, therefore, they should be counted as dead. But they are not brain-dead, and therefore would be diagnosed as alive. This is a contradiction which, in my view, makes untenable the current consensus in favor of an "integrated functions". To be sure, it has not escaped notice. The President's Commission report addressed the issue squarely, and its defense of its position is revealing. The text reads, in part: The (whole-brain-dead) lie with fixed pupils, motionless except for the chest movements produced by their respirators. Patients with intact brainstems ... can not only breathe, metabolize, maintain temperature and blood pressure, and so forth, on their own but also sigh, yawn, track light with their eyes, and react to pain or reflex stimulation. (italics in original). 4 In this passage and in the paragraphs which follow it, I locate three separate arguments. f

!hid., p.35.

81

82

Death, Dying and the Ending of Life, I

244 DANIEL WIKLER First, the Commission stands here right on that Achilles heel, whose presence is felt in the italicized phrase "on their own". The report quotes with approval the claim by Bernat, Culver, and Gert5 that despite the appearance oflife provided by the artificial support, what remains is not an integrated organism but "merely a group of artificially maintained subsystems". But the phrase "artificially maintained" reveals that this quotation merely shifts the weight to the other Achilles heel. Second: The Commission report concedes at this point that "Sometimes, of course, an artificial substitute can forge the link that restores the organism as a whole to unified functioning" 6 , citing kidney dialysis and the iron lung. In reply to the implicit critique of its thesis, it contrasts patients with these artificial aids to "the hypothetical of a decapitated body treated so as to prevent the outpouring ofblood and to generate respiration: continuation of bodily functions in that case would not have restored the requisites of human life" 7• Pallis has made the same point by displaying a gruesome photograph of a public beheading (by a Bangkok policeman, some time ago): from the headless body four columns ofblook spurt forth, indicating continuing heart function - which, according to Pallis, is not, even for those few seconds, constitutive of life. The headless body, however, is not an image which precisely supports the Commission's position. It lacks not only the brain stem, but also the higher brain and without further argument it is not clear which loss moves us to deny that the decapitated body is alive. If that body is maintained, as in the Commission's hypothetical, it arguably is alive, as a body if not as the patient himself. The third of the Commission's arguments lies in the list of observable differences between unconscious, respirator-dependent patients with intact brainstems and those without. The Commission insists that these differences, such as spontaneous yawning, are "startling". Compared to the differences between both classes of patients and cold corpses, I do not find this at all persuasive. In support of its claim, the Commission continues: "what is missing in the dead is a cluster of attributes, all of which form part of an ~rganism's responsiveness to its internal and external environment" 8 • But the well-maintained, brain-dead patient's body does ward off the relative cold of the intensive care unit, does keep blood circulating, 5 Bernat, J., Culver, C.M., and Gert, B., "On the Definition and Criterion of Death", Ann.lnt.Mtd. 369, 1981. 6 President's Commision, ibid., p.36. 7 Pallis, loc. cit. 8 President's Commission, loc.cit.

Death, Dying and the Ending of Life, I

BRAIN DEATH: A DURABLE CONSENSUS? 245 and so on. Not a healthy body, nor one with a long life ahead of it, but neither illness nor the imminence of death constitute death itself. III. DOES CONCEPTUAL COHERENCE MATTER? What difference does it make that the central argument for regarding brain death as death is incoherent? After all, even if my critique is justified, the points involved are subtle, and there is little evidence of any dissatisfaction with the brain death criterion on the part of the public in most countries. As working bioethicists, we may find these matters to be of academic interest; is there any further significance? The answer, I believe, should be: Not Yet. So long as the new definition of death saves lives by facilitating transplantation and by freeing space in intensive care units for patients who could benefit thereby, the work of the President's Commission and its counterparts abroad serves the public interest. However, we should keep our eyes on two potential problems. First, there is some evidence that the problems which I have claimed affiict the whole-brain-death criterion may be impeding the process of transplantation even as it supports it overall. Younger and his colleagues found that relatively few physicians and nurses involved in transplantation could state the central argument for the whole-braindeath criterion in any coherent manner. Their problems with the central argument - justified, in my view - may impede their ability to persuade families of brain-dead patients into regarding their loved ones' bodies, warm to the touch, as dead and therefore as suitable for organ donation. Of course, there are other reasons for this reluctance as well. Second, and eventually more importantly, we must consider what will happen if and when our neurologists acquire the ability to diagnose persistent vegetative state with complete certainty. At this point, we will have a new option: to re-redefine death as permanent loss of consciousness. As I have written elsewhere, I believe that a good case can be made for regarding these patients - though not their bodies - as dead. In effect, their bodies would constitute "living remains", rather like the donated heart which beats on in another person's body and this would be true whether these remains contained an intact brainstem and whether or not they were respirator-dependent. It is conceivable that some, or even most, of those members of the public who accept the whole-brain criterion of death do so for the "wrong" reason, viz., that these patients arc permanently unconscious. This was indeed the reason given by some of the doctors

83

Death, Dying and the Ending of Life, I

84

246 DANIEL WIKLER

and nurses questioned in Youngner's study9 • In terms of the current consensus, they are flatly mistaken. I think they are right. If and when we can diagnose PVS, their attitude would support the re-redefinition of which I have spoken. After presenting its arguments, the Commission rightly pointed out that "philosophical refinement beyond a certain point may not be necessary" 10 for its purposes. A definition of death based on permanent loss of consciousness, after all, does support the Commission's view that whole-brain-dead patients are dead. The Commission, moreover, framed its task with care: it was to address the problem of respirator-dependent patients, and sought the minimum amount of conceptual revision needed for that purpose. A definition of death based on permanent unconsciousness would have revised contemporary thinking unnecessarily. In the future, however, we may be addressing the problem of those patients who persist in an unconscious state for many years. Whether the degree of revision of current concepts of death which would define them as dead would be justified, and accepted by the public, may be the business of a future commission. As an issue in bioethics, the definition of death was put to rest, but it has not yet expired.

University

of Wisconsin

9 Youngner, S.J., Landefeld C.S., Coulton C.J., Juknialis B.W., Leary M., "'Brain Death'.and organ retrieval: a cross-sectional survey of knowledge and concepts among health professionals". JAMA 261: 2205-2210, 1989. ' 0 President's Commission, /oc. cit.

[6] The Dead Donor Rule: Should We Stretch It, Bend It, or Abandon It? Robert M. Arnold and Stuart]. Youngner The dead donor rule-that persons must be dead before their organs are taken- is a central part of the moral framework underlying organ procurement; Efforts to increase the pool of transplantable organs have been forced either to redefine death (e.g., anencephaly) or take advantage of ambiguities in the current definition of death (e.g., the Pittsburgh protocol). Society's growing acceptance of circumstances in which health care professionals can hasten a patient's death also may weaken the symbolic importance of the dead donor rule. We consider the implications of these efforts to continually revise the line between life and death and ask whether it would be preferable to abandon the dead donor rule and rely entirely on informed consent as a safeguard against abuse. ABSTRACT.

T

HREE IMPORTANT principles provide the moral framework for organ procurement in America. 1 First, the dead donor rule: persons must be dead before their organs are taken. Second, the prohibition against active euthanasia: although patients may be allowed to die under certain circumstances, they must never be actively killed. Third, the primacy of consent: patient or family consent must precede organ retrieval. These rules serve to reassure people that when they enter hospitals they will not be killed or mistreated so that their organs can be taken. By professing adherence to these principles, the Pittsburgh protocol denies that it breaks with existing policy (Caplan 1993 ). The scholars who write in this volume seem to accept these principles as well. For the most part, their judgments about non-heart-beating cadaver donor (NHBCD) protocols rest on whether they would increase the number of transplantable organs without violating existing ethical norms. We ask a more radical question-is it time to reassess the moral framework itself? By taking advantage of ambiguities in the definition and criteria of death, the Pittsburgh protocol pushes the dead donor rule to the limit.

86

Death, Dying and the Ending of Life, I KENNEDY INSTITUTE OF ETHICS JOURNAL • JUNE

1993

Would it be better, in our quest to fulfill the ever-greater promise of organ transplantation, to abandon rather than tamper with the dead donor rule, relying entirely on informed consent as a safeguard against abuse? 2 Or, instead, as Renee Fox (1993) suggests, is it simply time to reaffirm our commitment to basic values and stop pushing for more organs? These questions become more pressing as the prohibition against active euthanasia seems to be fading. THE DEAD DONOR RULE

Origins Although the term, dead donor rule, was first coined by John Robertson in 1988, it aptly describes an unwritten, uncodified standard that has guided organ procurement in the United States sinc;e the late 1960s (Robertson 1988). As DeVita et al. (1993) have noted, because of its informal nature, the reasons for it can only be inferred. There was undoubtedly discomfort with the isolated practice of taking kidneys from patients before they died following failed bypass surgery (Starzl 1992) because this appeared to violate health care professionals' admonition to "do no harm." More influential, perhaps, was the growing interest in severely braindamaged (but heart-beating) patients as a promising donor source (Starzl 1992). By introducing "brain death," our society designated a subgroup of severely brain-damaged patients, those with loss of all brain functions, as an acceptable organ source. By simultaneously embracing the prohibitions of the dead donor rule, our society seemed to reassure itself that other severely compromised patients would be protected from harmnamely, they would not be killed or used as organ sources prior to their death. Meanings of the Rule The prohibition that persons must be dead before their organs are taken has two distinct connotations. 1. Patients must not be killed by organ retrieval: The first connotation is that patients must not be killed by the process of removing their organs. For example, taking an unpaired vital organ such as the heart would kill a living donor. A slightly broader interpretation prohibits killing patients so that their organs can be taken, even if organs are removed after death occurs. Both meanings are consistent with homicide laws and the strong prohibition against "mercy" killing that has characterized our society.

Death, Dying and the Ending of Life, I ROBERT M. ARNOLD

&

STUART

J.

YOUNGNER • THE DEAD DONOR RULE

2. Organs must not be taken from patients until they die: The second connotation of the dead donor rule is that organs should not be taken from patients before they are dead, even if taking the organs does not kill them. The reasons for this prohibition are less obvious and less well articulated than those against killing, but include: a wish to prevent harm to or exploitation of the weak and vulnerable; mistrust of doctors; and concern that critically ill patients would be treated merely as a means to an end-as repositories of organs that could help others. Unlike the prohibition against killing, the second connotation is not absolute. For example, in the case of living related donors or partial transplants, we do allow organs to be taken from living persons. The organs are either paired (kidneys) or we take only part of them (liver or pancreas). While the practice of taking one of an organ pair or parts of organs from living related donors does not violate the first connotation of the dead donor rule-i.e., the patient is not killed by organ retrievalit does seem to violate the second connotation. Sentiment about this issue, however, is far from unanimous. Some people argue that even consenting living donors should not be exposed to the "unnecessary" risk of organ removal since cadaver organs are nearly as successful (Starzl 1992). Others point out that living related donorsespecialiy parents of dying children-are put in a coerced position simply by giving them the choice of either donating or letting their child die (Fox and Swazey 1978). Opposition to living donation grows when living donors are altruistic strangers (Fellner and Schwartz 1971), or the surgical risk is great, and becomes downright bellicose when consideration is given to buying organs from living strangers (Council 1986; Fox and Swazey 1992).3 In summary, the dead donor rule has ensured that patients will not be killed by or for organ removal. 4 It also limits nonlethal removal of organs from living persons to a narrow category-usually healthy rehitives of the potential recipient. LIVING WITH THE DEAD DONOR RULE: DONATING ORGANS PRIOR TO DECLARING DEATH

The public has a real fear that medical treatment will be compromised in order to increase organ procurement. Surveys reveal that concern about premature declaration of death and undertreatment are the leading reasons people give for not filling out organ donor cards (Robbins 1990). These fears appear in the public literature in books like Coma (Cook 1977), and [ 265 ]

87

88

Death, Dying and the Ending of Life, I KENNEDY INSTITUTE OF ETHICS JOURNAL • JUNE

1993

in a recent storyline in Dick Tracy (Locher and Collins 1991), where weak and defenseless persons are murdered for their organs. To allay these fears, the dead donor rule has become central to organ procurement; strategies for increasing the donor pool have tried to operate within its confines. One way to increase the organ supply without breaking the dead donor rule is simply to redraw the line between life and death so that it includes patients as dead who were previously considered alive. One could argue that the formulation of "brain death" was, all along, a well-intentioned effort to gerrymander the line between life and death in order to increase the donor pool without inflaming public opinion. 5 In fact, the Harvard Ad Hoc Committee that first endorsed irreversible loss of all bnrin function as a new criterion for determining death was quite direct in acknowledging its motivation to quell the "controversy in obtaining organs for transplantation" (Ad Hoc Committee 1968, p. 337). The whole-brain criterion for determining death was quickly and widely accepted for two reasons, neither of which had much to do with whether "brain death" was equivalent to death. First, irreversible loss of all brain function was easy to diagnose in adults, and thus seemed to offer, like the cardiopulmonary criterion, clear means for separating the living from the dead. 6 Second, once the diagnosis was evident, so too was the prognosis, which is as certain as anything in clinical medicine-patients did not regain consciousness and could not sustain a beating heart (on a respirator) for more than hours, days, or, in a few instances, weeks-despite the most aggressive treatment. 7 However as technology has advanced making it possible to sustain the heartbeat longer-in rare cases even for months (Parisi et al. 1982) -this reason may become less influential. It is these factors, plus the irresistible utilitarian appeal of organ transplantation, that greatly contributed to the acceptance of irreversible loss of all brain function as a justification for behaving as if patients were dead. 8 While the whole-brain criterion for death did facilitate an increase in transplantable organs, it has not been without problems. First, it has become clear that it lacks a widely accepted conceptual justification. Nearly a decade passed before any concept emerged to justify exactly why persons who have lost all brain function, but whose hearts continue to beat, are dead. This concept, offered by James Bernat et al. (1981), holds that such persons are dead because they have lost the ability to function as a wholei.e., "the spontaneous and innate activities carried out by the integration of all or most subsystems." Bernat's view has received little support in the philosophic and bioethics communities, where a higher-brain concept[ 266 ]

Death, Dying and the Ending of Life, I ROBERT M. ARNOLD

&

STUART

J.

YOUNGNER • THE DEAD DONOR RULE

loss of consciousness and cognition-finds greater acceptance (Gervais 1986; Green and Wikler 1980; Youngner and Bartlett 1983; Veatch 1976). Moreover, there is evidence of conceptual confusion in clinical practice. A study by Youngner and his colleagues found that one third of physicians and nurses involved in the management of "brain-dead" patients considered them to be dead because they had irreversibly lost consciousness and cognition. Youngner's study also found that another third of physicians and nurses did not perceive them as dead at all but accepted treating them as dead because they were irreparably damaged or death was imminent (Youngner et al. 1989; see also Tomlinson 1990). Confusion over the "real" time of death has been reported in surveys of families who donate their loved one's organs (Fulton, Fulton, and Simmons 1977). These findings, and the observation that many health professionals (and most journalists) describe "brain-dead" patients as having "died" only after either life-support machines or organs are removed, suggest that many persons disagree with or are, at a minimum, ambivalent about the conceptual status of persons whose brains have ceased functioning but whose hearts continue to beat. The whole-brain formulation thus seems less the result of intellectual "discovery" -i.e., we figured out that some people we said were alive are really dead-and more the product of conceptual gerrymandering to solve a social problem. Unfortunately, such utilitarian justification for defining death is inherently unstable and will result in attempts at redefinition whenever utility requires it. 9 The recent attempt to redefine death to include anencephalic infants is a good example. Summarizing the argument, Fost (1988, p. 7) says, "If resolution of a controversy that stands in the way of procuring organs for transplantation was a valid reason for redefining death in 1968, why would it not be a sufficient reason for another redefinition in 1988 ?" 10 Instead of grossly moving the line between life and death, the Pittsburgh protocol takes advantage of ambiguities in the way it has already been drawn. By declaring patients dead after two minutes of ventricular fibrillation, for example, the protocol makes organ retrieval possible at a time when: (1) the heart could almost certainly be restarted by medical intervention; (2) the medical evidence that it will not start up by itself (autoresuscitate) is incomplete (Lynn 1993); and (3) the functional status of the brain is unknown (Lynn 1993). Although "irreversible" has been part of the definition and criterion of death, as Cole (1993) and Lynn (1993) point out, its theoretical meaning and practical significance have never

89

90

Death, Dying and the Ending of Life, I KENNEDY INSTITUTE OF ETHICS JOURNAL • JUNE

1993

been adequately resolved. This leaves the distinction between almost dead, maybe dead, probably dead, or dead for sure, and the effect of the possibility of treatment on these states, indeterminate. The Pittsburgh protocol merely provides answers that improve the chances for obtaining organs.U The question for public policy is whether the continued readjustment of the definition of death and the criteria for determining it (which the Pittsburgh protocol exemplifies), will yield more organs or will foment suspicion and fear and turn the public against organ donation? Efforts to extend the line of death to include anencephalic infants failed after stirring considerable public controversy. How the public will react to the Pittsburgh protocol's interpretation of irreversible is unknown, but several of the scholars writing in this volume have expressed concern that it might backfire. LIVING WITH THE DEAD DONOR RULE: THE PROHIBITION AGAINST ACTIVE KILLING

As our society has accepted that, under the right circumstances, each and every life-sustaining intervention may be withheld or withdrawn, the timing of human death has become more and more a matter of deliberate human decision. At the same time, the prohibition against active killing has provided a "bright line" that must not be crossed (Robertson 1993, p. 250). Indeed, as our earlier discussion suggests, society may have redrawn one line (between life and death) precisely to avoid crossing another (between killing and allowing to die). · The Pittsburgh protocol superficially respects the bright line. Under its aegis, patients who are alive are "allowed to die," not killed, when they are removed from life support. The protocol even prohibits the use of fatal doses of morphine, leaving open the possibility that patients will experience discomfort as they are withdrawn from ventilatory support before they receive analgesia.

Conceptually Dim For better or for worse, the bright line is growing dimmer. Brock, for example, argues that, when it comes to turning off a ventilator, the distinction between killing and allowing to die is conceptually "confused and mistaken" (Brock 1992, p. 12). He gives the example of a terminally ill ALS patient who is mentally competent and completely respirator dependent. When the patient's physician removes the ventilator at the patient's request, "the common understanding is that the physician thereby [ z.68 ]

Death, Dying and the Ending of Life, I ROBERT M. ARNOLD

&

STUART

J.

YOUNGNER • THE DEAD DONOR RULE

allows the patient to die." Brock then changes the case so that, instead of the physician, a greedy and hostile son removes the patient from the ventilator. The son's claim that he didn't kill her, but merely allowed her to die from her ALS, "would rightly be dismissed as transparent sophistry." Brock points out that both the physician and son performed the same action with the same result. Both killed the patient; both acted in ways that resulted in the patient dying sooner than she otherwise would have. It is the nature of the motivation that makes killing morally acceptable in the first instance and blameworthy in the second. It simply makes us more comfortable to label the behavior we don't accept with the pejorative term, killing. 12 In other cultures, even honorably-motivated removal of life-sustaining ventilator support is considered killing in a pejorative sense. While this view might be dismissed as morally "unsophisticated," perhaps the paradoxical root meaning of that word-i.e., without sophistry-is more in order. In our own hospital settings, health professionals (Solomon et al. 1993) and families sometimes have trouble accepting the explanation that removing a ventilator is merely allowing to die; 25 years ago, few people would have accepted the explanation. The advice of the Harvard Ad Hoc Committee, in its seminal1968 ]AMA article, is interesting in this regard. It should be emphasized that we recommend the patient be declared dead before any effort is made to take him off a respirator, if he is then on a respirator. This declaration should not be delayed until he has been taken off the respirator and all artificially stimulated signs have ceased. The reason for this recommendation is that in our judgment it will provide a greater degree of legal protection to those involved. Otherwise, the physicians would be turning off the respirator on a person who is, under the present strict, technical application of law, still alive. (Ad Hoc Committee 1968, p. 339)

Morally Dim Even if one insists that removing a ventilator is not killing, undeniable forms of killing-.,.e.g., physician-assisted suicide, in which the physician helps the patient kill herself, and lethal injection by the physician-are favored by 54 percent of Americans in recent opinion polls (Malcolm 1990). It is entirely possible, if not likely, that such killing will be legally sanctioned or at least tolerated in some U.S. jurisdictions over the next few years.

91

92

Death, Dying and the Ending of Life, I KENNEDY INSTITUTE OF ETHICS JOURNAL • JUNE

1993

IMPLICATIONS FOR THE FUTURE: AUTONOMY AND ORGAN PROCUREMENT

The implications of these observations for the future of organ procurement are profound. The lines that forbid us to kill people for their organs or to take organs from people before they are dead may be, as Robertson (1993, p. 249) suggests, important symbols of "the ethical concerns that arise around organ transplantation." However, these symbols lose their meaning for many people as society accepts circumstances in which health professionals may hasten patients' deaths and as transplant policymakers constantly tinker with death's boundaries. Under these circumstances, bright lines dim, and symbols become detached from the moral and psychological roots from which they arose. Consider a future in which we have abandoned the symbolic prohibitions of the dead donor rule. Killing, in and of itself, would no longer constitute a harm. Real harms would have to do with suffering, and violations of patient autonomy or interests. What if, instead of continually gerrymandering the line between life and death, we simply ask, "Are there some patients whose quality of life is so unacceptable and whose death is so imminent (by fate or their own decision) that we may take their organs before they die?" Instead of pretending that we can continue to develop more accurate definitions of death (that coincidentally expand the donor pool), we would allow organ procurement from patients with severe head injuries who were irreparably damaged and near death but who have not lost all brain function. Anencephalic newborns represent a subcategory of such patients. After all, as Robertson (1993, p. 249) says, there can be little harm to "a patient who is so near death, and so presumably without interests at that point, that no real harm to his interests could be shown." Fost (1983) has argued for reformulating the central tenet of organ procurement to be "violates no interest" rather than "is dead," and this seems more consistent with current practices. After all, our·society does not require that a patient be dead prior to the removal of life-sustaining technology; it requires only that the removal does not violate the patient's interest (as defined by either the patient or surrogate). From a patient's perspective, the consequences of organ procurement (death) may not be that different than those of forgoing life support, except that organ procurement may help others. It is therefore unclear why stricter rules should govern one than the other. 13 [ 270 ]

Death, Dying and the Ending of Life, I ROBERT M. ARNOLD

&

STUART

J.

YOUNGNER • THE DEAD DONOR RULE

How near death and how damaged a potential donor would have to be in order to be considered beyond harm is problematic. One might try to develop societal standards, although recent debates over rationing and the use of QALY's indicate that this will be quite difficult (Hadorn 1992; LaPuma and Lawton 1990). 14 More likely, caregivers will rely on surrogates' assessments (as they do in decisions to forgo life-sustaining treatment). We could minimize the problems of judging what constitutes a harm by insisting on clear and rigorous consent by the potential donors rather than their surrogates. 15 People could sign donor cards or living wills that designated the circumstances under which their organs could be takeni.e., define their own threshold for being harmed. For example, a patient could state that if she becomes severely demented, one kidney should be removed and donated before artificial nutrition and hydration are discontinued and she is allowed to die. Machine-dependent patients could give consent for organ removal before they are dead. For example, a ventilator-dependent ALS patient could request that life support be removed at 5:00 p.m., but that at 9:00 a.m. the same day he be taken to the operating room, put under general anesthesia, and his kidneys, liver, and pancreas removed. Bleeding vessels would be tied off or cauterized. The patient's heart would not be removed and would continue to beat throughout the surgery, perfusing the other organs with warm, oxygen- and nutrient-rich blood until they were removed.16 The heart would stop, and the patient would be pronounced dead only after the ventilator was removed at 5:00 p.m., according to plan, and long before the patient could die from renal, hepatic, or pancreatic failure. If active euthanasia-e.g., lethal injection-and physician-assisted suicide are legally sanctioned, even more patients could couple organ donation with their planned deaths; we would not have to depend only upon persons attached to life support. This practice would yield not only more donors, but more types of organs as well, since the heart could now be removed from dying, not just dead, patients. CONCLUSION

The irresistible utilitarian appeal of organ transplantation has us hellbent on increasing the donor pool. Giving up the dead donor rule, however, raises the question of how far we are willing to go to procure more organsand some point out, save more lives. Are we headed for the utilitarian [ 271 ]

93

94

Death, Dying and the Ending of Life, I KENNEDY INSTITUTE OF ETHICS JOURNAL • JUNE

1993

utopia espoused by Jack Kevorkian, where organ retrieval and scientific experimentation are options in every planned death, be it mercy killing or execution? 17 If a look into such a future hurts our eyes (or turns our stomachs), is our discomfort any different from what we would have experienced 30 years ago by looking into the future that is today? How are we to understand such feelings? Are they merely emotional reactions and cultural habits that stand in the way of social progress? Does the dead donor rule put our devotion to symbols ahead of the real interest dying patients have in transplantation? Why not drop the dead donor rule and replace it with a rigorous consent process and protocols that prevent real harms (e.g., the treatable discomfort which the Pittsburgh protocol allows during "weaning" from the ventilator) instead of symbolic ones (e.g., the intentional hastening of death in order to prevent pain, which the protocol forbids)? Do we really protect helpless and hopelessly damaged patients by calling them dead, by claiming that we merely allow them to die when we turn off their ventilators, or by being stingy with analgesics? Isn't this an example of respecting the symbol at the expense of the very values they represent (Feinberg 1985)? Proponents of the dead donor rule argue that this rationalistic interpretation of symbols "underestimates the cognitive and moral significance of symbols; it suggests that aesthetes alone traffic in them, while moralists enjoy a symbol free access to the real world" (May 1985). Symbols keep us attuned to what we consider important. As such, they shape our selfconception and how we organize our communities. According to this view, our revulsion to a Kevorkian future is important to our moral character. Giving up the dead donor rule risks the weakening and eventual vanishing of these sentiments and thus should be opposed (May 1973; Weisbard 1993; Fox 1993). 18 Given the difficulties our society is likely to experience in trying to openly adjudicate these disparate views, 19 why not simply go along with the quieter strategy of policy creep? It seems to be getting us where we seem to want to go, albeit slowly. Besides, total candor is not always compatible with the moral compromises that inevitably accompany the formulation of public policy (Calabresi and Bobbitt 1978). Calling a spade a spade has at least one advantage, however. By framing our choices in stark rather than obfuscated terms, we may be able to choose our path more clearly and be less surprised where it takes us. Our society is on the brink of a paradigm shift in which the production [ 272 ]

Death, Dying and the Ending of Life, I ROBERT M. ARNOLD

&

STUART

J.

YOUNGNER • THE DEAD DONOR RULE

of body parts will increasingly link the intentional ending of some lives with the salvaging of others. From fetal tissue transplant to non-heartbeating cadaver donors and beyond, these practices will inevitably pit our insatiable longing for better health and longer life against deep-seated notions of the sacred and profane. How we attempt to resolve this conflict will reveal a great deal about who we are and what we value. We would like to thank Norman Fost, M.D., for his insightful comments and suggestions.

NOTES 1. The practices differ in other societies; For example, the Netherlands procures organs from non-heart-beating cadaver donors (NHBCDs). Under limited conditions, active euthanasia is also permitted there. Japan does not accept "brain death." 2. Others have suggested getting rid of the requirement that consent be obtained prior· to procurement. Some European countries, for example, Belgium and Austria, have already adopted this policy option, called presumed consent (see Roels et al. 1991). 3. It is ironic that we do not allow donation from living unrelated donors, who can autonomously weigh the benefits and risks of donation, while allowing donation from family members, who may be less able to make an autonomous decision because they are under extreme pressure (Spital 1991). 4. "Killing" is an important distinction here. There are cases in which a surrogate's decision to "let a patient die" is influenced by organ donation. One example involves patients with severe head trauma. Families have declined aggressive treatment that would have resulted in the patients surviving in a vegetative state, when the alternative was "brain death" and organ donation. 5. Fost (1983), in an article that foresees many of the problems raised by NHBCDs, argues that it was not necessary for legislators to redefine death in order to give clinicians guidance regarding these difficult matters. 6. Truog and Fackler (1992) raise questions regarding this assertion. Reviewing empirical data collected over the last 10 years, they demonstrate that many patients who fulfill current clinical standards for determining death by neurologic criteria still retain selective integrative brain functions (see also, Halevy and Brody, unpublished). 7. Britain's initial justification for "brain death" was that it predicted, with a high degree of certainty, that the patient would soon meet traditional cardiopulmonary criteria for death. Of course, this justification confuses the near dead with the currently dead (Pallis 1983 ). [ 2 73

]

95

96

Death, Dying and the Ending of Life, I KENNEDY INSTITUTE OF ETHICS JOURNAL • JUNE 1993

8. A social history of the acceptance of neurologic criteria for death, both within the medical community and the public, is badly needed. 9. A deeper objection, both Cole (1993) and Fost (1988) suggest, is that death is "a natural event or process that occurs in nature quite independently of human activity and ken" (Cole 1993, p. 146). As such, definitions of death exist independently of purposes and cannot be modified to meet our social purposes. 10. The debate about anencephalies is instructive for other reasons. First, it shows the lengths people were willing to go to avoid breaking the dead donor rule. In addition to redefining death, others have suggested treating anencephalies as a "special category" from whom organs can be removed without violating the dead donor rule, or subjecting them to intensive care solely to allow them to be defined as dead so that their organs can be procured (Harrison 1986; Walters and Ashwal 1988). Only rarely did commentators point out that another possible solution would be to do away with the dead donor rule (Fost 1983, 1988; Caplan 1987). Second, consistent with the utilitarian justification for defining death, some might argue that the reason that attempts to morally justify organ procurement from anencephalies failed was the paucity of organs that could actually be procured (Shewmon 1988). 11. In this way, the changes suggested in the Pittsburgh protocol mirror the recent controversy regarding organ procurement from an anencephalic infant at Lorna Linda. Like Lorna Linda, the University of Pittsburgh Medical Center (UPMC) chose to act first and to use their actions as an attempt to force a societal response-either acceptance or rejection through. public criticism, legislation, or litigation. Fost (1988) criticized Lorna Linda for using this approach in the anencephaly debate, and Lynn (1993 ), in this issue, criticizes UPMC for the same reason. They argue that seeking societal consensus and approval prior to acting is more likely to ensure that all relevant facts, interests, and arguments are included. 12. In earlier writings, Brock and others pointed out that in the medical literature killing and letting die are moral descriptions of an act that can be described in a more neutral manner (President's Commission 1983). Both the doctor and the son acted in ways that resulted in the patient dying sooner than she otherwise would have. The act is described as a killing if it is morally wrong; letting die if it is morally right. Describing an act as killing thus begs the question as to whether it is morally permissible. To emphasize that our use of the term killing is neutral regarding its moral status we will italicize the word. 13. Some might argue that the stricter rules are needed for organ procurement because of the non-patient-related incentives to obtain organs-e.g., the desire to help others or the institutional prestige. [ 274 ]

Death, Dying and the Ending of Life, I ROBERT M. ARNOLD

&

STUART

J.

YOUNGNER • THE DEAD DONOR RULE

14. Where on this continuum one falls will depend on the answers to a series of questions: What is meant by violates? Must donation be risk-free or is the requirement merely that the burdens are outweighed by the benefits? Are there limits to altruism as a virtue? Is the weighting of risks and benefits solely subjective or should society impose certain standards? If so, what are they and how can they be justified in a pluralistic society? 15. Relying solely on autonomy as a guiding principle for organ procurement would solve the problem of defining what constitutes a harm, but it would raise other problems. What would be done about those who had never expressed an opinion regarding procurement? This problem would be particularly acute for the never-competent, e.g., children. The philosophical justification for surrogate decision making is not as strong as autonomous decision making (Buchanan and Brock 1989). One might, therefore, allow an autonomous patient to donate in some circumstances in which one would not allow a surrogate to consent to donation. 16. Of course, if one eliminated the first connotation of the dead donor rule, one could also take the patient's heart. 17. Some will argue that this scenario is outlandish and inflammatory. We can give up t~e dead donor rule, they may argue, and yet keep from sliding too far along the slippery slope. However, once we abandon the dead donor rule and depend on autonomy to guide organ procurement, allowing less severely ill patients to donate their organs seems inevitable. Moreover, if recent history is any guide, active euthanasia may become a reality. 18. Much more needs to be said about how and why this will occur if this argument is to be tenable. After all, our recent changes regarding the forgoing of life-sustaining treatment do not seem to have coarsened our moral character. 19. We are, for example, deeply ambivalent about whether the dead donor rule should be disregarded. On the one hand, we think that there is little theoretical justification for it, and that by forsaking it we might be able to procure more organs and thus save more lives. On the other hand, we are repulsed by where this move may lead us and have grave concerns about the effect on our moral characters of treating individuals as replaceable body parts (Fox and Swazey 1992).

REFERENCES Ad Hoc Committee of the Harvard Medical School to Examine the Definition of Brain Death. 1968. A Definition oflrreversible Coma. Journal of the American Medical Association 205: 337-40.

[ 275 ]

97

98

Death, Dying and the Ending of Life, I KENNEDY INSTITUTE OF ETHICS JOURNAL • JUNE

1993

Bernat, James L.; Culver, Charles M.; Gert, Bernard. 1981. On the Definition and Criterion of Death. Annals of Internal Medicine 94: 389-94. Brock, Dan W. 1992. Voluntary Active Euthanasia. Hastings Center Report 22 (2): l0-22. Buchanan, Allen E., and Brock, Dan W. 1989. Deciding for Others: The Ethics of Surrogate Decision Making. Cambridge: Cambridge University Press. Calabresi, Guido, and Bobbitt, Philip. 1978. Tragic Choices. New York: Norton. Caplan, Arthur L. 1987. Should Foetuses or Infants Be Utilized as Organ Donors? Bioethics 1: 119-40. - - ·1993. The Telltale Heart: Public Policy and the Utilization of Non-HeartBeating Donors. Kennedy Institute of Ethics Journal 3: 251-62. Cole, David. 1993. Statutory Definitions of Death and the Management of Terminally Ill Patients Who May Become Organ Donors after Death. Kennedy Institute of Ethics ]ournal3: 145-55. Cook, Robin. 1977. Coma. Boston: Little, Brown. Council of the Transplantation Society. 1986. Commercialization in Transplantation. Transplantation 41: 1-3. DeVita, Michael A.; Snyder, James V.; and Grenvik, Ake. 1993. History of Organ Donation by Patients with Cardiac Death. Kennedy Institute of Ethics Journal 3: 113-29. Feinberg, Joel. 1985. The Mistreatment of Dead Bodies. Hastings Center Report 15 (1): 31-37. Fellner, C. H., and Schwartz, S. H. 1971. Altruism in Disrepute: Medical versus Public Attitudes Toward the Living Organ Donor. New England Journal of Medicine 284: 582-85. · Fost, Norman. 1983. The New Body Snatchers: On Scott's The Body as Property. American Bar Foundation Research Journal 3: 718-3 2. - - - · 1988. Organs from Anencephalic Infants: An Idea Whose Time Has Not Yet Come. Hastings Center Report 18 (5): 5-11. Fox, Renee C. 1993. "An Ignoble Form of Cannibalism": Reflections on the Pittsburgh Protocol for Procuring Organs from Non-Heart-Beating Cadavers. Kennedy Institute of Ethics Journal 3: 231-39. ___,and Swazey, Judith P. 1978. The Courage to Fail: A Social View of Organ Transplants and Dialysis, 2d ed. Chicago: University of Chicago Press. ___, and Swazey, Judith P. 1992. Spare Parts. New York: Oxford Universicy Press. Fulton, Julie; Fulton, Robert; and Simmons, Roberta. 1977. The Cadaver Donor and the Gift of Life. In Gift of Life: The Effect of Organ Transplantation on Individual, Family and Social Dynamics, ed. Roberta G. Simmons, Susan Klein Marina, and Richard L. Simmons, pp. 338-77. New Brunswick: Transaction Books. Gervais, Karen G. 1986. Redefining Death. New Haven: Yale University Press. [ 276 ]

Death, Dying and the Ending of Life, I ROBERT M. ARNOLD

&

STUART

J.

YOUNGNER • THE DEAD DONOR RULE

Green, Michael B., and Wikler, Daniel. 1980. Brain Death and Personal Identity. Philosophy & Public Affairs 9: 105-33. Hadorn, David C. 1992. The Problem of Discrimination in Health Care Priority Setting. journal of the American Medical Association 268: 1454-59. Halevy, A., and Brody, B. Brain Death: Reconciling Definitions and Tests, unpublished paper. Harrison, Michael R. 1986. The Anencephalic Newborn as Organ Donor. Hastings Center Report 16 (2): 21-23. LaPuma, John, and Lawton, Edward F. 1990. Quality-Adjusted Life Years. Ethical Implications for Doctors and Policy Makers. Journal of the American Medical Association 263: 2917-21. Locher, Dick, and Collins, Max. 1991. Pittsburgh Press (4 August-25 September). Lynn, Joanne. 1993. Are the Patients Who Become Organ Donors under the Pittsburgh Protocol for "Non-Heart-Beating Donors" Really Dead? Kennedy Institute of Ethics journal 3: 167-78. Malcolm, Andrew H. 1990. Giving Death a Hand: Rending Issue. New York Times (9 June): A6. May, William. 1973. Attitudes Toward the Newly Dead. Hastings Center Studies 1 (1): 3:-13. - - - · 1985. Religious Justifications for Donating Body Parts. Hastings Center Report 15 (1): 38--42. Pallis, C. 1983. ABC of Brain Stem Death: The Arguments about the EEG. British Medical journal 286: 284--87. Parisi, Joseph E.; Kim, Ronald C.; Collins, George H.; and Hilfinger, M. F. 1982. Brain Death with Prolonged Somatic Survival. New England Journal of Medicine 306: 14-16. President's Commission for the Study of Ethical Problems in Medicine and Biomedical·and Behavioral Research. 1983. Deciding to Forego Life-Sustaining Treatment. Washington, DC: U.S. Government Printing Office. Robbins, Rosemary A. 1990. Signing an Organ Donor Card: Psychological Factors. Death Studies 14: 219-29. Robertson, John A. 1988. Relaxing the Death Standard for Organ Donation in Pediatric Situations. In Organ Substitution Technology: Ethical, Legal, and Public Policy Issues, ed. Deborah Mathieu, pp. 69-76. Boulder, CO: Westview Press. - - - · 1993. Policy Issues in a Non-Heart-Beating Donor Protocol. Kennedy Institute of Ethics Journal3: 241-50. Roels, L.; Vanrenterghem, Y.; Waer, M.; et al. 1991. Three Years of Experience with a "Presumed Consent" Legislation in Belgium: Its Impact on MultiOrgan Donation in Comparison with European Countries. Transplantation Proceedings 23: 903--4.

[ 277 l

99

100

Death, Dying and the Ending of Life, 1 KENNEDY INSTITUTE OF ETHICS JOURNAL • JUNE

1993

Shewmon, D. Alan. 1988. Anencephaly: Selected Medical Aspects. Hastings Center Report 18 (5): 11-19. Solomon, Mildred Z.; O'Donnell, Lydia; Jennings, Bruce; and Guilfoy, Vivian. 1993. Decisions Near the End of Life: Professional Views on Life-Sustaining Treatment. American Journal of Public Health 83: 14-23. Spital, Aaron. 1991. Living Organ Donation: Shifting Responsibility. Archives of Internal Medicine 151: 234-35. Starzl, Thomas E. 1992. The Puzzle People: Memoirs of a Transplant Surgeon. Pittsburgh, PA: University of Pittsburgh Press. Tomlinson, Tom. 1990. Misunderstanding Death on a Respirator. Bioethics 4: 253-:64. Truog, Robert D., and Fackler, James C. 1992. Rethinking Brain Death. Critical Care Medicine 20: 1705-12. Veatch, Robert M. 1976. Death, Dying, and the Biological Revolution. New Haven, CT: Yale University Press. Walters, James W., and Ashwal, Stephen. 1988. Organ Prolongation in Anencephalic Infants: Ethics and Medical Issues. Hastings Center Report 18 (5): 19-27. Weisbard, Alan J. 1993. A Polemic on Principles: Reflections on the Pittsburgh Protocol. Kennedy Institute of Ethics Journal 3: 217-30. Youngner, Stuart J., and Bartlett, Edward T. 1983. Human Death and High Technology: The Failure of the Whole-Brain Formulations. Annals of Internal Medicine 99: 252-58. --;Landefeld, C. Seth; Coulton, Claudia].; et al. 1989. "Brain Death" and Organ Retrieval. A Cross-Sectional Survey of Knowledge and Concepts Among Health Professionals. Journal of the American Medical Association 261: 2205-10.

[ 278 J

[7] SOME MUST DIE

by Stuart J Youngner

Abstract. The transplantation and procurement of human organs has become almost routine in American society. Yet, organ transplantation raises difficult ethical and psychosocial issues in the context of "controlled" death, including the blurring of boundaries between life and death, self and other, healing and harming, and killing and letting die. These issues are explored in the context of the actual experiences of organ donors and recipients, brain death, the introduction of non-heartbeating donor protocols, and the increasing reliance on living donors. The author draws on a thematic analysis of the way that organ transplantation is presented in the media, films, and science fiction and on his clinical experience as a psychiatrist working with transplant patients, their families, and the nurses and physicians who care for them.

Keywords:

cannibalism; culture; death; Sigmund Freud; myths; nurses; organs; self

With the advent of organ transplantation, the myths, fantasies, and nightmares of past generations have become not only a reality but also a banality in the clinical and economic life of modern medical institutions. Organ transplantation, today the standard of care for many life-ending diseases, has become a growth industry. Dying persons desperately want it, third parties pay for it, the news media promote it, and increasing numbers of medical centers clamor to join the no-longer-elite circle of transplant centers that provide it.

Stuart J. Youngner is Director of the Center for Biomedical Ethics and the Susan B. Watson Professor of Biomedical Ethics, Case Western Reserve University, 10900 Euclid Ave., Cleveland, OH 44106-4976; e-mail [email protected]. This article appeared in Organ Transplantation. Meanings and Realities, ed. Stuart J. Youngner, R. C. Fox, and L. J. O'Connell, and is reprinted by permission. Copyright 1996 by The Park Ridge Center for Health, Faith, and Ethics, all rights reserved. Portions of this article were presented in a paper at the Annual Lecture Series in Science and Religion at Aurora University, Aurora, Illinois, Spring 2002.

Death, Dying and the Ending of Life, 1

102

706

Zygon

However, human body parts, the natural resources necessary for transplantation, are in short supply. The tensions created by this shortage remind us that transplantation retains magical, irrational, and frightening aspects, which dog its progress like shadows-shadows that can darken when the bright lights of rationality and utilitarianism propel public discourse in the effort to obtain more and more organs. Thus, for example, while public service announcements and human-interest stories extolling transplantation multiply, so too do grisly cartoons, jokes, and tabloid horror stories about the same subject. Our society views organ transplantation simultaneously from two distinct and often contradictory perspectives. The first is the rational ethos of the Enlightenment, appropriated not only by medical science, academia, and public policy but also by mainstream religions. The second is the messier and less well articulated world of emotion, superstition, and magic. In the context of human psychology, neither perspective is right or wrong. They simply exist and function together, an important and unavoidable quality of the human condition. With cool rationality alone, we would be computers or robots. Without it, we would be prey to our emotions and fantasies, unable to cope. Our experience with human organ transplantation is an excellent illustration of the dynamic and often uneasy relationship between these two ways of seeing and understanding the world. PRIMARY- AND SECONDARY-PROCESS THINKING

Sigmund Freud's notions of primary- and secondary-process thinking provide a helpful framework for understanding society's apparently contradictory attitudes toward organ transplantation. Secondary-process thinking is the most immediately familiar of the two. AB Charles Brenner notes, "It is ordinary conscious thinking as we know it from introspection, that is, primarily verbal and following the usual laws of syntax and logic" (Brenner 1973, 52). Primary-process thinking, in contrast, characterizes unconscious or preconscious thinking and dreaming. Compared with the secondary process, it is less objective and organized and "is ruled by emotions and hence full of wishful or fearful misconceptions ... remote from any logic (Fenichel 1972, 47). Primary-process thinking is carried out more through pictorial, concrete images; representation by allusion or analogy is frequent; and a part of an object may be used to stand for the whole. Similarities are not distinguished from identities, and mutually contradictory ideas can coexist peacefully. Primary-process thinking is a magical type of thinking. Not only may wish be equated with deed and fantasy with action, but the perpetrator of a crime or misdeed will be punished with the same injury he or she inflicted. In primary-process thinking, there is no sense of time; past, present, and future are all one (Brenner 1973, 52). Primary-process thinking is easily recognizable in infants and small

Death, Dying and the Ending of Life, I

Stuart J Youngner

103

707

children, in seriously disturbed psychiatric patients, and in dreams, where fantasies and other internal stimuli are projected onto other persons or the environment as delusions or hallucinations, indistinguishable from reality. Psychoanalysts from Freud to the present have observed that primary-process thinking is an integral, though subordinate, part of normal adult mental life, where it becomes visible in dreams, games, jokes, and slips of the tongue. The primary and secondary processes are in dynamic rather than static equilibrium. When the capacity of the secondary process is attenuated by sleep, metabolic disturbances, drugs, or severe anxiety, the primary process breaks through. Under these circumstances, the wishes and fears of the unconscious are perceived as real, and rational thought is replaced with magical thinking. Conversely, the irrational fears and delusions of the unconscious can be diminished by strengthening the secondary process, for example, through education or the reassurance of a trusted person or authority. The inherent nature of transplantation provides both a stimulus and rich material for the primary process. In order for some to live and benefit from transplantation, others must die. When organs are transplanted, conventional boundaries between persons are violated. When organs are taken from newly dead bodies, bodies are mutilated. While the organ shortage grows, our increasingly desperate efforts to create new sources of organs inevitably provoke additional discomfort. For example, by recognizing brain death, we have created a new class of dead persons whose hearts continue to beat for days, weeks, even months. And by taking organs from baboons, chimpanzees, and pigs, we violate boundaries between species. I argue that, at an even more subterranean level, organ transplantation represents a form of non oral cannibalism. Circumstantial evidence of the impact from these issues can be found in the very fact of the organ shortage, which persists despite widespread medical and legal acceptance of organ transplantation, a radical expansion of criteria for determining death, laws that require health professionals to give families the donation option, and a federally funded nationwide organ procurement and sharing system. This essay provides a more direct exploration by examining (1) the experience of patients who are waiting for or who have received organ transplants, (2) the treatment of organ transplantation in popular culture, (3) the link between American society's rapidly evolving but highly controversial acceptance of mercy killing and new methods of organ transplantation, (4) the experience and behavior ofhealth professionals involved in procuring organs from brain-dead patients, and (5) the use of language in transplantation. CASE STUDIES

Mr. Anderson. I first met Mr. Anderson when I performed a routine psychiatric examination as part of his evaluation for heart transplantation.

Death, Dying and the Ending of Life, 1

104

708

Zygon

He was a large, well-built man in his late fifties who appeared surprisingly healthy considering he was in the end stages of a cardiomyopathy-a progressive weakening of his heart muscles, most likely the result of a viral infection. Mr. Anderson lived in central Ohio, a two-and-a-half-hour drive from Cleveland, where our transplant program was situated. He had been a career Marine Corps officer who retired at age fifty to become the manager of a small dry-cleaning business. He was married and had two children and four grandchildren. Mr. Anderson's exam was unremarkable. He denied any psychiatric or social problems and had a positive attitude about transplantation. He expressed no concerns about taking an organ from another person into his body. I saw no evidence of depression or anxiety. I felt almost embarrassed when I asked him a routine question about whether he had ever had hallucinations. He was silent for a minute and then said, "Not exactly." He then recounted the following story. One month before, when he had come to Cleveland for his first evaluation visit, he had been informed that, if he were put on the waiting list, he would have to carry an electronic pager so that he could be reached instantly if a heart became available. He was also told that in order to save critical time he would be flown by emergency helicopter to Cleveland. This information made him quite anxious. He had been in two helicopter crashes, once in Korea and once in Vietnam. The Vietnam crash had been the more serious. He was the gunner in a helicopter that was flying into a battle zone to pick up wounded soldiers. Coming in low over the trees, the helicopter had been shot out of the air and crashed in flames. Miraculously, he had been thrown clear of the wreck with "hardly a scratch," but the pilot and copilot had burned to death. Mr. Anderson denied any symptoms of posttraumatic stress disorder other than a fear of flying in helicopters. He had not been in one since the crash in Vietnam. While he drove back home, he had worried about flying again in a helicopter. When he arrived at home, his nine-year-old grandson greeted him, full of questions about the transplant. When the child asked him where the heart would come from, the patient somewhat uncomfortably explained that it would come from a person who has died "in an accident or something." His grandson was surprised. He had assumed that "they just have them on the shelf in the operating room." He then asked his grandfather if he would still love him once he had another person's heart in his body. "Of course I will," Mr. Anderson had replied. That night he went to bed ruminating about the questions his grandson had raised. Later he awoke with a start when someone tapped him on the shoulder. The patient assured me he had not been dreaming. "I was awake," he said. "No doubt about it." He had sat up in bed to find the pilot and copilot of the helicopter that had crashed in Vietnam in his room. They stood silently in full battle gear, stared at him for a few minutes, and then

Death, Dying and the Ending of Life, I

Stuart J Youngner

105

709

beckoned with their hands for him to come. Mr. Anderson was terrified and began sweating profusely. The two men gradually faded away, leaving him trembling as he sat on the side of the bed. Because of his Vietnam experience, Mr. Anderson was especially vulnerable to unconscious and primitive feelings of responsibility and guilt. As Renee Fox has remarked, organ transplantation is both lifesaving and deathridden (Fox 1988, 170). The vast majority of organs come from persons who have died from sudden, unexpected head injuries. Something must happen to one person for another to be saved. Of course, our rational minds tell us that these deaths are linked to organ procurement only after the fact. To Mr. Anderson the linkage became more problematic for several reasons. Like many survivors of catastrophic events, he harbored a tremendous sense of guilt and responsibility for the deaths of his helicopter pilot and copilot. The horror of this earlier episode both fueled and was fueled by the transplant situation, in which Mr. Anderson would have to take another fateful helicopter ride that would end with himself saved at the expense of another's death. Mr. Anderson's defenses were further weakened when he saw transplantation through the eyes of his grandson, who hoped that hearts were to be found like equipment, on the shelf of the operating room, and innocently feared that his grandfather's love for him might disappear when the elder's original heart was discarded. Under the mantle of sleep, his primitive thinking gained ascendancy, awakening him to a terrifying hallucinatory morality play.

Ms. jones. Ms. Jones, an uneducated woman in her sixties, began having a recurrent nightmare months after she had received a kidney transplant. In her dream, a dead man approached her calling her name, and, like Mr. Anderson's fellow Marines, beckoned her to join him in death. She would awaken terrified and remain anxious during the day. She was certain that the dead man in her dream was the person from whom she had received a kidney; she had been told he had died in a car accident. In our discussions, she revealed that twenty-five years before, her sixyear-old son had run out in traffic and was struck by an automobile. He had been taken to the hospital and placed on a mechanical ventilator, where it was determined that he had no brain function. This had occurred in the early 1960s, before brain death was widely accepted and before health professionals felt comfortable removing dying patients from ventilators (many are still uncomfortable doing this). Ms. Jones insisted that she had been told that there was no hope but that if she wanted the plug pulled, she would have to do it herself "I wanted to do it, but I was afraid," she told me. Her son died two days later, but she had always felt guilty-that she had not prevented the car accident and that she had wanted to turn off the ventilator but had not done so. After she verbalized her guilt about her son's tragic death, she was

Death, Dying and the Ending of Life, 1

106

710

Zygon

able to connect this guilt with her nightmares and the death of a donor. As a result, her nightmares subsided.

Ms. Smith. A twenty-five-year-old woman with a psychiatric diagnosis of borderline personality disorder was referred to me because of overwhelming anxiety. Ms. Smith had received a kidney transplant two years before and was now in the hospital because her body was rejecting it. When I spoke with her, she told me that she had become obsessed with the thought that her grandfather had been murdered and that she had received his kidney, which was now rejecting her. During our talk, Ms. Smith told me that when she was a child she had been removed from her parents' home because they were sexually abusing her. She had been sent to live with her grandfather and uncles. Her grandfather was a "very nice man" but did nothing to protect her when her uncles sexually abused her. She had loved her grandfather (who had died several years before), but she was also very angry at him for not protecting her. Her anger and guilt were reflected both in the fantasy that he had been killed to save her and the fear that his kidney was now rejecting her. For most patients, concerns about the death of the donor surface in a less disturbing manner. Many of the people I interview volunteer that they are distressed that someone has to die. Several patients have assured me, without my asking, that they had signed organ donor cards since finding out that they would need a transplant. Naturally, many persons suppress their fears and fantasies or are reluctant to share them with psychiatrists or others who are screening them for transplantation suitability. Evidence indicates that complete suppression of these concerns may be a useful coping strategy in the perioperative period (Mai 1986, 1159). However, when patients are bothered by their fears, health professionals can be a source of education and reassurance. For example, a liver transplant candidate in her middle thirties was worried that the donor of her organ would come back to "haunt" her. By talking with staff and with patients who had already received liver transplants, she was able to overcome her fears. Examples of the ambivalence and guilt about donor deaths can be found in the medical literature as well. James Levenson and Mary Ellen Olbrisch (1987, 399) report "gallows humor" among patients waiting for transplant: They not uncommonly "talk about fantasies of standing on the roof of the hospital with a rifle" or ask hospital staff "whether they have had any opportunities to run down pedestrians on their way to work." In several instances, candidates for transplantation or their relatives "have coped by identifYing 'healthy specimens' among the house staff and inquiring as to their body weights and blood types." Robert Frierson and Steven Lippmann (1987, 347) report that "patients often found themselves hoping for inclement weather because of a greater likelihood of fatal accidents to afford more organ donations. This so-called 'rainy day syndrome' was often accompanied by significant feelings of guilt."

Death, Dying and the Ending of Life, I

Stuart J Youngner

107

711

POPULAR CULTURE

The legacy of Burke and Hare, two Englishmen who killed people so that they could sell their bodies for anatomical dissection, and the modern myth of Frankenstein described so vividly by Ruth Richardson (1996) and Leslie Fiedler (1996) are very much alive in our collective consciousness. Stimulated by the steady diet of upbeat news stories, sermons, and publicservice announcements about transplantation and the need for more organs, fears that people will actually be killed for organs find free rein in popular culture. Horror books and movies tell tales of powerful and maniacal physicians (either functioning independently or as agents of even more powerful scoundrels) who take advantage of patients or other, weaker persons to obtain their organs. A recent series of the comic strip "Dick Tracy'' concerned a nefarious conspiracy in which homeless people were shanghaied from a city shelter, killed, and used as organ sources (Locherand and Collins 1991). As noted earlier, the fears are often expressed as humor. In the movie Monty Python's The Meaning of Life, a group of medical people ring the doorbell of a family home. When the father answers the door, they ask him if he has signed a card donating his liver. When he replies in the affirmative, they barge into the house, tie him to the kitchen table, and proceed to remove his liver-without anesthetic and in front of his horrified family. Fears about killing for organs are given impetus by proposals to allow criminals condemned to death to donate their organs as part of the execution process. Some prisoners serving life sentences have even asked to be executed so they could both end their miserable existence in prison and donate organs. In fact, allowing condemned inmates to donate their organs upon execution is a: pet project of Dr. Jack Kevorkian, known to the American public as a crusader for physician-assisted suicide (Kevorkian 1991). Of course, such killings can be rationalized as different from the killings in Coma (Cook 1977), a novel and movie in which patients are killed and kept in suspended animation so that their body parts and fluids can be used. After all, in the horror movies and books, the organ sources are unwilling victims of murder. In real life, the prisoners give consent or even initiate requests to have their organs taken. These subtleties, however, are lost on the unconscious mind, where killing and organ procurement are linked concretely. Primitive but powerful associations are only reinforced by stories from China about executed criminals whose organs are taken whether or not they wanted to be donors ("Grim Commerce in China'' 1994). Even more disturbing are stories in the press about innocent persons killed for organs. In 1992, newspapers reported that hundreds of patients at a mental hospital near Buenos Aires were allegedly killed by greedy staff

Death, Dying and the Ending of Life, 1

108

712

Zygon

members who sold their body parts and blood ("Patients Killed for Organs" 1992; "Journal: Patients at Argentine Hospital" 1992). For several years, rumors have circulated that children from various South and Central American countries are being kidnapped and taken to the United States to be "fattened up" and then killed for their organs. More recently, foreigners have been attacked in Guatemala in a panic fed by rumors that Americans were coming to kidnap children, cut out their vital organs, and ship them to the United States for transplantations ("Behind the Kidnapping" 1994; "Guatemala Gang" 1994). An article in Asiaweek tided "Life for Sale" (1994) described the extensive and legal sale of kidneys (single kidneys from live donors) in India. It went on to describe a disturbing story from Bangladesh: Last year 400 children disappeared in Chittagong city, according to police. It's suspected that most of them were kidnapped for their organs. In October police in suburban Dhaka rescued 49 people and arrested two human traffickers. The men confessed that they were going to smuggle the group into India under the guise of finding them work. Once there they planned to sell some of them to human organ traders who operated through some hospitals in Bombay and Madras. The victims would be forced to donate their kidneys, eyes, hair and blood. (1994, 49)

None of these stories has ever been verified by reliable sources, and the transplant community in the United States has vehemently denied their veracity. Furthermore, no one has seriously proposed that any of these ghoulish practices actually occurs in the United States. But the newspaper stories and rumors keep coming. Horror movies and novels continue to be produced and sold. The stories find their way to the collective unconscious, where they percolate, resonating with the popular myths that Fiedler describes (1996), breathing new life (or death) into the psychic legacy of body snatching for dissection, which Richardson so vividly portrays ( 1996), and playing on the fears not merely of the paranoid and insecure but of the poor and disadvantaged as well. Examining the historical relationship between organ transplantation, the determination of death, and treatmentlimitation decisions helps to explain the darker side of American society's reaction to organ procurement. PLANNED DEATH AND ORGAN PROCUREMENT

Aside from living donors (who provide about 20 percent of donated kidneys in the United States), solid vascularized organs such as the kidneys, heart, and liver come from patients who are brain dead but whose hearts continue to beat. Brain death is a legal and clinical concept introduced in the 1960s that subsequently achieved widespread acceptance throughout the United States. The reasons for introducing it were quite rational: (1) to facilitate organ procurement and (2) to avoid legal concerns about turning off ventilators (Ad Hoc Committee 1968, 337; Youngner 1992, 570).

Death, Dying and the Ending of Life, I

Stuart J Youngner

109

713

A patient who suffers a massive brain injury (for example, in a car accident or following a heart attack) can now be resuscitated and put on a mechanical ventilator. Moreover, physicians and nurses in the intensive care unit can now perform many of the integrative functions previously carried out automatically by the brain, such as regulating body temperature and blood pressure. Brain-dead patients are legally dead in every jurisdiction in the United States, yet they are a wonderful source of organs because their hearts beat spontaneously, pumping warm, richly oxygenated blood throughout their living bodies until their organs can be removed and quickly put on ice. Brain death raises a profound problem that is relevant to our general discussion, which I consider later-namely, the cognitive dissonance engendered by the overwhelming signs of life in these "dead" patients. For the moment, however, I turn to new methods organ transplanters have employed to increase the donor pool. NEW METHODS FOR OBTAINING ORGANS

As organ transplantation has become more popular, the supply of braindead patients has not kept pace. In response to the organ shortage, transplanters have developed new techniques for retrieving organs from traditional corpses-so-called non-heart-beating cadavers (NHBCs). The University of Pittsburgh Medical Center has implemented a protocol in which the time of death (not brain death, but death by cardiopulmonary criteria) is controlled so that it occurs in the operating room, where organs can be quickly removed before they are damaged (Youngner and Arnold 1993, 2769; Arnold and Youngner 1993a, 103). Patients who are potential candidates for organ donation under this protocol are ventilator dependent but not brain dead. They might be severely brain injured but still have identifiable brain function, or they might have completely intact cognitive function but are unable to breathe spontaneously (for example, patients who are quadriplegic from high spinal injuries). In either case, after the patient's family (or the patient when competent) asks that life support be removed, they can also request that organs be donated for transplantation. After an elaborate informed-consent process, the patient is taken to the operating room (if organ procurement were not in the picture, ventilator removal and death would occur in the intensive care unit), prepped for surgery, and the ventilator turned off. Two minutes after the heart stops beating, the surgeons come in and remove the organs as quickly as possible to reduce warm ischemia time, during which organs are damaged. Controlling the timing of death in this way would have been hardly imaginable a quarter century ago, when turning off machines that kept people alive was very controversial. To many persons, it seemed too dose to killing. Health professionals and health-care institutions were reluctant

Death, Dying and the Ending of Life, I

110

714

qgon

to turn off mechanical ventilators for fear they would be sued or prosecuted for murder. (In countries that have had less experience with medical technology, the level of controversy remains high.) Clinical practice and the law have undergone dramatic evolution over the past two and a half decades. Competent patients or their surrogates may refuse any form of life-sustaining treatment, including mechanical ventilators and artificially provided fluids and nutrition. Today we turn off the ventilators not only of brain-dead patients but also of still-living patients with clinical conditions ranging from terminal illness to quadriplegia. However, this evolution in attitude and behavior has likely not run its course and has special relevance to organ procurement under protocols like the one in Pittsburgh. At preconscious and unconscious levels, our society has accepted increasingly active forms of physician-assisted suicide and euthanasia, paving the way to open acceptance. Local newspapers and television news programs regularly feature stories about Kevorkian and his assisted suicides. National polls reflect a growing public acceptance of physicianassisted suicide and euthanasia, which have become widespread and socially sanctioned practices in the "civilized" Netherlands. Prestigious medical journals such as the New England journal ofMedicine have published articles in which physicians have openly admitted helping patients end their lives and set forth comprehensive guidelines for this practice (Quill1991, 691; Quill, Cassel, and Meier 1992, 1380). While voter initiatives to legalize physician-assisted suicide and euthanasia have been narrowly defeated in two states, one was recently approved by Oregon voters. In sum, then, it appears likely that physician-assisted suicide and euthanasia will be explicitly or implicitly tolerated in many jurisdictions in the United States within the next decade. If this prediction proves accurate, one can easily extrapolate the likely scenario for organ procurement. If we ask patients, as the Pittsburgh protocol does, to become donors when they ask that their ventilators be turned off, why would we not allow them the same prerogative when we help them to commit suicide or put them to death at their own request (Arnold and Youngner 1993b, 263)? Unless something unforeseen intervenes to disrupt it, the evolution from an ever-expanding array of treatment limitations to an equally expanding array of voluntary suicides and mercy killings seems inevitable. The decision to allow euthanized patients to donate their organs would follow quite naturally, once the more controversial practices of suicide and mercy killing were to become sanctioned. By controlling the time and place of death, the Pittsburgh protocol takes a critical symbolic step: It links the planned death of one human being to the procurement of organs for another. The transplantation of tissue from electively aborted fetuses is a second example. What makes the Pittsburgh protocol legally and morally acceptable is that both the death of the donor and the taking of his or her organs are voluntary. This same protection

Death, Dying and the Ending of Life, I

Stuart J Youngner

111

715

could exist if organ donation were linked to physician-assisted suicide or voluntary active euthanasia. An extensive informed-consent process and the total absence of coercion would be necessary to prevent the killing for organs, about which both Richardson and the modern myths of popular culture warn and for which they, perhaps, prepare us. PUBLIC POLICY PROTECTIONS

From early on, the government and the transplant community have tried to dispel concerns that people will be killed or exploited for their organs. The National Organ Transplant Act of 1984 made clear that care of the potential donor and care of the potential recipient must not be provided by the same persons and that transplant personnel should in no way be involved in treatment decisions before the donor's death. In contrast with the Organ Transplant Act, the dead-donor rule evolved as an informal policy, but, like the Organ Transplant Act, it is intended to reassure the public that people will not be taken advantage of so that their organs can be used (Arnold and Youngner 1993b, 263). The dead-donor rule has two aspects: first, people must not be killed by or for organ retrieval; and, second, with the exception of completely healthy family members, people can have their organs taken only after they are dead, even if taking the organs will not kill them. So, for example, we are unwilling, even with a family's (or the patient's prior) permission, to take a single kidney from a patient in a permanently unconscious state. Nonetheless, public opinion polls regularly reflect the fear of many persons that their care will be compromised so that they can become a source for organs (Prottas and Batten 1986; Prottas and Batten 1991, 121). Such fears are greatest among minority groups, who have good historical reasons for mistrusting organized medicine or for doubting that they will be protected by the law (Collender, Hall, Yeager, et al. 1991, 442). Already, demagogues in the African-American community have sought to play on these fears (see "Farrakhan Links Race" 1994). Appeals to informed consent and protection of the law are more likely to quell the irrational fears of those members of society who have the power, privilege, and experience to exploit those protections more successfully. BRAIN DEATH: ANOTHER LAYER OF CONFUSION

When news media report on brain death, they regularly refer to legally dead patients as being "kept alive'' on life-support systems or "in critical condition." Such ambiguous language is the rule rather than the exception. I recently witnessed a television anchorman describe a severely braindamaged patient as "in critical condition but technically brain-dead." Such confusion is common even among health professionals, who stubbornly persist in describing patients who have lost all brain function as

Death, Dying and the Ending of Life, I

112

716

Zygon

brain-dead rather than simply dead. These patients, who are considered legally dead in all fifty states, are also characterized as being kept alive on "life" support and as "dying" after that support is removed. Even though the legal time of death occurs when the patient is determined to have irreversibly lost all brain function, health professionals regularly fill in the death certificate with the time the heart stopped beating (after life support is stopped). I alluded earlier to the cognitive dissonance stimulated by the phenomenon of brain death. Before the advent of medical technology, numerous indicators of death occurred more or less at once. So, for example, a person with a fatal heart attack would lose consciousness, stop breathing, become motionless and unresponsive, and have no detectable pulse all at approximately the same time. All the vital signs of life would vanish together. Now, however, medical technology has forced us to choose which signs oflife are sufficiently important that their loss constitutes the death of the patient, while other signs of life persist. The traditional view rejects the notion of brain death altogether, arguing that vital fluid flow-the movement of air and blood through the body-indicates life. Thus, the persistence of cardiac and pulmonary function sufficiently demonstrates that the patient is still alive. In the United States, this view receives considerable support in the fundamentalist Christian and Orthodox Jewish communities. In contrast, the more recent view holds that even with a spontaneously beating heart and air flow in and out of the lungs (by means of a ventilator), irreversible loss of all brain function (brain death) signals the death of the patient. However, in the intensive care unit (ICU), where brain-dead patients are maintained, and in the operating rooms, where their organs are removed, nurses, house officers, and anesthesiologists are often confused and sometimes dismayed by these paradoxical patients. In the ICU, nurses and physicians must "treat" these dead patients quite aggressively in order to maintain them for transplantation. They must attach them to breathing machines, monitor them for heart rhythm and blood pressure, give them fluids and nutrition, and, when indicated, administer antibiotics and other medications. The medical staff must also closely monitor and adjust the patients' blood chemistry and oxygen levels. These dead patients are even candidates for full resuscitation should they suffer cardiac arrest. Yet, in the next bed may lie a completely conscious patient who, at her own request, does not want to be resuscitated. ICU nurses often talk to braindead patients and are concerned when they are subjected to painful procedures. Brain-dead patients pose an even greater emotional challenge for operating-room personnel, who must maintain them through organ retrieval surgery. Such a patient is prepped like any other surgical candidate and draped to reveal only the operative field. An anesthesiologist stands at the

Death, Dying and the Ending of Life, I

Stuart J Youngner

113

717

head of the patient to manage the mechanical ventilator and maintain homeostasis by giving fluids and drugs. The surgeons use sterile technique, tie off or cauterize bleeding blood vessels, and carefully cut and separate tissue planes. The patient's chest rises and falls with the rhythm of the ventilator. And it is not uncommon to give the dead patient a transfusion of fresh blood while his or her organs are being removed. These similarities between brain-dead and regular surgical candidates add emotional force to the differences as the process unfolds. Instead of diseased tissue, healthy organs are removed, and then the meticulous attention to detail comes to an end. It becomes concretely clear that the purpose of the surgery is for another patient, not the one in the room. The mechanical ventilator is turned off, and the patient's newly emptied body is closed in one pass with coarse retention sutures. Some nurses and anesthesiologists describe the turning off of the ventilator as the most emotionally intense moment. Some say that not until then does the patient's spirit leave the room. Others frankly describe it as the second death of the patient. It is most upsetting for the unprepared and the uninitiated-young nurses or anesthesiologists or those who work at smaller community hospitals where organ procurement and transplantation are rare events. The resulting confusion and cognitive dissonance were reflected in a study ofiCU and operating-room physicians and nurses. Almost all participants intellectually accepted that brain-dead patients were indeed dead. However, when asked what makes brain-dead patients dead, fully onethird gave answers indicating that they really believed such patients to be alive, for example, "the patient will die soon, no matter what we do," or "the patient's quality oflife is unacceptable" (Youngner, Landefeld, Coulton, et al. 1989, 2205). Cognitive dissonance was magnified in the case of an eighteen-year-old woman, whom I will call Janet, twenty-two weeks pregnant, who suffered a spontaneous ruptured cerebral aneurysm and was admitted to the ICU, where an unequivocal diagnosis of brain death was made within twentyfour hours. Here, nurses and physicians cared for dead mother and living fetus. Eight weeks later, a healthy baby was delivered by cesarean section, following which the young woman's heart, liver, pancreas, and kidneys were removed and transplanted into four waiting patients, three of whom were cared for in the same ICU that had maintained the dead mother. Because the care of Janet was going to be both clinically and emotionally challenging (the first such case for the ICU), a small group of nurses volunteered to provide it. Heavily identified with the tragedy, they became very attached to both the dead mother and the living fetus, who had already been named. For many of the staff, taking care of the patient was a religious experience. Its mystical nature was enhanced by the fact that Janet had anticipated her death a week before it occurred, when she told her family that "if anything happens to me, I want them to do everything

Death, Dying and the Ending of Life, I

114

718

Zygon

to save the baby." The ICU staff's mission was to bring a healthy baby out of the tragedy, and they constantly watched and worried over it. But they were also preoccupied with the baby's mother. One nurse described what the day-to-day care of Janet was like: "We kept her immaculately clean and neat, even had her mother bring in a silk robe in which to dress her. I washed her hair every week. It was long, beautiful red hair, and it grew for eight weeks. I could sense the presence of her soul in her body." Another nurse sensed the presence of a soul hovering over the body, "watching us." The nurses developed rituals, including putting a picture of the dead mother on the wall. They played music in the room, "for the baby," but were convinced that the mother's heart rate changed in response to it. The physicians, who spent much less time with the patient (the nurses worked one-on-one with the patient in ten-hour shifts), were less emotionally involved, although one of the obstetricians was clearly convinced that "the whole thing [had] a preordained purpose." Not surprisingly, the medical staff constantly used speech indicating that the patient was alive. "Our job was to keep the mother alive until the baby was born," one physician told me. A nurse said, "We all knew she was dead, but we folt she was alive." The patient's mother told a physician, "Every time I leave, I think she is going to finally die, and each time I return to the ICU, she is still with us." What is reality for family members who watched the living body of a dead girl nurture, grow, and issue forth a healthy baby? While Janet's hair grew longer and the nurses washed and combed it, was she really alive or dead to them? They could rationally say that she was dead and explain why; the signs of life told another story. THE LANGUAGE OF ORGAN TRANSPLANTATION

Richardson argues ( 1996) that the language of organ transplantation is sometimes intended to disguise its darker side. Language can also provide an unwitting representation of transplantation's more disturbing but subterranean aspects. The persistence of the term brain-dead to describe patients whose hearts continue to beat but who have lost all brain function, and the habit of referring to them as alive only to describe them as dying when their ventilators are turned off, does not reflect mere ignorance of the facts. The transplant community correctly perceives that, by using the term brain death instead of simply death, health professionals and journalists encourage the notion that a difference exists-that brain-dead patients are not dead but constitute some other category of being. They are mistaken, however, in thinking that mere education will either eliminate the use of the term or solve the problem. The fact that experienced transplant surgeons and nurses make the same "slip" on a regular basis is evidence that something more powerful than

Death, Dying and the Ending of Life, I

Stuart J Youngner

115

719

lack of knowledge is at work. The physicians and nurses who cared for Janet were intellectually aware that she was legally dead and that the fetus inside her was alive. They persisted in referring to both mother and fetus as alive because of the abundant signs of life emanating from both. As mythologist Wendy Doniger insightfully commented when I described this phenomenon to her, "Why, it's just the way we persist in describing a sunset, when we know full well on the intellectual level that the earth is moving around the sun, not vice versa." Other words are viewed as politically incorrect by the transplant community because of their unpleasant connotations. Procurement of organs has an unsavory association with commodities and commerce, even prostitution. Some persons have suggested organ retrieval as preferable, but others have objected that retrieval implies that the organs belonged to someone other than the donor patient and that we are merely taking them back. Terms such as heart-beating cadaver or neomort (Gaylin 1974, 123) are offensive to many persons because they seem ghoulish or crude. Another example supports the notion that transplant language can be sanitizing: most cadaver "donors" are, of course, no such thing. They often have given no indication of what they want done with their organs when they die. The organs are most often donated by their families. It is more comforting, of course, to think of them as organ donors than as organ sources. Rather than view the language of transplantation as a product of ignorance or intentional disrespect, however, we should understand it as an inevitable expression of the complex and dynamic interaction between the rational-utilitarian and emotional-symbolic ways in which we understand or try to understand this wonderful and terrifYing miracle of science. CANNIBALISM

Richardson suggests that the word harvesting presents too benign an image of organ procurement. Some persons disagree, arguing that the word is disrespectful to the dead donors, implying that they are vegetables rather than human beings. An even more disturbing interpretation is possiblethat harvesting imports the notion of cannibalism. Of course, in an entirely concrete sense, organ transplantation is a form of nonoral cannibalism, that is, the taking of the flesh and blood from one person into another. Historically, two general categories of cannibalism have existed: survival cannibalism and ritual cannibalism. Survival Cannibalism. Of the two categories, survival cannibalism is metaphorically closest to organ transplantation-taking in the flesh of another as the only means of preserving one's own life. Examples of survival cannibalism abound throughout history. Two that are part of our public consciousness concern the notorious Donner party, in which human

Death, Dying and the Ending of Life, I

116

720

Zygon

cannibalism was widespread among a group of pioneers trapped by a winter snowstorm in the Sierra Nevada in the late nineteenth century (Stewart 1986), and the more recent story of a group of Uruguayan rugby players and their families stranded in the Andes Mountains after an airplane crash. The latter story was published in a bestselling book (Read 1974) and has been made into two popular movies. The inherent drama and tragedy of these situations are not the only things that capture the public imagination; these situations also depict well-meaning people who overcome their own sense of repugnance and violate a strict social taboo in order to survive. In such cases, society seems to understand and forgive-as long as those consumed died naturally and were not unjustly killed. Examples of survival cannibalism also offer insight into how rituals can be created or adapted to help overcome the extraordinary and terrifYing nature of the act. For example, the rugby team (the Old Christian Club) used a cannibalistic ritual at the very heart of Christianity, communion, to make their own cannibalism more acceptable. "It's like Holy Communion," one of them said. "When Christ died, he gave his body to us so that we could have spiritual life. My friend has given us his body so we can have physical life" (Read 1974, 83). When death is planned and manipulated (this is, when people are killed for their flesh), the emotional and moral stakes are raised. Examples of cannibalism abound in British naval history and provide examples of how reasonable people in unreasonable circumstances (for example, starvation in a life boat) rationalized their choice of whom to kill (Simpson 1985). The "fairest" way was the drawing oflots-for both the person to be killed and the killer. Less judicious were decisions to kill individuals because of their race or because they were children. One way to make such deaths more tolerable was to identifY people who were "about to die anyway," thereby diminishing the harm to them. Sometimes these murders were rationalized as mercy killing. One could argue that this same reasoning underlies our current acceptance of brain death. That is, in patients who have lost all brain function, we have identified a group of severely injured and dying persons who are so "beyond harm" that we feel justified in killing them in order to obtain their organs (Arnold and Youngner 1993b, 263). Because we would rather not think that we are killing them, we simply gerrymander the line between life and death to include them in the latter category. Conceptual gerrymandering is even more apparent in proposals to label anencephalic infants (born with no cerebral hemispheres but entirely functioning brain stems) as dead, precisely so that their organs can be taken for transplantation (Fletcher and Truog 1989, 388). Recently, the American Medical Association took a more candid approach: "It is normally required that the donor be legally dead before permitting the harvesting of organs. The use of the anencephalic infant as a live donor is a limited exception to

Death, Dying and the Ending of Life, I

Stuart J Youngner

117

721

the general standard because of the fact that the infant has never experienced, and will never experience, consciousness" ("Anencephalic Infants" 1994; "Council: Use of Anencephalic" 1994, 9). The AMA did not suggest that anencephalic infants were dead but rather that they were beyond harm and, therefore, could be killed for their organs. Of course, the AMA did not use the word killed. Ritual Cannibalism. Ritual cannibalism is often performed with the intent ofincorporating desired qualities of the person who is eaten. While incorporation of the donor's personal characteristics (other than the health of his or her organs) is not the intent of organ transplantation, the early transplant literature is replete with examples of patients who either feared incorporating unwanted characteristics or were convinced it had actually happened. Renee Fox (1996) points out how little the current psychiatric and medical literature comments about this issue, but it has not been ignored in lay literature, novels, or movies. Writing in the New Yorker, a general surgeon, Sherwin B. Nuland, describes his interview with a man named Cretella, who had received a heart transplant: Toward the end of my visit, our conversation turned to a topic I had been hesitant to bring up. What does it feel like to live with another person's heart beating in your chest? It proved to be something Cretella was trying very hard not to think about. "I don't know yet," he said. "I really don't know yet. When I catch myself thinking about it, I try to forget about it. You know-I think, What is it? A female? A male? Black? Orange? White?" I asked him what he would want it to be. "I don't know that yet, either. I can't answer any questions like that at all. I even get upset talking to you about it. When I talk about it, I get paranoid. I think mainly it's because I don't know what's going to happen tomorrow, and the reason for that is that I can be sitting here feeling fine and all of a sudden something clicks and I get nervous and everything just starts going. Something in my body changes, as if somebody pushed a button. I talked to another transplant patient-he's in his fifth year-and he says it still happens to him ... you know, they tell you it doesn't make any difference what kind of heart you get. And I'm sitting there thinking, I don't believe that, I honestly don't believe it." (Nuland 1990, 82)

In my own experience, transplant candidates and patients often express fantasies and worries about taking on characteristics of the donor. Mr. Anderson's young grandson, for example, was afraid that his grandfather would no longer love him after the man received a new heart. His "mistake" was an innocent and unrationalized representation of the emotional significance given to the heart throughout our society, even though we know that the brain is the real seat of our emotions. The boy expressed concretely what usually remains in our subconscious or emerges in common metaphors when we speak of losing our hearts, breaking our hearts, and having good or evil hearts. These examples, to use Doniger's analogy, are like talking about the sun setting over a post-Copernican horizon.

Death, Dying and the Ending of Life, I

118

722

Zygon

Transplant candidates sometimes joke with me about adopting the sexual, ethnic, or other personal characteristics of donors. They rarely express their fears directly, either because they have successfully rationalized them or because the potential candidates are worried that candor would harm their chances for receiving an organ (in fact, it would not). Humor, of course, provides a window into the unconscious, deriving its energy from the partial exposure of what is usually hidden and forbidden. It is not surprising that cartoons and movies often present concerns about identity and transplantation in a comic as well as a horror mode. I learned of a dramatic example of donor-personality incorporation when I was called by a producer of"The Phil Donahue Show," a somewhat sensational TV talk show. An upcoming broadcast would feature a group of heart transplant recipients in the New York City area who claimed that they not only had taken on personality characteristics of the donors but also had come to know intimate details about the donors' lives-all without having received any information about them! Accompanying the heart recipients was a psychoanalyst who had them in a therapy group. I was asked to appear as "a voice of rationality and science," to provide "balance" for the show. (I declined.) While thoughts and fears about incorporating personal qualities of the donor (or losing one's own identity) characterize transplantation, the converse distinguishes ritual cannibalism; that is, people are eaten with the express purpose of incorporating desirable characteristics. For the unconscious mind, the concrete or literal similarities are as important as the differences. The term cannibalism is itself used commonly to describe taking working parts from one machine (for example, an automobile) to fix another that is broken. The association between organs and food becomes even more concrete when transplant surgeons are seen (by other health professionals, as well as by the public when viewing television) transporting organs in easily recognized brand-name coolers usually reserved for carrying picnic lunches. Transplanters also use Tupperware® to transport tissues or wrap them in the same brand-name cellophane that sits on the shelves of our supermarkets. The Japanese film crew that taped a total body harvest in an American hospital for a Japanese national television program debating the merits of transplantation missed none of these culinary details (Lock 1996). One of my own patients made an inescapable connection when he expressed regret that someone had to die so that he could get a liver. "I wish I could go to the grocery store and buy one off the shelf," he lamented. CONCLUSION

I have used the words and experiences of transplant patients, their families, and the health professionals who care for them to examine some of the

Death, Dying and the Ending of Life, I

Stuart J Youngner

119

723

powerful but often subterranean psychological forces that exist alongside the more rational, tidy, and "constructive" view of the official transplant and public-policy communities. Neither is right or wrong. Each must be understood on its own terms. To ignore the more rational side is to miss the wonderful opportunity transplantation offers to save, extend, and improve the lives of thousands of people. To ignore or dismiss the more subterranean side is to build transplant policy on an unrealistic view of the human psyche that is not only insensitive but also ineffective because of its naivete. The tragic and often violent death of others, the taking in of another's flesh to live, the confusion of boundaries, the mutilation of dead bodies, and the cognitive dissonance of brain death are all disturbing but inescapable aspects of transplantation and organ procurement. Society adjusts and accepts new practices, even when they stimulate powerful fears and taboos. Richardson has chronicled how it took England four centuries of dissection to become comfortable enough for voluntary donation to triumph. Her observations about the Anatomy Act of 1832 and her view that it set back voluntary donation by a century should give us pause while we pursue public policies that ignore the deeply held fears and taboos of our society. NOTE I thank Robert Arnold, Rebecca Dresser, Renee Fox, Laurence O'Connell, Rina Youngner, and Julius Youngner for their thoughtful comments and suggestions. REFERENCES Ad Hoc Committee of the Harvard Medical School to Examine the Definition ofBrain Death. 1968. ''A Definition of Irreversible Coma." journal ofthe American Medical Association 205:337-40. Arnold, Robert M., and Stuart]. Youngner, eds. 1993a. "Ethical, Psychosocial, and Public Policy Implication of Procuring Organs from Non-Heart-Beating Cadavers." Special Issue, Kennedy Imtitute ofEthics journal3(2): 103-277. 1993b. "The Dead Donor Rule: Should We Stretch It, Bend It, or Abandon It?" Kennedy Institute ofEthics journal3(2): 263-78. "Behind the Kidnapping of Children for Their Organs." 1994. Los Angeles Times (2 May), Opinion section, MI. Brenner, Charles. 1973. Elementary Textbook of Psychoanalysis. New York: International Universities Press. Callender, C. 0., L. E. Hall, C. L. Yeager, et al. 1991. "Organ Donation and Blacks: A Critical Frontier." New England journal ofMedicine 325:442-44. Cook, Robin. 1977. Coma. Boston: Litde, Brown. "Council: Use of Anencephalic Organ Donors Ethical." 1994. American Medical News (27 June), 9. "Farrakhan Links Race to Transplants." 1994. New York Times (2 May), A18. Fenichel, Otto. 1972. Psychoanalytic Theory ofNeurosis. New York: W. W. Norton. Fiedler, Leslie. 1996. "Why Organ Transplant Programs Do Not Succeed." In Organ Transplantation: Meanings and Realities, ed. S. J. Youngner, R. C. Fox, and L. J. O'Connell, 56-65. Madison: Univ. ofWisconsin Press.

Death, Dying and the Ending of Life, I

120

724

Zygon

Fletcher, John C., and Robert D. Truog. 1989. "Anencephalic Newborns: Can Organs Be Transplanted before Brain Death?" New England journal ofMedicine 321 (6): 388-91. Fox, Renee C. 1988. Essays in Medical Sociology. New Brunswick, N.J.: Transaction Books. 1996. 'Merthoughts: Continuing Reflections on Organ Transplantation." In Organ Tramplantation: Meanings and Realities, ed. S. J. Youngner, R. C. Fox, and L. J. O'Connell, 252-72. Madison: Univ. ofWisconsin Press. Frierson, Robert L., and Steven B. Lippmann. 1987. "Heart Transplant Candidates Rejected on Psychiatric Indications." Psychosomatics 28(7): 347-55. Gaylin, Willard. 1974. "Harvesting the Dead." Harper's 52 (September): 23-30. "Grim Commerce in China." 1994. New York Times (30 August), A20. "Guatemala Gang Kidnaps US Girl, 7." 1994. Boston Globe (12 April), 6. "Journal: Patients at Argentine Hospital Were Killed for Organs." 1992. American Medical News (18 May), 92. Kevorkian, Jack. 1991. Prescription Medicine. Buffalo: Prometheus Books. Levenson, James L., and Mary Ellen Olbrisch. 1987. "Shortage of Donor Organs and Long Waits." Psychosomatics 28(8): 399-403. "Life for Sale." 1994. Asiaweek (13 April), 49. Locherand, Dick, and Max Collins. 1991. "Dick Tracy." Pittsburgh Press (4 August-25 September). Lock, Margaret. 1996. "Deadly Disputes: Ideologies and Brain Death in Japan." In Organ Tramplantation: Meanings and Realities, ed. S. J. Youngner, R. C. Fox, and L. J. O'Connell, 142-67. Madison: Univ. ofWisconsin Press. Mai, Francois M. 1986. "Graft and Donor Denial in Heart Transplant Recipients." American journal ofPsychiatry 143(9): 1159-61. National Organ Transplant Act. 1984. Public Law 98-507, 98th Cong., 19 October. Nuland, Sherwin B. 1990. "Annals of Surgery: Transplanting a Heart." New Yorker (10 February), 82-94. Opinion of the American Medical Association's Council on Ethical and Judicial Mfairs. 1994. 'Mencephalic Infants as Organ Donors." CE]A Opinion 10-A-94. "Patients Killed for Organs." 1992. [London] Guardian (14 April), 10. Prottas, ]. M., and H. L. Batten. 1986. Attitudes and Incentive in Organ Procurement. Report to the Health Care Financing Administration. 1991. "The Willingness to Give: The Public and the Supply of Transplantable Organs." journal ofHealth Politics, Policy, and Law 16:121-34. Quill, Timothy E. 1991. "Death with Dignity-A Case of Individualized Decision Making." New England journal ofMedicine 324 (10): 691-94. Quill, Timothy E., Christine K. Cassel, and Diane E. Meier. 1992. "Care of the Hopelessly Ill: Proposed Clinical Criteria for Physician-Assisted Suicide." New EnglandJournal of Medicine 327 (19): 1380-84. Read, Piers Paul. 1974. Alive. New York: Avon Books. Richardson, Ruth. 1996. "Fearful Symmetry: Corpses for Anatomy, Organs for Transplantation?" In Organ Transplantation: Meanings and Realities, ed. S. J. Youngner, R. C. Fox, and L. J. O'Connell, 66-100. Madison: Univ. ofWisconsin Press. Simpson, A. W. 1985. Cannibalism and the Common Law. Chicago: Univ. of Chicago Press. Stewart, George R. 1986. Ordeal by Hunger: The Story ofthe Donner Party. Lincoln: Univ. of Nebraska Press. Youngner, Stuart]., C. Seth Landefeld, Claudia]. Coulton, et al. 1989. '"Brain Death' and Organ Retrieval: A Cross-Sectional Survey of Knowledge and Concepts among Health Professionals." journal ofthe American Medical Association 261 (15): 2205-10. 1992. "Brain Death: A Superficial and Fragile Consensus." Archives ofNeurology 49:570-72. Youngner, Stuart J., and Robert M. Arnold. 1993. "Ethical, Psychosocial, and Public Policy Implications of Procuring Organs from Non-Heart-Beating Cadaver Donors." Journal ofthe American Medica/Association 269 (21): 2769-74. Youngner, Stuart J ., R. C. Fox, and L. J. O'Connell, ed. 1996. Organ Transplantation. Meanings and Realities. Park Ridge, Ill.: The Park Ridge Center for Health, Faith, and Ethics.

Part II Decision-Making at the End of Life

Competent Patients

[8] Medical Paternalism ALLEN BUCHANAN

I

There is evidence to show that among physicians in this country the medical paternalist model is a dominant way of conceiving the physician-patient relationship. I contend that the practice of withholding the truth from the patient or his family, a particular form of medical paternalism, is not adequately supported by the arguments advanced to justify it. Beyond the issue of telling patients the truth is the distinction between "ordinary" and "extraordinary" therapeutic measures, a distinction which, I argue, both expresses and helps to perpetuate the dominance of the medical paternalist model. There are two main types of arguments against paternalism. First are the arguments that rely upon a theory of moral rights rooted in a conception of personal autonomy. These arguments are more theoretically interesting and perhaps in the end they are the strongest arguments against paternalism. Second are the arguments that meet the paternalist on his own ground and then attempt to cut it from beneath him by showing that his arguments are defective. I shall concentrate on the second type of antipaternalist argument because I wish my arguments to have some practical effect, and I believe that this goal can best be achieved if they are directed against paternalist justifications which are actually employed by the practitioners of medical paternalism. Further, the arguments I advance require a minimum of theoretical baggage. The strength of a rights-based attack on paternalism depends ultimately upon whether a rational founda-

Death, Dying and the Ending of Life, I

126

371

Medical Paternalism

tion for the relevant theory of rights can be produced. It would be unfortunate if successful attacks on medical paternalism had to await the development and defense of a full-blown theory of moral rights. By articulating the inadequacy of the justifications which the paternalist himself advances, however, one need rely only upon those moral views to which the paternalist himself subscribes. My goal, then, is to present effective criticisms of medical paternalist practices which rely upon a minimal base of moral agreement between the paternalist and his critic. II

Paternalism is usually characterized as interference with a person's liberty of action, where the alleged justification of the interference is that it is for the good of the person whose liberty of action is thus restricted. 1 To focus exclusively on interference with liberty of action, however, is to construe paternalism too narrowly. If a government lies to the public or withholds information from it, and if the alleged justification of its policy is that it benefits the public itself, the policy may properly be called paternalistic. On the one hand, there may be a direct connection between such a policy and actual interference with the citizen's freedom to act. In order to withhold information from the public, agents of the government may physically interfere with the freedom of the press to gather, print, or distribute the news. Or government officials may misinform the public in order to restrict its freedom to perform specific acts. The police, for example, may erect signs bearing the words "Detour: Maintenance Work Ahead" to route unsuspecting motorists around the wreckage of a truck carrying nerve gas. On the other hand, the connection between withholding of information and actual interference with freedom of action may be indirect at best. To interfere with the public's freedom of information the government need not actually interfere with anyone's freedom to act-it may simply not divulge certain information. Withholding information may preclude an informed decision, and it may interfere with attempts to reach an inr. See, for example, G. Dworkin's paper "Paternalism," in S. Gorovitz et al., Moral Problems in Medicine (Englewood Cliffs, NJ; Prentice-Hall, I976), p. r8s.

Death, Dying and the Ending of Life, I

372

Philosophy & Public Affairs

formed decision, without thereby interfering with a person's freedom to decide and to act on his decision. Even if I am deprived of information which I must have if I am to make an informed decision, I may still be free to decide and to act. Granted the complexity of the relations between information and action, it seems plausible to expand the usual characterization of paternalism as follows: paternalism is interference with a person's freedom of action or freedom of information, or the deliberate dissemination of misinformation, where the alleged justification of interfering or misinforming is that it is for the good of the person who is interfered with or misinformed. The notion of freedom of information is, of course, unsatisfyingly vague, but the political examples sketched above along with the medical examples to follow will make it clearer. We can now turn to a brief consideration of evidence for the claim that medical paternalism is a widespread phenomenon in our society. III The evidence for medical paternalism is both direct and indirect. The direct evidence consists of the findings of surveys which systematically report physicians' practices concerning truth-telling and decisionmaking and of articles and discussions in which physicians and others acknowledge or defend paternalistic medical practices. The indirect evidence is more subtle. One source of indirect evidence for the pervasiveness of medical paternalist attitudes is the language we use to describe physician-patient interactions. Let us now consider some of the direct evidence. Though there are many ways of classifying cases of medical paternalism, two distinctions are especially important. We can distinguish between the cases in which the patient is legally competent and those in which the patient is legally incompetent; and between those cases in which the intended beneficiary of paternalism is the patient himself and those in which the intended beneficiary is the patient's guardian or one or more members of the patient's family. The first distinction classifies cases according to the legal status of the patient, the second according to the object of paternalism. A striking revelation of medical paternalism in dealings with legally competent adults is found in Donald Oken's essay, ''What to Tell

127

Death, Dying and the Ending of Life, I

128

Medical Paternalism

373

Cancer Patients: A Study of Medical Attitudes.'' 2 The chief conclusion of this study of internists, surgeons, and generalists is that", .. there is a strong and general tendency to withhold" from the patient the information that he has cancer. Almost go percent of the total group surveyed reported that their usual policy is not to tell the patient that he has cancer. Oken also notes that "no one reported a policy of informing every patient." Further, Oken reports that some physicians falsified diagnoses. Some physicians avoid even the slightest suggestion of neoplasia and quite specifically substitute another diagnosis. Almost everyone reported resorting to such falsification on at least a few occasions, most notably when the patient was in a far-advanced stage of illness at the time he was seen. 3 The physicians' justifications for withholding or falsifying diagnostic information were uniformly paternalistic. They assumed that if they told the patient he had cancer they would be depriving him of all hope and that the loss of hope would result in suicidal depression or at least in a serious worsening of the patient's condition. A recent malpractice case illustrates paternalistic withholding of information of a different sort. As in the Oken study, the object of paternalism was the patient and the patient was a legally competent adult. A bilateral thyroidectomy resulted in permanent paralysis of the patient's vocal cords. The patient's formerly healthy voice became frail and weak. The damage suit was based on the contention that by failing to tell the patient of the known risks to her voice, the physician had violated his duty to obtain informed consent for the operation. The physician's testimony is clearly paternalistic. In court the physician was asked "You didn't inform her of any dangers or risks involved? Is that right?" Over his attorney's objections, the physician responded, "Not specifically. . . . I feel that were I to point out all the complications-or even half the complica2.

In Moral Problems in Medicine, p.

in 1967.

3· Oken, p. IIJ.

112.

Oken's. study was first published

Death, Dying and the Ending of Life, I

374

Philosophy & Public Affairs

tions-many people would refuse to have anything done, and therefore would be much worse off." 1 There is also considerable evidence of medical paternalism in the treatment of legally incompetent individuals through the withholding of information from the patients or their guardians or both." The law maintains that it is the parents who are primarily responsible for decisions concerning the welfare of their minor children.u Nonetheless, physicians sometimes assume primary or even total responsibility for the most awesome and morally perplexing decisions affecting the welfare of the child. The inescapable need to make such decisions arises daily in neonate intensive care units. The most dramatic decisions are whether to initiate or not initiate, or to continue or discontinue life-sustaining therapy. Three broad types of cases are frequently discussed in recent literature. First, there are infants who are in an asphyxiated condition at birth and can be resuscitated but may suffer irreversible braindamage if they survive. Second, there are infants with Down's syndrome (mongolism) who have potentially fatal but surgically correctable congenital cardiovascular or gastrointestinal defects. Third, there are infants with spina bifida, a congenital condition in which there is an opening in the spine and which may be complicated by paralysis and hydrocephaly. New surgical techniques make it possible to close the spine and drain the fluid from the brain, but a large percentage of the infants thus treated suffer varying degrees of permanent braindamage and paralysis. A. Shaw notes that some physicians undertake the responsibility 4· Malpractice Digest (St. Paul, MN: The St. Paul Property and Liability Insurance Company, July-August 1977), p. 6. 5· It is interesting to note that according to both the usual and the expanded characterization of paternalism stated above, only a person who has certain physical and mental capacities can be an object of paternalism, since it is only when these capacities are present that it is correct to speak of interfering with that individual's freedom of action, ;misinforming him, or withholding information from him. 6. For a helpful summary, see J, A. Robertson and N. Frost, "Passive Euthanasia of Defective Newborn Infants: Legal Considerations," The Journal of Pediatrics 88, no. 5 ( 1976): 883-889.

129

Death, Dying and the Ending of Life, I

130

375

Medical Paternalism

for making decisions about life and death for defective newborns in order to relieve parents of the trauma and guilt of making a decision. He cites the following comment as an example of this position. At the end it is usually the doctor who has to decide the issue. It is ... cruel to ask the parents whether they want their child to live or die .... 7 We have already seen that the information which physicians withhold may be of at least two different sorts. In the cases studied by Oken, physicians withhold the diagnosis of cancer from their patients. In the thyroidectomy malpractice case the physician did not withhold the diagnosis but did withhold information about known risks of an operation. The growing literature on life or death decisions for defective neonates reveals more complex paternalistic praetices. Some physicians routinely exclude parents from significant participation in decision-making either by not informing the parents that certain choices can or must be made, or by describing the child's condition and the therapeutic options in such a skeletal way as to preclude genuinely informed consent. A case cited by Shaw is a clear example of a physician withholding from parents the information that there was a choice to be made. Baby A was referred to me at 22 hours of age with a diagnosis of esophageal atresia and tracheoesophageal fistula. The infant, the firstborn of a professional couple in their early thirties had obvious signs of mongolism, about which they were fully informed by the referring physician. After explaining the nature of the surgery to the distraught father, I offered him the operative consent. His pen hesitated briefly above the form and then as he signed, he muttered, "I have no choice, do I?" He didn't seem to expect an answer and I gave him none. The esophageal anomaly was corrected in routine fashion, and the infant was discharged to a state institution for the retarded without ever being seen again by either parent. 8 7· Shaw, "Dilemmas of 'Informed Consent' in Children," The New England Journal of Medicine 289, no. 17 ( 1973): 886. 8. Shaw, p. 885.

Death, Dying and the Ending of Life, I

Philosophy & Public Affairs

The following description of practices in a neonate intensive care unit at Yale illustrates how parents may be excluded because of inadequate information about the child's condition or the character of various therapeutic options. Parents routinely signed permits for operation though rarely had they seen their children's defects or had the nature of various management plans and their respective prognoses clearly explained to them. Some physicians believed that parents were too upset to understand the nature of the problems and the options for care. Since they believed informed consent had no meaning in these circumstances, they either ignored the parents or simply told them that the child needed an operation on the back as the first step in correcting several defects. As a result, parents often felt completely left out while the activities of care proceeded at a brisk pace. 9 Not every case in which a physician circumvents or overrides parental decision-making is a case of paternalism toward the parents. In ignoring the parents' primary legal responsibility for the child, the physician may not be attempting to shield the parents from the burdens of responsibility-he may simply be attempting to protect what he perceives to be the interests of the child. These examples are presented, not as conclusive evidence for the claim that paternalist practices of the sorts discussed above are widespread, but as illustrations of the practical relevance of the justifications for medical paternalism, which I shall now articulate and criticize.

IV In spite of the apparent pervasiveness of paternalistic practices in medicine, no systematic justification of them is available for scrutiny. Nonetheless, there appear to be at least three main arguments which advocates of paternalism could and sometimes do advance in justification of withholding information or misinforming the patient or his g. R. Duff and A. Campbell, "Moral and Ethical Dilemmas in the SpecialCare Nursery," The New England Journal of Medicine 289, no. 17 ( 1973): p. 893·

131

Death, Dying and the Ending of Life, I

132

377

Medical Paternalism

family. Since withholding information seems to be more commonly practiced and advocated than outright falsification, I shall consider the three arguments only as justifications of the former rather than the latter. Each of these arguments is sufficiently general to apply to each of the types of cases distinguished above. For convenience we can label these three arguments (A) the Prevention of Harm Argument, (B) the Contractual Version of the Prevention of Harm Argument, and (C) the Argument from the Inability to Understand. The Prevention of Harm Argument is disarmingly simple. It may be outlined as follows. The physician's duty-to which he is bound by the Oath of Hippocrates-is to prevent or at least to minimize harm to his patient. 2. Giving the patient information X will do great harm to him. 3· (Therefore) It is permissible for the physician to withhold information X from the patient. I.

Several things should be noted about this argument. First of all, the conclusion is much weaker than one would expect, granted the first premise. The first premise states that it is the physician's duty to prevent or minimize harm to the patient, not just that it is permissible for him to do so. However, since the weaker conclusion-that withholding information is permissible-seems more intuitively plausible than the stronger one, I shall concentrate on it. Second, the argument as it stands is invalid. From the claims that (I) the physician's duty (or right) is to prevent or minimize harm and that ( 2) giving information X will do the patient great harm, it does not follow that (3) it is permissible for the physician to withhold information X from the patient. At least one other premise is needed: ( 2') giving information X will do greater harm to the patient on balance than withholding the information will. The addition of (2') is no quibble. Once (2') is made explicit we begin to see the tremendous weight which this paternalist argument places on the physician's powers of judgment. He must not only determine that giving the information will do harm or even that it will do great harm. He must also make a complex comparative judgment. He must judge that withholding the information will result in less harm

Death, Dying and the Ending of Life, I

Philosophy & Public Affairs

on balance than divulging it. Yet neither the physicians interviewed by Oken nor those discussed by Shaw even mention this comparative judgment in their justifications of withholding information. They simply state that telling the truth will result in great harm to the patient or his family. No mention was made of the need to compare this expected harm with harm which might result from withholding the information, and no recognition of the difficulties. involved in such a comparison was reported. Consider two of the cases described above: a terminal cancer case and the thyroidectomy case. In order to justify withholding the diagnosis of terminal cancer from the patient the physician must not only determine that informing the patient would do great harm but that the harm would be greater on balance than whatever harm may result from withholding information. Since the notion of "great harm" here is vague unless a context for comparison is supplied, we can concentrate on the physician's evidence for the judgment that the harm of informing is greater on balance than the harm of withholding. Oken's study showed that the evidential basis for such comparative judgments was remarkably slender. It was the exception when a physician could report known examples of the unfavorable consequences of an approach which differed from his own. It was more common to get reports of instances in which different approaches has turned out satisfactorily. Most of the instances in which unhappy results were reported to follow a differing policy turned out to be vague accounts from which no reliable inference could be drawn. Oken then goes on to focus on the nature of the anticipated harm. It has been repeatedly asserted that disclosure is followed by fear and despondency which may progress into overt depressive illness or culminate in suicide. This was the opinion of the physicians in the present study. Quite representative was the surgeon who stated, "I would be afraid to tell and have the patient in a room with a window." When it comes to actually documenting the prevalence of such ontoward reactions, it becomes difficult to find reliable evi-

133

Death, Dying and the Ending of Life, I

134

379

Medical Paternalism

dence. Instances of depression and profound upsets came quickly to mind when the subject was raised, but no one could report more than a case or two, or a handful at most. . . . The same doctors could remember many instances in which the patient was told and seemed to do well. 10 It is not simply that these judgments of harm are made on the basis of extremely scanty evidence. The problem goes much deeper than that. To say simply that physicians base such judgments on extremely weak evidence is to overlook three important facts. First, the judgment that telling the truth would result in suicidal depression is an unqualified psychiatric generalization. So even if there were adequate evidence for this generalization or, more plausibly, for some highly qualified version of it, it is implausible to maintain that ordinary physicians are in a position to recognize and assess the evidence properly in a given case. Second, it is doubtful that psychiatric specialists are in possession of any such reliable generalization, even in qualified form. Third, the paternalist physician is simply assuming that suicide is not a rational choice for the terminally ill patient. If we attempt to apply the Prevention of Harm Argument to cases in whiCh the patient's family or guardian is the object of paternalism, other difficulties become apparent. Consider cases of withholding information from the parents of a neonate with Down's syndrome or spina bifida. The most obvious difficulty is that premise (I ) states only that the physician has a duty (or a right) to prevent or minimize harm to the patient, not to his family. If this argument is to serve as a justification of paternalism toward the infant patient's family, the advocate of paternalism must advance and support one or the other of two quite controversial premises. He must either add premise (I') or replace premise (I) with premise (I"): (I') If X is a guardian or parent of a patient Y andY is the patient of physician Z, then X is thereby a patient of physician Z as well. (I") It is the duty of the physician to prevent or minimize harm to his patient and to the guardian or family of his patient. 10.

Oken, "What to Tell Cancer Patients," pp.

II2, 113.

Death, Dying and the Ending of Life, I

Philosophy & Public Affairs

Since both the law and common sense maintain that one does not become a patient simply by being related to a patient, it seems that the best strategy for the medical paternalist is to rely on (I") rather than on (I'). Reliance on (I"), however, only weakens the case for medical paternalism toward parents of defective neonates. For now the medical paternalist must show that he has adequate evidence for psychiatric predictions the complexity of which taxes the imagination. He must first determine all the relevant effects of telling the truth, not just on the parents themselves, but on siblings as well, since whatever anguish or guilt the parents will allegedly feel may have significant effects on their other children. Next he must ascertain the ways in which these siblings-both as individuals and as a peer group-will respond to the predicted anguish and guilt of their parents. Then the physician must determine how the siblings will respond to each other. Next he must consider the possible responses of the parents to the responses of the children. And, of course, once he has accomplished all of this, the physician must look at the other side of the question. He must consider the possible harmful effects of withholding information from patients or of preventing them from taking an active part in decision-making. The conscientious paternalist must consider not only the burdens which the exercise of responsibility will allegedly place upon the parents, and indirectly upon their children, but also the burdens of guilt, self-doubt, and shame which may result from the parents' recognition that they have abdicated their responsibility. In predicting whether telling the truth or withholding information will cause the least harm for the family as a whole, the physician must first make intrapersonal comparisons of harm and benefit for each member of the family, if the information is divulged. Then he must somehow coalesce these various intrapersonal net harm judgments into an estimate of the total net harm which divulging the information will do to the family as a whole. Then he must make similar intrapersonal and interpersonal net harm judgments about the results of not telling the truth. Finally he must compare these totals and determine which course of action will minimize harm to the family as a whole. Though the problems of achieving defensible predictions of harm

135

136

Death, Dying and the Ending of Life, I

Medical Paternalism

as a basis for paternalism are clearest in the case of defective neonates, they are in no way peculiar to those cases. Consider the case of a person with terminal cancer. To eliminate the complication of interpersonal net harm comparisons, let us suppose that this person has no relatives and is himself legally competent. Suppose that the physician withholds information of the diagnosis because he believes that knowledge of the truth would be more harmful than withholding the truth. I have already indicated that even if we view this judgment of comparative harm as a purely clinical judgment-more specifically a clinical psychiatric judgment-it is difficult to see how the physician could be in a position to make it. But it is crucial to note that the notions of harm and benefit appropriate to these deliberations are not exclusively clinical notions, whether psychiatric or otherwise. In taking it upon himself to determine what will be most beneficial or least harmful 'to this patient the physician is not simply making ill-founded medical judgments which someday might be confirmed by psychiatric research. He is making moral evaluations of the most basic and problematic kind. The physician must determine whether it will be better for the patient to live his remaining days in the knowledge that his days are few or to live in ignorance of his fate. But again, this is a gross simplification: it assumes that the physician's attempt to deceive the patient will be successful. E. Ki.ibler-Ross claims that in many, if not most, cases the terminally ill patient will guess or learn his fate whether the physician withholds the diagnosis from him or not. 11 Possible harm resulting from the patient's loss of confidence in the physician or from a state of uncertainty over his prospects must be taken into account. Let us set aside this important complication and try to appreciate what sorts of factors would have to be taken into account in a wellfounded judgment that the remainder of a person's life would be better for that person if he did not know that he had a terminal illness than if he did. Such a judgment would have to be founded on a profound knowledge of the most intimate details of the patient's life history, his characteristic ways of coping with personal crises, his personal and u.- Kiibler-Ross, excerpts from Death and Dying, quoted in Moral Problems in Medicine, p. 122.

Death, Dying and the Ending of Life, I

Philosophy & Public Affairs

vocational commitments and aspirations, his feelings of obligation toward others, and his attitude toward the completeness or incompleteness of his experience. In a society in which the personal physician was an intimate friend who shared the experience of families under his care, it would be somewhat more plausible to claim that the physician might possess such knowledge. Under the present conditions of highly impersonal specialist medical practice it is quite a different matter. Yet even if the physician could claim such intimate personal knowledge, this would not suffice. For he must not only predict, but also evaluate. On the basis of an intimate knowledge of the patient as a person, he must determine which outcome would be best for that person. It is crucial to emphasize that the question which the physician must pose and answer is whether ignorance or knowledge will make possible a life that is better for the patient himself. The physician must be careful not to confuse this question with the question of whether ignorance or knowledge would make for a better life for the physician if the physician were terminally ill. Nor must he confuse it with the question of whether the patient's life would be a better life-a life more valuable to others or to society-if it ended in ignorance rather than in truth. The question, rather, is whether it would be better for the patient himself to know or not to know his fate. To judge that a certain ending of a life would be best for the person whose life it is, is to view that life as a unified process of development and to conclude that that ending is a fitting completion for that process. To view a human life as a unified process of development, however, is to view it selectively. Certain events or patterns of conduct are singled out as especially significant or valuable. To ascertain the best completion of a person's life for that person, then, is to make the most fundamental judgments about the value of that person's activities, aspirations, and experiences. It might be replied that we do make such value judgments when we decide to end the physiologic life of a permanently comatose individual. In such cases we do make value judgments, but they are not judgments of this sort. On the contrary, we believe that since this individual's experience has ended, his life-process is already completed. When the decision to withhold information of impending death is

137

138

Death, Dying and the Ending of Life, I

Medical Paternalism

understood for what it is, it is difficult to see how anyone could presume to make it. My conjecture is that physicians are tempted to make these decisions in part because of a failure to reflect upon the disparity between two quite different kinds of judgments about what will harm or benefit the patient. Judgments of the first sort fall within the physician's competence as a highly trained medical expert. There is nothing in the physician's training which qualifie~ him to make judgments of the second sort-to evaluate another human being's life as a whole. Further, once the complexity of these judgments is appreciated and once their evaluative character is understood, it is implausible to hold that the physician is in a better position to make them than the patient or his family. The failure to ask what sorts of harm/benefit judgments may properly be made by the physician in his capacity as a physician is a fundamental feature of medical paternalism. There is a more sophisticated version of the attempt to justify withholding of information in order to minimize harm to the patient or his family. This is the Contract Version of the Prevention of Harm Argument. The idea is that the physician-patient relationship is contractual and that the terms of this contract are such that the patient authorizes the physician to minimize harm to the patient (or his family) by whatever means he, the physician, deems necessary. Thus if the physician believes that the best way to minimize harm to the patient is to withhold information from him, he may do so without thereby wronging the patient. To wrong the patient the physician would either have to do something he was not authorized to do or fail to do something it was his duty to do and which was in his power to do. But in withholding information from the patient he is doing just what he is authorized to do. So he does the patient no wrong. First of all, it should be noted that this version is vulnerable to the same objections just raised against the non-contractual Argument from the Prevention of Harm. The most serious of these is that in the cases of paternalism under discussion it is very doubtful that the physician will or even could possess the psychiatric and moral knowledge required for a well-founded judgment about what will be least harmful to the patient. In addition, the Contract Version is vulnerable to other objections. Consider the claim that the patient-physician relationship is a contract in which the patient authorizes the physician to

Death, Dying and the Ending of Life, I

Philosophy & Public Affairs

prevent or minimize harm by whatever means the physician deems necessary, including the withholding of information. This claim could be interpreted in either of two ways: as a descriptive generalization about the way physicians and patients actually understand their relationship or as a normative claim about the way the physician-patient relationship should be viewed or may be viewed. As a descriptive generalization it is certainly implausible-there are many people who do not believe they have authorized their physician to withhold the truth from them, and the legal doctrine of informed consent supports their view. Let us suppose for a moment that some people do view their relationship to their physician as including such an authorization and that there is nothing morally wrong with such a contract so long as both parties entered into it voluntarily and in full knowledge of the terms of the agreement. Surely the fact that some people are willing to authorize physicians to withhold information from them would not justify the physician in acting toward other patients as if they had done so. The physician can only justify withholding information from a particular patient if this sort of contract was entered into freely and in full knowledge by this patient. What, then, is the physician to do? Surely he cannot simply assume that all of his patients have authorized him to withhold the truth if he deems it necessary. Yet if in each case he inquires as to whether the patient wishes to make such an authorization, he will defeat the purpose of the authorization by undermining the patient's trust. There is, however, a more serious difficulty. Even the more extreme advocates of medical paternalism must agree that there are some limits on the contractual relationship between physician and patient. Hence the obligations of each party are conditional upon the other party's observing the limits of the contract. The law, the medical profession, and the general public generally recognize that there are such limits. For example, the patient may refuse to undergo a certain treatment, he may seek a second opinion, or he may terminate the relationship altogether. Moreover, it is acknowledged that to decide to do any of these things the patient may-indeed perhaps must-rely on his own judgment. If he is conscientious he will make such decisions on

139

140

Death, Dying and the Ending of Life, 1

Medical Paternalism

consideration of whether the physician is doing a reasonable job of rendering the services for which he was hired. There are general constraints on how those services may be rendered. If the treatment is unreasonably slow, if the physician's technique is patently sloppy, or if he employs legally questionable methods, the patient may rightly conclude that the physician has not lived up to the implicit terms of the agreement and terminate the relationship. There are also more special constraints on the contract stemming from the special nature of the problem which led the patient to seek the physician's services in the first place. If you go to a physician for treatment of a skin condition, but he ignores that problem and sets about trying to convince you to have cosmetic nose surgery, you may rightly terminate the relationship. These general and special constraints are limits on the agreement from the patient's point of view. Now once it is admitted that there are any such terms-that the contract does have some limits and that the patient has the right to terminate the relationship if these limits are not observed by the physician-it must also be admitted that the patient must be in a position to discover whether those limits are being observed. But if the patient were to authorize the physician to withhold information, he might deprive himself of information which is relevant to determining whether the physician has observed the limits of the agreement. I am not concerned to argue that authorizing a physician to withhold information is logically incompatible with the contract being conditional. My point, rather, is that to make such an authorization would show either that (a) one did not view the contract as being conditional or that (b) one did not take seriously the possibility that the conditions of the contract might be violated or that (c) one simply did not care whether the conditions were violated. Since it is unreasonable to expect a patient to make an unconditional contract or to ignore the possibility that conditions of the contract will be violated, and since one typically does care whether these conditions are observed, it is unreasonable to authorize the physician to withhold information when he sees fit. The Contract Version of the Argument from the Prevention of Harm, then, does not appear to be much of an improvement over its simpler predecessor.

Death, Dying and the Ending of Life, I

Philosophy & Public Affairs

There is one paternalist argument in favor of withholding of information which remains to be considered. This may be called the Argument from the Inability to Understand. The main premise is that the physician is justified in withholding information when the patient or his family is unable to understand the information. This argument is often used to justify paternalistic policies toward parents of defective infants in neonate intensive care units. The idea is that either their lack of intelligence or their excited emotional condition prevents parents from giving informed consent because they are incapable of being adequately informed. In such cases, it is said, "the doctrine of informed consent does not apply." 12 This argument is also vulnerable to several objections. First, it too relies upon dubious and extremely broad psychological generalizations -in this case psychological generalizations about the cognitive powers of parents of defective neonates. Second, and more importantly, it ignores the crucial question of the character of the institutional context in which parents find themselves. To the extent that paternalist attitudes shape medical institutions, this bleak estimate of the parental capacity for comprehension and rational decision tends to be a self-fulfilling prophecy. In an institution in which parents routinely sign operation permits without even having seen their newborn infants and without having the nature of the therapeutic options clearly explained to them, parents may indeed be incapable of understanding the little that they are told. Third, it is a mistake to maintain that the legal duty to seek informed consent applies only where the physician can succeed in adequately informing parents. The doctor does not and cannot have a duty to make sure that all the information he conveys is understood by those to whom he conveys it. His duty is to make a reasonable effort to be understood. 13 Fourth, it is important to ask exactly why it is so important not to tell parents information which they allegedly will not understand. If the reason is that a parental decision based on inadequate understanding will be a decision that is harmful to the infant, then the 12.

Duff and Campbell, "Moral and Ethical Dilemmas," p. 893.

13. I would like to thank John Dolan for clarifying this point.

141

Death, Dying and the Ending of Life, 1

142

Medical Paternalism

Argument from the Inability to Understand is not an argument for paternalism toward parents. So if this argument is to provide a justification for withholding information from parents for their benefit, then the claim must be that their failure to understand will somehow be harmful to them. But why should this be so? If the idea is that the parents will not only fail to understand but become distressed because they realize that they do not understand, then the Argument from the Inability to Understand turns out not to be a new argument at all. Instead, it is just a restatement of the Argument from the Prevention of Harm examined above-and is vulnerable to the same objections. I conclude that none of the three justifications examined provide adequate support for the paternalist practices under consideration. If adequate justification is to be found, the advocate of medical paternalism must marshal more powerful arguments.

v So far I have examined several specific medical paternalist practices and criticized some general arguments offered in their behalf. Medical paternalism, however, goes much deeper than the specific practices themselves. For this reason I have spoken of "the medical paternalist model," emphasizing that what is at issue is a paradigm, a way of conceiving the physician-patient relationship. Indirect evidence for the pervasiveness of this model is to be found in the very words we use to describe physicians, patients, and their interactions. Simply by way of illustration, I will now examine one widely used distinction which expresses and helps perpetuate the paternalist model: the distinction between "ordinary" and "extraordinary" therapeutic measures. Many physicians, theologians, ethicists, and judges have relied on this distinction since Pius XII employed it in an address on "The Prolongation of Life" in 1958. In reply to questions concerning conditions under which physicians may discontinue or refrain from initiating the use of artificial respiration devices, Pius first noted that physicians are duty-bound "to take the necessary treatment for the preservation of life and health." He then distinguished between "ordinary" and "extraordinary" means.

Death, Dying and the Ending of Life, I

Philosophy & Public Affairs

But normally one is held to use only ordinary means-according to circumstances of persons, places, times, and culture-means that do not involve any grave burden for oneself or another. 14 Though he is not entirely explicit about this, Pius assumes that it is the right of the physician to determine what will count as "ordinary" or "extraordinary" means in any particular case. In the context of the issue of when a highly trained specialist is to employ sophisticated life-support equipment, it is natural to assume that the distinction between "ordinary" and "extraordinary" means is a distinction between higher and lower degrees of technological sophistication. The Pope's unargued assumption that the medical specialist is to determine what counts as "ordinary" or "extraordinary" reinforces a technological interpretation of the distinction. After all, if the distinction is a technological one, then it is natural to assume that it is the physician who should determine its application since it is he who possesses the requisite technical expertise. In my discussions with physicians, nurses, and hospital administrators I have observed that they tend to treat the distinction as a technological one and then to argue that since it is a technological distinction the physician is the one who should determine in any particular case whether a procedure would involve "ordinary" or "extraordinary" means. 15 Notice, however, that even though Pius introduced the distinction in the context of the proper use of sophisticated technical devices and even though he assumed that it was to be applied by those who possess the technical skills to use such equipment, it is quite clear that the distinction he explicitly introduced is not itself a technological distinction. Recall that he defines "ordinary" means as those which "do not involve any grave burden for oneself or another." "Extraordinary" means, then, would be those which do involve a grave burden for oneself or for another. If what counts as "extraordinary" measures depended only upon 14. Pius XII, "The Prolongation of Life," in Reiser et al., Ethics in Medicine, (Cambridge, MA: MIT Press, 1977), pp. sox-so4. 15. These discussions occurred in the course of my work as a member of committee which drafted ethical guidelines for Children's Hospital of Minneapolis.

143

144

Death, Dying and the Ending of Life, 1

Medical Paternalism

what would constitute a "grave burden" to the patient himself, it might be easier to preserve the illusion that the decision is an exercise of medical expertise. But once the evaluation of burdens is extended to the patient's family it becomes obvious that the judgment that a certain therapy would be "extraordinary'' is not a technological or even a clinical, but rather a moral decision. And it is a moral decision regardless of whether the evaluation is made from the perspective of the patient's own values and preferences or from that of the physician. Even if one is to evaluate only the burdens for the patient himself, however, it is implausible to maintain that the application of the distinction is an exercise of technological or clinical judgment. For as soon as we ask what would result in "grave burdens" for the patient, we are immediately confronted with the task of making moral distinctions and moral evaluations concerning the quality of the patient's life and his interests as a person. When pressed for an explanation of how physicians actually apply the distinction between "ordinary" and "extraordinary'' therapeutic measures, the director of a neonate intensive care unit explained to me that what counts as "ordinary" or "extraordinary" differs in "different contexts." Surgical correction of a congenital gastrointestinal blockage in the case of an otherwise normal infant would be considered an "ordinary" measure. But the same operation on an infant with Downs' syndrome would be considered extraordinary. I am not concerned here to criticize the moral decision to refrain from aggressive surgical treatment of infants with Down's syndrome. My purpose in citing this example is simply to point out that this decision is a moral decision and that the use of the distinction between "ordinary" and "extraordinary" measures does nothing to help one make the decision. The use of the distinction does accomplish something though: it obscures the fact that the decision is a moral decision. Even worse, it is likely to lead one to mistake a very controversial moral decision for a "value-free" technological or clinical decision. More importantly, to even suggest that a complex moral judgment is a clinical or technological judgment is to prejudice the issue of who has the right to decide whether life-sustaining measures are to be initiated or continued. Once controversial moral decisions are misperceived as clinical or technological decisions it becomes much

Death, Dying and the Ending of Life, I

390

Philosophy & Public Affairs

easier for the medical paternalist to use the three arguments examined above to justify the withholding of information. For once it is conceded that his medical expertise gives the physician the right to make certain decisions, he can then argue that he may withhold information where this is necessary for the effective exercise of this right. By disguising complex moral judgments as medical judgments, then, the "ordinary /extraordinary" distinction reinforces medical paternalism.

VI In this paper I have attempted to articulate and challenge some basic features of the medical paternalist model of the physician-patient relationship. I have also given an indication of the powerful influence this model exerts on medical practice and on ways of talking and thinking about medical treatment. There are now signs that medical paternalism is beginning to be challenged from within the medical profession itself. 16 This, I believe, is all to the good. So far, however, challenges have been fragmentary and unsystematic. If they are to be theoretically and practically fruitful they must be grounded in a systematic understanding of what medical paternalism is and in a critical examination of justifications for medical paternalist practices. The present paper is an attempt to begin the task of such a systematic critique. 16. See, for example, A. Waldman, "Medical Ethics and the Hopelessly Ill Child," The Journal of Pediatrics 88, no. 5 ( 1976): 8go-8g2. I would like to thank Rolf Sartorius and the editors of Philosophy & Public Affairs for several helpful comments on an earlier draft of this paper.

145

[9] SPECIAL ARTICLE ARROGANCE* FRANZ

J.

INGELFINGER,

this George W. Gay Lecture, specifically desigI Nnated as "upon medical ethics," I shall focus on

three issues. The first, an example of intergroup tensions, deals with the common accusation that bioscientists are arrogant, i.e., that they are presumptuous and overweening in their attitudes, decisions, and goals; that they exhibit, in the fashionable noun ofthe day, hubris. I shall argue that the bioscientist may be arrogant, but no more so than any other group and perhaps just a little bit less so. The second issue bears on the personal encounter between physician and patient: Is it marked by authoritarianism, paternalism, and domination? My answer is not only "yes" but also that a certain meas.ure of these characteristics is essential to good medical care. In fact, if you agree that the physician's primary function is to make the patient feel better, a certain amount of authoritarianism, paternalism, and domination are the essence of the physician's effectiveness.

Thirdly, I shall maintain that many physicians are

*Based on the George W. Gay Lecture, delivered by Dr. lngelfinger at Harvard Medical School on May 5, 1977, shortly bef~n: he retired as Editor of the Journal. Dr. lngelfinger died on March 26, 1980, leaving his lecture notes still only partially edited. He evidently could not persuade himself t~at his lecture deserved publication and had allowed the unfinished ma.nuscnpt to languish in his files until his death. However, a recent r~readmg convinced us that this was vintage Ingelfinger and therefore emmently worth presenting to our readers. With the permission of his family, we present an abridged and slightly edited version of Dr. lngelfinger's remarks. The Editors

M.D.

indeed arrogant in their behavior toward patients, but in a way that is not even specifically identified by any of the dictionary definitions of the word "arrogance." Although no learned vocation is exempt from the accusation, the professional group most often belabored for arrogance is that which uses advanced and complex technology in its thinking and doing. It is the scientist, whether in physics or in molecular biology, or even the parascientist in medicine, who is seen as making policy decisions motivated by self-interest and acting with a total disregard for broad human needs. Almost reflexly the hoary Clemenceauism is trotted out that war must be not left to the generals. Because of this societal apprehension, decision-making bodies are being created with the express purpose of limiting the influence of specifically those scientists who possess expertise relevant to the questions that n:ust be answered. Physicians are well acquamted w1th the provisions of Public Law 93-641, enacted in 1974, which establishes a network of health-systems agencies. The majority of members appointed to bodies responsible for planning and implementing this syste.m, such as area governing boards and sub-area counCils, must consist of so-called consumers rather than providers of health care. The same principle is being applied to even more momentous issues, such as whether arbitrary limits should be placed on mankind's search for knowledge. Thus, many maintain that the current frenetic argument about further research in recombi-

148

Death, Dying and the Ending of Life, 1

1508

THE NEW ENGLAND JOURNAL OF MEDICINE

nan! DNA should be resolved by groups in which the representatives of science are not only restricted but eliminated altogether. Indisputable reasons exist why decisions affecting society as a whole should be made by persons broadly representative of that society and not by those who might be biased, perhaps by reasons of personal gain or by entirely honorable but nevertheless unbalanced and enthusiastic intellectual commitment to a discipline. Indeed, the principle is basic in most democratic societies. Whether in Britain or the United States, it is Parliament or Congress after all, that decides how much money should be spent on various programs of biomedical research and public health. The justness of society's control is not the issue. Nor is it deniable that many specialists in the biosciences entertain a hemianoptic view of the world. It is understandable, furthermore, that the bulk of biomedical scientists and parascientists will as a group make decisions grossly influenced by their own skills andresources. Under many a circumstance the surgeon will cut when the internist will use antibiotics. My concern is that the words "arrogant scientist" imply that the ignorant, whether lay or professional, have no such faults. But the arrogance of ignorance can be as devastating as the arrogance of expertise. Yet the arrogance of ignorance or the arrogance of the antiscientific or anti-intellectual activist is hardly ever mentioned, although it may be as flagrant as that of the most doctrinaire member of the conventional scientific establishment. Let us start with an example of relatively low emotional intensity, that of diet. Over the years various dietary practices have been endorsed or condemned by physicians, and the popularity with the public of this or that dietary fad is an ancient but continuing phenomenon. For years doctors and public alike were convinced that the consumption of some foods aggravated or even caused peptic ulcers of the stomach or duodenum, or that other foods had salubrious effects. Harmful staples were identified as spicy, coarse, and strongly colored; those designated as beneficial were bland, smooth, and colorless. White protein such as chicken and fish was permissible, but red meats such as beef and lamb forbidden. One school of thought indeed, both medical and nonmedical - exercised reductionism to its ultimate: the ulcer patient was simply told to eat a white diet. Farina, potatoes, and milk, I supppose, were intended, but I always wondered if the literal minded might not also include horseradish. The influence of superstition, mysticism, and religion on dietary beliefs is of course strong. In 1910, for example, a popular fad was Fletcherism, the practice of chewing each bite of food intensely - perhaps 30 times. The originator of the cult, Horace Fletcher, always wore white suits to emphasize the spiritual purity of his obsessive mastication. Perhaps similar occult powers enhance the appeal of the white diet in the treatment of peptic ulcer.

Dec. 25, 1980

Nowadays we still know little about the cause and cure of peptic ulcer. Indeed, many excellent scientific studies employing the resources of modern technology have, on the one hand, increased knowledge about the nature of peptic ulcer but, on the other hanq, increased the expert's recognition of how little he knows about the causes of this disorder. More knowledge has paradoxically highlighted non-knowledge. Correspondingly, the value of dietary regimens imposed, in retrospect with somewhat arbitrary ignorance, is questioned by the skepticism of scientific expertise. Currently another dietary vogue is attracting the faithful: the overly refined, low-roughage foods of the industrialized West are said to cause diverticulosis and perhaps cancer of the large bowel, hemorrhoids, varicose veins, and calcifications of pelvic veins. Perhaps these processed foods are harmful, but the categorical endorsement of high-roughage diets impresses me as another example of an arbitrary, authoritarian pronouncement based on half-knowledge and on an unacknowledged ignorance of overall effects - in other words, somewhat arrogant. In 1952, long before Burkitt and others became interested in the effects of diet on the pathogenesis of colonic diverticula, a pathologist in Lima, Peru told me that at autopsy he could always differentiate the colon of the Spanish descendant from that of the Andean Indian. The Spanish descendant's colon, to be sure, might have diverticula, but the Indian's did not. On the other hand, the Indian characteristically had a very long colon, particularly a long and convoluted sigmoid. Because of this characteristic, Indians frequently suffered fatal volvulus of this organ. Albert Schweitzer recorded the huge prevalence of hernia in the patients he saw. So high-roughage diets may have their unfavorable as well as favorable consequences, but one would never know it from the denunciation one hears of low-roughage diets. I even suspect that mystic and moralistic forces are at work here, for obviously a high-roughage diet is a return to the natural ctmditions of the noble savage and the endurance of hardships, whereas the refined diet indicates an effete and self-indulgent luxury. My dietary homilies are intended to illustrate the multifarious societal forces that underlie questionable policy decisions - in this case, policies about what to eat. The public, the doctor, the professor of physiology participated in decisions that were spun arbitrarily out of flimsy strands of evidence. Their reasoning depended on what Alvin Weinberg has called trans-science. "Many of the issues," Weinberg wrote in 1972, * "that arise from the interaction between science or technology and society . . . hang on the answers to questions that can be asked of science and yet which cannot be answered by science. I propose the term trans-scientific for these questions since, though they *Weinberg AM. Science & trans-science. In: Civilization and science: in conflict on collaboration?: a Ciba Foundation Symposium. Amsterdam: Elsevier Press, 1972:105-22.

Death, Dying and the Ending of Life, I Vol. 303

No. 26

THE NEW ENGLAND JOURNAL OF MEDICINE

may arise in or around science, and can be stated in the language of science, they are unanswerable by science - that is, they transcend science." Failure to differentiate between science and transscience, failure to recognize that a regulation or recommendation may be concocted in a vacuum of knowledge is to my mind a manifestation of the arrogance of ignorance. I must make a distinction here: I do not assert that conclusions reached in the absence of reliable fact are per se arrogant. Such conclusions unavoidably characterize politics, science, and medicine, especially when the need for action is urgent. Arrogance enters when those reaching various decisions in the absence of adequate data fail to recognize or to admit how empty their cupboard of information is. Superior scientists or doctors, I should like to believe, are always aware of how little they know. Doubt tempers arrogance, and for this reason perhaps some bioscientists might be credited with sophrosyne rather than condemned for its opposite, hubris. (I admit I never heard of sophrosyne before I prepared for this lecture, but it has all the prerequisites for becoming a stylish word.) sociologists, ethicists, and others like to speak disparagingly of what they call the doctor's authoritarianism, paternalism, or domination. Such a position, I submit, is unrealistic and untenable. The physician is a person to whom patients go because they need or think they need help. Let us assume that the physician they select is competent and compassionate. In spite of these virtues, there is usually little the physician can do physically, that is, by cutting or by a chemical manipulation, to eradicate the cause of the patient's distress. That is why epidemiologists keep pointing out - and the recent book by Thomas McKeown (The Role of Medicine: Dream, Mirage, or Nemesis?) is an outstanding example - that the physician's intervention has done little to prolong life or eliminate serious morbidity. The figure generally quoted - although it may be an arrogant figure in that the substantiating data are fragmentary- is that 90 per cent of the visits by patients to doctors are caused by conditions that are either self-limited or beyond the capabilities of medicine. In other words, if we assume that physicians do make patients feel better most of the time, it is chiefly because the physician can reassure the patient or give medication that is mildly palliative. Even an operation may once in a while make a patient feel better, although it does not prolong his life or eradicate the source of his problems. If the physician is to be effective in alleviating the patient's complaints by such intangible means, it follows that the patient has to believe in the physician, that he has confidence in his advice and reassurance, and in his selection of a pill that is helpful (though not curative of the basic disorder). Intrinsic to such a belief is the patient's conviction that his physician not only can be trusted but also has some special know!-

149 1509

edge that the patient does not possess. He needs, if the treatment is to succeed, a physician whom he invests with authoritative experience and competence. He needs a physician from whom he will accept some domination. If I am going to give up eating eggs for the rest of my life, I must be convinced, as an ovophile, that a higher authority than I will influence my eating habits. I do not want to be in the position of a shopper at the Casbah who negotiates and haggles with the physician about what is best. I want to believe that my physician is acting under higher moral principles and intellectual powers than a used-car dealer. I'll go further than that. A physician who merely spreads an array of vendibles in front of the patient and then says, "Go ahead and choose, it's your life," is guilty of shirking his duty, if not of malpractice. The physician, to be sure, should list the alternatives and describe their pros and cons but then, instead of asking the patient to make the choice, the physician should recommend a specific course of action. He must take the responsibility, not shift it onto the shoulders of the patient. The patient may then refuse the recommendation, which is perfectly acceptable, but the physician who would not use his training and experience to recommend the specific action to a patient- or in some cases frankly admit "I don't know" - does not warrant the somewhat tarnished but still distinguished title of doctor. Although I have subscribed for some time to the principle that the physician must be authoritarian and paternalistic to some degree, my experience as a patient has substantiated that belief in the strongest way possible. If you will forgive me for being both anecdotal and personal, let me tell you how the lack of authoritarian decision brought agony to me and my family. About a year and a half ago it was discovered that I had an adenocarcinoma, a glandular cancer, sitting astride the gastroesophageal junction. Ironically, this had been an area of the gut to which I had paid much attention in my professional career as a clinical investigator and consultant; therefore, I can hardly imagine a more informed patient. The need for surgery was indisputable if I hoped to continue to be able to swallow. But after a successful operation my real dilemmas began. The surgeon had found no visible evidence that the cancer had spread. But this proved nothing, because cancers can spread to form tiny nests of cancer elsewhere - micrometastases. The current medical practice is to assume that a patient who has had an operation for any of a variety of cancers (including the type I had) should also be given prophylactic treatment in an effort to eradicate the micrometastases before they enlarge. For this purpose both chemotherapy and radiotherapy are being used extensively. So one question was: Should I have chemotherapy, with all its side effects? And if chemotherapy, what kind? Even more debatable was the question of whether I should have radiotherapy.

150

Death, Dying and the Ending of Life, 1

1510

THE NEW ENGLAND JOURNAL OF MEDICINE

There is no generally acceptable evidence that residual nests of adenocarcinoma cells will respond. In addition, radiation would involve for me a number of complications, such as fibrosis of the lungs, and the possibility of a host of less frequent but nevertheless serious side effects. At that point I received from physician friends throughout the country a barrage of well-intentioned but contradictory advice. The question of prophylactic radiotherapy was particularly moot. As a result, not only I but my wife, my son and daughter-in-law (both doctors), and other family members became increasingly confused and emotionally distraught. Finally, when the pangs of indecision had become nearly intolerable, one wise physician friend said, "What you need is a doctor." He was telling me to forget the information I already had and the information I was receiving from many quarters, and to seek instead a person who would dominate, who would tell me what to do, who would in a paternalistic manner assume responsibility for my care. When that excellent advice was followed, my family and I sensed immediate and immense relief. The incapacity of enervating worry was dispelled, and I could return to my usual anxieties, such as deciding on the fate of manuscripts or giving lectures like this. If arrogance in the sense of paternalism and dominance is an ingredient of beneficial medical care, these qualities have to be used appropriately. To the extent that paternalism and dominance are infected by some of the other meanings of arrogance, a physician's conduct with patients is correspondingly worsened. Thus, if his paternalism is accentuated by insolence, vanity, arbitrariness, or a lack of empathy, the care he attempts to provide his patients is nullified. In other words, a physician can be beneficially arrogant, or he can be destructively arrogant. physicians as a class, I suspect, are probably no more vain or insolent than any other people. Some are presumptuous and condescending, others self-effacing and sympathetic. Although arrogance in some of its more nefarious meanings - vanity, insolence, and ruthlessness, for example, -cannot, I believe, be identified as a general characteristic of the medical profession, the profession as a whole is affected by a brand of arrogance subsumed under lack of empathy. Doctors for various reasons find it difficult to put themselves in the patient's place; they do not sufficiently appreciate, or perhaps do not have the time to appreciate, how the patient feels and how he reacts to the medical information and procedures to which he is exposed. The problem, discussed to some extent in the April, 1977 issue of the Annals of Internal Medicine, has two components - one intellectual and to some degree ameliorable, and the other emotional and requiring for its correction the utmost in medical art. The first component is simply a matter of language. The argot

Dec. 25, 1980

of the physician has not only been ridiculed extensively but criticized as one of the devices used by the medical profession to maintain its mystique. This explanation may have held true in Moliere's time, but I doubt its validity today. The profession's addiction to its vocational jargon is probably a matter of habit, difficult to dispel because of the physician's training, his incessant intercourse with other doctors; his reading of The New England Journal of Medicine, and his relative ineptitude with ordinary English. Whatever the reasons, physicians tend to use terms that laymen either do not understand or misinterpret. Even one of the best communicators with the public I know, Dr. Timothy Johnson, may on occasion allow the guests on his popular program, "House Call," to lapse into language more appropriate for the medical amphitheatre than the public forum. "Endometrial carcinoma" is much easier for the medical tongue than "cancer of the lining of the womb." Many patients, to be sure, are acquainted with medical terms and use them, and it has been proposed that teaching the patient about medicine - enriching his vocabulary rather than profaning that of the doctor - might improve communication between the two parties. But do patients really understand medical words? We physicians certair,tly have problems with the principles and jargon of other skills, as is evident in the reproving letters I receive when we publish articles dealing with the arcane statistics or economics of medicine. Similarly, even if the patient uses words such as "myocardial infarction," does he really appreciate the spectrum of pathologic, diagnostic, prognostic, and therapeutic implications that this common expression conveys to the physician? All but the most medically sophisticated patients need to be informed, I suspect in nontechnical terms, and the physician who ignores this obligation is guilty of a form of arrogance. Even if a physician takes pains to use appropriate language, he may still lack empathy if he is not acutely sensitive to the emotional state of the patient seeking consultation. Distraught by anxiety, fear, and perhaps suspicion, the patient hears the sound but not the meaning of words; reassurances that cancer is an unlikely diagnosis, and a barrage of tests to prove this point, may convince the patient that the opposite is true. "We shall not need another operation" is recorded in the patient's mind as "another operation." Advice that antihypertensive drugs or insulin are in order, possibly for a lifetime, may give the patient the idea of incurability. Even advice on smoking and overeating may elicit negative instead of positive results in the susceptible. Perhaps one of the most flagrant examples of nonempathic arrogance today - an example not confined to the medical profession - is the pervasive idea that the failure of medical ministrations is the patient's fault. If he does not follow instructions and appears to disregard words of medical wisdom, the patient is labeled as noncompliant - another word in

Death, Dying and the Ending of Life, I Vol. 303

No. 26

THE NEW ENGLAND JOURNAL OF MEDICINE

vogue. Blaming the victim is currently a popular excuse for therapeutic failure, but to me it smacks of arrogance. It is a doctor's obligation, by explanation and persuasion, to get the patient to take his medication as prescribed. If the patient fails to do so, the blame is often as much the physician's as the patient's. It is of course easy enough for a speaker to moralize and to demand more empathy on the part of the physician. But what practical resources are available to the physician who is aware that the patient's emotional state may color and distort ordinary conversation? Getting to know the patient, his convictions and his problems, and the attitudes of his family, will of course help, but in these days of group practices, ancillary help, specialization, and mobile populations, "getting to know the patient" may be as difficult as containing medical costs. Currently popular measures to enhance medical efficiency also do not help. If a patient - whether an expectant mother, an alcoholic with early cirrhosis, or a heavy smoker with lung cancer - is first processed through a battery of questionnaires or computer terminals, then interrogated

151 1511

and examined by ancillary personnel, and finally seen by the doctor - to be delivered, to be subjected to liver biopsy, or to undergo pulmonary resection that patient will not know the doctor, and vice versa. How can the doctor under such circumstances be aware of his patient's thoughts and emotions? Efficient medical practice, I fear, may not be empathic medical practice, and it fosters, if not arrogance, at least the appearance of arrogance. In medical school, students are told about the perplexity, anxiety, and misapprehension that may affect the patient as he enters the medical-care system, and in the clinical years the fortunate and sensitive student may learn much from talking to those assigned to his supervision. But the effects of lectures and conversations are ephemeral and are no substitute for actual experience. One might suggest, of course, that only those who have been hospitalized during their adolescent or adult years be admitted to medical school. Such a practice would not only increase the number of empathic doctors; it would also permit the whole elaborate system of medical-school admissions to be jettisoned.

[10] Depression, Competence, and the Right to Refuse Lifesaving Medical Treatment Mark D. Sullivan, M.D., Ph.D., and Stuart]. Youngner, M.D. Objective: The authors explore the possibility that psychiatrists inappropriately extend their views on suicide by the medically well to refusal of lifesaving treatment by the seriously medically ill. Method: The legal and bioethics literature on competence to refuse lifesaving treatment and the possible impact of depression on this refusal is reviewed. Results: Over thf!' past 20 years, the burden of proof concerning the mental competence of seriously medically ill patients who refuse lifesaving treatment has shifted to the persons who seek to override these refusals. However, in psychiatry a patient's desire to die is generally considered to be evidence of an impaired capacity to make decisions about lifesaving treatment. This contrast between ethical traditions is brought into clinical focus during the evaluation and treatment of medically ill patients with depression who refuse lifesaving treatment. The clinical evaluation of the effect of depression on a patient's capacity to make medical decisions is difficult for several reasons: 1) depression is easily seen as a "reasonable" response to serious medical illness, 2) depression produces more subtle distortions of decision making than delirium or psychosis (i.e., preserving the understanding of medical facts while impairing the appreciation of their personal importance), and 3) a diagnosis of major depression is neither necessary nor sufficient for determining that the patient's medical decision making is impaired. Conclusions: Depression can be diagnosed and treated in patients with serious medical illness. But after optimizing medical and psychiatric treatment and determining that the patient is competent to make medical decisions, it may be appropriate to honor the patient's desire to die. (Am J Psychiatry 1994; 151:971-978)

I

n the context of severe and unbearable medical illness, refusals of medical treatment that hasten death are generally viewed as "allowing to die" or "letting nature take its course." In medical settings, legal competence to refuse life-sustaining treatment is presumed; the burden rests with the physician to prove otherwise. In psychiatric settings, a desire to die is not only considered prima facie evidence of mental disorder and critically flawed decision-making capacity but, by its very existence, provides a justification for treatment. In this article we explore the origins of these very different clinical and ethical perspectives and the moral assumptions that underlie them. We examine in depth one clinical phenomenon in which the clash between the two perspectives comes into sharpest focus: refusal of lifesustaining treatment by patients with serious concurrent medical illness and depression. Received June 9, 1993; revision received Oct. 4, 1993; accepted Nov. 18, 1993. From the Department of Psychiatry and Behavioral Sciences, University of Washington Medical School; the Clinical Eth-

ics Program, University Hospitals of Cleveland; and the Department of Medicine, the Department of Psychiatry, and the Center for Bio-

medical Ethics, Case Western Reserve University, Cleveland. Address

reprint requests to Dr. Sullivan, Department of Psychiatry and Behavioral Sciences, RP-10, University of Washington Medical School~ Se-

attle, WA 98195.

REFUSAL OF LIFE-SUSTAINING MEDICAL TREATMENT IN THE MEDICAL SETTING

Longstanding traditions in medicine and tort law have opposed treattnent refusal that will result in the death of the patient. However, the past three decades have brought our nation to a medical, legal, and social consensus that under the right circumstances, all types of lifesustaining medical interventions may be withheld or withdrawn. The ideal circumstance is when a fully competent patient who refuses treatment after adding up the burdens and benefits, decides, all things considered, that death is the best of the available alternatives. While patients' refusal of treatment may be questioned when health professionals are concerned about impaired decision-making capacity, among the medically ill, competence to make medical decisions is presumed until proven otherwise. The burden of proof concerning impaired patient competence is on the physician. Indeed, courts and legislatures across the country have articulated a "right to die" for seriously ill medical patients. This has been done through both court cases and natural death acts. While no absolute right has been recognized, no court has rejected the right to die in a wholesale fashion (1, p. 45).

154

Death, Dying and the Ending of Life, 1

DEPRESSION AND RIGHT TO REFUSE TREATMENT

The recognition of the right to die has taken place in the wake of advances in medical technology that make it possible to prolong life beyond the prospect of return to a satisfying and functional existence. The case of Karen Quinlan was exemplary. As legal scholar Alan Meisel has stated: "Prior to the Quinlan case, there was a great reluctance in case law to permit patients to refuse treatment when that refusal would probably lead to their death. Quinlan changed that by making clear to the courts what had long been apparent in the health professions, namely, that there is an increasing range of situations in which treatment will keep patients alive but not restore them to health" (1, p. 44). Although the legal precedent set by early cases such as Quinlan was limited to patients with extreme or terminal conditions, or to those who were being kept alive by "extraordinary" means (e.g., mechanical ventilators or renal dialysis machines), more recent cases have extended the right to refuse treatment. Courts have recognized treatment refusals by competent patients whose conditions were incurable but who were not near death-for example, patients with advanced neurological diseases such as amyotrophic lateral sclerosis or patients left quadriplegic after accidents who decided that their quality of life was unacceptable. Moreover, less extraordinary life-sustaining treatments such as tube feeding and hydration have become legitimate targets for treatment refusals (2). In all of these instances, the wishes of competent persons have been considered to be definitive, not the values of health professionals or health care institutions. The court rulings that have legitimated patients' decisions to limit treatment have not framed these refusals as suicide. Although decisions to turn off ventilators or remove feeding tubes resulted in quicker deaths for patients, these decisions were almost universally characterized as allowing the patient to die or letting nature take its course. It was the disease process, not the decisions and behavior of patients, families, or health professionals, that "caused" the patient to die. Some opponents of treatment limitation have regularly branded it as suicide or killing (3). Majority opinions in court cases have authorized treatment withdrawals and characterize them as allowing the patient to die, while some minority opinions angrily brand the same behavior as suicide and killing (4). Nevertheless, a clear prevalence of ethical and legal opinion now supports the right to refuse lifesaving treatment in the medical setting. REFUSAL OF LIFE-SUSTAINING TREATMENT BY THE PSYCHIATRIC PATIENT

While there have been important developments concerning the rights of psychiatric patients to refuse treatment in some circumstances, this has not generally included the right to refuse lifesaving treatment. Patients' wishes to die are still assumed to be rooted in psycho972

pathology and are routinely characterized as suicidal on the basis of the wish (5). Although legal competence traditionally has been considered a global property of persons, there has been a trend in the law toward a more specific understanding of competence. Alan Meisel commented: "The general incompetence approach has long been the dominant one used by the courts. However, it has fallen into disflicit standards, they approve withholding care in particular cases.271 Hence, if 275. See Capron & Kass, A Statutory Definition of the Standards for Determining Hr~man Death: An Appraisal and A Proposal, 121 U. PA. L. REv. 87 (1972). But in practice such mechanisms may not provide as much protection as desired because of the intrahospit:al familiarity and informal professional norms that often make fellow physicians reluctJ.nt to criticize or override the wishes of other physicians. See STAPF REPORT OF SENATll Co:r.w. ON FINANCE, MEDICARE AND MEDICAID; PROBLEMS, ISSUES, A'ND ALTERNATIVES, 91St Cong., ISt Sess. 105-12 (1970). 276. While explicit legislative approval of the criteria would be helpful, it would not be necessary for the legislature to draft or enact the criteria. A better solution might be to constitute formally a decisionmaking body with medical, community, and lay representation along the lines of the National Commission for the Protection of Human Subjects of Biomedical and Behavioral Research, created by the National Research Service Award Act of 1947, 42 U.S.C. § 289l, and delegate to it the development of the appropriate criteria and relevant clinical indicators. To prevent obsolescence in the criteria, and to monitor the extent of nontreatment and the effectiveness of developed standazds, the body could be permanently constituted and perform similar functions with regard to other biomedical problems. Alternatively, the courts could give judicial recognition to criteria developed by authoritative medical or professional groups. In either case it is essential that the criteria be perceived as the expression of a community consensus concerning the limits of protecting human life. 277. McCormick, in arguing for a definition of personhood based on a capacity for human relationships, also argues for explicit formulation of criteria for identifying defective newborns from whom treatment may justifiably be withheld: "'Broad guidelines,' 'substi.Dtive standards: There is the middle course, and it is the task of a community broader than the medical community. A gUideline is not a slide rule that makes the decision. It is far less than that. But it is far more than the concrete decision of the parents and physician, however seriously and conscientiously this is made. It is more like a light in a room, a light that allows the individual objects to be seen in the fullness of their context. Concretely, if there are certain infants that we agree ought to be saved in spite of illness or deformity, and if there are certain infants that we agree should be allowed to die, then there is a line to be drawn. And if there is a line to be drawn, there ought to be some criteria, even if very general, for doing this. Thus, if nearly every commentator has disagreed with the Hopkins decision, should we not be able to distill from such consensus some general wisdom that will inform

Death, Dying and the Ending of Life, I

January 1975]

INVOLUNTARY EUTHANASIA

considered thought produces a consensus that, for example, profoundly retarded, nonambulatory hydrocephalies who are blind and deaf or infants who are unlikely to survive beyond a year are not owed ordinary treatment, a basis for developing criteria exists. Initially a common law approach might serve to identify certain categories of defects, with cases added or removed by their similarity to clear-cut categories. On the other hand, a common-law-like evolution may be less desirable than a forthright state· ment of the standards implicit in our considered judgments of clear-cut cases. The venture is risky, but considerable light may be shed by attempt· ing to frame standards based on minimal capacity for social interaction, personality development, consciousness, and the like. Perhaps acceptable criteria cannot be articulated, because factual vari· ations are too many and justifiable ethical judgments cannot always be rationally explained. Nevertheless, the attempt should be made because the process of developing such criteria will at least reveal whether the judg. ments by which we presently decide to withhold care are principled or ar. bitrary, and thus whether they deserve to be honored at all.

b. Specified procedures. If we grant the impossibility of articulating reliable and useful criteria for decisionmaking, the danger of arbitrary decisions by parents and physicians still may be reduced by mandating procedures that assure access to all relevant information and time for reflection. If decisionmaking could be structured so that all considerations enter into the calculus, the risk of arbitrariness by parents and physicians would lessen. Thus, if the postpartum hours are fraught with emotion, no treatment should be withheld for, say, 7 days. Alternatively, the parents' decision could be disregarded until they have one or more counseling sessions with a social worker or some other suitable adviser. Similar measures could assure exposure to the widest range of information relevant to their decision: for example, the parents expressly can be made aware of the benefits and costs of institutional care and the availability of state aid for home care. Attending physicians also can be required to follow specified procedures before giving effect to parental choice. They can be required to state in writing their reasons for withholding treatment. A post hoc review, along the lines of a tissue committee or other forms of peer review, could be estab. lished. As we proliferate ideas for assuring impartial consideration of relevant factors, we move toward a committee structure. If we can achieve those advantages without undue bureaucracy, such an approach has appeal. and guide future decisions?" McCormick, supra note 205, at 172, I73· See also Carney, Medical Standards for Deformed Newborns, Wash. Post, Mar. 20, 1974, at Axs, col. x.

399

400

Death, Dying and the Ending of Life, I

STANFORD LAW REVIEW

III.

[Vol. 27: Page 213

CoNcLUSION

The pervasive practice of withholding ordinary medical care from defective newborns demonstrates that we have embarked on a widespread program of involuntary euthanasia. This practice has not resulted from a careful consideration of public policy alternatives, nor has it been arrived at by a public or collective decision. Formal public policy, in fact, condemns the practice, and until recently, the medical profession rarely acknowledged its existence. But now, as a result of new-found technological skills and perhaps changing attitudes toward social-utility assessments of human life, the practice has come to be accepted in the interstices of medical and legal practice. Given this situation, the crucial question is whether nontreatment of defective neonates is the opening wedge in expanding involuntary euthanasia, or whether its scope and impact can be limited. How we practice involuntary ~uthanasia thus becomes as important as the practice itself. The problem of treating defective newborns may usefully be viewed as a problem of the proper limits of discretion, and is therefore amenable to traditional legal controls on discretion-rules, procedures, and review. Nonenforcement of existing criminal laws grants parents and physicians effective discretion to decide the fate of infants born with a range of defects. Their decisions inevitably reflect their perception of the child's, the family's, and perhaps society's interests. In short, they implicitly or explicitly constitute judgments as to when social costs outweigh the benefits of treatment. But these criteria are rarely articulated, and their judgments do not undergo the close scrutiny that decisions of such magnitude warrant. In nearly all cases in which the attending physician concurs in the parental decision to withhold care, neither parents nor physicians are required to justify their choice, nor is the decision reviewed by a disinterested party. The absence of due process for the infant is all the more striking given the emotional circumstances of the parental decision and the lack of publicly certified guidelines or criteria for withholding care. We thus have a situation in which interests other than the infants can dominate, and in which arbitrary and unjustified killings can and have occurred. It is highly unlikely that full enforcement of present laws can correct the imbalances of the present situation. Some prosecutors might begin to enforce the law as the practice is more widely publicized. Yet if such enforcement occurs at all, it is likely to be scattered or sporadic. Although the law clothes the defective infant with a right to life, and a corresponding duty of care from those in certain relations with him, many people think that that right ends when it conflicts with the interests of parents, the medical profession, and the infant's own potential for full development. The

Death, Dying and the Ending of Life, I

January 1975]

INVOLUNTARYEUTHANA~A

law in action is likely to reflect this view, and, if the law in theory differs, this difference probably will be ignored. If the nontreatment of defective newborns has become deeply engrained in medical practice, one can only hope that it will be confined to those cases in which the clearest and most indisputable grounds for withholding care exist. The attending physician is a partial check on parents who would unjustifiably deny treatment, even if the criminal law is not. But the peculiarities of his role, the risk of conflicting interests, and the lack of special ethical skill or training, make the physician an unreliable protector of the infant. In fact, granting the physician this authority may only foster present tendencies of the medical profession to assume decisionmaking authority over issues that are not theirs by law, training, or expertise to make. Unless physicians are to be final arbiters of all social policy and ethical issues with a medical component, they should not have such authority here. By requiring that certain procedures be followed in deciding whether parents' nontreatment decision is to be final, the exercise of discretion can be structured to maximize the possibility of providing a disinterested decisionmaker who is equipped with relevant information and is sensitive to all interests. However, enforcing these procedures may present major problems, and even if administratively feasible, undoubtedly will require a relatively unfocused judgment as to whether care can be justifiably withheld. Thus, in addition to procedures, the decisionmaking should be confined or limited by specific criteria, identifying the cases in which treatment can be justifiably withheld. Drafting criteria creates problems and difficult ethical choices, but these are no more significant than those that faced the Harvard Medical School committee that produced a definition of brain death. Rather, the problem will be in deciding whether rules such as "no anencephalies need be treated" or "no hydrocephalic can be denied treatment for this defect alone" provide justifiable differentiations in deciding the need for treatment. The use of criteria thus confronts us with the question of whether there are classes of infants who may justifiably be allowed to die when parents so choose, and whether we can openly acknowledge the criteria which inform our decisionmaking. For while we may save lives and limit discretion by formalizing decisionmaking, we risk establishing a precedent that once loosed, is not easily cabined. Given the current acceptance of involuntary euthanasia of defective newborns, further danger from formalizing the precedent appears small. Indeed, subjecting the nontreatment decision to rules and procedures may demonstrate the solemn nature of a difficult situation. Since the power to cause the death of a defective newborn is an awesome one, it is essential that such decisions be carefully confined by law.

401

[23] Toward an Ethic of Ambiguity by JOHN D. ARRAS

W.en

it comes to matters of life and death, our society prefers procedure to substance. Instead of asking, "What is the right thing to do?" we ask, "Who should decide?" Sometimes this preference derives from the sober acknowledgment of a problem's intractability. We focus on a woman's right to choose, for example, because the debate on the moral merits of abortion, especially on the moral status of the fetus, has been interminable and fruitless. More often, however, our penchant for substituting process for substance rests on confusion or, worse still, on intellectual laziness. Nowhere is this tendency more apparent today than in the debate over gravely impaired newborns. Although the major protagonists are clearly committed, implicitly or explicitly, to certain substantive moral principles bearing on the treatment of such children, most of the public debate has been monopolized by procedural concerns. The controversy over the Reagan administration's policy, for example, has focused on such items as the telephone "hotline," the location and even the size of posters in neonatal nurseries, and the roving "Baby Doe Squads"-not on the policy's equally dubious conception of the moral and legal rights of seriously impaired newborns. Likewise, the report of the President's Commission for the Study of Ethical Problems in Medicine and Biomedical and Behavioral Research is better known for its cautious advocacy of hospital bioethics committees than for its substantive standard based on the "best interests" of the child. The attractiveness of such committees probably derives in large measure from their potential for transmuting a hard question (Who shall live?) into a more tractable one (Who shall sit on the committee?). I do not intend to belittle procedural concerns. The proper allocation of responsibility for decision making is an extremely important task. Moreover, it might well turn out JOHN D. ARRAS teaches in the department of social medicine, Montefiore Medical Center, Bronx, New York; and the department of philosophy. Barnard College.

that, when confronted by the hard cases, our substantive principles will founder on ambiguity and uncertainty. At such a juncture, we may not know what the "right thing" is, and we will have no choice but to "solve" the problem by permitting someone to cut the Gordian knot 1 Meanwhile, however, we ought to debate the merits of rival substantive principles hearing on the treatment of impaired children. Why should this be so? Perhaps the best reason is that there is simply no alternative. No matter who is ultimately given the authority to make decisions for these infantswhether it be the parents, doctors, a judge, or the Department of Health and Human Services-they will have to ask themselves whether certain babies should live or die. In deciding on a course of action or inaction, the agent will inevitably appeal, implicitly or explicitly, to one or more principles that ought to govern cases of this sort. Although we as a society may decide to lodge the authority for making such decisions in private hands, someone must decide and, in·so doing, appeal to standards of right and wrong. The issue is not whether we will employ substantive standards; rather, the question is which standards will guide us. Second, no matter who decides, their authority or discretion will not (or should not) be unbounded. As Ronald Dworkin reminds us, the notion of "discretion" resembles the hole in a doughnut: it only exists against a background of surrounding restriction. 2 Thus, even if parents are given discretion to refuse life-sustaining treatments for their children in certain situations, there will be limits to their freedom of choice. Presumably, the freedom to kill a child of the "wrong" sex, or of the wrong astrological sign, lies beyond the pale of legitimate discretion. And the proper bounds of this discretion can only be established by referring to substantive moral principles. Finally, we need to he more self-conscious about matters of principle in order to avoid inconsistency in our choice of procedures. The choice of a moral criterion conditions the range of acceptable means by which to carry it out. Given any particular moral principle, some procedures will appear fitting, while others will strain against the principle, generating a policy at odds with itself. For example, were we to adopt the President's Commission's "best-interest-of-thechild" standard-where "best interest" would not always mean continued existence-then the choice of a commjttee with significant lay representation seems to he ideal. Wf1en quality of future life is the issue, an exclusively medical committee would be out of place. Conversely, if we were to follow the Reagan administration in its rejection of any and all quality-of-life judgments--sanctioning nontreatment only for children on the brink of inevitable death-then it would be pointless, and indeed contradictory, to advocate a significant lay presence on these committees. Given the Reagan administration's "nondiscrimination on the basis of handicap" principle, there would appear to he no useful role for committees of any sort--except perhaps for prognosis committees com-

404

Death, Dying and the Ending of Life, I

posed exclusively of physicians attuned to the signs of imminent death. Thus, the administration's attempt to graft hospital "infant care committees," which will include "people with nonmedical perspectives," onto its "nondiscrimination" principle must be viewed as a product of deep intellectual confusion, or else as a desperate maneuver to extricate itself from an untenable political situation-' The Best Interests of the Child The Reagan administration's "nondiscrimination principle" is fatally flawed by an altogether superficial understanding of the concept of "discrimination.'"' By judging each and every failure to render equal treatment to normal and handicapped children alike as an instance of morally pernicious discrimination, the administration overlooks the very real possibility that there may well be morally significant differences between such children that justify such differential treatment. Rather than mandating equal treatment for handicapped and nonhandicapped alike, the moral principle of nondiscrimination requires simply that persons not be denied benefits on the ground of morally irrelevant reasons. Thus, while a person's sex is irrelevant to her qualifications to study law, it would be highly relevant in selecting someone for the position of wet nurse. Although the administration's conception of discrimination is woefully inadequate in theory, and thus as a basis for public policy, it at least raises the proper questions: How are we to show equal concern and respect for gr.1vely impaired newborns? Are there any differences between otherwise normal children and certain gravely impaired babies that might justifY allowing the latter to die through the intentional withholding of medical treatments? One important and plausible candidate for a morally relevant distinction between normal and certain anomalous newborns resides in the notion of their respective "best interests. "5 Continued life is obviously in the interests of the normal child, but what of the child impaired hy profound brain damage, whose days will be measured by operations and whose pain will be unrelieved hy the communication of human sympathy? Would it not be in this child's best interests to die, rather than endure a life of meaningless suffering? And if so, would that not provide us with a morally relevant distinction? The case for allowing the deaths of anomalous newborns on ihe ground of their "best interests" is by far the most compelling argument for infant euthanasia. By focusing on the child's interests, this standard avoids morally dubious utilitarian justifications based on the well-being, of other interested parties, such as parents, siblings, or even of society at large. And by giving full weight to the child's interests, this approach can justify nontreatment in certain cases without having to make the (usually pernicious) assumption that the child is a nonperson with no standing in the human community. Despite these advantages, however, the best-inter-

26

ests standard presents staggering problems of interpretation and application. A Preliminary Objection. Before coming to these problems, we must confront a direct challenge to the legitimacy of the best-interests standard itself. CertaL'1 theologians and jurists who oppose any resort to "quality-of-life" judgments on behalf of voiceless patients have argued that the bestinterests standard forces us to make an impossible comparison. Although we can and should compare better life against worse life, they argue, we cannot compare the advantages of life, even with severe disabilities, against the state of non being initiated by death. Since all our everyday evaluations take place against the backdrop of the assumed value of life, it is said, we cannot make the comparison of life against nonlife demanded by the best-interests standard without foundering on conceptual incoherence. 6 While this is not the occasion to launch a lengthy metaphysical explanation of how a state of nonexistence might be "preferable" to certain states of existence-how can someone be better off dead?-two points can be made in response to those who claim that death can never be in the best interests of a nondying child. First, we can concede that a belief in the desirability of life grounds our everyday judgments of "better" and "worse," while still maintaining that the moral choices posed by the plight of the impaired newborn are anything but ordinary. Although we usually profess an implicit faith in the conceptual connection hetween our notions of "life" and "good," the profoundly anomalous and severely brain-damaged child shatters this everyday faith and severs the assumed connection. In such cases, it is not at all obvious that mere life remains a good to the afflicted person; rather, it seems as though the child's illness has eclipsed the possibility of enjoying those normal human goods that ordinarily predispose us to say that life is good. 7 Moreover, while the existence of severe disabilities threatens to sever our everyday connection between "life" and "good," another item of everyday faith remains unshaken: We continue to believe that pain and suffering are evil, and that beneficence largely consists in abolishing, or at least ameliorating, these twin scourges of humankind. If an act of mercy killing were to deliver a child from a life of unmitigated pain-if only to a state of nonexistence rather than beatitude-it would take a profoundly dogmatic temperament to deny that death was in the child's best interest. Opponents of infant euthanasia could perhaps find other reasons to oppose such an act-for example, the belief that humans have no right to dispose of the.ir own or anyone else's life-but merely on the grounds of individual well-being, there is no denying that death would be better than a life of intense and constant pain. Second, if it were impossible for us to detach ourselves from our everyday grounding in the belief that life is good, if we could not wish someone else dead for his or her own sake, then the same difficulty would attach to first-person judgments as well. If it were in principle impossible to say The Hastings Centet Report, April 1984

Death, Dying and the Ending of Life, I that death would serve the best interests of any child, then it would be equally impossible for any rational adult to conclude that, given a bleak prognosis and the prospect of burdensome treatments, he or she would be better off dead. I doubt that the more thoughtful opponents of euthanasia would wish to embrace such a result. Even conservative theologians such as Paul Ramsey restrict their rejection of quality-of-life reasoning-and hence their rejection of "excessive burdensomeness" as a test of extraordinary means-to the class of voiceless patients who have not developed a value system that would allow them to assess the quality of future life for themselves.s Ramsey explicitly grants to the competent adult the right to refuse treatments that would render continued life a burden. ln order to assume this stance, he must admit the possibility that death would be in the best interest of certain persons; but this admission pulls the rug out from under the claim that, in principle, we cannot judge death to be in a child's best interest. The most that can be said is that no one has the right to make such judgments, or that allowing adults to make them for children will result in mistaken choices or downright abuse. We cannot say that, in principle, life (even with crushing impairments and constant pain) cannot even be compared with death. At best, such a denial is an unfortunate overstatement; at worst, it could be placed in the service of an overbearing medical paternalism. Indeed, what better way to thwart the choices of mature adults than to deny them the very capacity rationally to choose between an earlier death without treatment and a life of suffering with treatment"? In sum, we can say that death is not always and everywhere the greatest evil that could befall a person. Afflictions of the body and mind might conspire to render continued life a burden for certain individuals, for whom death might be viewed as a benefit-or at least as the lesser of two evils. As the President's Commission recently concluded, the best interests of the child-rather than a spurious nondiscrimination principle--should be our guide. Normally, of course, treatment will clearly be in a child's best interest. But the Commission granted the possibility that some pennanent handicaps might be "so severe that continued existence would not be a net benefit to the infant. " 9 Our next problem, then, is to determine with more precision how a best-interest standard might be applied and to discuss some weighty problems of interpretation. Prognostic Uncertainry. One problem will plague effons to implement any proposed criterion of nontreatment: medical and prognostic uncertainty. 10 Decisions must often be made regarding life or death in the absence of reliable data regarding a particular child's future life prospects. No matter where we draw the line separating "meaningful" from "excessively burdensome" life, we will often have to decide under conditions of great uncertainty. The spectmm of neonatal disabilities is very wide, as is the snectrum of disability within particular disease categories. The mere presThe Hastings Center

405

ence of Down syndrome or of spina bifida does not by itself indicate how disabled a child will be. Some Down children are profoundly retarded, while others turn out to be sufficiently intelligent to Jive at home with their families or in group homes and eventually to work productively in a sheltered setting. The problem is that, except for a few cla"es of disease, physicians cannot accurately predict the degree of a child's eventual impairment. This problem is compounded by our ability to diminish a child's eventual degree of disability by some means of vigorous medical and educational interventions. In the absence of reliable prognostic indicators of future quality of life, the rationality of nontreatment decisions for defective neonates is bound to be diminished. Any medical ethic that places a great value on human life must look with suspicion upon nontreatment decisions taken in the absence of reliable medical information. While medical uncertainty threatens to undermine the application of any set of criteria for nontreatment based on quality of lite, the best-interests standard is subject to further difficulties of interpretation. It may well turn out that, all things considered, death may serve the best interests of a severely afflicted infant; but before we can reach this conclusion, two important questions must be resolved. First, we must know what burdens to the child are admissible in our moral calculations of benefit and burden. Do any and all burdens to the child count, or only certain kinds? Second, we must have a clearer idea of the point of view from which these burdens will be assessed. Shall it be the viewpoint of the gravely impaired child or of the normal lldult? Which burdens shall count? So far, we have listed physical ailments that might befall a child and render excessively burdensome a prolongation of his or her life. Now we must consider whether burdens attributable to the child's socioeconomic status can be counted alongside strictly medical burdens as contributing to an excessively burdensome lite and, if so, whether such burdens morally ought to be so counted. In addition to suffering from the retardation associated with Down syndrome, the repeated surgeries associated with spina bifida, or the iatrogenic toll of respirator therapy for premature infants, a child may suffer from a host of nonmedical problems. She may, if allowed to live, suffer from parental rejection. Plagued by frustration, guilt, and disappointment, parents can and do neglect their impaired children-or, worse yet, abuse them. Or the parents may love and wish to care for their child, yet be too ill-equipped to care for her at borne and too poor to place her in an acceptable institution. Such parents face the dilemma of keeping their child at home, where the demands of caring for her disabilities are likely to drain their economic and emotional resources, or handing her over to an institution that is likely to be underfinanced and understaffed. At such an institution, the severely impaired child is likely to be relegated to some back ward where she will suffer from lack of love, human contact, and even basic care."

27

406

Death, Dying and the Ending of Life, I

"Clearly," it might be argued, "subjecting the anomalous child to this sort of institutionalized deprivation amounts to a harm of considerable proportion. 1n the absence of better alternatives to institutionalization, such as it is today in many economically deprived areas, these children should be relieved of potentially crushing burdens by means of an early, merciful death." This is a plausible argument. Unquestionably the abuse and neglect attributable to a family's inability to cope or to pay for humane care can amount to substantial burdens for the child. And so long as burdensomeness to the child remains the crucial criterion involved in the best-interest standard, why shouldn't social as well as purely medical burdens count in our overall calculations? Indeed, if socially induced burdens can make continued life excessively burdensome for a severely retarded child relegated to some back ward, logic would seem to compel us to weigh them as heavily as any other kind of burden. The important thing, it might be said, is not the source of the burden, but its effect on the child. If parental neglect, social ostracism, and degrading conditions can make life unbearable, why shouldn't they be a part of our moral deliberations? In spite of its initial plausibility, this line of reasoning runs afoul of the nondiscrimination principle. If it would be in a wealthy child's best interest to receive life-sustaining therapies, justice demands that a poor child with a similar medical condition and prognosis also receive the same treatment. Like cases should be treated alike. (The same can be said on behalf of children born to parents rich and poor alike who are unable or unwilling to nurture them. The parents' inadequacies are not the fault of their children.) True, in our previous discussion of medically induced burdens we suggested that certain disabilities could be so severe as to render an anomalous child significantly different from a normal child for purposes of moral analysis. We said then that such disabilities could prompt us to deny the premise that the severely impaired child was still "like" the normal child in all relevant respects, and thus to deny the conclusion that such children must receive "like" treatment. We argued, finally, that the severely impaired child was unlike the nonnal child precisely in the sense that her best interests would not be served by further treatment or continued existence. Why not repeat the same argument here with regard, this time, to socially related burdens? The answer to this eminently sensible question is that, whereas an extremely poor medical prognosis and quality of life can make for a significant moral difference between two babies, the same cannot be said of differences in wealth. Degree of wealth is never a morally significant difference between two potential recipients of basic health care. If a certain medical procedure is deemed to be truly beneficial and offered to a wealthy child with severe disabilities, it must also be offered to the similarly situated poor child. The latter is not responsible for the lack of concern of her parents, or for their inability to afford dignified and humane

28

follow-up care. To say that the wealthy child should live (because her anomalies are not so severe as to make life too burdensome) but that the poor child with similar prognosis should die (because of further burdens imposed by her poverty) is to indulge in the rankest kind of discrimination. For purposes of moral analysis, these two patients are identical, and they should therefore receive similar treatment. That is the demand of simple justice-unless we hold that being rich in itself makes the wealthy impaired child more deserYing of health care than the poor child, an unlikely concession for us to make. A policy that allowed us to consider socially induced burdens would not only be profoundly unjust; it would also most likely impede efforts to improve the plight of the institutionalized, poor, imperiled child. Historically, one of the most potent arguments in favor of social reform has rested on the existence of human suffering and deprivation. Could we expect vigorous efforts to reform public institutions housing severely impaired children once the suffering and neglect that characterize such institutions have become publicly recognized as good reasons for killing these newborns? My sympathies lie entirely with Paul Ramsey's response to the suggestion that a child's poverty ought to count in her constellation of publicly recognized burdens. "That's one way," says Ramsey, "to remove every evening the human debris that has accumulated since morning." For the great majority of these handicapped infants, radical social reform-not "beneficent" euthanasia-is the answer. The procedural implication of this argument about justice is that we should attempt to distinguish clearly between "treatment decisions" and decisions regarding the child's ultimate placement. 12 In other words, worries about the child's future life with a potentially abusive parent or in a hopelessly inadequate state institution should not be allowed to influence treatment decisions. Even though socially induced burdens can join forces with strictly physical disabilities to make a life excessively burdensome to its bearer, we must base our treatment decisions solely on the extent of medical disabilities. To take social factors into account is to act unjustly toward the child. At this point in the argument, we have clearly entered a "moral blind alley"-that is, a situation so structured that whatever course we take, we end up doing something morally unacceptable. 13 Ordinarily this structure takes the form of a conflict between two disparate moral concernsnamely, doing good and doing justice. Morality, through the principle of beneficence, requires us to increase people's happiness and mitigate their unhappiness. But another moral principle enjoins us to respect persons' rights and to do justice. When these conflicting demands meet head-on in a concrete situation, the stage is set for an ethical dilemma. How can they be reconciled? As Ronald Dworkin has contended, a concern for rights and justice can function as a "trump card" in moral and political argument with regard to competing utilitarian conThe Hastings Center Report, April 1984

Death, Dying and the Ending of Life, I siderations. 14 When there is a conflict between the demands of utility and the requirements of justice, we usually say that the former must simply yield to the latter. Pointing to the good or bad consequences said to flow from the performance or omission of an act is usually said to be an insufficient (or even irrelevant) justification if that act or omission happens to be unjust. For example, we say that a citizen's right to speak freely must be respected, even though rather serious consequences (such as public disturbances) may ensue from the exercise of that right. In this sort of case, the right course is clear-we should let the individual speak-and there is no serious moral dilemma. One moral concern has simply eclipsed another. Sometimes circumstances may be so extreme and the consequences so dreadful that the priority of justice can no longer be maintained. When this happens, an agent enters a moral blind alley. He must choose, but all of his alternatives are morally unacceptable. If he does the "right" thing, the act required by strict justice, then catastrophic results will follow; if he violates the demands of justice, he avoids terrible consequences at the cost of doing a deed that should never be done. In such cases, one may prudently decide to violate rights and justice in order to avoid catastrophe, but this does not always count as a complete justification. "Moral traces" of the demands of justice remain, even when circumstances make it plain that justice must be violated. Simply put, it does not become all right: we are left with dirty hands." Winston Churchill faced such a moral dilemma in World War 11-should he launch rocket attacks on German civilian populations, thereby killing hundreds of innocent persons, or risk losing the war? I suspect that we face a similar dilemma in the case of the impoverished impaired newborn. If we take the child's socially caused burdens into account-such as the likelihood of being neglected in a state institution-we may rightly decide that death would be better than such a fate. Allowing such a child to die would be a merciful act, averting an unbearable life for the child; but it would also be unjust. As we have seen, it is profoundly unjust to afford differential treatment to impaired newborns on the basis of socioeconomic factors. However, if we avert our eyes from the child's subsequent fate in a barren state institution, we act justly but possibly consign the child to a misemble life in a warehouse for discarded humans-a catastrophic result, to be sure. It would appear, then, that we are confronted by a genuine moral dilemma occasioned by social conditions of poverty and inequity. Here, as elsewhere in the field of biomedical ethics, straightforward appeals to notions of right and justice-abstracted from a social context of profound injustice--can lead to morally disastrous results. 16 So long as these unjust social conditions persist, so long as the rich gravely impaired child enjoys a bearable quality of life denied to a poor counterpart, the door of the neonatal nursery will continue to open onto a moral blind alley. Rather The Hastings Center

407

than serve as an invitation to resignation, however, this conclusion underscores the necessity of increased social (governmental) support of the handicapped. Surely, any society that insists upon a controversial strict standard of justice concerning nontreatrnent decisions-a standard that will weigh heaviest upon those parents most in need of psychological and financial support-surely, such a society thereby incurs a corresponding duty to support those who must bear the burden it imposes. What standard will measure the burden? There are essentially two possibilities: we can adopt either the point of view of the rational adult or that of the anomalous child. According to some philosophers, we ought to adopt the viewpoint of the mature adult or the so-called "reasonable person. " 17 "Since incompetent creatures are incompetent," writes Joseph Margolis, "we judge as rational adults of normal sensibilities would. It is in this sense that the decision to bring to an end the lives of selected fetuses and infants . . . is conceptually parasitic on the decision of the competent to end their own lives . . . . " Although this approach is flawed, it has a certain appealing modesty. It does not require us to plumb the depths of the incompetent person's psyche to discern there what the incompetent patient "really wants." Such modesty is becoming when the patient for whom we are to decide has not left us any evidence of how he or she would decide; it is especially becoming when the patient lacks the very capacity to develop a value system in the first place. Although courts have recently ascribed to themselves or to guardians the right to make so-called "substituted judgments" for voiceless patients, and although the proxy decision makers have been legally charged with the (impossible?) task of deciding as the patient would decide, were he or she miraculously to become competent for an instant, I suspect that they end up attempting to judge as any rational adult of normal sensitivities would. And for good reason. In the absence of explicit or implicit value judgments from the incompetent patient, what else do we have to go on? How do we know whether or not the permanently incompetent patient would judge life under present circumstances to be worth the candle? Far from being endowed with the Herculean powers of discernment required by such a tam 53 (1946). For an eloquent presentation of the problem raised by the defective infant see id. at 53-64. It is difficult to discuss the consultation feature of Williams' proposal for affirmative "mercy-killing'' because Williams himself never discusses it. This fact, plus the fact that Williams' recurrent theme is to give the general practitioner a free hand indicates that he himself does not regard consultation as a significant feature of his plan. The attending physician need only consult another general practitioner and there is no requirement that there be any concurrence in his diagnosis. There is no requirement of a written report. There is no indication as to what point in time there need be consultation. Probably there need be consultation only as to diagnosis of the disease and from that point on the e.""~:tent and mitigatory nature of the pain, and the firmness and rationality of the desire to die is to be judged solely by the attending physician. For the view that even under rather elaborate consultation requirements, in many thinly staffed communities the consulted doctor would merely reflect the view of the attending physician see Life aml Death, Time, March 13, 1950, p. 50. Mtt>r reviewing eleven case histories of patients wrongly diagnosed as having advanced cancer, diagnoses that stood uncorrected over long periods of time and after several admissions at leading hospitals, Doctors Laszlo, Colmer, Silver and Standard conclude: fErrors i1~ Diagnosis Atul Management of Cat~cer, 33 Annals Int. Med. 670 (1950) 1: [I] t became increasingly clear that the original error was one easily made, but that the continuation of that error was due to an acceptance of the original data without exploring their verity and completeness. doub~

Death, Dying and the Ending of Life, II

170

984

MINNESOTA LAW REVIEW

[Vol4Z:969

Evidently, the presumption is that the general practitioner is a sufficient buffer between the patient and the restless spouse or overwrought or overreaching relative, as well as a depository of enough general scientific know-how and enough information about current research developments and trends, to assure a minimum of error in diagnosis and anticipation of new measures of relief. Whether or not the general practitioner will accept the responsibility Williams would confer on him is itself a problem of major proportions.'= 42. In taking the Hippocratic Oath, the oldest code of professional ethics, the physician promises, of course, to "give no deadly medicine to any one if asked, nor suggest any such counsel." Many doctors have indicated they would not accept the role in whiF. legalized euthanasia would cast them. See, e.g., Frohman, supra note 37, at 1221 ("I could never deliberately choose the time of another's dying. The preservation of human life is not only the primary but the all-encom.yassing general law underlying the code of the physician•••• Do not ask life s guardian to be also its executioner.") ; Gumpert, A False- Mercy, 170 The Nation 80 (1950). ("As a physician, I feel I would have to reject the power and resJlOnsibility of the ultimate decision'') ; Lord Baden-Guest, 169 HL. Deb. [5th ser.] 551, 586 [1950]) (''You are asking the medical profession to do it. Ask somebody else.''), Kennedy, Euthanasia: To Be or Not To Be, Colliers, May 20, 1939, pp, 15, 57, reprinted in Colliers, A1>ril 22, 1950, pp. 13, 50 ("Who is going to carry out the sentence of death? I am sure not I .... too grisly a notion for the profession of medicine to stomach"). In 1950, a baimer year for mercykilling trials (see the Mo.br casei: supra note 17, and the Sander, Paiglat and Braunsdorf cases at notes 172- 76, 183 infra and accompanying text) the General Assembly of the World Medical Association approved a resolution recommending to all national associations that they "condemn the practice of euthanasia under any circumstances." New York Times, Oct. 18, 1950, p. 22, col. 4. Earlier that year1 the Medical Society of the State of New York went on record as being ''UllalterablJ," opposed to euthanasia and to any legislation which will legalize euthanasia.' New York Times, May 10, 1950, p. 29, col. 1. On the other hand, euthanasiasts claim their movement finds great support in the medical profession. The most impressive and most frequently cited piece of evidence is the formation, in 1946, of a committee of 1,776 physicians for the legalization of voluntary euthanasia in New York See Williams at 331 ; FletCher, oP. cit. supra note 3, at 187. Williams states that of 3;272 physicians who replied to a questionnaire in New York State in 1946, 80 per cent approved voluntary euthanasia and the Committee of 1)76 came from among this favorable group. I have been unable to find any authority for the 80 per cent figure, and Williams cites none. Some years ago, Gertrude Anne Edwards, then editor of the Euthanasia Society Bulletin, claimed 3,272 physicians-apparently all who replied-favored legalizing voluntary euthanasia. Edwards, Mercy Deat!J For Incterables Should Be Made Legal, The Daily Compass, Aug. 24, 1949, p. 8, col. 1 (issue of the day). Presumably, as in the case of the recent New Jersey questionnaire discussed below, every physician in New York was sent a questionnaire. If so, then the figures cited, whether Williams or Edwards, would mean a great deal more (and support the euthanasiasts a great deal less) if it were added that 88 or 89 per cent of the physicians in the state did not reply at all. In 1940, there were over 26,000 physicians in the State of New York, U.S. Department of Commerce, Bureau of the Census, The Labor Force, Part 4 at 366; in 1950 there were over 30,000, U.S. Dep't of Commerce, Bureau of the Census, Characteristics of the Population, Part 32 at 260. The most recent petition of physicians for legalized euthanasia was that signed by 166 New Jersey physicians early in 1957 urgiJlg in effect the adoption of the American Society's Bill See Anderson, Who Signed for

171

Death, Dying and the Ending of Life, II

1958}

"MERCY-KILLING" LEGISLATION

985

Putting that question aside, the soundness of the underlying premises of Williams' "legislative suggestion" will be examined in the course of the discussion of various aspects of the euthanasia problem.

B. The ((Choice." Under current propoals to establish legal machinery, elaborate or otherwise, for the administration of a quick and easy death, it is not enough that those authorized to pass on the question decide that the patient, in effect, is "better off dead." The patient must concur in this opinion. Much of the appeal in the current proposal lies in this so-called uvoluntary" attribute. But is the adult patient43 really in a position to concur? Is he truly able to make euthanasia a "voluntary'' act? There is a good deal to be said, is there not, for Dr. Frohman's pithy comment that 96 America 573 (1957). According tc• this article, the American Society had sent a letter to all the doctors in the state asking them to sign such a petition. The doctors were asked to check either of two places, one indicating that their name could be used, the other that it could not. The 1950 census records over 7,000 physicians in New Jersey. Characteristics of the Population, Part 30 at 203. Thus, about 98 per cent of the state medical Jlrofession declined to sign such a petition. The Medical Society of New Jersey immediately issued a statement that "euthanasia has been and continues to be in conflict with accepted principles of morality and sound medical practice." See Anderson, .s~~Pra. When their names were published in a state newspaper, many of the 166 claimed they had not signed the petition or that they had misunderstood its purpose •Jr that, unknown to them, some secretary had handled the matter in a routine manner. See Anderson, Et~tlzauasia-f

supra.

Cf. Paragraph 27 of the Memorandum submitted by the Council of the British Medical Association (Royal Commission, Minutes of Evidence at p. 3118n) :the optmon · · . . .. of the A ssoctation, no medical practitioner sbould b e asked to take part in bringing about the death of a convicted murderer. The Association would be most strongly opposed to any proposal to introduce, in place of judicial hanging, a method of execution which would require the services of a medical practitioner, either in carrying out the actual process of killing or in instructing others in the technique of the process. Examination of medical witnesses disclosed that they opposed execution by intravenous injection as "a matter of professional ethics" since "under oath we are bound to promote life . • • whereas any action which has as its object the termination of life, even directly, we feel is undesirable." Id. at para. 4041 (Feb. 3, 1950). See also para. 4 of the Memorandum of the Association of Anaesthetists to the effect that if intravenous injection is adopted as an alternative method of e.'l:ecution "the executioner should have no connection or association with the medical profession." Id. at p. 678A. For a general discussion of the problem and the views of the medical profession on the matter, see Royal Commission Report at paras. 737-748. Apparently the American medical profession has the same reluctance to participate in e.'l:ecution by intravenous injection. See Weihofen, The Urge to Punish 168 (1956). 43. It should be noted that under what mi;~ht be termed the "family plan" feature of Williams' proposal, minors may be euthanatized, too. Their fate is to be "left to the good sense of the doctor, taking into account, as he always does, the wishes of the parents as well ru;. those of the child." Williams, p. 340, n. 8. The dubious quality of the "v•Jluntariness" of euthanasia in these circumstances need not be labored.

Death, Dying and the Ending of Life, 1/

172

986

MINNE,SOTA. LAW REVIEW

[Vol. 42:969

the "voluntary" plan is supposed to be carried out "only if the victim is both sane and crazed by pain."H By hypothesis, voluntary euthanasia is not to be resorted to until narcotics hav~ long since been administered and the patient has developed a tolerance to them. When, then, does the patient make the choice? While heavily drugged ?45 Or is narcotic relief to 44. Frohman, V e~it:g Problems In Forensic Medicine: A. Physician's View, 31 N.Y.UL. Rev. 1215, 1222 (1956). 45. The disturbing mental effects of morphine, "the classic opiate for the relief of severe pain," Schiffrin and Gross, Systematic Analgetics, in Management Of Pain In Cancer p. Z2 ( Scbiffrin ed. 1956) and "still the most commonly used potent narcotic analgesic in treatment of cancer pain," Bonica, The Management of Cancer Pain, G.P., Nov. 1954, pp. 35, 39, have been described in considerable detail by Drs. Wolff, Hardy and Goodell in Sttulics on Pain: Measurement of the Effect of Morphine, Codeine, and other Opiates otJ the Pain Threshold and an Analys:s of their Relaliot~ to the Pait' E~perience, 19 J, Clinical Investig. 659, 664 (1940). It is not easy to generalize about the psychological effects of drugs for there is good reason to believe that the type of drug reaction is correlated with "differential personality dynamics, primarily in terms of the balance of mature, soctally oriented controls over impulsive, egocentric emotionality," von Felsinger, Lasagna and Beecher, Dn1g-Indteced Mood Changes in Man, 157 AM.A.J. 1113, 1119 (1955), that for example, persons with atypical reactions to drugs are likely to be those with pre-existing immaturity, anxiety and hostility, id. at 1116. See also Lindemann and Clark, Modifications In Ego Struct11rc at:d Personality Reactions Under the In/lfemce of the Effects of Dnegs, 108 Am. ]. Psychiatry 561 (1952). It would seem, however, that the severely ill person would be likely to experience substantially more pronounced effects than those described by Wolff, Hardy and Goodell, supra, because in that instance the "subjects" studied were the authors themselves, representing both sexes and different body types, experiencing various degrees of pain by exposing portions of their skin surfaces to thermal radiation, but in the case of an illness due to a malignancy or suspected malignancy, we start with a situation where "all kinds of irrational attitudes come to the fore". Zarling, Psychological Aspects of Pain In Terminal Malignancies, in Management of Pain in Cancer 205 (Schiffrin ed. 1956). The increasing use of ACTH or cortisone therapy in cancer palliation, see notes 98-101, ittfra and accompanyinft text, presents further problems. Such therapy "frequently'' leads to a 'severe degree of disturbance in capacity for rational, sequential thought" Lindemann and Clark, supra at 566. Clark, et. al., Preliminary Observations On MeMal Disturbances Occnrrillg In Patients tmder Therapy With Cortiso11e and ACTH, 246 N. Eng. J. Med. 205, 215 (1952) describe six case histories of "major disturbances" where "delusions of depressive, paranoid and grandiose types occurred" and "affective disturbances, also invariably present, varied from depression to hypomania and from apathy to panic; they included ill-defined states that might be described as bewilderment or turmoil." In a subsequent paper, the authors conclude, Oark, et. al., Ftirlher Observations On Mental Disturbances Associated With Corliso1te and ACTH Therapy, 249 N. Eng. J. Med. 178, 182 (1953) that the clinical course of psychoses associated with ACTH and cortisone is "more remarkable for its variability and unpredictability than any other feature," that, for example, mental disturbances may be separated by "intervals of relative lucidity," that "patients may have tolerated previous courses of ACTH or cortisone without complications and yet become psychotic during a subsequent course of treatment with comparable or ~ven smaller doses." For an extensive review of the many hypotheses purporting to explain mental disturbances associated with ACTH and cortisone see Quarton, ct. al., Mental Disturbances Associated with ACTH and Cortisone: A. Review of

Death, Dying and the Ending of Life, II

1958]

"MERCY-KILUNG" LEGISLATION

173

987

be withdrawn for the time of decision? But if heavy dosage no longer deadens pain, indeed, no longer makes it bearable, how overwhelming is it when whatever relief narcotics offer is taken away, too? "Hypersensitivity to pain after analgesia ~ wom off is nearly always noted."48 Moreover, "the mental side-effects of narcotics~ unfortunately for anyone wishing to suspend them temporarily without unduly tormenting the patient, appear to outlast the analgesic effect" and "by many hours."41 The situation is further complicated by the fact that "a person in terminal stages of cancer wlio had been given morphine steadily for a matter of weeks would certainly be dependent upon it physically and would probably be addicted to it and react with the addict's response/'"'8 The narcotics problem aside, Dr. Benjamin Miller, who probably has personally experienced more pain than any other commentator on the euthanasia scene,'9 observes: Anyone who has been severely ill knows how distorted his judgment became during the worst moments of the illness. Pain and the toxic effect of disease, or the violent reaction to certain surgical procedures may change our capacity for rational and courageous thought.50 If, say, a man in this plight were a criminal defendant and he were to decline the assistance of counsel would the courts hold that he had Espla11atory Hypotheses, 34 Med. 13 (1955). The authors emphasize the inadequacy of present knowledge of mental disturbances associated with this therapy, but believe, "because of the clinical and experimental studies which suggest it," that "it is useful to assume" "cortisone and ACTH produce a ['probably reversible'] specific pattern of modified nervous system function which is invariably present when a gross mental disturbance occurs," id. at41. 46. Goodman and Gilman, The Pharmacological Basis of Therapeutics 235 (2d eel. 1955). To the same effect is Seevers and Pfeiffer, A Study of the

A1uzlgesia~ S11bjective Depression, and Euphori~ Prodttced by Morphi~~e, Heroine, uilaudid and Codeine In the Normal Hmnan Snbject, 56]. Pharm. & E..'CJ)er. Therap. 166, 182, 187 (1936). 47. Sharpe, 111edication As A Threat To Testamentary Capacity, 35

N.C. L. Rev. 380, 392 (1957) and medical authorities cited therein. In the case of cortisone or ACTH therapy, the situation is complicated by the fact that "a frequent pattern of recovery'' from PsYchoses induced by such therapy is "by the occurrence of lucid inten'als of increasing frequency and duration, punctuated by relapses into psychotic behavior." Clark, et. al.,

Further Obscrvatio1~ On Mental Disturba1tces Associated With Cortisone a1ul ACTH Therapy, 249 N. Eng. J. Med. 178, 183, (1953). 48. Sharpe, stiPra, note 47, at 384. Goodman and Gilman, op. cit., supra,

note 46 at 234, observe that while "different jndividuals require varying periods of time before the repeated administration of morphfue results in tolerance, .•• as a rule ••• after about two to three weeks of continued use of the same dose of alkaloid the usual depressant effects fail to a\'pear'' whereupon "phenomenally large doses may be taken." For a discussion of "the nature of addiction," see Maurer and Vogel, Narcotics and Narcotic Addiction 20-31 (1954). 49. See note 77 infra and accompanying text. SO. Miller, Why I Oppose Mercy Killings, 1Voman's Home Companion, June 1950, pp. 38, 103.

Death, Dying and the Ending of Life, 1/

174

988

MINNESOTA LAW REVIEW

[Vol. 42:969

"intelligently and understandingly waived the benefit of counsel ?"51 Undoubtedly, some euthanasia candidates will have their lucid moments. How they are to be distinguished from fellow-sufferers who do not, or how these instances are to be distinguished from others when the patient is exercising an irrational judgment is not an easy matter. Particularly is this so under Williams' proposal, where no specially qualified persons, psychiatrically trained or otherwise, are to assist in the process. Assuming, for purposes of argument, that the occasion when a euthanasia candidate possesses a sufficiently clear mind can be ascertained and that a request for euthanasia is then made, there remain other problems. The mind of the pain-racked may occasionally be clear, but is it not also likely to be uncertain and variable? This point was pressed hard by the great physician, Lord Harder, in the House of Lords debates: During the morning depression he [the patient] will be found to favour the application under this Bill, later in the day he will think quite differently, or will have forgotten all about it. The mental clarity with which noble Lords who present this Bill are able to think and to speak must not be thought to have any counterpart in the alternating moods and confused judgments of the sick man. 52 The concept of "voluntary" in voluntary euthanasia would have a great deal more substance to it if, as is the case with voluntary admission statutes for the mentally ill,53 the patient retained the right to reverse the process within a specified number of days after he gives written notice of his desire to do so- but unfortunately this cannot be. The choice here, of course, is an irrevocable one. The Iiklihood of confusion, distortion or vacillation would appear to be serious draw-backs to any voluntary plan. Moreover, Williams' proposal is particularly vulnerable in this regard, since, as he admits, by eliminating the fairly elaborate procedure of the American and English Societies' plans, he also eliminates a time period which would furnish substantial evidence of the patient's 51. Moore v. Michigan, 355 U.S. 155, 161 (1957). 52. 103 H. L. Deb. (5th ser.) 466,492-93 (1936). To the same effect is Lord Horder's speech in the 1950 debates, 169 H. L. Deb. (5th ser.) 551, 569 (1950). See also Gumpert, A False Mercy, 170 The Nation 80 (1950): Even the incapacitated, agonized patient in despair most of the time, may still get some joy from existence. His mood will change between longing for death and fear of death. Who would want to decide what should be done on such unsafe ground? For a recent layman's account of the self-pity and fluctuating desires for life and death of a seriously ill person, see the reflections of the famous sports broadcaster Ted Husing in My Friends Wouldn't Let Me Die, Look, Feb. 4, 195S, p. 64. 53. See Guttmacher and Weihofen, Psychiatry and the Law 307 (1952).

Death, Dying and the Ending of Life, II

19581

"MERCY-KILLING" LEGISLATION

175

989

settled intention to avail himself of euthanasia.11' But if Williams does not always choose to slug it out, he can box neatly and parry gingerly: [T] he problem can be exaggerated. Every law has to face difficulties in application, and these difficulties are not a conclusive argument against a law if it has a beneficial operation. The measure here proposed is designed to meet the situation where the patient's consent to euthanasia is clear and incontrovertible. The physician, conscious of the need to protect himself against malicious accusations, can devise his own safeguards appropriate to the circumstances; he would normally be well advised to get the patient's consent in writing, just as is now the practice before operations. Sometimes the patient's consent will be particularly clear because he will have expressed a desire for ultimate euthanasia while he is still clear-headed and before he comes to be racked by pain; if the expression of desire is never revoked, but rather is reaffirmed under the pain, there is the best possible proof of full consent. I£, on the other hand, there is no such settled frame of mind, and if the physician chooses to administer euthanasia when the patient's mind is in a variable state, he will be walking in the margin of the law and may find himself unprotected.55 If consent is given at a time when the patient's condition has so degenerated that he has become a fit candidate for euthanasia, when, if ever, will it be "clear and incontrovertible?" Is the suggested alternative of consent in advance a satisfactory solution? Can- such a consent be deemed an informed one? Is this much different from holding a man to a prior statement of intent that if such and such an employment opportunity would present itself he would accept it, or if such and such a young woman were to come along he would marry her? Need one marshal authority for the proposition that many an "iffy" inclination is disregarded when the actual facts are at hand? 68 Professor Williams states that where a pre-pain desire for "ultimate euthanasia" is "reaffirmed" under pain, "there is the best 54. Williams, pp. 343-44. 55. I d. at 344. 56. Dr. James J. Walsh in Life Is Sacred, 9•~ The Forum, 333, 333-34, recalls the following Aesop's fable: It was a bitter-cold day in the wintertime, and an old man was gathering broken branches in the forest to make a fire at home. The branches were covered with ice, many of them were fr·Jzen and had to be pulled apart, and his discomfort was intense. F'mall:r the poor old fellow became so thoroughly wrought up by his suffering that he called loudly upon death to come. To his surprise, Death came at once and asked what he wanted. Very hastily the old man replied, 'Oh, nothing; nothing except to help me carry this bundle of sticks home so that I may make a fire!

Death, Dying and the Ending of Life, 1/

176

990

MINNESOTA LAW REVIEW

[Vol. 42:969

possible proof of full consent." Perhaps. But what if it is alternately renounced and reaffirmed under pain? What if it is neither affirmed or renounced? What if it is only renounced? Will a physician be free to go ahead on the ground that the prior desire was 'frational", but the present desire uirrational"? Under Williams' plan, will not the physician frequently "be walking in the margin of the law"just as he is now? Do we really accomplish much more under this proposal than to put the euthanasia principle on the books? Even if the patient's choice could be said to be uclear and incontrovertible," do not other difficulties remain? Is this the kind of choice, assuming that it can be made in a fixed and rational manner, that we want to offer a gravely ill person? Will we not sweep up, in the process, some who are not really tired of life, but think others are tired of them; some who do not really want to die, but who feel they should not live on, because to do so when there looms the legal alternative of euthanasia is to do a selfish or a cowardly act? Will not some feel an obligation to have themselves ueliminated" in order that funds allocated for their terminal care might be better used by their families or, financial worries aside, in order to relieve their families of the emotional strain involved? It would not be surprising for the gravely ill person to seek to inquire of those close to him whether he should avail himself of the legal alternative of euthanasia. Certainly, he is likely to wonder about their attitude in the matter. It is quite possible, is it not, that he will not exactly be gratified by any inclination on their parthowever noble their motives may be in fact- that he resort to the new procedure? At this stage, the patient-family relationship may well be a good deal less than it ought to be: Illness, pain and fear of death tend to activate the dependent longings [for the family unit] . Conflict can easily arise, since it may be very difficult for the individual to satisfy his need for these passive dependent needs and his previous concept of the necessity for a competitive, constructive individuality. Our culture provides few defenses for this type of stress beyond a suppression of the need. If the individual's defenses break down, he may feel angry toward himself and toward the members of his family. liT And what of the relatives? If their views will not always influence the patient, will they not at least influence the attending physician? Will a physician assume the risks to his reputation, if not his pocketbook, by administering the coup de grace over the 57. Zarling, supra note 45, at 215.

Death, Dying and the Ending of Life, II

1958}

~JlERCY-KILLINGnLEGISLATION

177

991

objection- however irrational- of a close relative ?58 Do not the relatives, then, also have a t'choice?" Is not the decision on their part to do nothing and say nothing itself a ''choice ?"511 In many families there will be some, will there not, who will consider a stand against euthanasia the only proof of love, devotion and gratitude for past events? What of the stress and strife if close relatives differas they did in the famous Sander case60 - over the desirability of euthanatizing the patient? At such a time, as the well-known Paight case clearly demonstrates,61 members of the family are not likely to be in the best state of mind, either, to make this kind of decision. Financial stress and conscious or unconscious competition for the family's estate aside: The chronic illness and persistent pain in terminal carcinoma may place strong and excessive stresses upon the family's emotional ties with the patient. The family members who have strong emotional attachment to start with are most likely to take the patient's fears, pains and fate personally. Panic often str.ikes them. Whatever guilt feelings they may have toward the patient emerge to plague them. If the patient is maintained at home, many frustrations and physical demands may be imposed on the family by the advanced illness. There may develop extreme weakness, incontinence and bad odors. The pressure of caring for the individual under these 58. The medical profession is apparently already quite sensitive about the "sue consciousness" on the part of the public. See Caswell, A Surgeo1fs Tlzoug!Jts 01J Jlalpractice, 30 Temple L.Q. 391 (1957) (symposium). There is good reason to think that "the greater incidence of suits and claims against physicians alleging medical malpractice and a greater financial success in prosecuting these" bas led to "insecurity'' on the part of many physicians, and "the insecure physician is going to play it safe." Wachowski and Stronach, The Radiologist awl Projessiotwl Jleciical Liability, 30 Temple L.Q. 398 (1957). Apparently, in some fields fear of claims and litigation has already set "the psychological stage for undertreatment." Id. at 399. 59. Cf. the examination of Sir Harold Scott, Commissioner of Police of the Metropolis by the Royal Commission on Capital Punishment, Minutes of Evidence 151 (Oct. 7, 1949): 1599. Nobody at present, except the law, bas to decide that a particular person should be sentenced to death, no individual?-No individual at present, except the Home Secretary, has to decide that a particular person sentenced to death must hang. 1600. The Home Secretary is in a different pc-sition, is be not? He does not primarily prescribe the death penalty; the law does that. The Home Secretary says whether or not he deems it right to interfere with the cou:~e of the .Jaw?-Yes1 that is the legal :;>osition. It is a different posttion, techmcally, but 1t seems to me that morally there really is no difference. The responsibility upon the Home Secretary is really to decide whether this man shill die or not die. The machinery may be by interference or non-interference \vitb the Jaw, but the responsibility to me seems the same. 60. See note 172, infra. See also the Jlohr case; supra note 17, where two brothers testified against a third who had eutbanatized a fourth. 61. See note 176, it•fra.

Death, Dying and the Ending of Life, 1/

178

992

MINNESOTA LAW REVIEW

[Vol. 42:969

circumstances is likely to arouse a resentment and, in turn, guilt feelings on the part of those who have to do the nursing.62 Nor should it be overlooked that while Professor Williams would remove the various procedural steps and the various personnel contemplated in the American and English Bills and bank his all on the "good sense" of the general practitioner, no man is immune to the fear, anxieties and frustrations engendered by the apparently helpless, hopeless patient. Not even the general practitioner: Working with a patient suffering from a malignancy causes special problems for the physician. First of all, the patient with a malignancy is most likely to engender anxiety concerning death, even in the doctor. And at the same time, this type of patient constitutes a serious threat or frustration to medical ambition. As a result, a doctor may react more emotionally and less objectively than in any other area of medical practice .•.• His deep concern may make him more pessimistic than is necessary. As a result of the feeling of frustration in his wish to help, the doctor may have moments of annoyance with the patient. He may even feel almost inclined to want to avoid this type of patient.68 The only Anglo-American prosecution involving an alleged mercy-killing physician seems to be the case of Dr. Herman Sander. The state's testimony was to the effect that, as Sander had admitted on various occasions, he finally yielded to the persistent pleas of his patient's husband and pumped air into her veins "in a weak moment." 6 ~ Sander's version was that he finally "snapped" under the strain of caring for the cancer victim,6 G bungled simple tasks,~111 and became "obsessed" with the need to "do something" for her- if 62. Zarling, supra, note 45 at 211-12. 63. I d. at 213-14. See also Dr. Benjamin Miller to the effect that cancer "can be a 'horrible experience' for the doctor too'' and that "a long difficult illness may emotionally exhaust the relatives and physician even more than the patient." Miller, .ntpra note 50, at 103; and Stephen, M11rder from the Best of Motives, 5 L.Q. Rev. 188 (1889), commenting on the disclosure by a Dr. Thwing that he had practiced euthanasia: "The boldness of this avowal is made particularly conspicuous by Dr. Thwing's express admission that the only person for whom the lady's death, if she had been allowed to die naturally, would have been in any degree painful was not the lady herself, but Dr. Thwing." 64. N.Y. Times, Feb. 24, 1950, p. 1, col. 6. 65. "As I looked at her face and all of the thoughts of the past went through my mind, something snapped in me, and I felt impelled or possessed to do something, and why I did it, I can't tell. It doesn't make sense." N.Y. Times, March 7, 1950, p. 19, col 1. 66. "I didn't use a tourniquet, which is also rather a ridiculous thing, because ordinarily in a normal patient we put on a tourniquet to bring up the vein so that we can see it. Her veins were collapsed anyhow and I couldn't have been thinking the way I ordinarily do at the time. Otherwise I wouldn't have acted this way." Ibid.

Death, Dying and the Ending of Life, II

1958)

"MERCY-KILLING" LEGISLATION

179

993

only to inject air into her already dead body.61 Whichever side one believes- and the jury evidently believed Dr. Sander68 - the case well demonstrates that at the moment of dedsion the tired practitioner's "good sense" may not be as good as it might be. Putting aside the problem of whether the good sense of the general practitioner warrants dispensing with other personnel, there still remains the problems posed by any voluntary euthanasia program : the aforementioned considerable pressures on the patient and his family. Are these the kind of pressures we want to inflict on any person, let alone a very sick person? Are these the kind of pressures we want to impose on any family, let alone an emotionallyshattered family? And if so, why are they not also proper considerations for the crippled, the paralyzed, the quadruple amputee, the iron lung occupant and their families? Might it not be said of the existing ban on euthanasia, as Professor Herbert Wechsler has said of the criminal law in another connection : It also operates, and perhaps more significantly, at anterior stages in the patterns of conduct, the dark shadow of organized disapproval eliminating from the ambit of consideration alternatives that might othenvise present themselves in the final competition of choice.89

C. The uHopelessly Incttrable'1 Patient and the Fallible Doctor. Professor Williams notes as "standard argument" the plea that "no sufferer from an apparently fatal illness should be deprived of his life because there is always the possibility that the diagnosis is wrong, or else that some remarkable cure will be discovered in time." 70 But he does not reach the issue until he has already dismissed it with this prefatory remark: 67. "[]Just the appearance of her face and the combination of all the thoughts of her long suffering and of her husband's suffering also-this expression on her face might have just touched me off and made me feel obsessed that I had to do something and what I did docs not make sense." Ibid. 68. See note 172 iiJ/ra, and accompanying text. 69. Wechsler, The Issues of the Nttremberg Tt·ial, 62 Pol. Sci. Q. 11, 16 (1947). Cf. Cardozo, What MediciJfe Ca" Do for Law, in Law and Literature 88-89 (1931) : Punishment is necessary, indeed, not only to deter the man who is a criminal at heart, who has felt the criminal impulse, who is on the brink of indecision, but also to deter others who in o•Jr existing social organization have never felt the criminal impulse and shrink from crime in horror. Most of us have such a scorn and loathing of robbery or forgery that the temptation to rob or forge is never within the range of choice; it is never a real alternative. There can be little doubt, however, that some of this repugnance is due to the ignominy that bas been attached to these and like offenses through the sanctions of the criminal law. If the ignominy were withdrawn, the horror might be dimmed. 70. Williams, p. 318.

Death, Dying and the Ending of Life, 1/

180

994

MINNESOTA LAW REVIEW

[Vol. 42:969

It has been noticed before in this work that writers who object to a practice for theological reasons frequently try to support their condemnation on medical grounds. With euthanasia this is difficult, but the effort is made.n Does not Williams, while he pleads that euthanasia not be theologically prejudged, at the same time invite the inference that nontheological objections to euthanasia are simply camouflage? It is no doubt true that many theological opponents employ medical arguments as well, but it is also true that the doctor who has probably most forcefully advanced medical objections to euthanasia of the so-called incurables, Cornell University's worldrenowned Foster Kennedy, a former president of the Euthanasia Society of America, advocates euthanasia in other areas where error in diagnosis and prospect of new relief or cures are much reduced, i.e., the "congenitally unfit". 72 In large part for the same reasons, Great Britain's Dr. A. Leslie Banks, then Principal Medical Officer of the Ministry of Health, maintained that a better case could be made for the destruction of congenital idiots and those in the final stages of dementia, particularly senile dementia, than could be made for the doing away of the pain-stricken incurable. 73 71. I d. at 317-18.

72. "What to do with the hopelessly unfit? I had thought at a youn~er time of my life that the legalizfug of euthanasia-a soft gentle-sounding word-was a thing to be encouraged; but as I pondered, and as my experience in medicine grew, I became less sure. Now my face is set against the legalization of euthanasia for any person, who, having been well, has at last become ill, for however ill they be, many get well and help the world for years after. But I am in favor of euthanasia for those hopeless ones who ·should never have been hom-Nature's mistakes. In this category it is, with care and knowledge, impossible to be mistaken in either diagnosis or prognosis." Kennedy, The Problem of Social Control of the Congenital Defective, 99 Am. J. Psycliiatry, 13, 14 (1942). "We doctors do not always know when a disease in a previously healthy person has become entirely incurable. But there are thousands and tens of thousands of the congenitally unfit, about whom no diagnostic error would be possible. •.. with nature's mistakes ••• there can be, after five years ••• of life, no error in diagnosis, nor any hope of betterment." Kennedy, Euthanasia: To Be or Not To Be, Colliers, May 20, 1939, pp, 15, SB; reprinted in Colliers, April 22, 1950, pp. 13, 51. At the February, 1939, meeting of the Society of Medical Jurisprudence, Charles E. Nixdorff, treasurer and board chairman of the Euthanasia Society of America stated that the case of a 19-year-old girl in Bellevue, with a broken back and paralyzed legs, who "prayed for death every night'' was sufficient reason for the Euthanasia Society "to carry on the fi~ht." "Dr. [Foster] Kennedy [then President of the Euthanasia Society], m conversation, said later he did not think that was a particularly gOod example. He sald he had known many such cases where the patients 'got around' and only recently he had 'danced with one.'" N.Y. Times, Feb. 14, 1939, p. 2, col. 6. 73. Banks, Etethanasia, 161 Practitioner 101, 106 (1948). According to him, neither "pain" nor "incurability'' "is capable of precise and final definition, and indeed if each case had to be argued in open court there would be conflict of medical opinion in practically every instance." ld. at 104.

Death, Dying and the Ending of Life, II

1958}

"MERCY-KILLING" LEGISLATION

181

995

Surely, such opponents of voluntary euthanasia cannot be accused of wrapping theological objections in medical dressing! Until the euthanasia societies of England and America had been organized and a party decision reached, shall we say, to advocate euthanasia only for incurables on their request, Dr. Abraham L. Wolbarst, one of the most ardent supporters of the movement, was less troubled about putting away "insane or defective people [who] have suffered mental incapacity and tortures •Jf the mind for many years" than he was about the "incurables".r4 He recognized the "difficulty involved in the decision as to in~bility" as one of the "doubtful aspects of euthanasia." Doctors are only human beings, with few if any supermen among them. They make honest mistakes, like ether men, because of the limitations of the human mind.76 He noted further that "it goes without saying that, in recently developed cases with a possibility of cure, euthanasia should not even be considered," that "the law might establish a limit of, say, ten years in which there is a chance of the patient's recovery." 76 Dr. Benjamin Miller is another who is unlikely to harbor an ulterior theological motive. His interest is more personal. He himself was left to die the death of a "hopeless" tuberculosis victim only to discover that he was suffering from a rare malady which affects the lungs in much the same manner but seldom kills. Five years and si."\.i:een hospitalizations later, Dr. Miller dramatized his point by recalling the last diagnostic clinic of the brilliant Richard Cabot, on the occasion of his official retirement : He was given the case records [complete medj:cal histories and results of careful examinations] of two patients and asked to diagnose their illnesses•... The patients had died and only the hospital pathologist lmew the exact diagnosis beyond doubt, for he bad seen the descriptions of the postmortem findings. Dr. Cabot, usually very accurate in his diagnosis, that day missed both. The chief pathologist who had selected the cases was a wise person. He had purposely chosen two of the most deceptive to remind the medical students and young physicians that even at the end of a long and rich experience one of the greatest diagnosticians of our time was still not infallible.77 74. Wolbarst, Legalize Euthanasia/, 94 The Forum 330, 332 (1935). But see Wolbarst, The Doctor Looks at Euthanasia, 149 Medicat Record, 3S4 (1939).

75. Wolbarst, Legalize Euthanasia!, 94 The Forum 330, 331 (1935). 76. Id. at 332, 77. Miller, .supra note 50, at 39.

'

Death, Dying and the Ending of Life, 1/

182

996

MINNESOTA LAW REVIEW

[Vol. 42:969

Richard Cabot was the John W. Davis~ the Jolln Lord 0 1Brian, of his profession. When one reads the account of his last clinic, one cannot help but think of how fallible the average general practitioner must be, how fallible the young doctor just starting practice must be--and this, of course, is all that some small communities have in the way of medical care--how fallible the worst practitioner, young or old, must be. If the range of skill and judgment among licensed physicians approaches the wide gap between the very best and the very worst members of the bar-and I have no reason to think it does not-then the minimally competent physician is hardly the man to be given the responsibility for ending another's life.78 Yet, under Williams' proposal at least, the marginal physician, as well as his more distinguished brethren, would have legal authorization to make just such decisions. Under Williams' proposal, euthanatizing a patient or two would all be part of the routine day's work. 79 Perhaps it is not amiss to add as a final note, that no less a euthanasiast than Dr. C. Killick Millard80 had such little faith in the average general practitioner that as regards the mere administering of the cott·P de grace, he observed : In order to prevent any likelihood of bungling, it would be very necessary that only medical practitioners who had been specially licensed to euthanise (after acquiring special knowledge and skill) should be allowed to administer euthanasia. Quite possibly, the work would largely be left in the hands of the official euthanisors, who would have to be appointed specially for each area.st True, the percentage of correct diagnosis is particularly high in cancer.82 The short answer, however, is that euthanasiasts most emphatically do not propose to restrict mercy-killing to cancer cases. Dr. Millard has maintained that "there are very many diseases be78. As to how bad the bad physician can be, see generally, even with a

grain of salt, 3 Belli, Modern Trials §§ 327-353 (1954). See also Regan, Doctor and Patient and the Law 17-40 (3d ed. 1956).

79. See note 41 .supra, and accompanying text 80. As Williams points out, p. 330, Dr. Millard introduced the topic of euthanasia into public debate in 1932 when he advocated that mercykilling should be legalized in his presidential address to the Society of Medical Officers of Health. In moving the second reading of the voluntary euthanasia bill, Lord Ponsonby stated that "the movement in favour of drafting a Bill" had "originated" with Dr. Millard. 103 H.L. Deb. 466-67 (1936). 81. Millard, The Case For Buthatzasia, 136 Fortnightly Review 701, 717 (1931). Under his proposed safeguards (two independent doctors, followed by a "medical referee") Dr. Millard viewed error in diagnosis as a non-deterrable "remote possibility." Ibid. 82. Euthanasia opponents readily admit this. See e.g., Miller, .supra note 50, at 38.

Death, Dying and the Ending of Life, II

1958]

"MERCY-KILLING" LEGISLATION

183

997

sides cancer which tend to kill 'by inches', and where death, when it does at last come to the rescue, is brought about by pain and e.-dlaustion." 83 Furthermore, even if mercy-killings were to be limited to cancer, however relatively accurate the diagnosis in these cases, here, too, "incurability of a disease is never more than an estimate based upon experience, and how fallacious experience may be in medicine only those who have had a great deal of experience fully realize." 84 Dr. Daniel Laszlo, Chief of Division of Neoplastic Diseases, Montefiore Hospital, New York City, and three other physicians have observed : The mass crowding of a group of patients labeled 'terminal' in institutions designated for that kind of care carries a grave danger. The experience gathered from this group makes it seem reasonable to conclude that a fresh evaluation of any large group in mental institutions, in institutions for chronic care, or in homes for the incurably sick, would unearth a rewarding number of salvageable patients who can be returned to their normal place in society.•.. For purposes of this study we were especially interested in those with a diagnosis of advanced cancer. In a number of these patients, major errors in diagnosis or management were encountered.85 The authors then discuss in considerable detail the case histories of eleven patients admitted or transferred to Montefiore Hospital alone with the diagnosis of "advanced cancer in its terminal stage/' none of whom had cancer at all. In three cases the organ suspected to be the primary site of malignancy was unaffected; in the other eight cases it was the site of some nonmalignant disease. The impact of these findings may be gleaned from a subsequent comment by Doctors Laszlo and Spencer: "Such cases [of mistaken diagnosis of advanced cancer] are encountered even in large medical centers and probably many more could be found in areas poorly provided with medical facilities."S 8 Only recently, Dr. R. Ger, citing case histories of false cancer 83. Millard, .mpra note 81, at 702. 84. Frohman, Vexing Problems in Forensic M'edicitJe: A Physiciat(s View, 31 N.Y.U.L. Rev. 1215, 1216 (1956). Dr. Frohman added: We practice our art with the tools and information yielded by laboratory and research scientists, but an ill patient is not subject to experimental control, nor are his reactions always predictable. A good physician employs his scientific tools whenever they are useful, but many are the times when intuition, chance, and faith are his most successful techniques. 85. Laszlo, et. al., Errors In Diagnosis Atul Ma11agement Of Cancer, 33 Annals Int. Med. 670 (1950). 86. Laszlo and Spencer, Medical Problems ln The Ma1wgement Of Cancer, 37 Med. Clin. N.A 869, 873 (1953).

Death, Dying and the Ending of Life, 1/

184

998

MINNESOTA LAW REVIEW

[Vol. 42:969

diagnoses to buttress his point, had occasion to warn his colleagues~

Students are often told, and one is exhorted repeatedly in textbooks to do so, to regard signs and symptoms aJ>pearing over the age of 40 years as due to carcinoma [malignant epithelial tumor] until proved otherwise. While it is true that carcinoma should take first place on grounds of commonness, it must not be forgotten that there are other conditions which may mimic carcinoma clinically, radiologically and at operation, and which are essentially benign. There is danger, moreover, when presented with a case simulating carcinoma to assume it to be carcinoma without proving or disproving the diagnosis. This may give rise to unnecessary fatalities by either denying treatment because of a hopeless prognosis or carrying out unnecessary procedures.81 Even more recently, Doctors De Vet and Walder scored the "extremely dangerous" tendency on the part of general practitioners and specialists alike "when a neoplasm becomes manifest in a patient previously operated on for a malignant tumour . . • to presume that the new growth is a metastasis [a transfer of the malignant disease] ."88 Their studies demonstrated that it is ''by no means a rare occurrence" for patients to develop ''another, benign tumour after having been operated upon for a malignant one."80 De Vet and Walder also stress the "remarkable similarity'' in symptoms, including "violent pain" in both cases, between metastases and benign processes of the spinal column and the spinal cord.00 Faulty diagnosis is only one ground for error. Even if the diagnosis is correct, a second ground for error lies in the possibility that some measure of relief, if not a full cure, may come to the fore within the life e.'qlectancy of the patient. Since Glanville Williams does not deign this objection to euthanasia worth more than a pass-

87. Ger, Diagnosis a1Jd Misdiag~:osis of Carci1J011UJ., 28 So. Afr. Med. J, 670 (1954). 88. De Vet and Walder, Pseudo-Metastases, 7 Archivium Chirurgicum Neerlandicum 78 (1955). 89. I d. at 83. 90. Id. at 82. Consider also the following: At the 1951 annual meeting of the American Cancer Society, devoted to cytologic diagnosis of cancer, Dr. Henry M. Lemon noted: Proceedi1Jgs, Symposium 01~ Esfoliativc Cytology at 106 (Oct 23-24, 1951): The problem of false positive diagnoses has always been a difficult one. About 5 per cent of the 541 non-cancer patients in whom cancer secre· tions have been studied in the past had false positive diagnosis made, and in our experience, gastritis has been a common cause of false positive diagnosis. At the same meeting, Dr. William A. Cooper told of "fifteen misses'' in X-ray gastric cancer diagnosis out of one hundred cases (id. at 102) : Four of the twenty-five cases of cancer were said to have benign lesions, while eleven of the seventy-five benign lesions were said to have cancer.

Death, Dying and the Ending of Life, II

1958]

"MERCY-KILLING" LEGISLATION

185

999

ing reference, 01 it is necessary to turn elsewhere to ascertain how it has been met. One answer is : It must be little comfort to a man slowly coming apart from multiple sclerosis to think that, fifteen years from now, death might not be his only hope.112 To state the problem this way is of course, to avoid it entirely. How do we know that fifteen days or fifteen /uJurs from now, "death might not be [the incurable's] only hope?" A second answer is: [N]o cure for cancer which might be found 'tomorrow' would be of any value to a man or woman 'so far advanced in cancerous toxemia as to be an applicant for euthanasia'.93 As I shall endeavor to show, this approach is a good deal easier to formulate than it is to apply. For one thing, it presumes that we know today what cures will be found tomorrow. For another, it overlooks that if such cases can be said to exist, the patient is likely to be so far advanced in cancerous toxemia as to be no longer capable of understanding the step he is taking and hence beyond the stage when euthanasia ought to be administered.9 ' A generation ago, Dr. Haven Emerson, then President of the American Public Health Association, made the point that "no one can say today what will be incurable tomorrow. No one can predict what disease will be fatal or permanently incurable until medicine becomes stationary and sterile." Dr. Emerson went so far as to say that "to be at all accurate we must drop altogether the term 'in91. See Williams, p. 318. 92. Pro & C01•: Shall We Legalize "Mercy Killing"!, Readers Digest, Nov. 1938, pp. 94, 96. 93. James, Etttlzamzsicv-Right or Wrong!, Survey Graphic, May, 1948, pp. 241, 243; Wolbarst, The Doctor Looks at Eutlunw.sia, 149 Medical Record, 354,355 (1939). 94. Thus, Doctor Millard, in his leading article, sttpra note 81, at 71(), states: A patient who is too ill to understand the sigiificance of the step he is taking has got beyond the stage when euthanasia ought to be administered. In any case his sufferings are probably nearly over. Glanville Williams similarly observes (pp. 342-44) : Under the bill as I have proposed to word it, the consent of the patient would be required, whereas it seems that some doctors are now accustomed to give fatal doses without consulting th·~ patient. I take it to be clear that no legislative sanction can be accorded to this practice, in so far as the course of the disease is deliberately anticipated. The essence of the measures proposed by the two societies is that euthanasia should be voluntarily accepted by the patient. ••• The measure here proposed is designed to meet the situation where the patient's consent to euthanasia is clear and incontrovertible.

Death, Dying and the Ending of Life, 1/

186

1000

MINNESOTA LAW REVIEW

[Vol. 42:969

curables' and substitute for it some such tenn as 'chronic illness'."811 That was a generation ago. Dr. Emerson did not have to go back more than a decade to document his contention. Before Banting and Best's insulin discovery, many a diabetic had been doomed. Before the Whipple-Minot-Murphy liver treatment made it a relatively minor malady, many a pernicious anemia sufferer had been branded "hopeless." Before the uses of sulfanilimide were disclosed, a patient with widespread streptococcal blood poisoning was a condemned man. 96 Today, we may take even that most resolute disease, cancer, and we need look back no further than the last decade of research in this field to document the same contention.97 Three years ago, Dr. William D. McCarthy presented the results to date, of an effort begun in 1950 to open a new approach in cancer palliation,98 a report whose findings of "remarkable improvement" in nearly a third of the cases invoked strong editorial comment in the New England Journal of Medicine. 99 At the time of Dr. McCarthy's report, 100 "hopeless" patients with a wide variety of neoplasms had been treated with a combination of nitrogen mustard and ACTH or cortisone. "All patients in the series were in advanced or terminal phases of disease, and were accepted for treatment only after the disease was determined to be progressive after adequate surgery or radiation therapy."100 Dr. McCarthy summarizes the results : In several of these cases there was associated tumor regression or arrest, with definite prolongation of life in increased comfort. This group constituted 15 per cent of the series. Reserved for the classification as excellent response were 16 additional patients (16 per cent) whose subjective and objective remissions were striking, often accompanied with tumor regression or arrest, and whose improvement persisted for six months or longer. These patients represent the true temporary remissions of the series. They are, however, temporary remissions and not

95. Emerson, Who Is Incurable? A Query and Repl:y, N.Y. Times, Oct. 22, 1933, § 8, p. 5, col. 1. 96. Ibid., Miller, supra note 50, at 39. 97. This is not to say that progress in the treatment of cancer cases has been limited to the last decade. Over twenty years ago1 Lord Horder, 103 H.L. Deb. 466, 492 (1936), opposing the euthanasia bill m the House of Lords debates, observed: [A]lthough it is common lmowledge that the essential causative factors of cancer still elude us, there are patients to-day suffering from this disease, not only living but free from pain, who would not have been living ten years ago, and this as the result of advances made in treatment 98. McCarthy, The Palliati01• and Remission of Cancer WitT• Combilled Corticosteroid and Nitroge1• M1tstard Therapy, 252 N. Eng. J. Med. 467 (1955). 99. Treatmmt of Advanced Cancer, 252 N Eng. J. Med. 502 (1955). 100. McCarthy, snpra note 98 at 468.

Death, Dying and the Ending of Life, II

1958]

"MERCY KILLING" LEGISLATION

187

1001

permanent remissions or so called 'cures.' Nevertheless, as a group originally considered hopeless, each has been afforded longer life, acceptable health and freedom from pain. Fortunately, prolongation of life appeared to occur only in patients who received good palliation. • • . Unusual temporary remissions for inter-,rals as long as three years were obtained..••101 Needless to say, a number of those who received substantial bene-

fits from this particular therapy were suffering from great pain and

appeared to be leading candidates for voluntary euthanasia. In 1950, the year the new combination therapy investigation was initiated, a swift death appeared to be their only hope. Instead they resumed full and useful lives for a considerable period of time.102 Since February, 1951, in a new effort to inhibit certain cancer growth,1°3 a number of advanced cancer patients at the Memorial Center for Cancer and Allied Disease have had their adrenal glands removed.10 i Of a total of ten patients with cancer of the prostate adrenalectomized at the time of the 1952 report, three died in the immediate postoperative period of various causes, leaving seven 101. I d. at 470, 475. Some of the results were little short of spectacular. See e.g., Case 1, id. at 470, the case of a woman whc·se reticulum-cell sarcoma "was considered too disseminated for radiation therapy" who responded so well to therapy that she returned to employment as a nurse for three years ; Case 3, ibid., that of a man taken to the hospital "in a terminal state" with "a massive lymphosarcoma of the pelvis" which had received X-ray therapy and which was increasing rapidly in size, who returned to his occupation and but for a short interval when he undenvent a second- course of therapy "continued working up to the time of his death ••• eighteen months after the 1st course of combination therapy'' ; Case 11, id. at 472-73, that of a stomachcancer victim "in a terminal condition, unable to re12in solids or fluids" who, after three months of the therapy, regained her normal weight, returned to her occupation and enjoyed excellent health for a full year. On the other band, some 40 per cent of the group were considered failures (those who died within a month and those who survived longer but received little benefit) ; 29 per cent were classed as fair in response (moderate but brief palliation). Id. at 470. 102. See also Ra.vich, Eutha1rasia aud Pai1~ Iu Cancer, 9 Unio Internationalis Contra Cancrum 397 (1953), a report of the promising experi· mental chemotherapeutic measures (n-Butanol, glycerine and sodium tbirosulfate) of Dr. Emanuel Revici and the staff of the Institute of Applied Biology. A number of patients whose cancers "had advanced beyond the point where any help was to be anticipated from surgery, X-ray or radium, according to the opinions of the attending physicians," id. at 398, returned to their normal occupations after the onset of treatment and remained on the job for several years. 103. Drs. Huggins and Scott bad reported the first total bilateral adrenalectomies in patients with prostatic carcinoma in 1945, but smce cortisone was not then available all patients died in adrenal insufficiency. The authors therefore concluded at that time that the operation was not practical and temporarily abandoned this approach. See Huggins and Scott, Bilateral Adre11alectom:v In Prostatic Cancer: Clinical Features and Uritwry E~cretion of 17 Ketosteroids and Estroge1•, 122 Annals of Surgery 1031 (1945). 104. West, et. al., The Effect of Bilateral Adre1wlectom:v Upon Neoplastic Disease i1J Man, 5 Cancer 1009 (1952).

Death, Dying and the Ending of Life, 1/

188

1002

MINNESOTA LAW REVIEW

[Vol. 42:969

effective cases for evaluation : The most striking beneficial response to adrenalectomy was relief of pain. Three of the patients were confined to bed with pain prior to surgery and were taking narcotics frequently.... All three had striking relief of pain postoperatively and became ambulatory. One (J.W.) was in a stuporous condition preoperatively, confined to bed, and unable to feed himself. Following adrenalectomy his general condition improved remarkably. He became ambulatory and was able to return home to live a relatively normal life. This improvement has been maintained until the present, 218 days after surgery.... Summarizing the prostatic cancer cases, all seven effective cases had striking subjective improvement. Only two cases showed objective improvement. Improvement was temporary in all cases.105 From all indications "J.W." was a most attractive target for the euthanasiasts. He was suffering from "severe pain requiring frequent injections of narcotics for relief ... was extremely lethargic and relatively unresponsive . . . had to be fed by the nursing staff." 106 If he, to use Dr. Wolbarst's words, was not "so far advanced in cancerous toxemia as to be an applicant for euthanasia," when will anybody be? I am not at all sure that at this point J.W. was still capable of consenting to his death. If he were, he certainly had reached the very brink. As it turned out, however, to have put J.W. out of his misery at the time would have been to deprive him of over seven months of a "relatively normallife." 107 Adequate quantities of cortisone and other active corticoids had just become available. The postoperative problem of adrenal insufficiency had just been solved. Breast cancer, the most common cancer in woman, 108 has also yielded substantially to adrenalectomy. A recent five-year evaluation of 52 consecutive patients with metastatic mammary cancer who underwent adrenalectomy disclosed that significant objective 105. Id. at 1012-13. Dr. M. P. Reiser of the University of Minnesota Medical School and his colleagues have planted radon-filled seeds of gold into the prostate area in an effort to save patients with "inoperable" cancer of the prostate gland. As a result, thirteen of twenty-five patients have lived at least a year; six have lived from three to seven years. Radon is the gas of radium. See Cohn, 'U' Reports Victories Over Cancer, Minneapolis Morning Tribune, April 4, 1958, p. 13, col. 4. 106. West, SIIPra note 104 at 1010. 107. An addendum to the report discloses that J.W.'s post-operative "subjective improvement" lasted 220 days and that he survived for 294 days, id. at 1016-17. What pain J. W. suffered in his last days is not revealed, but in general discussion the authors state that ". . • [I]n the majority of the cases, the pain never did return to its preoperative intensity even though the patient later died of cancer." Id. at 1015. 108. American Cancer Society, 1958 Cancer Facts and Figures 17.

Death, Dying and the Ending of Life, II

1958]

"J,fERCY-KILLING" LEGISLATION

189

1003

remJssJons of varying lengths of time occurred in 20 patients.109 Prolonged survival-from three years to 68 months-occurred in seven of these patients, all of whom had been suffering from advanced stages of the disease, had failed to respond to various other types of therapy and were incapacitated. After treatment, «all of them were able to resume their normal physical activities."110 One of the seven had had such extensive metastases that she "appeared to be moribund," but she survived, with great regression of the neoplasm, more than five years after adrenalectomy.111 The pituitary gland, as well as the adrenal glands, has had an increasing apparent role in the control of breast cancer. Since 1951, the availability of ACTH and cortisone has allowed an intensive investigation of the effects of hypophysectomy, i.e., surgical removal of the pituitary body. The results have been most gratifying. Arecent report, for example, discloses that o£ twenty-eight patients with advanced breast cancer who underwent total hypophysectomy, "eighteen ... have demonstrated striking objective clinical regressions" up to twenty months while an additional four who showed no objective evidence of regression experienced "striking relief of pain."112 The dynamic state of current cancer r~earch would appear to be amply demonstrated by the indication, already, that in the treatment of advanced breast cancer, adrenalectomy, itself still in the infant stages, may yield to hypophysectomy.113 109. Dao and Huggins, Metastatic Cancer of the Breast Treated by Admzalectomy, 165 A.M.A.]. 1793 (1957). Furthermore, an additional nine patients who tmderwent no demonstrable regression experienced marked objective impro1•ement in relief of bone pain, disappearance of respiratory symptoms and return of a sense of wellbeing, An earlier report on adrenalectomy disclosed that of five "effective" breast carcinoma cases, a sixth having died of other causes a short time after undergoing the operation, "all had severe pain preoperatively, and all had either partial or complete relief of pain follo,ving adrenalectomy." West, s11pra note 101,. at 1014. 110. Id. at 179o. 111. Ibid. 112. Kennedy, French and Peyton, Hypophysectomy in Advm~ced Breast Cancer, 255 N. Eng. J. Med., 1165, 1171 (1956). See also Kennedy, The Prcse1~t Status of Hormone Therapy it~ Advanced Breast Cancer, 69 Radiology 330, 333-34 (1957). For earlier reports, see Luft and Olivecrona, llypophysectomy 11J Mat~: Erpcricnces it~ Metastatic Cancer of the Breast, & Cancer 261 (1955) (13 of 37 patients showed subjective or objective improvement for from 3 to Z1 months); Pearson, et. al., Hypophysectomy n~ Treatment of Advanced Cancer, A.M.A.]. 17 (1956) (over half of 41 patients who could be evaluated underwent objective remissions). 113. "In view of the favorable responses after hypophysectomy, the concomitant adrenal atrophy and the ease in managing the patient, it appears that hypophysectomy is to be preferred over adrenalectomy in the treatment of advanced breast cancer." Kennedy, French and Peyton, supra note 112, at 1171.

190

Death, Dying and the Ending of Life, 11

1004

MINNESOTA LAW REVIEW

[Vol 42:969

Tru~ many types of cancer still run their course virtually unhampered by man's arduous efforts to inhibit them. But the nwnber of cancers coming under some control is ever increasing. With medicine attacking on so many fronts with so many weapons who would bet a man's life on when and how the next type of cancer will yield, if only just a bit ('1U Tru~ we are not betting much of a. life. For even in those areas where gains have been registered, the life is not "saved," death is only postponed. Of cours~ in a sense this is the case with every "cure" for every ailment. But it may be urged that after all there is a great deal of difference between the typical "cure" which achieves an indefinite postponement, more or less, and the cancer respite which results in only a brief intermission, so to speak, of rarely more than six months or a year. Is this really long enough to warrant all the bother? Well, how long is long enough? In many recent cases of cancer respit~ the patient, though experiencing only temporary relief, underwent sufficient improvement to retake his place in society.1111 Six or twelve or eighteen months !s long enough to do most of the

114. In addition to the various approaches to the cancer problem discussed .supr~ consider, e.g., the followmg items which have appeared in the daily newspapers the past few months : (1). In April of 19581 s?entists uncovered a new chemical compoundfluorine combined with a oooy compound used by cancer cells for growthwhich inhibits the growth of cancer cells. The discovery was bailed as a major step in the search for a medical "magic bullef' which can kill cancer cells outright. New York Times, Apn1 4, 1958, p. 23, col. 7; Minneapolis Morning Tribune, April 4, 19581 p. 14, col. 5. (2). Neutron radiation on Drain cancer ~atients has led to "significant'' increases in length of life, according to Dr. William H. Sweet of the Harvard Medical School. This September, Dr. Sweet will use an atomic reactor in an unprecedented effort to remove all remnants of brain cancer from a patient. Cohn, Brain Cancer Surgeons Will Use Atomic Reactor, Minneapolis Morning Tribune, March 30, 1958, p. 1, col. 1. (3). There is reason to think that neurosome surgery, sound waves focussed on precise spots inside the brain may prove valuable in treating brain cancers--with a dosage devised to klll only cancer cells. Palsy victims for as long as 35 years hlive been relieved by such treatment. New York Times, April 2, p. 33, col. 8; Minneapolis Morning Tribune, April 2, 1958, p.8, col. 5. {4). Dr. Roy Hertz an expert of the National Cancer Institute, has disclosed that a drug called methOtrexate has suppressed all evidence of a type of cancer occurring in woman during pregnancy, but the ''full value of the treatment remains to be determined." New York Times, Feb. 29, 1958, p. 62, col.4. (5). Dr. L. M. Tocantins of Jefferson Medical College has been conducting experiments to combat leukemia with whole-body X-ray doses calculated to kill the sick bone marrow cells that are producing the illness. Good marrow, taken from the bones of volunteers, is then injected into the patients. Such a technique has reversed leukemia's course in mice and given some of them normal life spans. Cohn, The'J Give Ribs to Fight Leukemia, Minneapolis Morning Tribune, March 26, 1958, p. 1, col. 4. 115. See notes 101, 102, 105, 109 .supra.

191

Death, Dying and the Ending of Life, II

1958]

"MERCY-KILLING" LEGISLATION

1005

things which socially justify our existence, is it not? Long enough for a nurse to care for more patients, a teacher to impart learning to more classes, a judge to write a great opinion, a novelist to write a stimulating book, a scientist to make an important discovery and, after all, for a factory hand to put the wheels on another year's Cadillac.

D. uMistakes Are Always Possible''. Under Professor Williams' "legislative suggestion" a doctor could "refrain from taking steps to prolong the patient's life by medical means" solely on his own authority. Only when disposition by affirmative "mercy-killing" is a considered alternative need he do so much as, and only so much as, consult another general practitioner.ue There are no other safeguards. No "euthanasia referee/' no requirement that death be administered in the presence of an official witness, as in the English society's bill. No court to petition, no committee to investigate and report back to the court, as in the American society's bill. Professor Williams' view is: It may be allowed that mistakes are always possible, but this is so in any of the affairs of life. And it is just as possible to make a mistake by doing nothing as by acting. All that can be expected of any moral agent is that he should do his best on the facts as they appear to him.m

-

That mistakes are always possible, that mistakes are always made, does not, it is true, deter society from pursuing a particular line of conduct-if the line of conduct is c01npelled by needs which override the risk of mistake. A thousand Convicting the Innocent'sl18 or Not Guilty's uo may stir us, may spur us to improve the administration of the criminal. law, but they cannot and should not bring the business of deterring and incapacitating dangerous criminals or would-be dangerous criminals to an abrupt and complete halt. Professor Williams points to capital punishment, as proponents of euthanasia are fond of doing,120 but defenders of this practice do not-as, of course, they cannot-rest on the negative argwnent that "mistakes are always possible." Rightly or wrongly, 116. For a discussion of the legal significance of "mercy-killing'' br, omission and Williams' consultation feature for affirmative "mercy-killing, ' see note 41 supra. 117. Williams p. 318. 118. Borchard, Convicting the Innocent (1932). 119. Frank and Frank, Not Guilty (1957). 120. See, e.g., Fletcher, Morals and Medicine, 181, 195-96 (1954) ; Millard, The Case For Eutlwuasia, 136 Fortnightly Review 701, 717 (1931); Potter, The Case for Et~tlwuasia, Reader's Scope, May 1947, pp. 111, 113.

Death, Dying and the Ending of Life, 1/

192

1006

MINNESOTA LAW REVIEW

[Vol. 42:969

they contend that the deterrent value of the death penalty so exceeds that of life-imprisonment or long-term imprisonment that it is required for the protection of society, that it results in the net gain of a substantial number of human lives.121 This is generally regarded as the "central" or "fundamental" question in considering whether the death penalty should be abolished or retained.122 This, as Viscount St. Davids said of a House of Lords debate on capital punishment which saw him advocate abolition, "was what the whole debate was about. " 128 Presumably, when and if it can be established to the satisfaction of all reasonable men that the deterrent value of capital punishment as against imprisonment is nil or de minimus, mistakes will no longer be tolerated and the abolitionists will have prevailed over the few remaining retentionists who would still defend capital punishment on other grounds.124 In any event, it is not exactly a show of strength for euthanasiasts to rely on so battered and shaky a practice as capital punishment.125 121. See generally Royal Commission on Capital Punishment, Report, Cmd. No. 8932, paras. 55-68 (1949-53); Michael and Wechsler, Criminal Law And Its Administration, 235-62 (1940). 122. H. L. A Hart, Murder And The Principle Of PrmishtnC?~t: England And The United States, 52 N.W. U. L. Rev. 433, 446, 455 (1957). See also, e.g., Bye, Capital Punishment in the United States 31 {1919) ; Gardiner, Capital Punislunent as a Deterrent: And the Alternative 17, 22 (1956); Caldwell, Why Is the Death Penalty Retained! 284 Annals Am. Acad. Pol. & Soc. Sci. 45, 50 (1952). 123. "I believed that the figures showed that if you abolish capital punislunent you do not, in fact, lose more human lives. Other noble Lords took the opposite view; they believed that if capital punislunent were abolished we should lose more lives. Both sides, however, believed that there is an ultimate value in human life. That was what the whole debate was about." 169 H.L. Deb. (5th ser.) 551, 591 (1950). 124. The remaining pockets of resistance would be manned by those who would utilize the death penalty as an instrument of vengeance, as a device for placing a special stigma on certain crimes, and as a means of furnishing the criminal with an extraordinary opportunity to repent before e.xecution. See the discussion in the Royal Commission Report, snpra note 121, at paras. 52-54. 125. Books attacking the utilization of the death penalty include Bye, Capital Punislunent In the United States (1919) ; Calvert, Capital Punish~ ment In The Twentieth Century (4th ed. 1930); Frank and Frank, Not Guilty Z48 (1957); Gardiner, Capital Punislunent As A Deterrent: And the Alternative (1956) ; Koestler, Reflections on Hanging (1956) ; Lawes, Twenty Thousand Years In Sing Sing, 291-337 (1932); Weihofen, The Urge To Punish 146-70 (1956). In February, 1956, the House of Commons on a free vote of 292 to 246 passed a resolution calling for the abolition or suspension of the death penalty which stated in part that "the death penalty for murder no longer accords with the needs or true interests of a civilized society'' 548 H.C. Deb. (5th ser.) 2556, 2652, 2655, (1956). The House of Lords, however, rejected the legislation passed in the spirit of this resolution. See H.L.A Hart, Sltpra note 122, at 434. Bertrand Russell recently commented (SitPra note 2, at 385) : I have not the relevant statistics, but I think if a poll had been taken [of the House of Lords in 1936] it would have been found that most of those who objected to euthanasia favoured capital punislunent, the dominant

Death, Dying and the Ending of Life, II

1958]

"MERCY-KILLING" LEGISLATION

193

1007

A relevant question, then, is what is the need for euthanasia which leads us to tolerate the mistakes, the very fatal mistakes, which will inevitably occur? What is the compelling force which requires us to tinker with deeply entrenched and almost universal128 precepts of criminal law? Let us first examine the qualitative need for euthanasia: Proponents of euthanasia like to present for consideration the case of the surgical operation, particularly a highly dangerous one: risk of death is substantial, perhaps even more probable than not; in addition, there is always the risk that the doctors have misjudged the situation and that no operation was needed at all. Yet it is not unlawful to perform the operation.121 The short answer is the witticism that whatever the incidence of death in connection with different types of operations "no doubt, it is in all cases below 100 per cent, which is the incidence rate for euthanasia."128 But this may not be the full answer. There are occasions where the law permits action involving about a 100 per cent incidence of death, for example, self-defense. There may well be other instances where the law should condone such action, for e....:ample, the "necessity" cases illustrated by the overcrowded lifeboat,129 the starving survivors of a ship-wreck,130 and-perhaps best of all-by Professor Lon Fuller's penetrating and fascinating tale of the trapped cave explorers.131 In all these situations, death for some may well be excused, if not justified, yet the prospect that some deaths ·will be unnecessary is a real one. He who kills in self-defense may have misjudged the facts. They who throw passengers overboard to lighten the load may no sooner do so than see umasts and sails of rescue • . . emerge out of the fog". 132 But no human being will ever find himself in a situation where he knows for an absolute certainty that one consideration in each case being faithfulness to tradition. Perhaps, but I would speculate further that if such a poll had been taken, it may well have been found that most of those who favored euthanasia objected to capital punishment And on such grounds as the irrevocability of the death sentence and the inevitable incident of error in the selection of its victims, the insufficient showing that such a drastic method is needed, and, perhaps, the sanctity of life. 126. See Silving, S11Pra. note 7. 127. See, e.g., Fletcher, op. cit. Stepra note 3, at 198; Euthanasia Society of America, Merciful Release, art. 7; Millard, Sttpra note 81 at 717. 128. Rudd, Entlzanasia, 14 ]. Clin. & Exper. Psychopath. 1, 4 (1953). 129. See United States v. Holmes, 26 Fed. Cas. No. 15,383 (C.C.E.D. Pa.1842). 130. See Regina v. Dudley & Stephens, 14 Q.E:.D. 273 (1884). 131. Fuller, The Case of the Speltmcean Explorers, 62 Harv. L. Rev. 616 (1949). 132. Cardozo, Wlzat Medicine Ca1~ Do For Law, in Law and Literature 113 (1931).

Death, Dying and the Ending of Life, 1/

194

1008

MINNESOTA LAW REVIEW

[Vol. 42:969

or several must die that he or others may live. "Modem legal systems ..• do not require divine lmowledge of human beings.''18 ' Reasonable mistakes, then, may be tolerated if as in the above circumstances and as in the case of the surgical operation, these mistakes are the inevitable by-products of efforts to save one or more human lives.18' The need the euthanasiast advances, however, is a good deal less compelling. It is only to ease pain. Let us next examine the quantitative need for euthanasia: No figures are available, so far as I can determine, as to the number of say, cancer victims, who undergo intolerable or overw whelming pain. That an appreciable number do suffer such pain, I have no doubt. But that anything approaching this number whatever it is, need suffer such pain, I have-viewing the many sundry 133. Hall, General Principles of Criminal Law, 399 (1947). Cardozo, on the other hand, seems to say that absent such certainty it is wrong for those in a "necessity" situation to escape their plight by sacrificing any life. Cardozo, supra note 132, at 113. On this point, as on the whole question of "necessity," his reasoning, it is submitted, is paled by the careful and inten· sive analyses found in Hall, snpra, at 377-426, and Williams, op. cit. supra note 4, at 577-586. See also Cahn, The Moral Decision (1955). Although he takes the position that in the Holmes' situation, "if none sacrifice themselves of free will to spare the others-they must all wait and die together," Calm rejects Cardozo's view as one which "seems to deny that we can ever reach enough certainty as to our factual beliefs to be morally justified in the action we take." Id. at 70-71. Some time after this paper was in galley, Section 3.02 of the Model Penal Code (Tent. Draft No. 8 1958) made its appearance. This section provides (unless the legislature has otherwise spoken) that certain "necessity'' killings shall be deemed justifiable so long as the actor was not "reckless or negligent in bringing about the situation requiring a choice of evils or in appraising the necessity for his conduct." The section only applies to a situation where "the evil sought to be avoided by such conduct is greater than that sought to be prevented by the law," e.g. killing one that several may live. The defense would not be available, e.g., '1to one who acted to save himself at the expense of another, as by seizing a raft when men are shipwrecked." Cotmmmt to Section 3.02, id. at 8. For "in all ordinary circumstances lives in being must be assumed • • • to be of equal value, equally deserving the protection of the law." Ibid. 134. Cf. Macauley, Notes on the Indian Penal Code, Note B, p. 131 (1851), reprinted in 7 The Miscellaneous Works of Lord Macauley 252 (Bibliophile ed.): It is often the wisest thing that a man can do to expose his life to great hazard. It is often the greatest service that can be rendered to him to do what may very probably cause his death. He may labor under a cruel and wasting malady which is certain to shorten his life, and which renders his life, while it lasts, useless to others and a torment to himself. Suppose that under these circumstances he, undeceived, gives his free and intelligent consent to take the risk of an operation which in a large proportion of cases has proved fatal, but which is the only method by which his disease can possibly be cured, and which, if it succeeds, will restore him to health and vigor. We do not conceive that it would be expedient to punish the surgeon who should perform the operation, though by performing it he might cause death, not intending to cause death, but knowing himself to be likely to cause it

Death, Dying and the Ending of Life, II

1958]

"MERCY-KILLING" LEGISLATION

195

1009

palliative measures now available135-considerable doubt. The whole field of severe pain and its management in the terminal stage of cancer is, according to an eminent physician, "a subject neglected far too much by the medical profession."1:16 Other well-qualified commentators have recently noted the "obvious lack of interest in the literature about the problem of cancer pain"181 and have scored "the deplorable attitude of defeatism and therapeutic inactivity found in some quarters."188 The picture of the advanced cancer victim beyond the relief of morphine and like drugs is a poignant one, but apparently no small 135. The management of intractable pain in cancer may be wouped under two main categories : (1) measures which check, decrease or eliminate the growth itself, (2) symptomatic treatment, i.e., control of the pain without affecting the growth. In the first category are palli~tive operations for cancers no longer curable; radiation, roentgen and X-ray therapy; administration of endocrine substances, steroids, nitrogen mustards, and radioactive iodine and iron. See text, at notes 98-113 supra. In the second category are nonnarcotic analgesics such as cobra venom, hypnotics and sedatives ; narcotic analgesics, such as morphine, codeine, methadone and, recently, chlorpromazine j neurosurgical operations, such as rhizotomy, the technique of choice in the management of cancer pain of the head and neck, spinothalmic tractotomy and chordotomy, for relief of pain at or below the nipple line; and prefrontal lobotomy. The various measures sketched above are discussed at considerable length in Bonica and Backup, Control of Cancer Pain, 54 Nw. Med. 22 (1955); Bonica, The Matusget1rent of Cancer Pai1~, G. P., Nov. 1954, p. 35, and more extensively by Doctors Scbiffrin and Gross (Systematic Analgetics), Sadove and Balogot (Nerue Blocks For Pain In Malignancy), Sugar (Neurosurgical Aspects of Pai1~ Management'), Taylor and Schiffrin (Hftmoral and Chemical Palliation of Maligtzancy) Schwarz (Surgical Procedures In CotJtrol Of Pain In Advatu:ed Cancer) and Carpender (Radiation Therapy In The Relief Of PaitJ 11~ Malignant Disease) in The Management Of Pain In Cancer (Schiffrin ed. 1956). Relief of pain by nerve blocking "has a great deal more to offer' than prolonged narcotic therapy. Effective blocks produce adequate relief of pain and enable these sufferers to receive more intensive radiation therapy and other forms of medical treatment which otherwise could not be tolerated.'' Bonica and Backup, supra at Zl; Bonica, supra., at 43. "A recent analysis of cases reported in the literature revealed that of the many patients treated by alcohol nerve blocking, 63 per cent obtained complete relief, 23.5 per cent obtained partial relief, and only 13.5 per cent received no benefits from the blocks!' Bonica, suPra, at 43. "Chordotomy is perhaps the most useful and most effective neurosurgical operation for the relief of cancer pain. When skillfully carried out in properly selected patients, it produces complete relief in about 65 per cent of the patients, partial relief in another 25 per cen!, and no relief in approximately 10 per cent" Bonica and Backup, supra at 2;,. Prefrontal lobotomy is a radical procedure which many regard as a last resort Bilateral prefrontal lobotomy almost always produces striking changes in the patient's personality, frequently impairing judgment and causing apathy; the mental changes produced by unilateral lobotomy are much less marked, but pain is likely to recur if the patient survives more than several months. See Sugar, .nepra, at 101-104; Bonica, .nepra at 41-42. 136. Foreword by Dr. Warren H. Cole in Management of Pain In Cancer (Schiffrin ed.l956). 137. Bonica and Backup, .nepra. note 135, at 20. The senile psychoses constituted almost a 400% numerical increase and a 155% increase in the proportion of total first admissions since 192(). Malzberg, A Statistical Review of Mental Disorders in Later Life, in Mental Disorders in Later Life 13 (Kaplan ed. 1956). Dr. George S. Stevenson classes both psychoses together as "mental illness of aging" : "As a rule these patients have very limited prospect of recovery. Iti fact, they die on the average within fifteen months after admission to a mental hospital." Stevenson, Mental Health Planning For Social Action 41 (1956). 200. U.S. Dep't of Health, Education and Welfare, Patients in Mental Institutions 1955, Part II, Public Hospital for the Mentally Ill 21. Some 13,972 were cerebral arteriosclerosis cases; 9,589 had senile brain diseases. 201. See note 143 supra. 202. Banks, Euthatwsia, 26 Bull. N.Y. Acad. Med. 297, 305 (1950). 203. "Mental diseases are said to be responsible for as much time lost in hospitals as all other diseases combined." Boudreau, Mental Health: The New Public Health Froutier, 286 Annals Am. Acad. Pol. & Soc. Sci. 1 (1953). As of about ten years ago, there were "over 900,000 patients under the care and supervision of mental hospitals." Felix and Kramer, Estent of the Problem of Met~tal Disorders, id. at 5, 10. Taking only the figures of persons sufficiently ill to warrant admission into a hospital for long-term care of psychiatric disorders, "at the end of 1950 there were 577,000 patients .•• in all long-term mental hospitals." Id. at 9. This figure represents 3.8

Death, Dying and the Ending of Life, II

1958]

"MERCY-KILLING" LEGISLATION

213

1027

When Williams turns to the plight of the "hopelessly defective infants/' his characteristic vim and vigor are, as in the senility discussion, conspicuously absent: While the Euthanasia Society of England has never advocated this, the Euthanasia Society of America did include it in its original program. The proposal certainly escapes the chief objection to the similar proposal for senile dementia: it does not create a sense of insecurity in society, because infants cannot, like adults, feel anticipatory dread of being done to death if their condition should worsen. Moreover, the proposal receives some support on eugenic grounds, and more importantly on humanitarian grounds-both on account of the parents, to whom the child will be a burden all their lives, and on account of the handicapped child itself. (It is not. howev•:!r, proposed that any child should be destroyed against the wishes of its parents.) Finally, the legalization of euthanasia for handicapPed children would bring the law into closer relation to its pracrtical adminis~ tration, because juries do not regard parental mercy-killing as murder. For these various 'reasons the proposal to legalize humanitarian infanticide is put forward from time ;to time by individuals. They remain in a very small minority, and the proposal may at present be dismissed as politically insignifi-

cant,2o'

It is understandable for a reformer to limit his present proposals for change to those with a real prospect of success. But it is hardly reassuring for Williams to cite the fact that only "a very small minority" has urged euthanasia for "hopelessly defective infants" as the only reason for not pressing for such legislation now. If, as Williams sees it, the only advantage voluntary euthanasia has over the involuntary variety lies in the organized movements on its bebal£, that advantage can readily be wiped out. In any event, I do not think that such "a very small minority" has advocated "humanitarian infanticide." Until the organization of the English and American societies led to a concentration on the voluntary type, and until the by-products of the Nazi euthanasia program somewhat embarrassed, if only temporarily, most proponents of involuntary euthanasia, about as many writers urged one type as another. 20G Indeed, some euthanasiasts have taken considerper 1,000 population, and a "fourfold increase in number of patients and a

twofold increase in ratio of patients to general population since 1903." Ibid, "During 1950, the state, county and city mental hospitals spent $390,000,000 for care and maintenance of their patients. I d. at 13. 204. Williams, pp. 349-50. 205. In Turano, Murder by Req2eest, 36 Am. :M:ercury 423 (1935), the author goes considerably beyond the title of his paper. He scores the "barbarous social policy'' which nurtures "infant monstrosities and hopelessly injured children for whom permanent suffering is the sole joy of living" and "old men and women awaiting slow extinction from the accumulated

Death, Dying and the Ending of Life, 11

214

1028

MINNESOTA LAW REVIEW

[Vol. 42:969

able pains to demonstrate the superiority of defective infant euthanasia over incurably ill euthanasia.206 ailments of senility:' id. at 424, and notes in his discussion of "permissive statutes" that "when the sufferer is not mentally competent, the decision could be left to n~ relatives," id. at 428. In Should They Live!, 7 American Scholar 454 (1938) Dr. W. G. Lennox refers to the congenital idiots, the incurably sick, the mentally ill and the aged as "that portion of our population which is a heavy and permanent liability," id. at 457, and agrees with others that "there is somewhere a biological limit to altruism, even for man," id. at 458. Dr. Lennox would presently eliminate "only the idiots and monsters, the criminal permanently insane and the suffering incurables who themselves wish for death," id. at 464. W. W. Gregg similarly advocates euthanasia for all "criminally or hopelessly insane," The Right to Kill, 237 No. Am. Rev. 239, 247 (1934) and concludes (id, at Z49): With the coming of a more rational social order ••• it is possible to foresee the emergence of a socialized purpose to eliminate such human life as shows itself conspicuously either inhuman, or unhuman, or unable to function happily; in order thereby to help bring about a safer and fuller living for that normal humanity which holds the hope of the future. W. A. Shumaker, in Those U1~fit to Live, 29 L.N. 165, 166-67 (1925) comments: Could we but devise an acceptable formula, ten thousand idiots annually put to death by state boards of health would mean no more to us than ten thousand pedestrians annually put to death by automobilists do now.

*

* *sense reason why an absolute idiot It is impossible to give a common should be permitted to live. His life is of no value to him or to anyone else, and to maintain its existence absorbs a considerable part of the life of a normal being. Of course one shrinks at the thought of putting him to death. But why is it that we shrink? And why, though we shrink from such an act, do we find it possible to excuse him who does it?

*

*

Is the balance swinging too * far toward overconsideration not only for the idiot but for the moron and the lunatic and too little consideration for the normality on which civilization must rest? In 1935, Dr. Alexis Carrel, the Rockefeller Institute's famed Nobel Prize winner, took the position that "not only incurables but kidnapers, murderers, habitual criminals of all kinds, as well as the hopelessly insane, should be quietly and painlessly disposed of." Newsweek Nov. 16, 1935, p. 40; Time, Nov.18,1935, p. 53; Pro and C01~: Right a1w Wrong of Mercy Killing, 1 The Digest ZZ (1937). Another debate on mercy killing, Pro & Con: Shall We Legali::e "Mercy Killi1111", 33 Readers Digest, Nov. 1938, p, 94 similarly embraced involuntary situations. The "question presented'' was : Should physicians have the legal privilege of putting painlessly out of their sufferings unadjt4stably defective i1~fants, patients suffering from painful and incurable illness and the hopelessly insane and jeeble-mi1uled provided, of course, that maximum legal and professional safeguards against abuse are set up, including the consent of the patient whC1J ratio1wl and adult? (Emphasis added.) The proponents of euthanasia made the pitch for voluntary euthanasia, then shifted (p. 95): Euthanasia would also do away with our present savage insistence that some of us must live on incurably insane or degraded by the helplessness of congenital imbecility. For the results of a 1937 national poll on the question which covered the problem of "infants born permanently deformed or mentally handicapped" as well as "persons incurably and painfully ill" see note 207 infra, and accompanying text. 206. Dr. Foster Kennedy believes euthanasia of congenital idiots has two major advantages over voluntary euthanasia (1) error in diagnosis and

Death, Dying and the Ending of Life, II

1958]

"MERCY-KILLING" LEGISLATION

215

1029

As for dismissing euthanasia of defective infants as "politically insignificant," the only poll that I know of which measured the public response to both types of euthanasia revealed that 45% favored eutlza11asia for defective infants ttnder certain conditions while only 37.3% approved euthanasia for the incttrably and painfully ill under any co1ulitions.201 Furthermore, of those who favored the mercy killing cure for incurable adults, some 40% would require only family permission or medical board approval, but not the patient's permission.208 Nor do I think it irrelevant that while public resistance caused Hitler to yield on the adult euthanasia front, the killing of malformed and idiot children continued unhindered to the end of the war, the definition of "children" expanding all the while. 209 Is it the embarrassing experience of the Nazi euthanasia program which has rendered destruction of defective infants presently "politically insignificant"? If so, is it any more of a jump from the incurably and painfully ill to the unorthodox political thinker than it is from the hopelessly defective infant to the same "unsavory character?" possibility of betterment by unforseen discoveries are greatly reduced; (2) there is not mind enough to hold any dream or hope which is likely to be crushed by the forthright statement that one is doomed, a necessary communication under a voluntary euthanasia program. Kennedy's views are contained in Entha11asia: To Be Or Not To Be, Colliers, May 20, 1939, p. 15, reprinted, with the notation that his views remain unchanged, in Colliers, April 22, 1950, p. 13; The Problem of Social Co1drol of the Congenital Defective, 99 Am.]. Psychiatry 13 (1942). See also t•!.'tt at'notes 72-74 supra. Dr. Wolbarst also indicates that error in diagnosis and possibilities of a cure are reduced in the case of insane or defective people. See text at notes 74-76, srtpra. 207. Tile Forttme Qttarterly Survey: IX, Fortune, July 1937, pp. 96, 106. Actually, a slight majority of those who took a position on the defective infants favored euthanasia under certain circumstances since 45% approved under certain circumstances, 40.5% were unconditionally opposed, and 14.5% were undecided. In the case of the incurably ill, only 37.3% were in favor of euthanasia under any set of safeguards, 47.5% were flatly opposed, and 15.2% took no position. Every major poll taken in the United States on the question has shown popular opposition to voluntary euthanasia. In 1937 and 1939 the American Institute of Public Opinion polls found 46% in favol", 54% opposed. A 1947 poll by the same group found only 37% in favor, 54o/o opposed and 9% of no opinion. For a discussion of these and other polls by various newspapers and a breakdown of the public attitude on the question in terms of age, sex, economic and educational levels see Note, JudicUil Determi11ation of Moral Co11dttct ~~~ Citize1~Mp Hearings, 16 U. of Chi. L. Rev. 138, 141-42 and n. 11

( 19~s· Williams notes, however, at 332, a 1939 British Institute of Public Opinion poll found 68% of the British in favor ·:>f some form of legal euthanasia. 208. The Fortttne Q11arterly S1ervey, note 207, suPra, at 106. 209. Mitscherlich and Mielke, Doctors of Infamv 114 (1949). The Reich Committee for Research on Hereditary Diseases and·Constitutional Susceptibility to Severe Diseases" originally dealt only with child patients up to the age of three, but the age limit was later raised. to eight, twelve, and apparently even sixteen or seventeen years. !d. at 116.

Death, Dying and the Ending of Life, 11

216

1030

MINNESOT.A. LAW REPlEW

[Vo1.4Z:969

Or is it not so much that the euthanasiasts are troubled by the Nazi experience as it is that they are troubled that the public is troubled by the Nazi experience? I read Williams' comments on defective infants for the proposition that there are some very good reasons for euthanatizing defective infants, but the time is not yet ripe. When will it be? When will the proposal become politically significant? After a voluntary euthanasia law is on the books and public opinion is sufficiently "educated?" Williams' reasons for not extending euthanasia- once we legalize it in the narrow "voluntary" area- to the senile and the defective are much less forceful and much less persuasive than his arguments for legalizing voluntary euthanasia in the first place. I regard this as another reason for not legalizing voluntary euthanasia in the first place.

B. The Parade of Horrors. Look, when the messenger cometh, shut the door, and hold him fast at the door; is not the sound of his master's feet behind bim]2IO

_ This is the "wedge principle," the "parade of horrors" objection, if you will, to voluntary euthanasia. Glanville Williams' peremptory retort is: This use of the 'wedge' objection evidently involves a particular determination as to the meaning of words, namely the words 'if raised to a general line of conduct.' The author supposes, for the sake of argument, that the merciful extinction of life in a suffering patient is not in itself inunoral. Still it is immoral, because if it were permitted this would admit 'a most dangerous wedge that might eventually put all life in a precarious condition.' It seems a sufficient reply to say that this type of reasoning could be used to condemn any act whatever, because there is no human conduct from which evil cannot be imagined to follow if it is persisted in when some of the circumstances are changed. All moral questions involve the drawing of a line, but the 'wedge principle' would make it impossible to draw a line, because the line would have to be pushed farther and farther back until all action became vetoed. 211 I agree with Williams that if a first step is "moral" it is moral wherever a second step may take us. The real point, however, the 210. II Kingsf VI, 32, quoted and applied in Sperry, The Case .A.gain.rt Mercy Killing, 70 i\J1L Mercury 271, 276 (1950). 211. Williams, p. 315. At this point, Williams is quoting from Sullivan, Catholic Teaching on the Moral1ty of Euthanasia 54-55 (1949). This thorough exposition of the Catholic Church's position on euthanasia was originally published by the Catholic University of America Press, then republished 1)y the Newman Press as The Morality of Mercy Killing (1950).

Death, Dying and the Ending of Life, II

1958]

"MERCY~KILUNCU

LEGISLATION

217

1031

point that Williams sloughs, is that whether or not the first step is precarious, is perilous, is worth taking, rests in part on what the second step is likely to be. It is true that the "wedge" objection can always be advanced, the horrors can always be paraded. But it is no less true that on some occasions the objection is much more valid than it is on others. One reason why the "parade of horrors" cannot be too lightly dismissed in this particular instance is that Miss Voluntary Euthanasia is not likely to be going it alone for very long. Many of her admirers, as I have endeavored to show in the preceding section, would be neither surprised nor distressed to see her joined by Miss Euthanatize the Congenital Idiots and Miss Euthanatize the Permanently Insane and Miss Euthanatize the Senile Dementia. And these lasseswhether or not they themselves constitute a "parade of horrors"certainly make excellent majorettes for such a parade: Some are proposing what is called euthanasia; at present only a proposal for killing those who are a nuisance to themselves; but soon to be applied to those who are a nuisance to other people.212 Another reason why the "parade of horrors" argument cannot be too lightly dismissed in this particular instance, it seems to me, is that the parade has taken place in our time and the order of procession has been headed by the killing of the "incurables" and the ''useless" : Even before the Nazis took open charge in Germany, a propaganda. barrage was directed against the traditional compassionate nineteenth-century attitudes toward the chronically ill, and for the adoption of a utilitarian, Hegelian point of view.••• Lay opinion was not neglected in this campaign. Adults were propagandized by motion pictures, one of which, entitled 'I Accuse', deals entirely with euthanasia. This film depicts the life history of a. woman suffering from multiple sclerosis; in it her husband, a. doctor, finally kills her to the accompaniment of soft piano music rendered by a sympathetic colleague in an adjoining room. Acceptance of this ideology was implanted even in the children. A widely used high-school mathematics text ••• included problems stateCl in distorted terms of the cost of caring for and rehabilitating the chronically sick and crippled. One of the problems asked, for instance, how many ne\v housing units could be built and how many marriage-allowance loans could be given to newly wedded couples for the amount of money it cost the state to care for 'the crippled, the criminal and the insane••• .' The beginnings at first were merely a subtle shift in emphasis in the basic attitude of the physicians. It started, with the ac212. Chesterton, Etcthanasia and Mtcrder, 8 Am. Rev. 486, 49() (1937).

Death, Dying and the Ending of Life, 11

218

1032

MINNESOTA LAW REVIEW

[Vol. 42:969

ceptance of the attitude, basic in the euthanasia ntovement, that there is such a thing as life not worthy to be lived. This attitude in its early stages concerned itself merely with the severely and chronically sick. Gradually the sphere of those to be included in this category was enlarged to encompass the socially unproductive, the ideologically unwanted, the racially unwanted and finally all non-Germans. But it is important to realize that the infinitely small wedged-in lever from which this entire trend of mind received its impetus was the attitude toward the nonrehabilitatable sick.213 213. Alexander, Medical Science Under Dictatorship, 241 New England Journal of Medicine 39, 44, 40 (1949) (emphasis added). To the same effect is Ivy, Nasi War Crimes of a Meclteal Nature, 139 ].A.M.A. 131, 132 (1949), concluding that the practice of euthanasia was a factor which led to "mass killing of the aged, the chronically ill, •useless eaters' and the politically Widesirable," and IyX, Nasi War Crimes of a Medical Natt1re, 33 Federation Bulletin 133, 142 (1947), noting that one of the arguments the Nazis employed to condone their criminal medical experiments was that "if it is right to take the life of useless and incurable persons, which as they point out has been suggested in England and the United States, then it is right to take the lives of persons who are destined to die for political reasons." Doctors Leo Alexander and A. C. Ivy were both expert medical advisors to the prosection at the Nuremberg Trials. See also the November 25, 1940 entry to Shirer, Berli1~ Diary 454, 458-

59 (1941):

I have at last got to the bottom of these 'mercy killings'. It's an evil tale. The Gestapo, with the knowledge and approval of the German government, is systematically putting to death the mentally deficient population of the Reich. ••• X, a German told me yesterday that relatives are rushing to get their kin out of private asylums and out of the clutches of the authorities. He says the Gestapo is doing to death persons who are merely suffering temporary derangement or just plain nervous breakdown. What is still unclear to me is the motive for these murders. Germans themselves advance three:

*

*

*

3. That they are simply the result of the extreme Nazis deciding to carry out their eugenic and sociological ideas.

The third motive seems most*likely *to me. * For years a group of radical Nazi sociologists who were instrumental in putting through the Reich's sterilization laws have pressed for a national policy of eliminating the mentally Wifit They say they have disciples among many sociologists in other lands, and perhaps they have. Paragraph two of the form letter sent the relatives plainly bears the stamp of the sociological thinking: 'In view of the nature of his serious, incurable ailment, his death, which saved him from a lifelong institutional sojourn, is to be regarded merely as a release.' This contemporaneous report is supported by evidence Wicovered at the Nuremberg Medical Trial. Thus, an August, 1940 form letter to the relatives of a deceased mental patient states in part: "Because of her grave mental illness life was a torment for the deceased. You must therefore look on her death as a release." This form letter is reproduced in Mitscherlich and Mielke, Doctors of Infamy 103 (1949). Dr. Alexander Mitscherlich and Mr. Fred Mielke attended the trial as delegates chosen by a group of German medical societies and Wiiversities. According to the testimony of the chief defendant at the Nuremberg Medical Trial, Karl Brandt, Reich Commissioner for Health and Sanitation and personal physician to Hitler, the Fuhrer has indicated in 1935 that if

Death, Dying and the Ending of Life, II

1958]

"MERCY-KILLING" LEGISLATION

219

1033

The apparent innocuousness of Germany's "small beginnings" is perhaps best shown by the fact that German Jews were at first excluded from the program. For it was originally conceived that "the blessing of euthailasia should be granted only to [true] Gennans." 2 u Relatively early in the German program, Pastor Braune, Chairman of the Executive Committee of the Domestic Welfare Council of the German Protestant Church, called for a halt to euthanasia measures nsince they strike sharply at the moral foundations of the nation as a whole. The inviolability of human life is a pillar of any social order."215 And the pastor raised the same question which euthanasia opponents ask today, as well they might, considering the disinclination of many in the movement to stop at voluntary "mercy killings" : Where do we, how do we, draw the line? The good pastor asked: How far is the destruction of socially unfit life to go? The mass methods used so far have quite evidently taken in many people who are to a considerable degree of sound mind. . • • Is it intended to strike only at the utterly hopeless cases-the idiots and imbeciles? The instruction sheet, as already mentioned, also lists senile diseases. The latest decree by the same authorities requires that children with serious congenital disease and malformation of every kind be registered, tc• be collected and processed in special institutions. This necessarily gives rise to grave apprehensions. Will a line be drawn at the tubercular? In the case of persons in custody by court order euthanasia measures, have evidently already been initiated. Are other abnormal or anti-social persons likewise to be included? Where is the borderline? Who is abnormal, antisocial, hopelessly sick ?216 Williams makes no attempt to distinguish or minimize the Nazi Germany experience. Apparently he does not consider it worthy of mention in a euthanasia discussion. There are, however, a couple of obvious argum~ts by which the Nazi experience can be minimized. war came- he would· effectuate the policy of euthanasia since in the general upheaval of war the open resistance to be anticipated on the part of the church would not be the potent force it might otherwise be. Mitscherlich and Mielke supra at 91. . Certain petitions to Hitler by parents of malformed clu1dren requesting authority for "mercy deaths" seem to have played a part in definit~y making up his mind. Ibid. 214. Defendant Viktor Brack. Chief Admini:;trative Officer in Hitler's private chancellory, so testified at the Nuremberg Medical Trial, 1 Trials of War Criminals Before the Nuremberg Military Tribunal Under Control Council Law No. 10, 877.80 (1950) ("The Medical Case"). 215. Mitscherlich and Mielke, op. cit. sttpra note 213, at 107. 216. Ibid. According to testimony at the Nuremberg Medical Trial, although they were told that "only incurable patients, suffering severely, were involved," even the medical consultants tc• the program were "not quite clear on where the line was to be drawn." Id. at 94.

Death, Dying and the Ending of Life, II

220

1034

MINNESOTA LAW REVIEW

[Vot42:969

One goes something like this : It is silly to worry about the prospects of a dictatorship utilizing euthanasia "as a pretext for putting inconvenient citizens out of the way. Dictatorships have no occasion for such subterfuges. The firing squad is less bother."21 ' One reason why this counter argument is not too reassuring, ho\vever, if again I may be permitted to be so unkind as to meet speculation with a concrete example to the contrary, is that Nazi Germany had considerable occasion to use just such a subterfuge. Thus, Dr. Leo Alexander observes: It is rather significant that the German people were considered by their Nazi leaders more ready to accept the exterminations of the sick than those for political reasons. It was for that reason that the first exterminations of the latter group were carried out under the guise of sickness. So-called 'psychiatric experts' were dispatched to survey the inmates of camps with the specific order to pick out members of racial minorities and political offenders from occupied territories and to dispatch them to killing centers with specially made diagnoses such as that of 'inveterate German hater' applied to a number of prisoners who had been active in the Czech underground.

*

*

*

A large number of those marked for death for political or racial reasons were made available for 'medical e.'Cperiments involving the use of involuntary human subjects/218 The "hunting season" in Germany officially opened when, Hitler signed on his own letterhead, a secret order dated September 1, 1939, which read : Reichsleiter Bouhler and Dr. Brandt, M.D., are charged with the responsibility of enlarging the authority of certain physicians, to be designated by name, in such a manner that persons who, according to human judgment, are incurable can, upon a more careful diagnosis of their condition of sickness, be accorded a mercy death. 219 217. Pro & Con: Shall We Legalize "Mercy Killing", 33 Readers Digest, Nov. 1938, p. 94 at 96. 218. Alexander, supra note 213, at 41. Dr. Alexander Mitscherlich and Mr. Fred Mielke similarly note: The granting of 'dying aid' in the case of incurable mental patients and malformed or idiot children may be considered to be still within the legitimate sphere of medical discussion. But as the 'winnowing process' continued, it moved more and more openly as purely political and ideological criteria for death, whether the subjects were considered to be 'undesirable racial groups,' or whether they had merely become incapable of supporting themselves-. The camouflage around these murderous intentions is revealed· especially by proof that in the concentration camps prisoners wete selecteaby the same medical consultants who were simultaneously sitting in judgment over the destiny of mental institution inmates. Mitscherlich and Mielke, op. cit. .s~epra note 213 at 41. 219. This is the translation rendered in the judgment of Military Tribunal 1, 2 Trials of War Criminals Before The Nuremberg Military Tri-

Death, Dying and the Ending of Life, II

1958]

"MERCY-KILLING" LEGISLATION

221

1035

Physicians asked to participate in the program were told that the secrecy of the order was designed to prevent patients from becoming "too agitated" and that it was in keeping with the policy of not publicizing home front measures in time of war. 220 About the same time that aged patients in some hospitals were being given the "mercy" treatment, 221 the Gestapo was also "systematically putting to death the mentally deficient population of the Reich. 222 The courageous and successful refusal by a Protestant pastor to deliver up certain cases from his asylum228 well demonstrates that even the most totalitarian governments are not always indifferent to the feelings of the people, that they do not always feel free to resort to the firing squad. Indeed, vigorous protests by other ecclesiastical personalities and some physicians, numerous requests of various public prosecutors for investigation of the circumstances surrounding the mysterious passing away c•f relatives, and a generally arous~d public opinion finally caused Hitler to yield, if only temporarily, and in August of 1941 he verbally ordered the disbunals Under Control Council Law No. 10, 195 (1950) ("The Medical Case"). A slightly different but substantially identical translation appears in Mitscherlich and Mielke, op. cit. szepra note 213 at 92. The letter, Document 630-PS, Prosecution Exhibit 330, as written in the original German, may be found in 26 Trial of Major War Criminals Before the International Military Tribunal169 (1947). For conflicting views on whether or not the order was back-dated, compare Mitscherlich and Mielke, op. cit. szepra with Koessler, Euthanasia It~ The Hadamar Sanatorium and lntematiomu Law, 43 J.' Crim. L., C. & P.S. 735, 737 (1953). 220. Mitscherlich and Mielke, op. cit. szepra note 213, at 93-94. 221. In the fall of 1940, Catholic priests at a large hospital near Urach "notices that elderly people in the hospital were dying in increasing numbers, and dying on certain days." Straight, Germany Esectetes Her "Unfit'', 104 New Republic 627 (1941). Such incidents led a German bishop to ask the Supreme Sacred Congregation whether it is right to kill those "who, although they have committed no crime deserving death, yet, because of mental or physical defects, are no longer able to benefit the nation, and are considered rather to burden the nation and to obstruct its energy and strength." Ibid. The answer was, of course, in the negative, ibid., but "it is doubtful if the mass of German Catholics, even if they learned of this statement from Rome, which is improbable1 understood what it referred to. Only a minority in Germany knew of the mercy deaths'." Shirer, op. cit. szepra note 213, at 459 n. 1. 222. Shirer; op. cit. szepra note 213, at 454. 223. "Late last summer, it seems Pastor von Bodelschwingh was asked to deliver up certain of his worst cases to the authorities. Apparently he got wind of what was in store for them. He refused. The authorities insisted. Pastor von Bodelschwingh hurried to Berlin to protest. Pastor von Bodelschwingh returned to Bethel. The local Gauleiter ordered him to tum over some of his inmates. Again he refused. Berlin then ordered his arrest. This time the Gauleiter prote:;ted. The pastor was the most popular man in his province. To arrest him in the middle of war would stir up a whole world of unnecessary trouble. He himself declined to arrest the man. Let the Gestapo take the responsibility; he wouldn't. This was just before the night of September 18, [1940]. The bombing of the Bethel asylum followed. Now I understand why a few people wondered as to who dropped the bombs." Shirer, op. cit. stepra note 213 at 454-55.

Death, Dying and the Ending of Life, II

222

1036

MINNESOTA LAW REVIEW

[Vo1.42:969

continuance of the adult euthanasia program. Special gas chambers in Hadamar and other institutions were dismantled and shipped to the East for much more extensive use of Polish J ews.:m Perhaps it should be noted, too, that even dictatorships fall prey to the inertia of big government: It is ... interesting that there was so much talk against eutha" nasia in certain areas of Germany, particularly in the region of Wiesbaden, that Hitler in 1943 asked Himmler to stop it. But, it had gained so much impetus by 1943 and was such an easy way in crowded concentration camps to get rid of undesirables and make room for newcomers, that it could not be stopped. The wind had become a whirlwind.225 Another obvious argument is that it just can't happen here. I hope not. I think not. But then, neither did I think that tens of thousands of perfectly loyal native"born Americans would be herded into prison camps without proffer of charges and held there for many months, even years, because they were of "Japanese blood"226 and, although the 224. Mitscherlich and Mielke, op. cit. supra note 213, at 113-114; Koessler, supra note 219, at 739. 225. Ivy, Nasi W.ar Crimes of a Medical Nature, 33 Federation Bulletin 133, 134 (1947). 226. As Justice Murphy pointed out in his dissenting opinion in Korematsu v. United States, 323 U.S. 214, 241-42 (1944) : No adequate reason is given for the failure to treat these Japanese Americans on an individual basis by holding investigations and hear.ings to separate the loyal from the disloyal, as was done in the case of persons of German and Italian ancestry. It is asserted merely that the loyalties of this group 'were unlmown and time was of the essence.' Yet nearly four months elapsed after Pearl Harbor before the first exclusion order was issued; nearly eight months went by until the last order was issued; and the last of these 'subversive' persons was not actually removed until almost eleven months had elapsed. Leisure and deliberation seem to have been more of the essence than speed. And the fact that conditions were not such as to warrant a declaration of martial law adds strength to the belief that the factors of time and military necessity were not as urgent as they have been represented to be. Moreover, there was no adequate proof that the Federal Bureau of Investigation and the military and naval intelligence services did not have the espionage and sabotage situation well in hand during this long period. Nor is there any denial of the fact that not one person of Japanese ancestry was accused or convicted of espionage or sabotage after Pearl Harbor while they were still free, a fact which is some evidence of the loyalty of the vast majority of these individuals and of the effectiveness of the established methods of combatting these evils. It seems incredible that under these circumstances it would have been impossible to hold loyalty hearings for the mere 112,000 persons involved-or at least for the 70,000 American citizens-especially when a large part of this number represented children and elderly men and women. Justice Murphy then went on to note that shortly after the outbreak of World War II the British Government examined over 70,000 German and Austrian aliens and in six months freed 64,000 from internment and from any special restrictions. 354 U.S. 242 n. 16. See generally Rostow, The Japanese AmericatJ Cases-A Disaster, 54 Yale L.J. 489 (1945), a tale well calculated to keep you in anger and shame.

Death, Dying and the Ending of Life, II

1958]

"MERCY-KILLING" LEGISLATION

223

1037

general who required these measures emitted considerable ignorance and bigotry,227 his so-called military judgment would be largely sustained by the highest court of the land. The Japanese American experience of World War II undoubtedly fell somewhat short of first-class Nazi tactics, but we were getting warm. I venture to say it would not be too difficult to find American citizens of Japanese descent who would maintain we were getting very warm indeed. In this regard, some of Justice JacksOJl's observations in his Kore-matsu, dissent228 seem quite pertinent: All who observe the work of courts are familiar with what Judge Cardozo described as 'the tendency of a principle to expand itself to the limit of its logic.' [Nature of the Judicial Process, p. 51.] A military commander may overstep the bounds of constitutionality, and it is an incident. But if we review and approve, that passing incident becomes the doctrine of the Constitution. There it has a generative power of its own, and all that it creates will be in its own image. Nothing better illustrates this danger than does the Court's opinion in this case. It argues that we are bound to uphold the conviction of Korematsu because we upheld one in Hirabayashi v. United States, 320 U.S. 81, when we sustained these orders in so far as they applied a curfew requirement to a citize11 of Japanese ancestry. I think we should learn something from that experience. In that case we were urged to consider only the curfew feature, that being all that technically was involved, because it was the only count necessary to sustain Hiraba.yashi's conviction and, sentence. We yielded, and the Chief Justice guarded the opinion as carefully as language will do.... H)wever, in spite of our 227. See e.g., General J. L. Dewitt's Final Recommendation to the Secretary of War, U.S. Army~ Westen. Defense Command, Final Report, Japauese Evacttatio1~ From the West Coast, 1942 (1943) at 32 ("The Japanese race is an enemy race and while many second and third generation Japanese born on United States soil, possessed of United States citizenship, have become 'Americanized,' the racial strains are undiluted. • • ."), and his subsequent testimony, Hearillgs Before S1wcommittee of House Committee ots Naval Affairs 01~ H.R. 30, 78th Cong., 1st Sess. (1943) at 739-40 ("You needn't worry about the Italians at all except h certain cases. Also, the same for the Germans except in individual cases. But we must worry about the JaP.anese all the time until he is wiped off the map. Sabotage and espionage will make problems as long as he is allowed in this area-problems which I don't want to have to worry about''.) After a careful study, Professor (now Dean) RostO\v took this position: The dominant factor in the development of this policy was not a military estimate of a military problem, but familiar V'l est Coast attitudes of race prejudice. The program of excluding all persons of Japanese ancestry from the coastal area was conceived and put through by the organized minority whose business it has been for forty-five years to increase and e.'Cploit racial tensions on the West CoaSt. The Native Sons and Daughters of the Golden West and their sympathizers, were lucky in their general, for General DeWitt amply proved himself to be one of them in opinion and values. Rostow, supra note 226, at 496. 228. See note 226, .stepra.

Death, Dying and the Ending of Life, II

224

1038

MINNESOTA LAW REVIEW

[Vol. 4Z:969

limiting words we did validate a discrimination on the basis of ancestry for mild and temporary deprivation of liberty. Now the principle of racial discrimination is pushed from support of mild measures to very harsh ones, and from temporary deprivations to indeterminate ones. And the precedent which it is said requires us to do so is Hirabayashi. The Court is now saying that in Hirabayashi we did decide the very things we there said we were not deciding. Because we said that these citizens could be made to stay in their homes during the hours of dark, it is said we must require them to leave home entirely; and if that, we are told they may also be taken into custody for deportation ; and if that, it is argued they may also be held for some undetermined time in detention camps. How far the principle of this case would be extended before plausible reasons would play out, I do not know.2 29 It can't happen here. Well, maybe it cannot, but no small part of our Constitution and no small number of our Supreme Court opinions stem from the fear that it can happen here unless we darn well nzake sztre that it does not by adamantly holding the line, by swiftly snuffing out what are or might be small beginnings of what we do not want to happen here. To flick off, as Professor Williams does, the fears about legalized euthanasia as so much nonsense, as a chimerical "parade of horrors," is to sweep away much of the ground on which all our civil liberties rest. Boyd, 280 the landmark search and seizure case which paved the way for the federal rule of exdusion,281 a doctrine which now pre-' vails in over twenty state courts as well,282 set the mood of our day in treating those accused of crime: It may be that it is the obnoxious thing in its mildest and least repulsive form; but illegitimate and unconstitutional practices get their first footing in that way, namely, by silent approaches and slight deviations from legal modes of procedure. • . • It is the duty of courts to be watchful for the constitutional rights of the citizen, and against any stealthy encroachments thereon. Their motto should be obsta principiis. ...238 Z29. 323 U.S. at 246-47. 230. Boyd v. United States, 116 U.S. 616 (1886). 231. See e.K., Weeks v. United States, 23Z U.S. 283 (1914); G-

vided by a physician for that purpose. A paradigm case of voluntary active euthanasia is a physician's administering the lethal dose, often because the patient is unable to do so. The only difference that need exist between the two is the person who actually administers the lethal dose--the physician or the patient In each, the physician plays an active and necessary causal role. In physician-assisted suicide the patient acts last (for example,JanetAdkins herselfpushed the button after Dr. Kevorkian hooked her up to his suicide machine), whereas in euthanasia the physician acts last by perfornting the physical equivalent of pushing the button. In both cases, however, the choice rests fully with the patient. In both the patient acts last in the sense ofretaining the right to change his or her mind until the point at which the lethal process becomes irreversible. How could there be a substantial moral difference between the two based only on this small difference in the part played by the physician in the causal process resulting in death? Ofcourse, it might be held that the moral difference is clear and important-in euthanasia the physician kills the patient whereas in physician-assisted suicide the patient kills him- or herself. But this is misleading at best In assisted suicide the physician and patient together kill the patient To see this, suppose a physician supplied a lethal dose to a patient with the knowledge and intent that the patient will wrongfully administer it to another. We would have no difficulty in morality or the law recognizing this as a case ofjointaction to kill for which both are responsible. If there is no significant, intrinsic moral difference between the two, it is also difficult to see why public or legal policy should permit one but not the other; worries about abuse or about giving anyone dominion over' the lives of others apply equally to either. As a result, I will take the arguments evaluated below to apply to both and will focus on euthanasia. My concern here will be with voluntary euthanasia only-that is, with the case in which a clearly competent patient makes afullyvoluntaryand persistent request for aid in dying. Involuntary euthanasia, in which a competent patient explicitly refuses or opposes receiving euthanasia, and nonvoluntary

230

Death, Dying and the Ending of Life, II Hastings Center Report, Marcb-April1992

euthanasia, in which a patient is incompetent and unable to express his or her wishes about euthanasia, will be considered here only as potential unwanted side-effects ofpermitting voluntary euthanasia. I emphasize as well that I am concerned with active euthanasia, not withholding orwithdrawing life-sustaining treatment, which some commentators characterize as "passive euthanasia." Finally, I will be concerned with euthanasia where the motive of those who perform it is to respect the wishes of the patientandto provide the patient with a "good death," though one important issue is whether a change in legal policy could restrict the performance of euthanasia to only those cases. A last introductory point is that I will be examining only secular arguments about euthanasia, though of course manypeople's attitudes to it are inextricable from their religious views. The policy issue is only whether euthanasia should be permissible, and no one who has religious objections to it should be required to take any part in it, though of course this would not fully satisfy some opponents. The Central Ethical Argument for Voluntary Active Euthanasia

The central ethical argument for euthanasia is familiar. It is that the very same two fundamental ethical values supporting the consensus on patient's rights to decide about life-sustaining treatment also support the ethical permissibility of euthanasia. These values are individual self-determination or autonomy and individual well-being. By self-determination as it bears on euthanasia, I mean people's interest in making important decisions about their lives for themselves according to their own values or conceptions of a good life, and in being left free to acton those decisions. Self-determination is valuable because it pennits people to form and live in accordance with their own conception of a good life, at least within the bounds of justice and consistent with others doing so as well. ln exercising self-detennination people take responsibility for their lives and for the kinds of persons they become. A central aspect of human dignity lies in people's capacity to direct their lives in this way. The value of exercising self-de-

termination presupposes some minimum of decisiomnaking capacities or competence, which thus limits the scope of euthanasia supported by selfdetermination; it cannot justifiably be administered, for example, in cases of serious dementia or treatable clinical depression. Does the value of individual self-detemiination extend to the time and manner of one's death? Most people are very concerned about the nature of the last stage of their lives. This reflects not just a fear of experiencing substantial suffering when dying, but also a desire to retain dignity and control during this last period of life. Death is today increasingly preceded by a long period of significant physical and mental decline, due in part to the technological interventions of modem medicine. Many people adjust to these disabilities and find meaning and value in new activities and ways. Others find the impairments and burdens in the last stage of their lives at some point sufficiendy great to make life no longer worth living. For many patients near death, maintaining the quality of one's life, avoiding great suffering, maintaining one's dignity, and insuring that others remember us as we wish them to becomeofparamountimportanceand outweigh merely extending one's life. But there is no single, objectively correct answer for everyone as to when, if at all, one's life becomes all things considered a burden and unwanted If selfdetermination is a fundamental value, then thegreatvariabilityamongpeople on this question makes it especially im- portant that individuals control the manner, circumstances, and timing of

their dying and death. The other main value that supports euthanasia is individual well-being. It might seem that individual well-being conflicts with a person's self-determination when the person requests euthanasia. Ufe itself is commonly taken· to be a central good for persons, often valued for its own sake, as well as necessary for pursuit of all other goods within a life. But when a competent patient decides to forgo all further life-sustaining treatment then the patient, either explicitly or implicitly, commonly decides that the best life possible for him or her with treatment is of sufficiently poor quality that it is worse than no further life at all. Ufe

11

is no longer considered a benefit by the patient, buthasnow become a burden. The same judgment underlies a request for euthanasia: continued life is seen by the patient as no longer a benefit, but now a burden. Especially in the often severely compromised and debilitated states of many critically ill or dying patients, there is no objective standard, but only the competent patient's judgment of whether continued life is no longer a benefit Of course, sometimes there are con-

ditions, such as clinical depression, that call into question whether the patient has made a competent choice, either to forgo life-sustaining treatment or to seek euthanasia, and then the patient's choice need not be evidence that continued life is no longer a benefitfor him or her.Just as with decisions about treatment, a determination of incompetence can warrant not honoring the patient's choice; in the case of treatment, we then transfer decisional authority to a surrogate, though in the case of voluntary active euthanasia a determination that the patient is incompetent means that choice is not possible. The value or right of self-detennination does not entitle patients to compel physicians to act contrary to their own moral or professional values. Physicians are moral and professional agents whose own self-determination or integrity should be respected as well If performing euthanasia became legally permissible, but conflicted with a particular physician's reasonable understanding of his or her moral or professional responsibilities, the care of a patient who requested euthanasia should be t:taJ:1S. ferred to another. Most opponents do not deny that there are some cases in which the values of patient self-determination and wellbeing support euthanasia. lnstead, they commonly offer two kinds of arguments against it that on their view outweigh or override this support The first kind of argument is that in any individual case where considerations of the patient's self-ld age as a time of new, open possibilities. That campaign has met with only partial success. Despite great progress, the elderly are still subject to discrimination and stereotyping. The struggle against ageism is hardly over. Three major concerns have, nonetheless, surfaced over the past few years. They are symptoms that a new era has arrived. The first is that an increasingly large share of health care is goi~g to the elderly in comparison with benefits for children. The federal government, for instance, spends six times as much on health care for those over 65 as for those under 18. As the demographer Samuel Preston observed in a provocative 1984 presidential address to the Population Association of America: There is surely something to be said for a system in which things get better as we pass through life rather than worse. The great leveling off of age curves of psychological distress, suicide and income in the past two decades might simply reflect the fact that we have decided in some fundamental sense that we don't want to face futures that become continually bleaker. But let's be clear that the transfers from the working-age population to the elderly are also transfers away from children, since the working ages bear far more responsibility for childrearing than do the elderly. 1

Preston's address had an immediate impact. The mainline aging advocacy groups responded with pained indignation, accusing Preston of fomenting a war between the generations. But led by Dave Durenberger, Republican Senator from Minnesota, it also stimulated the formation of Americans for Generational Equity (AGE), an organization created to promote debate on the burden to future generations, but particularly the Baby Boom generation, of "our major social insurance programs."z These two developments signalled the outburst of a struggle over what has come to be called "lntergenerational equity" that is only now gaining momentum. The second concern is that the elderly dying consume a disproportionate share of health care costs. Stanford economist Victor Fuchs has noted: At present, the United States spends about I percent of the gross national product on health care for elderly persons who are in their last year of life•• ' •• One of the biggest challenges facing policy makers for the rest of this century will be how to strike an appropriate balance between care of the [elderly) dying and health services for the rest of the population.•

The third concern is summed up in an observation by Jerome L. Avorn, M.D., of the Harvard Medical School: With the exception of the binh-control pill, each of the medical-technology interventions developed since the 1950s bas its most widespread impact on people who are past their fifties-the further past their fifties, the greater the impact.•

Death, Dying and the Ending of Life, II CALLAHAN: AGING & TilE ENDS OF MEDICINE

435

127

Many of these interventions were not intended for the elderly. Kidney dialysis, for example, was originally developed for those between the age of IS and 45. Now some 30CI/o of its recipients are over 65. These three concerns ha\·e not gone unchallenged. They have, on the contrary, been strongly resisted, as has the more general assertion that some form of rationing of health care for the elderly might become necessary. To the charge that the elderly receive a disproportionate share of resources, the response has been that what helps the elderly helps every other age group. It both relieves the young of the burden of care for elderly parents they would otherwise have to bear and, since they too will eventually become old, promises them similar care when they come to need it. There is no guarantee, moreover, that any cutback in health care for the elderly would result in a transfer of the savings directly to the young. Our system is not that rational or that organized. And why, others ask, should we contemplate restricting care for the elderly when we wastefully spend hundreds of millions of dollars on an inflated defense budget? The charge that the elderly dying receive a large share of funds hardly proves that it is an unjust or unreasonable amount. They are, after all, the most in need. As some important studies have shown, moreover, it is exceedingly difficult to know that someone is dying; the most expensive patients, it turns out, are those who are expected to live but who actually die. That most new technologies benefit the old more than the young is perfectly sensible: most of the killer diseases of the young have now been conquered. These are reasonable responses. It would no doubt be possible to ignore the symptoms that the raising of such concerns represents, and to put off for at least a few more years any full confrontation with the overpowering tide of elderly now on the way. There is little incentive for politicians to think about, much less talk about, limits of any kind on health care for the aged; it is a politically hazardous topic. Perhaps also, as Dean Guido Calabresi ofthe Yale Law School and his colleague Philip Bobbitt observed in their thoughtful 1978 book Tragic Choices, when we are forced to make painful allocation choices, "Evasion, disguise, temporizing ••• (and] averting our eyes enables us to save some lives even when we will not save all."5 Yet however slight the incentives to take on this highly troubling issue, I believe it is inevitable that we must. Already rationing of health care under Medicare is a fact of life, though rarely labeled as such. The requirement that Medicare recipients pay the first $500 of the costs of hospital care, that there is a cutoff of reimbursement of care beyond 60 days, and a failure to CO\'er long-term care, are nothing other than allocation and cost-saving devices. As sensitive as it is to the votes of the elderly, the Reagan administration only grudgingly agreed to support catastrophic health care costs of the elderly (a benefit that will not, in any event, help many of the aged). It is bound to be far more resistant to long-term care coverage, as will any administration. But there are other reasons than economics to think about health care for the elderly. The coming economic crisis provides a much-neede(f opportunity to ask some deeper questions. Just what is it that we want medicine to do for us as we age? Earlier cultures believed that aging should be accepted, and that it should be in part a time of preparation for death. Our culture seems increasingly to reject that view, preferring instead, it often seems, to think of aging as hardly more than another disease, to be fought and rejected. Which view is correct? To ask that question is only to note that disturbing puzzles about the ends of medicine and the ends of aging lie behind the more immediate financing worries.· Without some kind of answer to them, there is no hope of finding a reasonable, and possibly even a humane, solution to the growing problem of health care for the elderly. Let me put my own view directly. The future goal of medicine in the care of the aged should be that of improving the quality of their life, not in seeking ways to extend that life. In its longstanding ambition to forestall death, medicine has in the care of

Death, Dying and the Ending of Life, II

436 128

ANNALS NEW YORK ACADEMY OF SCIENCES

the aged reached its last frontier. That is hardly because death is absent elsewherechildren and young adults obviously still die of maladies that are open to potential cure-but because the largest number of deaths (some 70%) now occur among those over the age of 65, with the highest proportion in those over 85. If death is ever to be humbled, that is where the essentially endless work remains to be done. But however tempting that challenge, medicine should now restrain its ambition at that frontier. To do otherwise will, I believe, be to court harm to the needs of other age groups and to the old themselves. Yet to ask medicine to restrain itself in the face of aging and death is to ask more than it, or the public that sustains it, is likely to find agreeable. Only a fresh understanding of the ends and meaning of aging, encompassing two conditions, are likely to make that a plausible stance. The first is that we-both young and old-need to understand that it is possible to live out a meaningful old age that is limited in time, one that does not require a compulsive effort to turn to medicine for more life to make it bearable. The second condition is that, as a culture, we need a more supportive context for aging and death, one that cherishes and respects the elderly while at the same time recognizing that their primary orientation should be to the young and the generations to come, not to their own age group. It will be no less necessary to recognize that in the passing of the generations lies 'the constant reinvigoration of biological life. Neither of these conditions will be easy to realize. Our culture has, for one thing, worked hard to redefine old age as a time of liberation, not decline. The terms "modern maturity" or "prime time" have, after all, come to connote a time of travel, new ventures in education and self-discovery, the ever-accessible tennis court or golf course, and delightfully periodic but gratefully brief visits from well-behaved grandchildren. This is, to be sure, an idealized picture. Its attraction lies not in its literal truth but as a widely-accepted utopian reference point. It projects the vision of an old age to which more and more believe they can aspire and "vhich its proponents think an affluent country can afford if it so chooses. That it requires a medicine that is singleminded in its aggressiveness against the infirmities of old age is of a piece with its hopes. But as we have come to discover, the costs of that kind of war are prohibitive. No matter how much is spent the ultimate problem will still remain: people age and die. Worse still, by pretending that old age can be turned into a kind of endless middle age, we rob it of meaning and significance for the elderly themselves. It is a way of saying that old age can be acceptable only to the extent that it can mimic the vitality of the younger years. There is a plausible alternative: that of a fresh vision of what it means to live a decently long and adequate life, what might be called a natural life span. Earlier generations accepted the idea that there was a natural life span- the biblical norm of three score years and ten captures that notion (even though, in fact, that was a much longer life span than was then typically the case). It is an idea well worth reconsidering, and would provide us with a meaningful and realizable goal. Modern medicine and biology have done much, however, to wean us away from that kind of thinking. They have insinuated the belief that the average life span is not a natural fact at all, but instead one that is strictly dependent upon the state of medical knowledge and skill. And there is much to that belief as a statistical fact: the average life expectancy continues to increase, with no end in sight. But that is not what I think we ought to mean by a natural life span. We need a notion of a full life that is based on some deeper understanding of human need and sensible possibility, not the latest state of medical technology or medical possibility. We should instead think of a natural life span as the achievement of a life long enough to accomplish for the most part those opportunities that life typically affords people

Death, Dying and the Ending of Life, II CALLAHAN: AGING & TilE ENDS OF MEDICINE

437

129

and which we ordinarily take to be the prime benefits of enjoying a life at all-that of loving and living~ of raising a family, of finding and carrying out work that is satisfying, of reading and thinking, and of cherishing our friends and families. If we envisioned a natural life span that way, then we could begin to intensify the devising of ways to get people to that stage of life, and to work to make certain they do so in good health and social dignity. People will differ on what they might count as a natural life span; determining its appropriate range for social policy purposes would need extended thought and debate. My own view is that it can now be achieved by the late 70s or early 80s. That many of the elderly discover new interests and new facets of themselves late in life-my mother took up painting in her seventies and was selling her paintings up until her death at 86-does not mean that we should necessarily encourage a kind of medicine that would make that the norm. Nor does it mean that we should base social and welfare policy on possibilities of that kind. A more reasonable approach is to ask how medicine can help most people live out a decently long life, and how that life can be enhanced along the way. A longer life does not guarantee a better life-there is no inherent connection between the two. No matter how long medicine enabled people to li\-e, death at any time-at age 90, or 100, or 110-would frustrate some possibility, some as-yet-unrealized goal. There is sadness in that realization, but not tragedy. An easily preventable death of a young child is an outrage. The death from an incurable disease of someone in the prime of young adulthood is a tragedy. But death at an old age, after a long and full life, is simply sad, a part of life itself. As it confronts aging, medicine should have as its specific goal that of averting premature death, understood as death prior to a natural life span, and the relief of suffering thereafter. It should pursue those goals in order that the elderly can finish out their years with as little needless pain as possible, and with as much vigor as can be generated in contributing to the welfare of younger age groups and to the community of which they are a part. Above all, the elderly need to ha\·e a sense of the meaning and significance of their stage in life, one that is not dependent for its human value on economic productivity or physical vigor. What would a medicine oriented toward the relief of suffering rather than the deliberate extension of life be like? We do not yet have a clear and ready answer to that question, so long-standing, central, and persistent has been the struggle against death as part of the self-conception of medicine. But the Hospice movement is providing us with much helpful evidence. It knows how to distinguish between the relief of suffering and the extension of life. A greater control by the elderly over their dying- and particularly a more readily respected and enforceable right to deny aggressive life-extending treatment-is a long-sought, minimally necessary goal. What does this have to do with the rising cost of health care for the elderly? Everything. The indefinite extension of life combined with a never-satisfied improvement in the health of the elderly is a recipe for monomania and limitless spending. It falls to put health in its proper place as only one among many human goods. It fails to accept aging and death as part of the human condition. It fails to present to younger generations a model of wise stewardship. How might we devise a plan to limit health care for the aged under public entitlement programs that is fair, humane, and sensitive to their special requirements and dignity? let me suggest three principles to undergird a quest for limits. First, government has a duty, based on our collective social obligations to each other, to help people live out a natural life span, but not actively to help medically extend life beyond that point. Second, government is obliged to develop under its research subsidies, and pay

Death, Dying and the Ending of Life, II

438

130

ANNALS NEW YORK ACADEMY OF SCIENCES

for, under its entitlement programs, only that kind and degree of life-extending technology necessary for medicine to achieve and serve the end of a natural life span. The question is not whether a technology is available that can save the life of someone who has lived out a natural life span, but whether there is an obligation for society to provide them with that technology. I think not. Third, beyond the point of natural life span, government should provide only the means necessaey for the relief of suffering, not life-extending technology. By proposing that we use age as a specific criterion for the limitation of life-extending health care, I am challenging one of the most revered norms of contemporary geriatrics: that medical need and not age should be the standard of care. Yet the use of age as a principle for the allocation of resources can be perfectly valid, both a necessary and legitimate basis for providing health care to the elderly. There is not likely to be any better or less arbitrary criterion for the limiting of resources in the face of the open-ended possibilities of medical advancement in therapy for the aged. Medical "need," in particular, can no longer work as an allocation principle. It is too elastic a concept, too much a function of the state of medical art. A person of 100 dying from congestive heart failure "needs" a heart transplant no less than someone who is 30. Are we to treat both needs as equal? That is not economically feasible or, I would argue, a sensible way to allocate scarce resources. But it would be required by a strict need-based standard. Age is also a legitimate basis for allocation because it is a meaningful and universal category. It can be understood at the level of common sense. It is concrete enough to be employed for policy purposes. It can also, most importantly, be of value to the aged themselves if combined with an ideal of old age that focuses on its quality rather than its indefinite extension. I have become impressed with the philosophy underlying the British health care system and the way it meets the needs of the old and the chronically ill. It has, to begin with, a tacit allocation policy. It emphasizes improving the quality of life through primary care medicine and well-subsidized home care and institutional programs for the elderly rather than through life-extending acute care medicine. The well-known difficulty in getting dialysis after 55 is matched by like restrictions on access to open heart surgery, intensive care units, and other forms of expensive technology. An undergirding skepticism toward technology makes that a viable option. That attitude, together with a powerful drive for equity, "explains," as two commentators have noted, "why most British put a higher value on primary care for the population as a whole than on an abundance of sophisticated technology for the few who may benefit from it."6 That the British spend a significantly smaller proportion of their GNP (6.2%) on health care than Americans (10.811/o) for an almost identical outcome in health status is itself a good advertisement for its priorities. Life expectancies are, for men, 70.0 years in the U.S. and 70.4 years in Great Britain; and, for women, 77.8 in the U.S. arid 76.7 in Great Britain. There is, of course, a great difference in the ethos of the U.S. and Britain, and our individualism and love of technology stand in the way of a quick shift of priorities. Yet our present American expectations about aging and death, it turns out, may not be all that reassuring. How many of us are really so certain that high-technology American medicine promises us all that much better an aging and death, even if some features appear improved and the process begins later than in earlier times? Between the widespread fear of death in an impersonal ICU, cozened about machines and invaded by tubes, on the one hand, or wasting away in the back ward of a nursing home, on the other, not many of us seem comforted.

Death, Dying and the Ending of Life, II CALLAHAN: AGING & THE ENDS OF MEDICINE

439 131

Once we have reflected on those fears, it is not impossible that most people could be persuaded that a different, more limited set of expectations for health care could be made tolerable. That would be all the more possible if there was a greater assurance than at present that one could live out a full life span, that one's chronic illnesses would be better supported, and that long-term care and home care would be given a more powerful societal backing than is now the case. Though they would face a denial of life-extending medical care beyond a certain age, the old would not necessarily fear their aging any more than they now do. They would, on the contrary, know that a better balance had been struck between making our later years as good as possible rather than simply trying to add more years. This direction would not immediately bring down the costs of care of the elderly; it would add new costs. But it would set in place the beginning of a new understanding of old age, one that would admit of eventual stabilization and limits. The time has come to admit we can not go on much longer on the present course of open-ended health care for the elderly. Neither confident assertions about American affluence, nor tinkering with entitlement provisions and cost-containment strategies will work for more than a few more years. It is time for the dream that old age can be an infinite and open frontier to end, and for the unflagging, but self-deceptive, optimism that we can do anything we want with our economic system be put aside. The elderly will not be served by a belief that only a lack of resources, or better financing mechanisms, or political power, stand between them and the limitations of their bodies. The good of younger age groups will not be served by inspiring in them a desire to Jive to an old age that will simply extend the vitality of youth indefinitely, as if old age is nothing but a sign that medicine has failed in its mission. The future of our society will not be served by allowing expenditures on health care for the elderly endlessly and uncontrollably to escalate, fueled by a false altruism that thinks anything less is to deny the elderly their dignity. Nor will it be served by that pervasive kind of self-serving that urges the young to support such a crusade because they will eventually benefit from it also. We require instead an understanding of the process of aging and death that looks to our obligation to the young and to the future, that recognizes the necessity of limits and the acceptance of decline and death, and that values the old for their age and not for their continuing youthful vitality. In the name of accepting the elderly and repudiating discrimination against them, we have mainly succeeded in pretending that, with enough will and money, the unpleasant part of old age can be abolished. In the name of medical progress we have carried out a relentless war against death and decline, failing to ask in any probing way if that will give us a better society for all age groups. The proper question is not whether we are succeeding in giving a longer life to the aged. It is whether we are making of old age a decent and honorable time of life. Neither a longer lifetime nor more life-extending tech~ology are the way to that goal. The elderly themselves ask for greater financial security, for as much self-determination and independence as possible, for a decent quality of life and not just more life, and for a respected place in society. The best way to achieve those goals is not simply to say more money and better programs are needed, however much they have their important place. We would do better to begin with a sense of limits, of the meaning of the human life cycle, and of the necessary coming and going of the generations. From that kind of a starting point, we could devise a new understanding of old age.

Death, Dying and the Ending of Life, II

440 132

ANNALS NEW YORK ACADEMY OF SCIENCES REFERENCES

1.

PRESTON, S. H. 1984. Children and the elderly: divergent paths for America's dependents. Demography 21: 491-495. 2. Americans for Generational Equity. Case Statement. May 1986. 3. Fucus, V. R. 1984. Though much is taken: reflections on aging, health, and medical care. Milbank Mem. Fund Q. 62: 464-465. 4. AvoRN, J. L. 1986. Medicine, health, and the geriatric transformation. Daedalus 115: 211-225. 5. CALABRESI, G. & P. BoBBITT. 1978. Tragic Choices. W. W. Norton. New York, NY. 6. MILLER, F. H. & G. A. H. MILLER. 1986. The painful prescription: a procrustean perspective. N. Engl. J. Med. 314: 1385.

[51] Am I Iviy

Parents~

Keeper? 1

NORMAN DANIELS

Must no one at al1, then, be called happy while he lives; must we, as Solon says, see the end? Aristotle, Nichomacbean Ethics, I: 10

1. OPPORTUNITY, AGE·BIAS, AND COMPETITION FOR RESOURCES Tf an acceptable general theory of distributive justice requires us to guarantee fair of opportunity, then a principle for the distribution of health care seems to follow? Institutions delivering health care services, both preventative and curative, should be governed by the fair equality of opportunity principle. On this view, health care is "special" because of its connection to the special social good, opportunity. Impairments of normal functioning mean that an individual might not enjoy the range of opportunities normal for his society. Health care needs, however, are things we need to maintain, restore, or compensate for the loss of normal species functioning. But then meeting health care needs is as important as guaranteeing individuals that their opportunity is within the normal range for their society. Although there are some problems in thus extending the notion of opportunity ,3 as we shall see, the distributive theory for health care that emerges is both rich and suggestive. Still, some people may raise an interesting objection to this approach. Any distributive theory for health care, we might suppose, should account for the importance of meeting the health care needs of young and old alike. In our system, people over age 65 use health care services at roughly 3.5 times the rate (in dollars) of those below 65. In 1977, per capita expenditures for those over 65 were $1,745; they were $661 for those age 19-64 and $253 for those under 19.4 But, so goes the objection, if fair equality of opportunity were the relevant principle of distributive

.1. equality

Death, Dying and the Ending of Life, II

442 518

NORMAN DANIELS

justice, we could not justify such extensive use of services for the elderly. Their opportunities lie in the past and are no longer a matter of pressing social concern, especially if we concentrate on opportunities to enter jobs and careers (which I do not). Consequently the opportunity principle cannot possibly provide a justification for our current distribution of health care services. Like the discounted future earnings approach to the problem of valuing lives, the fair equality of opportunity approach seems age-biased; it would discriminate against the elderly in a morally unacceptable way. Of course, it might be possible to counter this accusation of age-bias by arguing that it is our current practice, not the opportunity principle, which is agebiased. Some might urge that we spend-too much on health care for the elderly, for example in a last-ditch effort to extend life marginally or to prolong dying. One study shows that 50 percent of all hospital billings are to some 13 percent of the patients, the seriously chronically ill. About 40 percent of these "high-cost" patients are over 65, whereas only 15 percent of the low-cost patients are.5 We seem compelled to employ life-prolonging technologies whenever we can, which is more frequently among the elderly. Yet we are blind to the impact of such a policy on the health prospects of the young. For example, in a context of rising costs and scarce resources, and in any case, of tightened public budgets, we are more willing to impose stricter eligibility requirements and budget ceilings on Medicaid, most of whose recipients are young women and·children,!h-afi to alter our practices with regard to the dying elderly. Such rationing, so the reply goes, is age-biased against the young, as is, it might be added, the special immunity shielding Social Security benefits when all other social welfare programs are being drastically cut. I believe both the objection and the reply are so crudely drawn that they do not help us assess the status of the opportunity approach to health care distribution. But they do serve to raise a prior, indeed, more interesting and general ques· tion: When is a distributive system, such as a health care system, age-biased? Moreover, they highlight the importance of the question by raising the specter of age groups competing for scarce resources, pitting father and son and mother and daughter against grandfather and grandmother. To be sure, honor thy father and thy mother! But how much? and how long? Once raised, the specter of age-group competition for scarce resources threatens traditional values, like duties to the elderly, by eroding our confidence that we understand their limits. Similarly, the view that the elderly are entitled to support and deserve it, because of their past contributions to cooperative, productive schemes, also gives little guidance in answering the question. how much? 6 And other moral notions, like the injunction to respect persons equally, seem to give less guidance than we might hope.' Yet, these questions about competition for resources must be answered, and they will be answered, either by principle or by default. So we must look for a principled way to tell when distributive schemes are age-biased or fair. I try to formulate more clearly the general question about age-bias and compare it to other sorts of distribution problems in sections 2 and 3. Because birth cohorts are transformed successively into different age groups, and because we are

Death, Dying and the Ending of Life, II

AM I MY PARENTS' KEEPER?

443

519

concerned with distributive principles that govern social institutions over time, I formulate a modified "prudent saver" model. Allocations such savers approve will be morally acceptable distributions between age groups (sec, 4 ). I argue in section 5 that the notion of an age-relativized opportunity range will play an important part in the deliberations of such a saver, and this suggests how my opportunity approach to health care distribution can be spared the charge of age-bias. Indeed, when properly understood, we gain an edge on important resource allocation issues that underlie many criticisms of the U.S. health care system and its treatment of the elderly. In section 6, I draw some lessons about the problem of equity between birth cohorts from a brief glance at the Social Security system. Section 7 concludes with some qualifications.

2. WHEN ARE ACTS, POLICIES, OR INSTITUTIONS AGE-BIASED? It is tempting to think about age-bias, or agism, on the model of race and sex bias. Although philosophers have not written much about aging or age-bias, the gerontological literature has drawn some clear parallels. The elderly are portrayed as aminority that is treated in unfavorable, even discriminatory, ways by the more powerful majority. Some, using methods standardly employed to measure the effects of racism, e.g., measures of economic inequality, have suggested that agism induces even greater inequality than racism or sexism.8 Others have pointed out ideological similarities. Crude age-related stereotypes are generated. These may have a psychological appeal, based on fear of aging and death, and a rationalizing function, "justifying" policies, like compulsory retirement, that favor younger cohorts. These ster· eotypes notoriously interfere with the delivery of considerate, quality medical care and are reflected in derogatory hospital and nursing home jargon (and practices). Similarly, special problems arise from agist attitudes in mental health-care contexts.9 It is easy to think of cases in which appeals to age are morally wrong in much the way that certain appeals to race or sex are morally objectionable. A policy that cut off voting privileges for the elderly, or required them to take a competency test (on the model of driver recertification tests), would be morally objectionable, though, of course, we do allow age to play a role in assigning voting rights to the young. Practices that excluded the elderly from certain kinds of housing would be similarly objectionable. Job discrimination against the elderly-or against a protected age group (say those over 40)- has received attention, though not to my knowledge any sustained philosophical examination. Specifically, hiring practices or other job assignment and wage practices that appeal to age criteria and not competence seem morally objectionable in the way sex or race criteria do.10 Of course, there is an asymmetry here: we do exclude the young (say those under 16) from job eligibility, presumably because there is an overriding social concern that there are better things than working which the young should be doing for themselves. But this exception points us to the general issue: age, like race or sex, seems to be a morally irrelevant criterion for a broad range of contexts.

444

Death, Dying and the Ending of Life, II

520

NORMAN DANIELS

To be sure, we have to unpack the notion of moral relevancy here if we are to get a useful explanation, and this difficult task is not one I can undertake here. Still, many cases are clear. Race is not an indicator of competency to perform a job, and so it is morally objectionable to use it as a guide to hiring practices (except, possibly, in the context of certain compensatory practices, like affirmative action). Age is not an index of the likelihood of being a good tenant, and so is morally irrelevant to rental practices. And where some associate a relevant trait (industriousness, intelligence, crankiness) with a generally irrelevant one (race, age), the associations usually are, in the important cases, false; that is, they are part of a racist or sexist myth.11 At best they are crude, statistical generalizations which will clearly be unfair (by denying fair opportunity) to individuals, about whom the generalization is quite wrong. Although these cases and considerations explain the temptation to draw parallels between the use of age, sex, and race criteria, other cases challenge the analogy. Consider the question in a rationing context that has been criticized as agebiased by many, namely the policy that existed (at least implicitly) in the British National Health Service of not giving renal dialysis to those over age 65.12 Let us suppose that dialysis is medically effective for elderly patients, permitting relatively normal functioning, so that the age criterion is not merely a guide to medical suitability. Does the appeal to such an age criterion in rationing constitute an age-bias, by which I mean a morally unacceptable discrimination? Our earlier considerations suggest it does. If the sole difference between two persons, one age 64 and the other age 66, is their age, and that is the basis for deciding who gets dialysis, then it surely looks like the rationing scheme is age-biased in a morally objectionable way. But the rationing case is more complicated; contrary considerations come to mind. Consider two rationing schemes. Scheme A involves a direct appeal to an age criterion: No one over age 70 is eligible to receive any of several high-cost, life-extending technologies, e.g., dialysis, bypass operations, or angioplasty. Because age rationing greatly reduces the utilization of each ·technology, there are resources available for developing all of them. Scheme B rejects age rationing and allocates life-extending technology solely by medical need. As a result, it can either develop just one such technology, say dialysis, making it available to anyone who needs it, or it can develop several and ration them by lottery. Given our earlier discussion, Scheme A seems age-biased in a way that B is not. The effect of B, however, is to reduce the likelihood of people under 65 reaching a normal life-span (say "three score and ten"). Some would contend that Scheme B, though it lacks reference to an explicit age criterion, has a systematic negative effect on younger age groups and is in that sense age-biased in a morally objectionable wayP Of course, the contention depends on showing that maximizing the likelihood of reaching a normal lifespan is morally preferable to merely extending life wherever we can (without any reference to age). Considered moral judgments differ on this and related issues {and in ways that may reflect our interests given our ages); moreover, there are strong considerations and arguments inclining us in opposing directions. The problem is made to seem even more intractable because these moral disagreements are set in

Death, Dying and the Ending of Life, II

AM I MY PARENTS' KEEPER?

445

521

the context of a distributive framework that makes one group's gains look like another group's losses. I would like to su~gest a different di~tributive framework for conceiving the problem, one that permtts a fresh theorettcal perspective. The perspective can be introduced by observing an important fact about certain health care insurance schemes. Suppose we have a health care financing scheme that guarantees substantial access to medically needed health care services for the elderly. The details of the scheme do not much matter here. It could be a universal national health insurance scheme with subsidization for those who cannot afford premiums,14 or it could be a composite financing system that included private as well as publicly subsidized programs. But even a scheme that does not redistribute income raises the same issues provided that it is "community rated" and incorporates all ages into one riskpool. The central fact is that health care needs vary with age, so that the elderly will use certain health resources at a higher per capita rate than the working· age population. In 1977, the 10.8 percent of the population over 65 incurred 29 percent of the total personal medical services bill. Consequently, any such insurance scheme involves a transfer of wealth from later birth cohorts to earlier ones, from younger age groups to older ones. But if the insurance scheme continues over a long period of time, birth cohorts that are now transferring wealth-aid-in-kind-to their elders will eventually be the beneficiaries of such transfers from later birth cohorts. Consequently, any such health insurance scheme can be viewed over time as a savings scheme: Participation in the scheme transfers resources, in the form of contingent claims on health_ CaJ"e services_, from one's youth to one's old age. Of course, such savings are not "vested" assets, like money in the bank; but we are deferring resources from one point in our lives to another, and so have a kind of savings scheme. Notice how focusing on an institution-the insurance scheme-that operates through time forces a shift in our perspective on the rationing problem. We are driven to converting the synchronic or time-slice distribution problem we first raised, namely, how to ration health care resources between competing groups while avoid· ing age-bias, into a diachronic perspective in which we are concerned with the treatment of the same people through the various stages of their lives. From this perspective-from my perspective-three questions about the design of the institution, here the insurance scheme, arise in a quite natural way: {a) At what rate of savings should I defer the use of health care resources within my life? (b) What do I most need and want by way of health care benefits at each stage of my life? (c) How can I be sure that my participation in the scheme involves equitable transfers between my birth cohort and both earlier and later ones, given the fact that economic and population growth rates vary through time? The last question is familiar, of course, because of current worries about intercohort inequities in the Social Security system, and I shall return to these matters in section

Death, Dying and the Ending of Life, II

446

522

NORMAN DANIELS

6. The answer to question (b) will have a bearing on the answer to (a). I have raised these questions in the first person. But, because we are concerned with cooperative social schemes, it may be necessary to answer the questions from a more general perspective, that of a prudent saver, or even some more hypothetical construction in which the saver operates behind a "veil of ignorance" of appropriate thickness. I return to this issue in sections 3 and 4. But first I want to return to the rationing problem I posed earlier, using the perspective suggested here. Suppose I know I have available to me a lifetime health care allocation, say in the form of an insurance benefit package. However, it is up to me to budget, once and for all, that allocation or benefit package so that it is used to meet my needs and preferences- over my lifetime ...How would it .be rational for me to budget itgiven all the uncertainties about my future health, wealth, and family situation? One plausible proposal might be for me to reserve certain life-extending technologies for my younger years, reasoning that my doing so maximizes the chances of my living a normal life-span. I then might use the "savings" embodied in that restriction to provide myself with more social support services in my old age. I might reason that such services could vastly imporve the quality of my years in old age and that such an improvement is worth the increased risk of a slightly shortened old age. I might then instruct-through my benefit package-providers to treat me accordingly, that is, to appeal to an age criterion in their utilization decisions concerning me. This package is intended to resemble the age-rationing scheme the British NHS apparently used for hemodialysis, and a rationale for the NHS scheme could be modeled on my reasoning about my package. Under this scenario, although age is used as a criterion in the utilization decisions involving me and everyone else who joins the same insurance scheme, there is a minimal basis for suggesting my treatment is age-biased in a morally objectionable way. It might be thought that there is no "bias" here merely because I consentbuy into-the scheme. But the fact of my actual consent to the scheme is not the main issue here. Consent does not quite count for everything: blacks or women might consent to race- and sex-biased treatments without thereby overriding all claims that the treatments are morally objectionable.15 As I suggest in the next section, there is an important difference between the age and race or sex distribution problems, and it is this difference that explains why age-rationing in such schemes is not morally objectionable in the way race or sex rationing would be. We are not in a position to answer the question, when are acts, policies, or institutions age-biased? But we have seen strong reason to think that not every appeal to an age criterion for rationing is as morally unaceeptable as comparable appeals to sex or race would be. We must explore further why the cases are different.

3. DOES AGING POSE A DISTINCT DISTRIBUTION PROBLEM? The distribution problem between age groups is usefully contrasted with two distribution problems it somewhat resembles. Consider first whether the age-group

447

Death, Dying and the Ending of Life, II

AM I MY PARENTS' KEEPER?

523

problem is just a special case of the problem of obligations to future generations. After all, age cohorts are commonly referred to as 'generations'. And both seem to raise the issue of competition for resources: present and future,generations-just like age groups-compete with each other. The problem of obligations to future generations is also sometimes formulated as the problem of finding a just savings principle. So too the issue of a fair savings principle arises in the age-group problem, at least if we view schemes that transfer income or health care benefits from young· er cohorts to older ones as a kind of "savings institution." What rate of transfer, what savings rate, is just? Moreover, there is another point of similarity: transfer schemes operating through extended periods must be concerned that different birth cohorts enjoy equitable "replacement ratios." (The replacement ratio is the ratio of benefits to contributions.) This problem of equity does strongly resemble the just savings problem between generations} 6 Nevertheless, I think the differences between the aging and future generations problems are greater than their similarities. The major difference is that young birth cohorts are transformed in time into elderly cohorts: they age. But no current generation becomes a future generation. It follows that certain special features of the future generations problem do not arise in the aging problem. We do not have to consider the great uncertainties about conditions of life in the very distant future; we do not have to worry about the puzzling conceptual problems that may attend positing obligations by existing people to nonexistent ones. But the most important consequence of this difference between the problems is that some form of prudential reasoning is naturally appropriate to solving the age-group distribution problem in a way that it is not for the future generations problem. I know I will grow old, or at least that I must prepare for the eventuality of growing old. So, I have a concern for the structure of institutions that will help me defer the use of resources in a ptudent way. In no such direct way does prudence make me concerned about saving resources for future generations. (Of course this contrast will have to be qualified somewhat in the next section since we will ultimately be talking about prudence in a more hypothetical context.) A further difference between the aging and future generations problems is that different age groups coexist and compete politically for social goods distributed in cooperative arrangements. Future generations are not here to fight for their interests. This difference may be an important psychological and political fact; it is less clear how relevant it is to the problem of deciding what arrangements are in principle just. It does, however, introduce some issues about the implementation of cooperative schemes and the contrast between ideal and nonideal arrangements, to which I return in section 6. The fact that different age groups coexist and share in a distribution scheme contemporaneously may make the age-group problem resemble more closely the distribution problems among other demographic groups (races, sexes, classes) than it does the future generation problem. The similarity enhances the sense that we are concerned with a synchronic distribution problem, a time-slice in which competition rages. Moreover, as we saw in section 2, many of the same issues arise in all

Death, Dying and the Ending of Life, II

448

524

NORMAN DANIELS

these demographic competitions. The moral arbitrariness of certain appeals to age resembles the arbitrariness of appeals to race or sex. Similarly, we are concerned that our treatment in cooperative schemes should reflect the fact that we are all equally to be considered as persons, regardless of age, sex, race, class, and so on. But the very same fact that makes the aging problem different from the future generations _problem also _makes it different from the distribution problem involving other demographic groups. Young birth cohorts age and are transformed into older age groups. We become old, but we do not change generations, race, sex, or (usually) class. As Zeckhauser and Viscusi put it, "the elderly comprise a minority group we can all hope to join." 17 This basic fact points to the naturalness of the suggestion that we think about cooperative social schemes that bear on aging in prudential terms, even though we may have to abstract somewhat from the perspective of a real individual, that is, from the economist's "prudent saver." It is now possible to explain why appeal to an age criterion in some rationing schemes works differently from appeals to race or sex criteria. From the perspective of institutions operating over time, the age criterion operates within a life and not between lives. One important criticism of utilitarianism, advanced by Rawls, is that it extends a principle of individual rational choice appropriate to distributions within a life into a social principle of rational choice that crosses personal boundaries.111 Thus it is rational and prudent that I take from one stage of my life to give to another; in order to make my life as a whole better. But it is morally problematic just when society can take from one person to give to another in order to maximize, say, total happiness. Rawls's point would explain the deep problem facing any at· tempt to ration life-extending resources by race or sex: in this case, taking from some to make society as a whole better off would fail to respect the difference between persons. But now we see the difference between the race and aging cases. Rationing by age criteria looks like a case of crossing personal boundaries only if we take a "time-slice" perspective. Once we take the perspective of institutions operating through time, the appearance of crossing boundaries between persons fades and we are concerned primarily with distribution through the stages of a life. No comparable point is true for rationing by race or sex over time. This general point is not to deny there are some irreducible interpersonal aspects of the cohort issue. For example, the question of equity in replacement ratios between birth cohorts raises an issue of equity between persons in the face of changing economic growth rates or birth rates. This issue aside, however, the core of the age-group problem has a different philosophical texture from either of the other distribution problems we have considered. The distribution problem between age groups must, of course, be set within a framework that takes more general issues of distributive justive into account. This might suggest there is no special problem of distribution between age groups. One might, for example, think Rawls proceeds as if there is no special problem of justice. His Difference Principle requires that the worst-off groups are to be made maximally well of as measured by an index of primary social goods, which includes basic liberties, opportunity, income, wealth, powers, and self-respect. But the value of

Death, Dying and the Ending of Life, II

AM I MY PARENTS' KEEPER?

449

525

the index for a representative individual is determined by his share of primar1 goods over his lifetime; thus it is to include what social or economic mobility the individual will enjoy or regret. This lifetime index assignment might suggest that Rawls ig· nores the problem of distribution between age groups, perhaps dismissing it as a problem for individual savings. But Rawls's simplifying assumption involving the index is not a slight of hand that makes the problem of distribution between age groups disappear. The problem we are concerned with reemerges as soon as one tries to arrange basic social institutions that embody the more general principles of justice over time. For then, the problem of rationing income or health care benefits throughout the stages of a life arises again, and this problem requires the establishment of cooperative schemes or institutions of a rather basic sort. The Difference Principle, to continue with the Rawlsian example, maximizes the index level of representative, worst-off individuals over their lifetimes. But several cooperative "savings" schemes might be compatible with satisfying the Difference Principle. That is, the more general theory of distributive justice is silent on the age-group distribution problem except where intercohort transfer or savings schemes interfere with the Difference Principle. Moreover, if I am right that health care institutions should be governed by ·the fair equality of opportunity principle, as I later extend it, then at least this "savings institution" constrains the Difference Principle.

4. PRUDENCE AND AGING I have been suggesting that we approach the problem of competition-or distribution-between age groups from the perspective of institutions that operate through time to defer resources from one stage of life to another. But in converting what began as an interpersonal distribution problem, with all its attendant worries about age-bias, into an intralife problem of rational or prudent savings, my approach encounters an objection, one with a paradoxical air about it. The objection is that the shift to talk about prudential allocations of resources within a life, far from telling us when distributions between age groups are just or age-biased, prohibits us from raising the question about age-bias at all. For, when a person favors one stage of life over another, even the person's inadequate or inappropriate allocation to a stage of his life-or, in a cooperative scheme, to an age group-is not viewed as immoral or unjust. It is merely imprudent, at worst. Consider some examples. Olga is a figure skater who has invested very heavily in the development of certain talents and skills while neglecting others. She has ignored the development of critical social skills, acquired only the narrowest educa· tion, and led an austere, even grim, childhood and youth. If she achieves wealth and fame in her career-becomes a star of the Ice Capades-she may feel the gamble has paid off. Later stages of her life will reap benefits from the sacrifices of earlier stages. But what can we say about the extreme hardship and sacrifice she experi· ences in her childhood? They might be imprudent (even if the gamble pays off). But it seems merely metaphorical to say she was "unfair" to her childhood. So, if

Death, Dying and the Ending of Life, II

450

526

NORMAN DANIELS

the intralife model precludes saying that the plan is unfair to a life-stage or biased against it, it looks like we are not solving the problem we thought existed. Of course, it does make a difference whether the plan for Olga is hers or is imposed by ambitious parents and skating instructors, perhaps even with "false consciousness" on Olga's part. But here the unfairness to Olga is both that she was denied resources and opportunities thought normal to development and that she was denied auton· omy, the chance to design her own life plan. Consider a case in which Olga's plan is writ large into a social policy, a kind of initiation rite. From age 20 through 30, people are given just a living stipend for the work they do. They accumulate no property and lead austere lives. After the initiation or "social indenture" period, they are presented with an annuity polity that enhances their income at a later stage of life, or they are given some other award, perhaps just acceptance as full-fledged citizens who benefit from the labor of the next birth cohort going through the process. Is such a scheme. age-biased?. Depending on details, a central complaint might be that the system too severely restricts certain liberties, which we may see as a social good that should not be rationed or "saved" in this way. B1..1t, liberties aside, suppose the system were stable, seemed to reflect a shared conception of a rational plan of life, and appeared to be as voluntary as any well-entrenched social custom involving initiation rites. We might be inclined to say it is imprudent for the indentured cohort to "save" in this fashion; they might disagree. Of course, we might not be able to say even that if the "return on investment" for participation is higher than in alternative schemes. In any case, does our ability to complain about age-bias disappear? The examples really raise two issues. The first is primarily terminological. Ordinarily, we do not import moral notions, like fairness and justice, into prudential contexts, viz. allocations within a lifetime. But the proposal here is that our proper standard for judging the fairness of distributions between age groups-which do exist as distinct groups of people, in contrast to stages within a life-is prudential. We are to view the different age groups as if they were but stages of one life, for, from the perspective of cooperative "savings" schemes operating through time, each person is treated at the different stages of his life in just the same way the different groups are. Accepting the proposal would be grounds for ignoring the suggestion that the language of prudence bars us from raising issues of justice. But the examples also point out that what is prudent from the perspective of one rational person or group of persons may not be from that of another. The ap· peal to a prudential or rational savings model usually carries with it the notion of an individual with a given set of preferences or "conception of the good." What is prudent is so from the individual's conception of the good. How, then, can we use the suggestion that prudential reasoning is the key to solving the distribution problem for different age groups? The social institutions that bear on saving encompass people with different conceptions of the good. There are two main strategies. One is the proposal that we rely on market me· chanisms to allow people every chance to express their own prudent preferences. The social task, then, is to make sure such markets function properly and that in-

Death, Dying and the Ending of Life, II

AM I MY PARENTS' KEEPER?

451

527

come distribution is initially just. Specifically, with regard to health care, one mighr look to a market for insurance schemes that differed from one another in their "rate of savings" for later stages of life. That is, some might have lower premiums and offer less coverage in later years; others might defer more resources, in the form of contingent benefits. People would then buy the packages it is prudent for them to buy, given many facts about their situations, including their conceptions of the good, risk averseness, and so on. This approach converts with a vengeance the problem of age-group distribution into an individual savings problem: the social concern is to provide a setting in which individual rationing within lives can take place. But there is little room for social institutions to guarantee that prudent allocation takes place. The second strategy is to modify the appeal to prudential reasoning by using a hypothetical agent, one which abstracts from certain features of individuals. Such an agent then seeks principles for the design of the relevant social institutions. I shall suggest a version of the second strategy which is appropriate to the "savings" problem for health care. In its general form the strategy is familiar as the hypothetical contractarian approach used by Rawls and others. But it is worth considering some limitations of the first strategy first. The strengths and limits of the first strategy are revealed if we consider the way in which a rational consumer might think about the problem of chronic illness or disability. The long-term care such conditions require is a focal point of criticism of the treatment of the elderly. In our system, chronically disabled or enfeebled persons tend to be institutionalized much more frequently and earlier than comparable persons in other systems, e.g., the British or Swedish. Moreover, they are often institutionalized at inappropriate levels of care, and possibly at higher cost than alternative forms of treatment or service would involve. The incentives for such institutionalization are built into Medicare and Medicaid reimbursement schemes. The effects of such "overmedicalization" on both the mental and the physical health of the elderly are serious.111 Yet, as Christine Bishop points out, the uncertainty facing the onset and costs of disability make it an obvious candidate for insurance schemes.20 The rational consumer would presumably try to buy a package that avoided the features of our current long-term care system. Any individual faces a significant, actuarially calculable chance of chronic illness or disability over his lifetime; the chance increases with age (let us leave aside those disabled from birth or facing a known genetic disposition to disability). Although only one in twenty persons over 65 is in a nursing home in a given year, one in four will at some time enter one.2 1 Chronic illness or disability may require large expenditures for medical, personal care, or social support services. Moreover, the size of the expenditures for a given disability will vary with other contingencies, such as family situation and preferences for living conditions. The uncertainty surrounding each of these contingencies and their joint risk suggests that rational consumers will enhance their well-being over their lifetime if they pay a modest insurance premium rather than keeping the money and risking a large loss. Specifically, we might expect rational consumers to want insurance schemes that offered them benefits flexible enough to meet their real needs. They would want alternatives to

Death, Dying and the Ending of Life, II

452 528

NORMAN DANIELS

nursing home institutionalization if they needed lower levels of care, or some family help, or modifiable living quarters. Thus they would buy contingency claims on the joint risks of disability and other factors, such as the absence of family support or the unsuitability of living arrangements. The connection between disability as an insurance problem and as a problem of "savings" becomes clear when we see, as Bishop notes, that short-term coverage faces special problems. If coverage is actuarially fair and we riskpool by age, high premiums will face the elderly, those most in need of the insurance and those least likely to be able to pay for it because of declining incomes. The prudent consumer, anticipating such higher premiums, would have to save, perhaps by buying an annu· ity to cover his later premiums. But since one does not know how long one will live, to 100 or to 66, it is hard to predict how much to save. Notice, however, that plans offering lifetime coverage with a fixed premium are equivalent to such savings: a community rated lifetime plan has a built·in savings feature because of the distri· bution of needs by age. Although these considerations suggest there should be a demand for such insurance, we find no market offering it. Bishop points to several reasons for marketfailure: (1) uncertainty about inflation adds to the insurer's risk, where real benefits and not fixed dollar amounts are involved, so private coverage would be discouraged; (2) administration costs are high if coverage of the population is not extensive; (3) some current public programs would partially undercut the market for such insurance; (4) "adverse selection," which means too many high-risk people buy, driv· ing premiums up and low·risk people out, and "moral hazard," say in the form of overstating disability, are especially likely for these forms of insurance. From these facts, Bishop concludes that private marketing of such insurance is not likely to develop and that some form of universal, compulsory insurance should be instituted, different proposals for which she considers. Bishop's proposal for a unified national insurance scheme for long·term care, encompassing medical, personal care, and social support services, is surely a step in the right direction. Moreover, her discussion of the scope and content of the lifetime coverage scheme is informed by prudential considerations, which I earlier sug· gested were necessary to undercut the issue of age group competition. The scheme continues through time so that the young who pay higher premiums (or taxes) now in due time will be beneficiaries of such "savings," through the similarly higher payments of later birth cohorts. Still, there is a gap in her argument: the social obligation to provide such compulsory coverage does not follow from the fact of private market-failure alone. Nor does it follow from the fact that net well·being might be greater if the public scheme were instituted, for there are many public schemes that might enhance net well-being. We need some argument that the social good protected by such a cooperative social scheme is especially important, say because it is a social good that gives rise to claims of justice. Schemes such as the one Bishop discusses involve a significant income redistribution: entitlements to benefits, presumably at an "adequate" level, will be subsidized for those who cannot buy them. But what deter-

Death, Dying and the Ending of Life, II

AM I MY PARENTS' KEEPER?

453

529

mines that level? And why should those who are better off be willing to provide it? Moreover, as long as we are considering such insurance schemes from the perspective of the prudent agent who knows his full situation, we might find much reluctance on the part of some to enter community-rated schemes whose premiums involve subsidization of those with the worst risks. If I know I have several children whom 1 am likely to be able to prepare for lucrative careers, I might not want to be in a riskpool with childless people. My commitment to a community-rated scheme that is not actuarily fair to me would have to be based on considerations other than prudence alone. My knowledge of particular facts about me allows my individual interests to influence choice. Any bargains struck in the light of full knowledge then risk allowing the accidents of current age-group competition to unduly influence the arrangements governing long-term cooperative schemes. These considerations suggest that distributive principles which we need to embody in such cooperative social schemes are not likely to be derived- from the -prudential perspective of fu11y informed rational agents. These limits of the first strategy, and other issues in moral methodology which cannot be discussed here, incline me to the second strategy. That is, for the design of cooperative social schemes, we need a perspective that abstracts in a reasonable way from the full-blown rational consumer used by the economist, but -Which still permits some form of prudential reasoning about the "savings" problem for health care. At this point it is tempting to employ some version of Rawls's veil of ignorance.22 Thus the prudent agents deliberating about principles to govern their cooperative scheme should know nothing of their age, family situation, health status and genetic history, socioeconomic status, or their particular conception of the good. Such a device might be defended on the grounds that the constraints are procedurally "fair": they reflect the deliberators' status as "free, equal moral agents." Of course, the agents would have to have some "thin theory of the good," like Rawls's primary social goods, or they would have nothing to be prudent about. Any justification for such a hypothetical contractor model would carry me much too far afield.23 Clearly it is not enough to suggest that the constraints on knowledge seem to be but exaggerations of the considerable uncertainty we face outside the veil in planning health, family, and economic eventualities over a lifetime. But since I am not prepared to offer such a defense, l shall have to restrict myself to a suggestion. Prudent deliberators, appropriately constrained, would seek a health-care and longterm-care system that protected their normal opportunity range at each stage of their lives. The notion of an age-relativized opportunity range needs explanation, and I explore the merits of this suggestion in the next section.

5. EQUAL OPPORTUNITY AND HEALTH CARE FOR THE ELDERLY I suggested that meeting health care needs is of special moral importance because it enhances fair equality of opportunity; specifically, it helps guarantee individuals a fair chance to enjoy the normal opportunity range for their society. The normal

Death, Dying and the Ending of Life, II

454

530

NORMAN DANIELS

opportunity range for a given society is the array of "life plans" reasonable persons in it are likely to construct for themselves. The range is thus relative to key features of the society-its stage of historical development, its level of material wealth and technological development, and even important cultural facts about it. Facts about social organization, including the conception of justice regulating its basic institu· tions, will of course determine how that total normal range is distributed in the population. Nevertheless, that issue of distribution aside, normal species-typical functioning provides us with one clear parameter relevant to determining what share of the normal range is open to a given individual, holding constant, for the moment, the individual's skills and talents. Impairment of normal functioning through disease (and disability) constitutes a fundamental restriction on individual opportunity rel· ative to that portion of the normal range which the individual's skills and talents would ordinarily have made available to him. Of course, we also know that skills and talents can be undeveloped or misdeveloped because of social conditions (e.g., family background). So, if we are interested in having individuals enjoy a fair share of the normal opportunity range, we will want to correct (say through education) for special disadvantages here too. Still, restoring normal functioning has a particular and limited effect on an individual's enjoyment of the normal range. It lets the individual enjoy that portion of the range to which his full range of skills and talents would give him access, assuming that these too are not impaired by special social disadvantages. There is no presumption here of eliminating individual differences: these act as a baseline constraint on the degree to which individuals enjoy the normal range. 24 The notion of a normal opportunity range can be refined for its special appli· cation to our problem about distribution between age groups. Life plans, we might note, dearly have stages, which reflect important divisions in the life cycle. Without meaning to suggest a particular set of divisions as a framework, it is easy to observe that lives have phases in which different general tasks are central: nurturing and training in childhood and youth, pursuit of career and family in adult years, and the completion of life projects in later years. Of course, what it is reasonable to include in a life plan for a stage of one's life not only reflects facts about one's own talents and skills, tastes and preferences, but also depends in part on social policy and other important facts about the society. These qualifications already are present in the notion of normal opportunity range itself. The suggestion I want to explore is that prudent design of the institutions that affect us over the different stages of our lives requires reference to the notion of age·relativized opportunity range. Specifically, prudent deliberation about the design of such institutions, carried out with the degree of abstraction from individual perspective appropriate to the task, would attempt to assure individuals a fair chance at enjoyment of the normal opportunity range for each life stage. With this refinement, the fair equality of opportunity account I am proposing for health care will avoid the pitfall of age-bias. Consider now the perspective of designers of a health care system who are under an appropriate veil of ignorance. It keeps them from knowing their individual

Death, Dying and the Ending of Life, II

AM I MY PARENTS' KEEPER?

455

531

health status, conception of the good, age, income, and other important facts about themselves as individuals. At the same time it lets them know important facts about the disease/age profile for their society, its technological level, and even that life· span has been increasing, largely as a result of other features of social policy. One feature of their problem emerges as critical: in choosing principles for institutions that defer the use of resources, they must assume life-span is normal. Since they cannot appeal to any very special conception of the good, which might lead them to discount the importance of their projects or plans at a certain stage of their lives, they must treat these stages as of comparable importance. Here they are simply in compliance with Sidgwick's account of rationality: each moment of life is equally valuable and m1.1st not be discounted merely because it comes at one point in our lives rather than another. Of course, there are standard problems facing the Sidgwickian view. Even -though it-disallows "p1.1re time preferences," it does not block some "impure" rea· sons for discounting the value of certain moments. Nevertheless, given our problem of design from behind a veil of ignorance, some special views people might hold are not available to them. Still, there are other problems. The concept of rationality itself does not determine which moments are to count as "ours." If I refuse to plan for "my" care when "I" have advanced senility, am I being imprudent? Am I being irraj~onal if I insist that that senile person (if he is one) is not really me and that I do not care what happens to him? I ignore these worries here. From their perspectives, prudent deliberators will not know just what their situation is, what preferences or projects they might have at a given stage of their lives. Still, they do know that they will have a conception of the good and that it will determine what is meaningful for them in their lives. But then it is especially important for them to make sure social arrangements give them a fair chance to enjoy the normal range of opportunities open to them at any stage. This protection of the range of opportunities they enjoy is doubly important because they know they may want to revise their life plans; consequently they have a high·order interest in guaranteeing themselves the opportunity to pursue such revisions. But impairments of normal species functioning clearly restrict the portion of the normal opportunity range open to an individual at any stage of his life. Consequently, health care serv· ices should be rationed throughout a life in a way that respects the importance of age·relativized opportunity range. Let us consider two implications of this view for the design of health care systems, keeping in mind that these systems operate through time on all stages of one's life. The first implication is the suggestion that personal medical services have the same underlying rationale for their importance as various personal care and social support services for the disabled. Medical services are intended to preserve or restore normal functioning; in turn, normal functioning is important because of its impact on individual opporutnity range. But the same rationale makes personal care and support services for the disabled elderly person important: they compensate for losses of normal functioning in ways that enhance individual opportunity. A major criticism of the U.S. health care system, that it encourages premature

Death, Dying and the Ending of Life, II

456

532

NORMAN DANIELS

and overinstitutionalization of the elderly, should be assessed in this light. The issue becomes not just one of costs: is institutionalization more or less cost-effective compared with home care and social support services? Rather, opportunity range for many disabled persons will be enhanced if they are helped to function normally outside institutions. They will have more opportunity to complete projects and pursue relationships of great importance to them, or even to modify the remaining stage of their life plans within a greater range of options. Often the issue is discussed in terms of the loss of dignity and self-respect that accompanies premature institutionalization or institutionalization at inappropriate levels of care. My suggestion here is that the underlying issue is loss of opportunity range, which obviously has its effect on autonomy, dignity, and self-respect. Viewed in this light, the British system, in which extensive home care services exist, far more respects the importance of normal opportunity range for the elderly than does our system. The second implication is more controversial, and I am less sure of it. I believe that prudence would dictate giving greater emphasis to enhanCing individual chances of reaching a normal life-span than to extending the normal life-span. It might at first seem that such a contention runs counter to the earlier appeal to Sidgwick's principle, that it is irrational to entertain pure time preferences. But I am not urging that a given moment of life for a person older than the normal life-span is worth less than a comparable moment for a younger person. About that, the prudent deliberator can make no judgment. But he must acknowledge several important considerations. Assume, for the moment, that productivity and birth rate are held con· stant. Then, increasing life-span-here beyond the normal range-must compel us to save resources at a great rate in earlier stages. Where policies lead to greater longevity primarily because they reduce infant and childhood mortality rates, we are likely to have some increase in productivity, which may not necessitate greater rates of saving. But where the extension is due primarily to extending marginally the lives of elderly people, we clearly are required to save at an increased rate. To the extent that such increased savings undermine the ability to protect normal function in younger age groups, or even in the late stages of a normal span, we face an increased likelihood of not reaching a normal life-span. Prudence would thus urge us to pursue a different policy. Under the conditions imposed here on institutional design, we can abstract from what might merely be thought a matter of "personal taste," whether to live a longer life with fewer resources or enjoy a better chance of living a normal life-span. Consider the rationing schemes discussed in section 2. Recall that Scheme A permitted no one over age 65 or 70 to get certain high-technology, high-cost services. Such rationing by age permitted the development and use of more such services for younger people. Scheme B developed fewer such services and rationed them solely by criteria involving medical suitability and lotteries. I am suggesting that prudential considerations would incline our modified prudent deliberators choosing between such schemes to prefer an enhanced chance of reaching a normal life-span over an increased chance of living a life slightly longer than the normal span. If this conclusion is correct about where prudential considerations incline us,

Death, Dying and the Ending of Life, II

AM I MY PARENTS' KEEPER?

457

5B

then my strategy of using prudence to guide justice in distributions between age groups should lead us to think Scheme B is morally preferable. The whole point here is that the scheme works through time: each of us, not just a particular group of people, will enjoy the increased chance of reaching a normal life-span. And our gain in this regard is not made at the expense of another group, but at the expense of our reduced chance of living to a longer than normal life-span. The point brings to mind a rationing practice ascribed to the Aleuts: the elderly, or perhaps only the enfeebled elderly, are sent off to die, sparing the rest of the community from the burden of sustaining them. From descriptions of the practice, the elderly quite willingly accept this fate: and it is fair that they should. They were the beneficiaries of comparable sacrifices by their parents and grandparents. If prudence demands such a harsh rationing scheme in the conditions the Aleuts face, then we are blocked from any suggestion that the practice is age-biased in a morally unacceptable way. Yet this example should remind us that such rationing schemes are prudent only under certain explicit conditions. So the prudence of selecting Scheme A over B in the preceding deliberation is quite sensitive to assumptions about the scarcity of resources and the way in which policies involve explicit trade-offs. The argument is r;ot a general defense of all schemes for rationing by age. It is worth noting one last implication of these considerations and this strategy for approaching the age-group distribution problem. Where prudential considerations do not decide between alternative schemes-and some might reject my argument leading to the selection of A over B-we may not be dealing with a consideration of justice at alL More generally, several schemes may all appear prudent, and then we have no basis in considerations of justice for distinguishing among them. Where there are honest differences about what is prudent, we may be dealing with cases whose resolution call for a democratic political process, not transcendent principles of justice.

6. EQUITY, ERRORS, AND THE STABILITY OF "SAVINGS INSTITUTIONS" Thus far I have been ignoring an important question facing cooperative schemes in which saving is accomplished through a compact between birth cohorts. In such schemes, one birth cohort transfers resources for the use of earlier birth cohorts and receives similar transfers from its successors. (We support our parents and expect our childrens' support in return.) How can a given cohort be assured that its benefits from the scheme will be equitable when compared with the benefits enjoyed by other cohorts? If we call the ratio of benefits received to contributions made the 'replacement ratio', our problem is to determine when replacement ratios are equitable.25 The problem arises because we must operate such a savings scheme under conditions of considerable uncertainty. Most important, there is uncertainty about population growth rates, economic growth rates, and technological change, with its impact on productivity. Consequently, any such scheme must derive its stability

Death, Dying and the Ending of Life, II

458

534

NOR.'"AN DANIELS

from an underlying commitment to equity in replacement ratios enjoyed by successive birth cohorts. Errors are likely to abound, and inequities will arise, but the pre· sumption must be that these errors will be corrected. Still this presumption in favor of correcting errors does not mean that everything is up for renegotiation all the time. The basic institution must be stable. Clearly we need some theory about what equity involves here. I think this problem of equitable treatment between birth cohorts resembles the problem of a just savings rate between present and future generations, but I cannot say more about the connection here.Z6 Instead, I shall rely on the point just made about the higher-order interest parties have in ensuring the stability of the savings institution. Stability requires a belief in equity.If one cohort seeks terms too much in its favor, say when it is young, it will very likely pay the price when it is old; similarly, if it seeks too much when it is old, it will risk rebellion from the young. My guess is that there is a tendency to view equity as requiring approximate equality in replacement ratios. In any case, I shall make such an assumption, primarily for the sake of illus· trating a slightly different point. Suppose, then, that "mature" savings schemes, those in which beneficiaries have been longtime contributors, should treat different birth cohorts equally. They should aim for equal replacement ratios. In the "steady state" condition, where there is no economic growth and no population increase, the assumption is unproblematic. In favorable conditions, of positive economic growth and incre-asing population, we can in fact do better in the following sense: benefits can steadily increase, even if rates of contribution do not. Of course, equity considerations between cohorts might incline us to temper this "chain letter" effect: if we could project the economic and demographic trends, we might raise the replacement ratios of earlier cohorts somewhat to offset the anticipated increasing rates of later cohorts. Unfortunately, in our social security and health care contexts, we now face the opposite conditions: declining birth rates and poor economic growth. What this means in the Social Security system, for example, is that the current 3: 1 ratio of contributors to beneficiaries will decline to about 2:1 by 2030 (in the immature system of the 1950s, ratios were even more favorable). To maintain current benefit levels, contributors have to be taxed at tremendously high rates. (The same point applies to health care "savings" schemes.) What is worse, much of our recent planning, including the major benefit increases of 1972, seemed to ignore these shifts. To have planned for equity in replacement ratios, many critics argue, would have required (1) taxing earlier generations at a yet higher rate than they were taxed, (2) stunning increases in real wages through rapid economic growth, (3) reducing benefits substantially in the interim, or (4) some combination of these steps. One problem is that the presumption in favor of equitable treatment between cohorts encounters strong resistance in the political arena. Some cohorts are in a better position to protect their interests than others, undermining long-term stabil· ity of the scheme. A second problem is that not everyone believes or understands the problem, and there is often reason to think some factors are exaggerated,

Death, Dying and the Ending of Life, II

AM I MY PARENTS' KEEPER?

459

535

perhaps for ideological reasons. Thus many people point with alarm to the shifting ratios of contributors to beneficiaries. But there is a countervailing point: the total ratio of employed to nonemployed, lumping retirees with children, is not changing in such an alarming way. The ratio of total nonlabor to labor is higher in 1975 than for any year through 2040?7 The implication is that the smaller number of children will require fewer resources, which can then be diverted for use by the then elderly baby-boom cohort. What follows, then, is that we must not look too much at one distributive institution in isolation from others. Nevertheless, the general point remains. We operate an income or savings scheme in a nonideal context. It will always encounter various sorts of "interstitial" equity considerations which are generated by both great uncertainty and political expediency.28 A good example is the tremendous replacement ratio advantage offered the early entrants into our Social Security system: attempting to lower that ratio might have undermined political support for the Social Securi_cy system as a whole. Similarly, in the United States, no fund was ever generated which was significant enough to cushion the effects of our current decline in real wages and declining population growth rate. Politicians were afraid to raise tax rates without pairing the increases with benefit increases. More interesting details of this history are available and constitute an important case study of the contrast between ideal and nonideal contexts.29 It is interesting to note that health care savings schemes face comparable-or even more serious-problems of birth cohort equity. First, as with income support schemes, there will be a bias in favor of early entrants. Such a bias is hard to avoid in immature schemes. But there also is an opposite bias in the case of health care. Consider a scheme in which some form of age rationing of new technology is involved. Our Scheme A will do as an example. An elderly person might complain about A by saying it is not really fair to his cohort: his cohort never had the benefit of increasing its chances of reaching a normal life-span because the technology (say dialysis) now being denied it was also not available in its youth. Two points might be made in response to this complaint. First, it might be argued that each birth cohort is treated equally in the following way. At some point in its life, each cohort will be denied the best available life-extending technologies, but at all other points in its life it will have a better chance of receiving them. To be sure, the particular technology (dialysis) it is denied may not be the very one it had a better chance of receiving, but there is a fairness in the exchange. Still, if technology improves very rapidly, the bargain is not quite as favorable from a prudential perspective as it might have seemed when we ignored the fact or rate of technological advance. A second point is more general: some such changes, e.g., in technology, are at least as difficult to project as the other factors that lead to error (replacement ratio differences) in saving schemes. Indeed, it seems in general the case that we might be even more prone to error in the health care setting than in the income support setting. In such a context, given the overriding importance of stability in such schemes, considerable tolerance for error must obtain.

Death, Dying and the Ending of Life, II

460

536

NORMAN DANIELS

7. SOME QUALIFICATIONS I fear it is easy to misconstrue, at least to misapply, my argument. It does not in general sanction rationing by age. Such justification is possible only under very special circumstances. For example, it should not be taken as a hasty endorsement of age rationing as a convenient "cost-constraining" device in the context of current debates. Part of the temptation to misconstrue the force of my argument comes from failing to recognize the difference between "ideal" and "nonideal" moral con~ texts. It is important to see that many of the problems facing large numbers of the elderly in our society are consequences not of "age-bias" but of other inequalitiesand, I would argue, injustices. The worst-off among the elderly are usually the same people who were worst-off in earlier stages of their lives. Problems with social and individual savings schemes may exacerbate their plight, but their ultimate situation is largely determined by their earliest position in society. This is not, of course, to say they are getting what they deserve. It is to raise the more basic question about the justice of the underlying distributive institutions. In the context of such injustice, it is just "blaming the victims" to talk about the "inadequacy" or even the "imprudence" of "their" savings. No one could reasonably be expected to save prudently for old age from such inadequate income and wealth shares in their working years. Of course, we can rectify or adjust for underlying inequities by income and health care support in the later years, which we to some extent do. But this adjustment should not primarily be seen as an issue of "justice between age groups"; it is really a more basic issue of distributive justice which forces the correciton. Nevertheless, I have been claiming that there is a distinct problem of distributive justice here. The residual problem is to select principles of distributive justice that will govern the basic institutions responsible for distributing social goods through the various stages of life. My proposal has been that a form of prudential reasoning should guide the design of such institutions. From the perspective of such institutions, goods are distributed through the stages of a life, not between different persons in distinct age groups. With health care institutions, justice requires we allocate health care in a manner that assures individuals a fair chance at enjoying the normal opportunity range, and prudence suggests that it is equally important to protect individual opportunity range for each stage of life. Under certain assumptions, prudence would urge some forms of rationing by age. Similarly, prudence might suggest that some forms of nonmedical services that meet the health care needs of the elderly are more important than certain medical services, because they better protect the normal opportunity range for that stage of life. But suggesting that prudence is our guide to the design of "savings institutions" does not, of course, mean that these matters of design are not matters of justice. Here prudence guides justice; it does not prevent us from talking about it. In proposing that we use prudential considerations to determine the justice of distributions between age groups, I take for granted a background involving other just institutions. It is in this sense that I have been concerned with a problem in

Death, Dying and the Ending of Life, II

AM I MY PARENTS' KEEPER?

461

537

what is known as "ideal" moral theory: we are looking at principles and institutions operating in a society that is in general just. Remarks about the permissibility of rationing by age must thus be taken in this context. If the basic institutions of a given society do not comply with acceptable principles of distributive justice, rationing by age could make things even worse. Indeed, prudential considerations that might endorse rationing by age depend on what sort of resource scarcity exists. Moreover, it is important what che source of the scarcity is. If the scarcity is the result of unjust arrangements operating elsewhere in the system, the argument from prudence may well be undermined. My worries here are part of a more general problem. In another context, I have argued that the moral philosopher considering issues of public policy must take into account both the "framework" for the problem and the "context of com· pliance. " 30 A framework is determined by how much of the basic fundamental political, social, and economic institutions we take to be fixed and how much we allow to be revised in the social system under question. The more major changes of fundamental institutions we allow, the more basic the framework. The context of compliance is the degree to which the problem arises in a society that complies with acceptable principles of justice for its basic institutions. "Ideal" theory is full-com· pliance theory. Philosophers (myself included) have generally concentrated on basic frameworks and have for the most part concerned themselves with ideal theory. Yet pub· lie policy makers operate in less basic frameworks and, I believe, in contexts far from ideal. Too little philosophical attention has been paid to the problem of making points developed for basic frameworks and ideal theory relevant to the public policy maker. Philosophical vision is fine, but philosophers must focus their vision on the problem raised by policy makers. Unfortunately, myopic bureaucrats, whose vision has adapted to ensure their survival in nonbasic frameworks, may not see any relevance in philosophical vision. What is at once a philosophical problem, of con· necting ideal nonideal contexts, is also a political problem: philosophers must not let themselves become myopic as the price of seeing things in the relevant light. Notes 1. This paper is based on research funded by the National Center for Health Services Research, Grant Number HS 03097, OASH. I am indebted to Mary Anne Bailey, Hugo Bedau, Christine Bishop, Dan Brock, Allen Buchanan, Leslie Burckholder, Joshua Cohen. Daniel Den· nett, David Gauthier, Allan Gibbard, and Daniel Wilder for helpful discussion. This version was prepared for the President's Commission for the Study of Ethical Problems in Medicine and Biomedical and Behavioral Research. It forms part of a book in progress, Justice and Health Crwe Delwey, to be published by Cambridge University Press. 2. The approach is developed in my "Health Care Needs and Distributive Justice," Philosophy and Public Affairs 10, no. 2 (Spring 1981):146-79. 3. We are concerned with more than just opportunity to enter jobs and careers, but this weakens the argument Rawls offers for the fair equality of opportunity principle. See John Rawls, A Theory of justice {Cambridge, Mass., 1971), sec. 14. I note some of the problems in "Health Care Needs," pp. 169ff. For other criticisms, see Allen Buchanan, ''The Right to a

Death, Dying and the Ending of Life, II

462

538

NORMAN DANIELS

'Decent Minimum' of Health Care," The President's Commission for the Study of Ethical Problems in Medicine and Biomedical and Behavioral Research (Washington, D.C., 1981). 4. R. M. Gibson and C. R. Fisher, "Age Differences in Health Care Spending, Fiscal Year 1977," Social Security Bulletin 42, no. 1 Uanua.ry 1979):3-16. 5. c.]. Zook, and F. D. Moore, "High-cost Users of Medical Care" New England journal of Medicine 302, no. 18 (May 1, 1980):996-1002. The aged and chronically ill have become the major users of intensive care: 44 percent of all ICU patients at Massachusetts General Hospital were over 65 and 25 percent were over 75, yet only 17 percent of people in the hospital catchment area are over 65. Similarly, patients over 70 years old, a group excluded from some coronary care units in the past, were 32 percent of all admissions in another study. See E. W. Campion, A. G. Mulley, et a!., "Medical Intensive Care for the Elderly: A Study of Current Uti&a· tion, Costs, and Outcomes," journal of the American Medical Assockltion 246, no. 18 (Nov. 6, 1981):2052-56; G. E. Thibault, A. G. Mulley, et al., "Medical Intensive Care: Indications, Interventions, and Outcomes," New England journal ofMedicine 302, no.17 (April24,1980):938-42. 6.]. N. Morgan, "The Ethical Basis of the Economic Claims of the Elderly," in Soci4l Policy, Sock!/ Ethics, and the Aging Society, ed. B. L. Neugarten, and R. J. Hav:ighurst (Chicago, 1976), pp. 67-69. 7. A. R. Jonsen, "Principles for an Ethics of Health Services," in Sockll Policy, Social Ethics, and the Aging Society, pp. 97-105. 8. E. B. Palmore and K. Manton, "Ageism Compared to Racism and Sexism," journal of Gerontology 28, no. 3 (1973):363-69. 9. SeeR. N. Butler, "Age-Ism: Another Form of Bigotry," Gerontologist 9 (1969):243-46; and R. N. Butler and M. I. Lewis, Aging and Mental Health: Positi!Je Psychosocial Approaches (St. Louis, 1977), especially pp. 141-43. 10. A qualification may be needed here: age criteria may not function exa.ctly like race or sex criteria in such contexts. It might not seem imprudent for age-related practices, like seniority, to be given weight. Indeed, from the perspective of a prudent person allocating job opportunity over a lifetime, it might seem worth trading greater training options in his youth for greater job security in his later years. No such reference to race or sex criteria is plausible, largely because the prudential saver model allows greater freedom here than where distn'butions more clearly cross the boundaries between persons, See section 3. 11. See my "IQ, Heritability, and Human Nature," in Proceedings of the Philosophy of Science Association, ed. R. S. Cohen (Dordrecht, 1976), pp. 143-80. It is important to remember that the elderly are a nonhomogeneous group; cf. C. C. Pegels, Health Care and the Elde!'ly (Rockville, Md., 1980). 12. See A. Caplan, "What Are the Morals of Our Treatment of Renal Failure?" in Sock!/ Responsibility: journalism, La'W, and Medicine, ed. L. ]. Hodges (Lexington, Va., 1980), pp. 3250. 13. The allocation issues here are numerous. For example, the young might prefer investing in preventive efforts, like health hazard regulation, whereas the elderly may not benefit from such long-term investments. 14. Partly because of retirement policy and pa.rtly because of inequalities in income distribution throughout earning years, about 25 percent of the elderly are below or near the poverty line. The average older couple receives less than half the income of younger couples. See Pegels, Health Care and the Elderly, p. 4. 15. The issue is quite complicated and arises often in making moral judgments about race, caste, or sex practices in other cultures. Often we try to avoid the issue by discounting consent, sa.y by labeling it "false consciousness." But what if the consent seems genuine? Do such problems mske the appeal to Kantian views of the person and hypothetical contracts all the more problematic, or all the more attractive? 16. It is worth noting a point of contrast between the aging problem posed here and the just savings problem as it is discussed by John Rawls {see A Theory of justice, sees. 44, 45). Rawls

Death, Dying and the Ending of Life, II

AM I MY PARENTS' KEEPER?

463

539

is primarily concerned with preserving adequate capital and nonrenewable resources so that sue· cessive generations are in a position to maintain institutions of justice. In contrast, we are concerned with institutions that transfer income or aid-in-kind between age groups so that their consumption will yield just income-support and health-care distributions through our lifetimes. Rawls's approach to the just savings problem involves his device of a thick veil of ignorance: we do not know which generation we will be in when we are choosing our principles of justice. Moreover, he imposes a motivational constraint on parties making the hypothetical contract: they are concerned about the well-being of a generation or two in each direction (from their grandparents to their grandchildren). Contractors operating under such constraints would pru• dently grant each generation an equal claim on resources necessary to maintaining institutions of justice. In this way the just savings rate acts as a constraint on other principles of justice, such as the Difference Principle: no society can maximin unless it has set aside the resources required by the just savings principle, The question (c) posed in the last section, about equity in replacement ratios between birth cohorts participating in the scheme, may need for its resolution arguments based on such a hypothetical contractual apparatus. But I think that answers to .. qu~io~ (a)_ at1d (b),_!J._bout t]l~ .rate: of sa.:vings and the content of the benefits "saved," may be approached more directly, with a less "veiled" form of prudence. Still, the answer to (c) \\