New Directions in the Ethics of Assisted Suicide and Euthanasia [2 ed.] 3031253140, 9783031253140

Citation preview

The International Library of Bioethics 103

Michael Cholbi Jukka Varelius   Editors

New Directions in the Ethics of Assisted Suicide and Euthanasia Second Edition

The International Library of Bioethics Founding Editors David N. Weisstub Thomasine Kimbrough Kushner

Volume 103

Series Editor Dennis R. Cooley, North Dakota State University, History, Philosophy, & Religious Studies, Fargo, ND, USA Editorial Board Terry Carney, Faculty of Law Building, University of Sydney, Sydney, Australia Marcus Düwell, Philosophy Faculty of Humanities, Universiteit Utrecht, Utrecht, The Netherlands Søren Holm, Centre for Social Ethics and Policy, The University of Manchester, Manchester, UK Gerrit Kimsma, Radboud UMC, Nijmegen, Gelderland, The Netherlands Daniel P. Sulmasy, Edmund D. Pellegrino Center for Clinical Bioethics, Washington, DC, USA David Augustin Hodge, National Center for Bioethics, Tuskegee University, Tuskegee, AL, USA Nora L. Jones, Center for Urban Bioethics, Temple University, Philadelphia, USA

The International Library of Bioethics—formerly known as the International Library of Ethics, Law and the New Medicine comprises volumes with an international and interdisciplinary focus on foundational and applied issues in bioethics. With this renewal of a successful series we aim to meet the challenge of our time: how to direct biotechnology to human and other living things’ ends, how to deal with changed values in the areas of religion, society, and culture, and how to formulate a new way of thinking, a new bioethics. The International Library of Bioethics focuses on the role of bioethics against the background of increasing globalization and interdependency of the world’s cultures and governments, with mutual influencing occurring throughout the world in all fields. The series will continue to focus on perennial issues of aging, mental health, preventive medicine, medical research issues, end of life, biolaw, and other areas of bioethics, whilst expanding into other current and future topics. We welcome book proposals representing the broad interest of this series’ interdisciplinary and international focus. We especially encourage proposals addressing aspects of changes in biological and medical research and clinical health care, health policy, medical and biotechnology, and other applied ethical areas involving living things, with an emphasis on those interventions and alterations that force us to re-examine foundational issues.

Michael Cholbi · Jukka Varelius Editors

New Directions in the Ethics of Assisted Suicide and Euthanasia Second Edition

Editors Michael Cholbi School of Philosophy, Psychology, and Language Sciences University of Edinburgh Edinburgh, UK

Jukka Varelius Department of Philosophy, Contemporary History, and Political Science University of Turku Turku, Finland

ISSN 2662-9186 ISSN 2662-9194 (electronic) The International Library of Bioethics ISBN 978-3-031-25314-0 ISBN 978-3-031-25315-7 (eBook) https://doi.org/10.1007/978-3-031-25315-7 1st edition: © Springer International Publishing Switzerland 2015 2nd edition: © The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 This work is subject to copyright. All rights are solely and exclusively licensed by the Publisher, whether the whole or part of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission or information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed. The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use. The publisher, the authors, and the editors are safe to assume that the advice and information in this book are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or the editors give a warranty, expressed or implied, with respect to the material contained herein or for any errors or omissions that may have been made. The publisher remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. This Springer imprint is published by the registered company Springer Nature Switzerland AG The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland

Introduction to the Revised Edition

Since the publication of the first edition of New Directions in the Ethics of Assisted Suicide and Euthanasia in 2015, assisted suicide and euthanasia have remained central topics in public and political debates worldwide. Not all jurisdictions that have contemplated allowing physicians to play an active role in ending their patients’ lives have legalized such procedures. Yet since 2015, active forms of euthanasia or assisted suicide have become legal in Austria, Canada, New Zealand, Spain, the states of California, Colorado, Hawaii, Maine, New Jersey, New Mexico, and Washington DC in the USA, and the Australian states of Tasmania, Victoria, and Western Australia.1 Legalization of the procedures is currently under consideration in Chile, England, Italy, Ireland, Jersey, Portugal, Scotland, and the states of Arizona, Connecticut, Delaware, Indiana, Kansas, Kentucky, Massachusetts, Minnesota, New York, North Carolina, Rhode Island, Utah, Virginia, and Wisconsin in the USA. Simultaneously, and relatedly, scholarly discussion on assisted suicide and euthanasia has continued vigorously. Maintaining that the procedures are contentious remains as warranted as it was at the time when the first edition of New Directions in the Ethics of Assisted Suicide and Euthanasia appeared. While the topics examined in the first edition of the volume are still relevant to the current end-of-life discussions, both theoretical and practical, some pertinent themes did not receive attention in that edition and some novel ones have emerged since its publication. Among the most central of these are questions relating to medical futility, disability, gender, markets, and new technology in the end-of-life context. Moreover, concrete cases from jurisdictions that have allowed active forms of euthanasia and/or assisted suicide can provide information valuable to assessing and developing existing and proposed end-of-life practices. This motivates the compilation of this 1

The law in Tasmania is due to commence by 23 October 2022. In 2015, active forms of euthanasia or assisted suicide were already legal in Belgium, Colombia, Luxembourg, the Netherlands, Switzerland, and the states of Oregon, Washington, and Vermont in the USA. In Germany, Italy, and the state of Montana in the USA, court rulings create a defense for physicians’ participating in ending their patients’ lives.

v

vi

Introduction to the Revised Edition

revised edition of New Directions in the Ethics of Assisted Suicide and Euthanasia. Besides including ten of the 13 chapters of the first edition, this revised edition contains eight new chapters tackling emerging themes. In the first of the new chapters, Marc De Hert, Sigrid Sterckx, and Kristof Van Assche (“Due Care in the Context of Euthanasia Requests by Persons with Psychiatric Illness: Lessons from a Recent Criminal Trial in Belgium”) highlight difficulties arising in connection with euthanasia and assisted suicide for the mentally ill. In view of a recent highly visible court case in Belgium, the authors discuss questions related to legal competence, borderline personality disorders and Asperger Syndrome as grounds of ending life, and the role of physicians participating in euthanasia. Based on the discussion, De Hert, Sterckx, and Van Assche propose that the Belgian Euthanasia Law should be complemented with additional legal safeguards against physicians’ performing euthanasia when patient suffering could be alleviated without ending life. In Chap. 12, Nancy S. Jecker (“Medical Futility and Assisted Dying”) draws attention to the close connection between physicians’ withholding and withdrawing futile life-sustaining treatment and actively helping patients to die. Based on placing the concept of medical futility in the context of bioethical principles of autonomy, beneficence, non-maleficence, and justice, Jecker presents a prima facie defense of active voluntary euthanasia and physician-assisted suicide for terminally ill patients as a last resort. According to Jecker, besides having a duty to withhold or withdraw futile life-sustaining treatment, physicians also have a duty to help ensure that dying patients have a peaceful and dignified death. In Chap. 13, Christopher A. Riddle (“Medical Aid in Dying: The Case of Disability”) focuses on the perceived harms of permitting medical aid in dying for people with disabilities. He proposes that the harms pertinent in this context can be usefully divided into two kinds, person affecting and personhood affecting. Riddle considers whether there are adequate principled or empirical grounds to think that allowing medical aid in dying for people with disabilities is unacceptably harmful in either of these senses. He concludes that, on the contrary, denying disabled people access to aid in dying constitutes a serious personhood affecting harm. In Chap. 14, Jennifer A. Parks (“Feminist Approaches to Medical Aid in Dying: Identifying a Path Forward”) examines feminist perspectives on medical aid in dying. She contrasts justice-based feminist arguments for medical aid in dying with feminist care-based arguments against the practices and maintains that developing a robust feminist account of medical aid in dying presupposes adopting both justice and care approaches. Hence, while defending a woman’s right to medical aid in dying, Parks argues that ensuring just practices of medical aid in dying for women presupposes duly addressing pertinent concerns of care. In Chap. 15, Michael Cholbi (“Envisioning Markets in Assisted Dying”) questions the idea that assisted dying should be the exclusive province of physicians and unmotivated by pecuniary considerations. He suggests replacing the medical monopoly with a market in assisted dying, wherein those jurisdictions in which assisted dying is lawful allow its provision by physicians but also by clinically qualified non-physician providers who may charge fees for their services. Cholbi defends the new model against central criticisms and

Introduction to the Revised Edition

vii

maintains that it answers to legitimate interests individuals have in connection with how they die and would make assisted dying available to a wider population of those eligible for it. Chapters 16 and 17 relate to the role of new and expected technology in the end-of-life context. Ryan Tonkens (“Robots, AI, and Assisted dying: Ethical and Philosophical Considerations”) examines the possible applications of research in robotics and artificial intelligence to providing assistance in dying. He proposes that robotic systems and artificial intelligence could, among other things, provide patients contemplating ending their lives information about medical assistance in dying, assess their eligibility for receiving such assistance, and even administer pertinent medications. Tonkens also maintains that, in addition to playing such concrete roles, applications of advanced robotics systems and artificial intelligence could help to advance philosophical and ethical debates on assisted dying. Adam Buben (“Dying to Live: Transhumanism, Cryonics, and Euthanasia”) considers assisted dying from a transhumanist perspective. The technology for transcending human limitations transhumanists typically long for is not yet available. Accordingly, several transhumanists could be expected to be interested in cryonically preserving themselves until modifying themselves in their desired ways becomes possible. Buben examines whether there are cases in which such preservation, or assisted dying for the purpose of improved preservation quality, would be morally warranted. He proposes that, under particular conditions, transhumanists should be allowed to choose such death in order to improve their chances of leading the kind of existence they desire in the future. Finally, in Chap. 18, Michael Gill’s “When is Self-perceived Burden an Acceptable Reason to Hasten Death?” challenges the taboo that patients’ perception that their illnesses or care impose burdens on others cannot be a compelling reason for them to seek assisted dying. After offering a prime facie case for the acceptability of burden-based decisions, Gill addresses the objections that burden-based decisions are vitiated by mistaken beliefs (particularly regarding whether their care is in fact burdensome to their loved ones) and that burden-based decisions are vitiated by mistaken value judgments, including judgments about the significance of autonomy and interdependence, judgments potentially distorted by depression, and judgments that would adversely affect vulnerable groups such as the disabled. As we proposed in the introduction to the first edition of New Directions in the Ethics of Assisted Suicide and Euthanasia, we believe that taking a broader range of perspectives, evidence, and arguments into account in the end-of-life debates may enable us to see assisted suicide and euthanasia in a new light. Accordingly, our hope again is that the publication of this revised volume helps to take debates on assisted suicide and euthanasia a step forward. Yet, work is still to be done on end-of-life questions relating to race and climate change, for instance. Whether proponents and opponents of assisted suicide and euthanasia will ultimately reach a consensus or a general agreement on the moral and legal acceptability of the practices remains to be seen. Before proceeding to the chapters of this volume, a terminological point is in order. In discussing the practices we refer to as assisted suicide and euthanasia, different disciplines use different terms, and historically, the choice of terms has not been

viii

Introduction to the Revised Edition

theoretically neutral. Proponents of the practices tend to use terms such as ‘death with dignity’ or ‘assistance in dying,’ whereas opponents of the practices often prefer terms such as ‘doctor-assisted suicide.’ These terms have different connotations, and in some ways, the disputants are inviting audiences to think of the practices in a certain way by their chosen terminology. In our view then, while the disputants are focusing on the same practices, there is no universally shared terminology here. Accordingly, while continuing to speak about assisted suicide and euthanasia ourselves, we have allowed the authors of the following chapters to use their preferred nomenclature. Michael Cholbi [email protected] Jukka Varelius [email protected]

Introduction

Although euthanasia or assisted suicide has been practiced for at least two millennia, the current scholarly debate about assisted dying began to take shape in the 1950s and 1960s. Thanks to a confluence of developments during that period, death became increasingly ‘medicalized,’ i.e., the time or manner of a person’s death fell more and more under human technological control. One such development was changes in patterns of mortality wherein chronic degenerative illnesses such as cancer or organ failures supplanted accidents or infectious disease as the most common causes of death. Deaths of the former sort are typically preceded by lengthy courses of treatment and tend to culminate in dying that is medically supervised. Furthermore, intensive care and life support technologies enabled medical professionals to prolong, or even halt, the process of biological death. These developments tend to put death and dying within the scope of human agency: More and more, death occurs due to choices made near the end of life. The contemporary debate about assisted dying largely revolves around who may exercise such agency and under what conditions. Physician-assisted dying thus became, along with abortion and research abuse, one of the central topics driving the emergence of contemporary bioethics. Medical professionals, philosophers, ethicists, theologians, and legal scholars have generated a scholarly literature on assisted dying consisting of hundreds of books and thousands of articles, written in multiple languages. It is nevertheless possible to identify four central threads in the debates laid out in this vast literature. The first of these centers on the moral norms that regulate the behavior of medical professionals and the compatibility of those norms with physician-assisted death. Some organizations of medical professionals, such as the American Medical Association (AMA), maintain that medicine has been historically defined by a categorical moral norm against doctors killing their patients (or enabling their patients to kill themselves). In a similar vein, some authors insist that medical morality is absurd unless doctors have an indefeasible duty to abstain from killing. This view entails that whatever the other merits of assisted dying, permitting physicians either to assist in their patients’ suicides or to kill their patients is too great a deviation from this norm. Proponents of physician-assisted death maintain that the interpretation of medical morality that unconditionally prohibits doctors from ending their patients’ lives is a ix

x

Introduction

misinterpretation. Proper moral norms for medical professionals entail, advocates of physician-assisted death argue, that when a patient’s suffering is unbearable to him and there is no way of adequately alleviating it other than ending his life, a doctor ought to adhere to the patient’s request to be helped to die. There are no esoteric moral norms internal to the profession of medicine that could outweigh this duty (see, e.g., Garcia 2007; Pellegrino 2001; and Seay 2011). A second thread in this debate concerns the nature of a person’s involvement in bringing about another person’s death. As mentioned above, some opponents of assisted dying contend that physicians may not kill their patients. However, the current ethical and legal consensus is that physicians are morally obligated to honor patients’ decisions to forego or cease treatments that may extend their lives (e.g., cancer patients opting not to receive additional chemotherapy or terminally ill individuals who request removal of life support).1 On its face, these claims seem to be in tension with the aforementioned precept that doctors may not kill their patients. How, advocates of a right to physician-assisted dying ask, can doctors have a right knowingly to hasten death without also having a right to assist their patients in dying or to kill patients who request it? Here, opponents of physician-assisted dying invoke several different distinctions. One is the distinction between killing and letting die. Opponents of physician-assisted dying argue that doctors may let a patient die, but may not kill a patient. Physicianassisted dying is thus ruled out, but measures through which a patient dies due to the progress of her illness are not. A second apparently relevant distinction is between acts and omissions: When a doctor assists a patient to die, the death results from the doctor’s positive act, whereas (for instance) when a doctor does not administer potentially life-extending treatment to a patient who competently declines the treatment, the death results from the doctor’s omission, that is, from the doctor from refraining from acting. The former, it is argued, is morally impermissible, the latter morally permissible. A third distinction invoked in this regard, associated with the Doctrine of Double Effect, is between death being intended as an outcome of one’s actions and death’s being merely a foreseen outcome of one’s actions. According to this doctrine, it is permissible to act so that one foresees an otherwise morally impermissible outcome (e.g., the death of an innocent person) so long as (a) that outcome is not intended, and (b) a sufficiently good outcome is intended instead. Applied to practices such as abortion to save a mother’s life, the doctrine appears to permit doctors acting in the awareness that death will result so long as death is not intended. Assisted suicide and euthanasia are similarly precluded because their intent is death, whereas withdrawing treatment, etc., result in deaths that are merely foreseen. All three of these distinctions involve different ways in which an exercise of agency can bring about another person’s death. If plausible, such distinctions could explain why killing a patient (or assisting in her death) is impermissible although acts that knowingly lead to a patient’s death are not necessarily impermissible. Unsurprisingly, advocates of physician-assisted dying question both the coherence and moral 1

The latter, withdrawing life-sustaining measures, has come to be known as ‘passive’ euthanasia.

Introduction

xi

significance of these distinctions. To such advocates, delineating between morally permissible and morally impermissible measures all of which knowingly result in a patients’ death fails either because the distinctions themselves cannot be made out precisely or because the distinctions cannot bear the moral weight their advocates assign to them. While some arguments in defense of these distinctions merely restate them, some novel defenses of the divisions have been presented. Yet advocates of physicianassisted death have usually not found the new defenses more convincing than the older ones. Some of them have even claimed that the distinctions are nothing but ‘moral fictions,’ ad hoc metaphysical devices used to buttress otherwise groundless moral positions. Not unpredictably, opponents of physician-assisted death have denied this. Some of them have argued that the distinctions between acts and omissions and between foreseeing and intending can still be relied on in the context of end-oflife decision-making (cf., e.g., Miller, Truog and Brock 2009, Huddle 2012, and McLachlan 2011). The third thread in debates about assisted dying concerns how respect for patient autonomy and beneficence determine the moral and legal acceptability of physicianassisted death. The four principles approach introduced by Tom Beauchamp and James Childress (1979) has been highly influential in bioethics. Advocates of physician-assisted dying often appeal to two of these principles in justifying their position: the principle of respect for patient autonomy, that patients (at least when competent) have the primary right to determine which treatments or interventions they will be subject to, and the principle of beneficence, that medical professionals should act so as to benefit their patients. According to many advocates of physicianassisted dying, so long as it in a patient’s interests and the patient provides informed consent, patients have a right to physician-assisted dying no less forceful than their right to other medical interventions. Just as with the previous two threads, the third thread of the modern debate remains contentious. For advocates of physician-assisted death, the legalization of physicianassisted dying that has already occurred within some jurisdictions appears as “the triumph of autonomy” in bioethics and public policy. Yet opponents of physicianassisted death insist that the two principles are far too abstract and too few in number to adequately account for all the complexities, nuances, and ambiguities related to concrete end-of-life decision-making. With respect to autonomy, they dispute both whether the conditions for informed, rational consent to one’s own death can be met and whether the value of autonomy should be accorded primacy in medical decisionmaking. With respect to beneficence, opponents of physician-assisted dying question whether, given the value typically attached to human life, ending a person’s life can really be a beneficent act. Furthermore, they maintain that much more research concerning the principles, possible alternative starting points for assessing questions related to physician-assisted death, and the specific features of the kinds of cases in which end-of-life choices would be made is needed for it to be possible to adequately assess the moral and legal acceptability of physician-assisted death. Again, proponents of physician-assisted death have not been convinced by these strategies. Some of them have even proposed that the call for more research is simply a strategy by

xii

Introduction

which opponents of physician-assisted death try to retain the status quo in which the procedures remain illegal in most jurisdictions. In this view, the purpose of opponents of physician-assisted death is to ensure that the moral and legal acceptability of such procedures remains under “interminable scrutiny” (see, e.g., Beauchamp 2006 and Parker 2005). A final thread in these debates is less about physician-assisted dying per se than about the predicted consequences of its legalization or acceptance. Opponents of physician-assisted dying may concede that euthanasia or assisted suicide are morally permissible if undertaken voluntarily, etc., but nevertheless contend that the acceptance of such practices would weaken our general respect for human life and the prohibition on killing. As a consequence, other more morally objectionable practices, such as involuntary euthanasia or the killing of the disabled, would emerge. Such slippery slope worries reflect anxieties about societies’ ability to implement physician-assisted dying without inviting abuse or corrosion of moral standards. While the claims that permitting physician-assisted death would have devastating societal consequences were once difficult to evaluate, the legalization of the procedures within some jurisdictions has allowed for the possibility of conducting empirical studies by which to assess such claims. According to the pertinent empirical research done so far, the number of deaths has not dramatically increased in the jurisdictions that allow (some form of) physician-assisted death, members of vulnerable groups have not been pressured to die against their will, and no signs of general devaluation of human life can be detected within the jurisdictions examined. Yet these kinds of research results have not persuaded those who advocate slippery slope arguments against physician-assisted death. The advocates of the arguments have responded by maintaining that the research results from the so far rather few jurisdictions examined cannot be generalized so that the case for the procedures in some other areas, and still less in general, could be assessed on their basis. It has also been suggested that the examinations have been methodologically inadequate so that the data they provide does not reliably describe the situation even in the jurisdictions they studied. (Cf., e.g., Rietjens et. al. 2009 and Materstvedt 2009) Advocates of physician-assisted death may be inclined to interpret such doubts as nothing more than further strategic moves in the debate. While this overview gives a rather gloomy picture of the modern debate, it would be unfair to say that no progress has occurred. For instance, though the plausibility and usefulness of the results of the empirical studies on the effects of legalization of physician-assisted death remain debated, that such studies have been made and their results are being discussed is arguably a step forward. At least, the discussion can direct future research toward the right path so that adequate data relevant to assessing the slippery slope arguments is ultimately acquired. For another example, there have been some attempts to find novel solutions that would satisfy both proponents and opponents of physician-assisted death. Perhaps, the most noteworthy of them is the procedure called terminal sedation in which the decision to provide palliation to a suffering patient is accompanied by the decision to forgo the provision of nutrition and hydration and, accordingly, the ultimate result is the patient’s death. Yet insofar

Introduction

xiii

as terminal sedation is very similar to active euthanasia, opponents of physicianassisted death are unlikely to accept the procedure. And to the extent that it does not provide a patient with the kind of ending for her life that she would autonomously want, terminal sedation may not satisfy those proponents of physician-assisted death who emphasize the value of patient autonomy (cf., e.g., van Delden 2007). Why have the opposing camps failed to reach a consensus or even a general agreement on the moral and legal acceptability of physician-assisted death? There are three possible diagnoses. In a pessimistic view, the starting points and values of proponents and opponents of physician-assisted death are in so many respects so different from each other that their achieving any agreement on the matter is impossible. On this view, the controversy between advocates and opponents of physician-assisted death is interminable. A moderate view entails that although the two parties in the debate will continue to endorse differing points of departure and values related to end-of-life questions a practical compromise that satisfies both of them can be found. In the most optimistic view, further investigation into the pertinent questions will ultimately lead to a full rational resolution of the debate. We take no stance on which of these three diagnoses is correct. However, regardless of which of these diagnoses prove correct, it seems clear that the modern debate has stagnated, and as a result, the ethical and legal debates surrounding assisted suicide and euthanasia would benefit if they were to expand beyond their narrow focus on the four aforementioned threads. Taking a broader range of perspectives, evidence, and arguments into account may help us to see assisted dying in a new light and, more optimistically, help to determine which of the above three positions—the pessimistic, the moderate, or the optimistic—is the correct one. This volume was developed with precisely that purpose in mind, to showcase scholarship that investigates the ethics or legality of physician-assisted dying from fresh points of view. Our hope is that engaging with this scholarship will challenge those who believe that assisted dying remains a high stakes issue to think more imaginatively about it and refine their positions accordingly. The first three chapters raise challenges to what has become perhaps the most common defense of physician-assisted dying, that individuals are autonomous and so have the right to determine the manner or circumstances of their deaths, including a right to enlist the help of others to bring about their deaths. Emma C. Bullock (“Assisted Dying and the Proper Role of Patient Autonomy”) considers the role of patient autonomy in justifying physician-assisted dying. Bullock argues that the fact that a patient autonomously chooses to die does not entail that dying is in her best interests. On the contrary, evidence from psychology and behavioral economics indicates that we often fail to make rational self-determining choices. Bullock proposes that the place of autonomy is therefore not to serve as an index of a patient’s interest in dying. Rather, autonomy is better seen as a “side constraint” that permits physicianassisted dying when it is also objectively in the patient’s best interests. Of course, advocates of “strong” paternalism have long believed that we ought to prevent suicide and assisted dying when it is not in a person’s best interests. Thomas Schramme (“Preventing Assistance to Die: Assessing Indirect Paternalism Regarding Voluntary Active Euthanasia and Assisted Suicide”) addresses a variation of this paternalist

xiv

Introduction

argument: that there may be “indirect” paternalistic grounds for interfering with assistance in some suicides even if there are not legitimate grounds for interfering with the suicidal acts themselves. Schramme notes that the introduction of a thirdparty assistant into the equation complicates the underlying ethical issues, especially given the possibility that suicidal individuals may not have a justified entitlement to assistance in dying. Ultimately, Schramme concludes that the indirect paternalist argument does not succeed in showing that suicidal individuals cannot have a right to the assistance of others. For such individuals may genuinely need such assistance in order to exercise their right to die and those who provide it are enabling suicidal persons to pursue the morally legitimate end of minimizing their own suffering. Julian Savulescu (“Autonomy, Interests, Justice, and Active Euthanasia”) shares Bullock’s skepticism regarding the power of autonomy to justify physician-assisted dying in a straightforward way. In his estimation, appeals to autonomy do not show that those who wish to be helped to die have a claim to active euthanasia on the part of medical professionals. Simply having a preference for ending one’s life prematurely cannot, Savulescu argues, ground a claim to active euthanasia. Savulescu also rejects the other main arguments presented for physician-assisted dying so far. However, he contends that patients who voluntarily undergo palliated starvation, believing that this in their best interests, have a right to do so based on respect for autonomy. But in such cases, a person may well die more quickly, and thereby make use of fewer medical resources, if she were to undergo active euthanasia. The proper ethical basis for active euthanasia is therefore respect for autonomy together with distributive justice, according to Savulescu. Chapters 4 and 5 attempt to extend the existing debate into a highly controversial area, physician-assisted dying for those suffering from mental illness, decline, or distress. In “Mental Illness, Lack of Autonomy, and Physician-Assisted Death,” Jukka Varelius argues that physician-assisted suicide could be morally permissible for psychiatric patients facing unbearable and incurable suffering, even if those patients lack the autonomy to opt for assisted dying. Varelius proposes that lacking autonomy need not be a decisive reason to deny assisted suicide to such individuals, for here the value of patient autonomy is primarily instrumental, that is, that autonomy is not to be respected for its own sake but because autonomous choices are likely to reflect individuals’ conceptions of how their lives should go. He also suggests that the suffering of non-autonomous patients can be morally more significant than that of autonomous patients. Accordingly, non-autonomous patients with unbearable and incurable suffering have compelling reasons to seek to end their lives that an autonomous person would endorse. Dementia is another condition where questions about the justifiability of physician-assisted dying have been raised. Jocelyn Downie and Georgia Lloyd-Smith (“Assisted Dying for Individuals with Dementia: Challenges for Translating Ethical Positions into Law”) investigate whether those with dementia would be able to access assisted dying within the various jurisdictions that presently permit it. They find that despite many individuals reporting that they would want access to assisted dying were they to develop dementia, the very features of the laws that license physician-assisted dying for other conditions often preclude its being available for dementia. For example, dementia patients are not terminally ill

Introduction

xv

until late in the progression of their disease, have suffering that is primarily mental or existential rather than straightforwardly physical, often lack competence, and are unable to communicate their wishes. Downie and Lloyd-Smith conclude with a critical examination of a recent legislative proposal in Canada that may provide wider access to physician-assisted dying for those with dementia. The next three chapters raise novel questions about the ethical responsibilities of clinicians with respect to physician-assisted dying. David M. Adams (“Clinical Ethics Consultation and Physician-Assisted Suicide”) starts from the fact that nowadays, many key care decisions made in medical settings involve professional consulting ethicists. Consulting ethicists typically assist family members, patients, and medical personnel in deliberating about morally complex care decisions. Adams proposes that currently accepted conceptions of the aims and principles meant to guide clinical ethics consultation are deficient when it comes to consultation regarding physicianassisted suicide. According to these conceptions, clinical ethicists are to provide counsel on the basis of current law, institutional policies, professional standards, scholarly consensus, and prevailing practices. In the case of assisted suicide, current law and institutional policies are in flux, professional standards do not all align, and no scholarly consensus exists regarding the ethics of physician-assisted dying. In light of the unsettled legal and moral status of assisted suicide, Adams concludes that clinical ethicists must engage the relevant parties in a process of moral inquiry and deliberation aimed at reaching a shared agreement among them as to whether assisted suicide should be permitted in a given case. As noted above, much of the literature on physician-assisted dying is concerned with whether it is compatible with the established norms of the medical profession. Richard Huxtable and Jonathan Ives (“Licence to Kill: A New Model for Excusing Medically Assisted Dying?”) suggest a new way of thinking about this matter. Acknowledging that there are reasonable grounds both for permitting and prohibiting physician-assisted dying, they propose a compromise or “middle way” that is novel in two respects. First, they suggest that only medical specialists trained in assisted dying (“thanatologists”) be allowed to assist patients to die. Secondly, rather than conferring prospective legal immunity on thanatologists when they participate in assisted dying, Huxtable and Ives suggest that thanatologists may be retrospectively excused for assisting in dying if certain conditions are met. In their estimation, this compromise may satisfy both advocates and opponents of physician-assisted dying insofar as it retains the presumption that physician-assisted dying is wrong but permits thanatologists to rebut this presumption in particular cases. Chapter 8 relates physician-assisted dying to other medical technologies or techniques. David Shaw (“Saving Lives with Assisted Suicide and Euthanasia: Organ Donation After Assisted Dying”) explores the benefits of allowing those who engage in physician-assisted dying to donate their organs and the force that such a possibility has in augmenting the case for legalizing physician-assisted dying. Doing so, Shaw argues, is harmless and would increase the number of organs available for transplant

xvi

Introduction

while ensuring that healthy tissues do not go to waste. Donating one’s organs after assisted dying also enables the dying to bear the burdens of death more easily. The next two chapters bring two recent developments in philosophical methodology to bear on assisted dying. Adam Feltz (“Everyday Attitudes About Euthanasia and the Slippery Slope Argument”) applies the methods of experimental philosophy to ascertain some common attitudes toward the ethics of euthanasia. In a pair of studies, Feltz found that describing euthanasia differently (as ‘euthanasia,’ ‘aid in dying,’ etc.) has rather modest effects on experimental subjects’ judgments regarding whether euthanasia is ethically permissible. Feltz also found that subjects’ judgments about its ethical permissibility were significantly influenced by whether the treatment was perceived as voluntary, an influence greater than general demographic differences and differences in more general moral judgments. In other words, test subjects generally gave little credence to the distinction between active and passive euthanasia, but find euthanasia increasingly problematic the less voluntary it is. Feltz observes that subjects’ ability to distinguish among degrees of voluntariness with respect to euthanasia suggests that a premise common in ‘slippery slope’ arguments against voluntary euthanasia—that individuals struggle to differentiate among voluntary, non-voluntary, and involuntary euthanasia—appears unfounded. In “‘You Got Me Into This…’: Procreative Responsibility and its Implications for Suicide and Euthanasia,” Rivka Weinberg mines the emerging field of procreative ethics for insights regarding the ethics of assisted dying. More specifically, Weinberg asks, on the assumption that assisted dying is at least sometimes morally permissible, who has the responsibility to assist a suicidal individual to die? Weinberg advances the thesis that parents may have a special responsibility to assist in light of (a) the risks that parents impose on the children they create (risks of suffering, unhappiness, etc.), risks the children do not agree to be exposed to, and (b) parental responsibilities to meet fundamental needs that children cannot meet at their own. While Weinberg acknowledges that many parents would find assisting a child to die emotionally taxing, she concludes that doing so reasonably falls within the sacrifices that parents are often obligated to make on their children’s behalf. The chapters of this volume cover topics that so far have not received sufficient attention in the debate on assisted dying. But this still leaves many angles on the ethics of assisted dying largely unaddressed, including its relationship to general societal, environmental, and technological questions and developments, such as the aging and growth of the human population, climate change and the sufficiency of food, air, and water, and innovations as regards human enhancement and environmental technology. The most discouraging prospect is that, in the future, an increasingly older, bigger, and sicker population of humans has to make do with a continuously decreasing stock of natural and other resources. Perhaps, human enhancement or other technologies will ameliorate these problems. But even so, it is likely that the conceptions we now have about the value of human life and death will undergo some changes. Accordingly, these kinds of questions and developments can also be

Introduction

xvii

expected to affect our attitudes toward the moral and legal acceptability of physicianassisted dying. This volume demonstrates that plenty of philosophical and bioethical questions remain about assisted dying. Yet it is an issue that transcends the bounds of ‘medical ethics’ as it has traditionally been conceived. Michael Cholbi Jukka Varelius

References Beauchamp, Tom L., and James F. Childress. 1979. Principles of Biomedical Ethics (first edition). New York: Oxcord University Press. Beauchamp, Tom L. 2006. The right to die as the triumph of autonomy. Journal of Medicine and Philosophy 3: 643–654. Garcia, Jorge L. A. 2007. Health versus harm: Euthanasia and physicians’ duties. Journal of Medicine and Philosophy 32: 7–24. Huddle, Thomas S. 2012. Moral fiction or moral fact? The distinction between doing and allowing in medical ethics. Bioethics 27: 257–262. Materstvedt, Lars J. 2009. Inappropriate conclusions in research on assisted dying. Journal of Medical Ethics 35: 272. McLachlan, Hugh V. 2011. Moral duties and euthanasia: Why to kill is not necessarily the same as to let die. Journal of Medical Ethics 37: 766–767. Miller, Franklin G., Robert D. Truog, and Dan W. Brock. 2009. Moral fictions and medical ethics. Bioethics 24: 453–460. Parker, Malcolm. 2005. End games: Euthanasia under interminable scrutiny. Bioethics 19:523–536. Pellegrino, Edmund D. 2001. Physician-assisted suicide and euthanasia: Rebuttals of rebuttals—the moral prohibition remains. Journal of Medicine and Philosophy 26: 93–100. Rietjens, Judith A. C., Paul J. van der Maas, Bregje D. Onwuteaka-Philipsen, B.D., Johannes J. M. van Delden, and Agnes van der Heide, A. 2009. Two decades of research on euthanasia from the Netherlands. What have we learnt and what questions remain? Journal of Bioethical Inquiry 6: 271–283. Seay, Gary. 2011. Euthanasia and common sense: A reply to Garcia. Journal of Medicine and Philosophy 36: 321–327. van Delden, Johannes J. M. 2007. Terminal sedation: Source of a restless ethical debate. Journal of Medical Ethics 33: 187–188.

Contents

1

Assisted Dying and the Proper Role of Patient Autonomy . . . . . . . . . Emma C. Bullock

2

Preventing Assistance to Die: Assessing Indirect Paternalism Regarding Voluntary Active Euthanasia and Assisted Suicide . . . . . Thomas Schramme

1

17

3

Autonomy, Interests, Justice and Active Medical Euthanasia . . . . . . Julian Savulescu

4

Mental Illness, Lack of Autonomy, and Physician-Assisted Death . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Jukka Varelius

49

Assisted Dying for Individuals with Dementia: Challenges for Translating Ethical Positions into Law . . . . . . . . . . . . . . . . . . . . . . . Jocelyn Downie and Georgia Lloyd-Smith

67

5

31

6

Clinical Ethics Consultation and Physician Assisted Suicide . . . . . . . David M. Adams

93

7

License to Kill: A New Model for Excusing Medically Assisted Dying? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 117 Richard Huxtable and Jonathan Ives

8

Saving Lives with Assisted Suicide and Euthanasia: Organ Donation After Assisted Dying . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 137 David M. Shaw

9

Everyday Attitudes About Euthanasia and the Slippery Slope Argument . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 145 Adam Feltz

xix

xx

Contents

10 “You Got Me Into This …”: Procreative Responsibility and Its Implications for Suicide and Euthanasia . . . . . . . . . . . . . . . . . . . . . . . . . 167 Rivka Weinberg 11 Due Care in the Context of Euthanasia Requests by Persons with Psychiatric Illness: Lessons from a Recent Criminal Trial in Belgium . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 181 Marc De Hert, Sigrid Sterckx, and Kristof Van Assche 12 Medical Futility and Physician Assisted Death . . . . . . . . . . . . . . . . . . . 203 Nancy S. Jecker 13 Medical Aid in Dying: The Case of Disability . . . . . . . . . . . . . . . . . . . . 225 Christopher A. Riddle 14 Feminist Approaches to Medical Aid in Dying: Identifying a Path Forward . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 243 Jennifer A. Parks 15 Envisioning Markets in Assisted Dying . . . . . . . . . . . . . . . . . . . . . . . . . . 263 Michael Cholbi 16 Robots, AI, and Assisted Dying: Ethical and Philosophical Considerations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 279 Ryan Tonkens 17 Dying to Live: Transhumanism, Cryonics, and Euthanasia . . . . . . . . 299 Adam Buben 18 When Is Self-perceived Burden an Acceptable Reason to Hasten Death? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 315 Michael B. Gill

Editors and Contributors

About the Editors Michael Cholbi is Chair in Philosophy at the University of Edinburgh. He has published widely in ethical theory, practical ethics, and the philosophy of death and dying. His books include Suicide: The Philosophical Dimensions (Broadview, 2011) and Grief: A Philosophical Guide (Princeton University Press, 2021). He is Editor of several scholarly collections and Co-editor of the textbook Exploring the Philosophy of Death and Dying: Classic and Contemporary Perspectives (Routledge, 2020). Jukka Varelius is Research Fellow in the Department of Philosophy, Contemporary History, and Political Science at the University of Turku, Finland. His work focuses on questions of applied ethics.

Contributors David M. Adams Department of Philosophy, California State Polytechnic University, Pomona, USA Kristof Van Assche Research Group Personal Rights and Property Rights, Faculty of Law, University of Antwerp, Antwerp, Belgium Adam Buben Leiden University, Leiden, The Netherlands Emma C. Bullock Department of Philosophy, Central European University, Budapest, Hungary Michael Cholbi School of Philosophy, Psychology and Language Sciences, University of Edinburgh, Edinburgh, UK Jocelyn Downie Faculty of Medicine, Schulich School of Law, Dalhousie University, Halifax, Canada xxi

xxii

Editors and Contributors

Adam Feltz Department of Psychology, University of Oklahoma, Norman, OK, USA Michael B. Gill Department of Philosophy, University of Edinburgh, Edinburgh, UK Marc De Hert University Psychiatric Centre, Katholieke Universiteit Leuven, Kortenberg, Belgium; Department of Neurosciences, Centre for Clinical Psychiatry, Katholieke Universiteit Leuven, Leuven, Belgium; Antwerp Health Law and Ethics Chair, University of Antwerp, Antwerp, Belgium Richard Huxtable Centre for Ethics in Medicine, University of Bristol, Bristol, UK Jonathan Ives Centre for Ethics in Medicine, University of Bristol, Bristol, UK Nancy S. Jecker Department of Bioethics and Humanities, University of Washington School of Medicine, Seattle, WA, USA Georgia Lloyd-Smith West Coast Environmental Law, Vancouver, Canada Jennifer A. Parks Department of Philosophy, Loyola University Chicago, Chicago, IL, USA Christopher A. Riddle Utica University, Utica, NY, USA Julian Savulescu The Oxford Uehiro Centre for Practical Ethics, Faculty of Philosophy, University of Oxford, Oxford, UK Thomas Schramme Department of Philosophy, University of Liverpool, Liverpool, UK David M. Shaw Institute for Biomedical Ethics, University of Basel, Basel, Switzerland Sigrid Sterckx Department of Philosophy and Moral Sciences, Bioethics Institute Ghent, Ghent University, Ghent, Belgium Ryan Tonkens Centre for Health Care Ethics, Lakehead University, Thunder Bay, ON, Canada; NOSM University, Thunder Bay, Canada; Department of Philosophy, Lakehead University, Thunder Bay, Canada Jukka Varelius Department of Philosophy, Contemporary History, and Political Science, University of Turku, Turku, Finland Rivka Weinberg Department of Philosophy, Scripps College, Claremont, CA, USA

Chapter 1

Assisted Dying and the Proper Role of Patient Autonomy Emma C. Bullock

In this paper I use the umbrella term ‘assisted dying’ to cover both voluntary euthanasia (VE) and physician assisted suicide (PAS). This does not mean that the distinction between VE and PAS should be, or are being, elided. Indeed, one might argue that there is an important distinction between a doctor intentionally killing a patient (VE) and intentionally helping a patient to commit suicide (PAS). The decision to use this umbrella term is that the argument presented here applies to both VE and PAS.

Abstract A governing principle in medical ethics is respect for patient autonomy. This principle is commonly drawn upon in order to argue for the permissibility of assisted dying. In this paper I explore the proper role that respect for patient autonomy should play in this context. I argue that the role of autonomy is not to identify a patient’s best interests, but instead to act as a side-constraint on action. The surprising conclusion of the paper is that whether or not it is in the best interests for the patient to die is a morally objective matter. This allows for the possibility that it can be in the best interests of the patient to die even if she autonomously considers it to be in her best interest to continue living. I argue that concerns about ‘mandatory’ euthanasia can be met when patient autonomy is respected as a side-constraint on action. Ultimately, this means that assisted dying is permissible, not because the autonomous patient views her suffering to be unbearable, but because it is in her objective best interests and she permitted it via her consent. Keywords Best interests · Patient autonomy · Unbearable suffering · Objective list · Assisted dying

E. C. Bullock (B) Department of Philosophy, Central European University, Zrinyi u. 14, Budapest 1051, Hungary e-mail: [email protected] © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 M. Cholbi and J. Varelius (eds.), New Directions in the Ethics of Assisted Suicide and Euthanasia, The International Library of Bioethics 103, https://doi.org/10.1007/978-3-031-25315-7_1

1

2

E. C. Bullock

1.1 Introduction A governing principle in medical ethics is respect for patient autonomy. It is therefore unsurprising that debates on voluntary euthanasia and physician assisted suicide have focused on the relevance and scope of this principle. The aim of this paper is to argue for the proper role that patient autonomy ought to play in determining the permissibility of these two forms of assisted dying. In the following I recap arguments against an overemphasis on patient self-determination and argue that self-determination has a subsidiary role in determining whether assisted dying is permissible. I suggest that whether or not assisted dying is in the best interests1 of the patient ought to be determined objectively. The role that patient autonomy plays in determining the permissibility of assisted dying is as a permission or constraint on action. The advantage of this account is that it fits well with emerging models of medical decision-making2 that, with good reason, downplay the importance of patient self-determination but do not rule out the importance of patient autonomy entirely. To date, discussion on the moral permissibility of assisted dying has failed to catch up with this movement.3 This paper aims to bridge this gap by outlining the proper role of patient autonomy with regard to assisted dying. There might be good reasons to deny that assisted dying is morally justified, regardless as to whether it is determined to be in the patient’s best interests and the patient autonomously consents to it.4 I do not address these arguments here. Instead, the focus of this paper is to assess what the argument from respect for patient autonomy amounts to and to outline the role that patient autonomy should play in determining the permissibility of assisted dying.

1.2 Informed Consent and Respect for Patient Autonomy The central case for the moral permissibility of assisted dying is that it constitutes respect for individual autonomy (see Young 2014a, b).5 The argument from autonomy 1

Throughout this paper I use the term ‘best interests’ to narrowly refer to the promotion or protection of the patient’s well-being. A broader usage of the term might include values in addition to wellbeing, such as financial gains and legal obligations. 2 For models and defences of shared-decision making see Birchley (2014), Maclean (2006), Sandman and Munthe (2009, 2010). For defences of various forms of paternalism see Conly (2013), Loewy (2005), Scoccia (2008), Thaler and Sunstein (2008). 3 Little has been written on the diminishing relevance of patient autonomy in recent bioethics literature as related to assisted dying. The exception is Varelius (2006), which explores the proper ends of medicine in view of a growing distinction in the bioethics literature on objective characterisations of the goals of medicine and, on the other hand, respect for patient autonomy (Varelius 2006, 121–2). 4 For a comprehensive overviews of objections to assisted dying see Brock (1993). 5 This is not the only argument in favour of the moral permissibility of forms of assisted dying. A second important argument focuses on patient well-being and the moral demand to reduce patient suffering (Brock 1993, 206). However, this second argument is often directly linked to the argument

1 Assisted Dying and the Proper Role of Patient Autonomy

3

draws upon the claim that a person has a right to shape her own life through her choices and extends this right to include the right of patients to choose the manner of their death (Gray 1999, 21; see also Battin 2005, 20). Patient autonomy is argued to have a ‘pivotal role’ in end-of-life decision making: “…permitting people the opportunity to decide the timing and circumstances of their own demise if that is what they wish” (Biggs 2001, 96). The ‘right to die’ is thus derived from a more general right to direct the course of our own lives (Chetwynd 2004, 175).6 The appeal to patient autonomy in relation to the right to die can be traced to the rise of the doctrine of informed consent (Beauchamp 2006, 644). According to the doctrine, a patient has the right to refuse medical treatment on the condition that she is (A) fully informed of the nature, benefits and risk of each procedure, and (B) her consent or refusal is freely chosen (Declaration of Helsinki 2008). The main justification for informed consent procedures is the importance of respecting patient autonomy (Manson and O’Neill 2007, 17; see also Beauchamp and Childress 1989, 75), protecting: “…the [patient’s] right to decide how one is to live one’s life, in particular how to make critical life-decisions” (Feinberg 1986, 54). One way in which to unpack the role of informed consent as related to assisted dying is to reflect on the kind of autonomy that informed consent ought to protect. Problematically, the concept of ‘autonomy’ is vague and has various meanings in different contexts. As Faden and Beauchamp have noted, the term has been loosely associated with ideas as diverse as: “…privacy, voluntariness, self-mastery, choosing freely, the freedom to choose, choosing one’s own moral position, and accepting responsibility for one’s choices” (Faden and Beauchamp 1986, 7). The concept has also been used to bear connotations with freedom, independence and self-determination (Schermer 2002, 1; see also Feinberg 1986, 28). In the following I explore two key ways in which respect for patient autonomy as required by the doctrine of informed consent can be understood. First, respect for patient autonomy, understood as self-determination, may be required since this is the best means of protecting the patient’s well-being (Sect. 1.3). Secondly, respect for patient autonomy could be viewed as a side-constraint on action (Sect. 1.4), regardless as to whether this leads to a decline in patient well-being. Specifically, I argue, that the doctrine of informed consent protects patient autonomy as a sideconstraint on action rather than as a means for protecting patient well-being. By making this distinction we get a clearer idea of the role that patient autonomy ought to play in determining the permissibility of assisted dying.

from autonomy with the burden of the patient’s suffering being determined by the autonomous patient (Brock 1993, 207). 6 An important caveat to this argument is that respect for patient self-determination does not require health care practitioners to assist in patient dying at the patient’s request (Brock 1993, 207). The argument from autonomy should thus be seen as a constraint on morally permissible assisted dying; it does not amount to the claim that it is morally obligatory to carry out requests for assisted dying.

4

E. C. Bullock

1.3 Autonomy as Self-Determination Although moral philosophy has conceptualised autonomy in a variety of ways, respect for patient autonomy is commonly equated with respect for patient selfdetermination (see Foster 2009, 3; Macioce 2012, 101). On this view, a patient’s autonomy is identified with the patient’s ability to make choices that shape her life in accordance with her own conception of the good life (Brock 1993, 205–6). Various reasons have been given for the importance of respect for patient autonomy understood as patient self-determination. These arguments range from the claim that (a) the exercise of self-determination gives our life meaning (see Nozick 1974, 50; Varelius 2006, 379–380; Glover 1990, 81; Buchanan and Brock 1990, 38–9),7 to the claim that (b) respect for self-determination is the most effective means for promoting and protecting the patient’s well-being. The argument that (b) respect for self-determination equates to maximal wellbeing promotion has been defended on the basis that: “[w]hen patients are competent and have access to information, they are the best judge of what is in their interests and whether the expected benefits of a proposed treatment outweigh the burdens” (English et al. 2004, 108).8 Such arguments can be traced back to Mill, who argues that respect for individual choice is the best means for maximising utility: “…since the individual’s ‘choice of pleasure’ rests with her own judgement” (Mill 2008, 112; see also Erbay et al. 2010, 36; Savulescu 2003, 138–139). The view that individual selfdetermination ought to be respected because it allows for well-being maximisation has been generally supported within medical ethics, it being argued that we ought to respect patient self-determination because patients have a special expertise regarding their well-being (see, e.g., Veatch 2000, 704; Tännsjö 1999, 16). In the context of the debate on assisted dying, the connection between self-determination and well-being has also been emphasised: It might seem that individual well-being conflicts with a person’s self-determination when the person requests euthanasia […] 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…there is no objective standard, but only the competent patient’s judgment of whether continued life is no longer a benefit (Brock 1993, 206–7).

If it is correct that respect for patient self-determination is morally required because this is the best or only means of protecting or promoting her well-being, this means that that the: “…possible courses of action physicians can legitimately

7

Here I focus on arguments based on the connection between self-determination and maximisation of patient well-being only. 8 For a detailed overview of defences of the claim that respect for self-determination facilitates the attainment of maximal best interests see Bullock (2014, 4).

1 Assisted Dying and the Proper Role of Patient Autonomy

5

take is ultimately determined by the autonomous decisions of their patients” (Varelius 2006, 123).9 This would further entail that: “the moral acceptability of voluntary euthanasia and physician-assisted suicide is also dependent on the patients’ autonomous decisions” (Varelius 2006, 123–4). Whilst the view that respect for patient self-determination is the best means of protecting patient well-being is widespread, an emerging trend in the medical ethics literature is to question the focus on patient self-determination as a means for protecting and promoting patient well-being altogether. The movement away from an emphasis on the importance of respect for patient self-determination has been made on several fronts, ranging from empirical data indicating that the choices patients make are often not self-determined, to models that facilitate shared-decision making, and defences of paternalism. The shared conclusion of these positions is that a policy of respect for a patient’s self-determined choices can, in fact, pull apart from the protection and promotion of the patient’s well-being. A key argument for rejecting the importance of respecting patient selfdetermination, as a means for protecting patient well-being, draws upon evidence from behavioural economics that individuals are poor decision-makers rarely making decisions that direct their lives in a meaningful way (Thaler and Sunstein 2008). Work in behavioural economics suggests that patient decision-making is impaired by a vast number of cognitive influences (Conly 2013). Empirical evidence suggests that people are unable to predict their reactions to future emotional events, or how happy or unhappy an event will make them to the extent that they do not know their own preferences (Blumenthal-Barby 2013, 212). Patient’s thereby often lack what the proponent of respect for patient self-determination assumes: “…a set of preferences which are clearly-defined, well-understood, and rank-ordered so that people can make logical trade-offs among them” (Schneider 1998, 69). In addition to the hindrances faced by patients when making ‘self-determined’ medical decisions, a growing body of work indicates that patients would prefer to rescind their decisionmaking authority altogether. The desirability of making a medical decision decreases the more severely ill the patient is (Botti and Iyengar 2006, 32), and a number of studies have shown that preferences for decision making are generally weak (see Ende et al. 1989, 26–27; Robinson and Thomson 2001, i34; Strull et al. 1984). All of the above could simply be seen as a failure to respect self-determination in practice.10 The problem with this response is that influences on patient selfdetermination are endemic: irrelevant suggestions can alter a patient’s assessment of a decision and external influences on a patient’s decision are unavoidable (Thaler and Sunstein 2006, 250). Moreover, these cognitive influences are not easily avoided even when the patient is aware that her decision is subject to them (see Schiavone et al. 2014; Thaler and Sunstein 2008). It simply turns out that individuals are rarely,

9

Varelius (2006, 123) refers to this as ‘subjectivism’ about the goals of medicine. I would like to thank Jukka Varelius for raising this point in comments on a draft of this paper.

10

6

E. C. Bullock

if ever, capable of exercising self-determination in a way that is free from nonself-determined cognitive influences. The argument that patient’s often lack selfdetermination cannot therefore be addressed by facilitating better respect for patient self-determination in medical practice. In view of empirical research that patients lack self-determination, autonomy (understood as a means for protecting well-being) has been increasingly recognised as something that is overvalued (Conly 2013, 25). This has led to the development of decision-making models that explicitly move away from the focus on patient self-determination. The development of ‘nudge’ paternalism is one example of a new model of decision-making that reflects the general concern that patient self-determination is overvalued. Nudge paternalism interferes with the individual’s choices by making it more difficult for them to choose an option that they themselves judge to be detrimental to their best interests, and easier to choose the option that they judge to be favourable (Thaler and Sunstein 2008, 5). The position is justified on the basis that patients would probably make different decisions in the absence of their cognitive hindrances (Schiavone et al. 2014, 105). Whilst respect for patient selfdetermination is still treated as being important, a patient’s decisions are not taken at face-value. Rather than viewing patients as experts in the exercise of their selfdetermination the nudge paternalist argues that health care practitioners are justified in ‘nudging’ the patient towards the option that is likely to serve their true interests and preferences. In addition to models of nudge paternalism, ‘widely advocated’ models of shared decision making in healthcare practice have arisen in order to address similar concerns (see Birchley 2014; Sandman and Munthe 2009, 2010). The claim that the patient necessarily gets the decision right from the start is argued to be ‘implausible’ (Sandman and Munthe 2010, 73). Instead of ‘abandoning’ patients to the decision they make: “…regardless of the possibly catastrophic consequences that might follow” (Maclean 2006, 329) medical practitioners are encouraged to advise and support their patients, thereby increasing the chance that the patient will reach a decision that is consistent with the patient’s goals and values (Maclean 2006, 337). Once again, whilst still respected, the role of patient self-determination takes a lesser role, the priority changing to making the decision that best protects the patient’s well-being (see also Loewy 2005, 464; Callahan 1992).

1.4 A Revised Role for Patient Consent Acknowledging the limited role that patient self-determination has to play in protecting patient well-being leaves us with at least two options with respect to the argument from autonomy for the moral permissibility of assisted dying. One option would be to deny that assisted dying is ever permissible given a patient’s limited ability to make decisions that protect her well-being. A second option is to revise the role that autonomy plays in end-of-life decision making. This option allows for the permissibility of assisted dying on the basis of a principle for respect for autonomy as

1 Assisted Dying and the Proper Role of Patient Autonomy

7

a side constraint on action, rather than out of respect for patient self-determination. It is this second option that I explore in the following. I want to suggest the following two necessary (although not sufficient) conditions for the moral permissibility of assisted dying: · assisted dying is in the objective best interests of the patient · assisted dying has been consented to by the patient.11 Given that we ought to be suspicious that a patient can exercise her selfdetermination in a way that protects her best interests I argue that the determination of whether it is in the patient’s best interests to die ought to be determined objectively. Secondly, I argue that the role patient autonomy plays in determining the permissibility of assisted dying is as a permission or constraint on action. Whether or not assisted dying is permissible thus depends on objective facts about the patient’s best interests and the presence of the patient’s consent. Whilst these conditions do not justify assisted dying, they account for the role that patient autonomy ought to play in end-of-life decision-making. Importantly, they provide a way of thinking about the role of patient autonomy in a way that is in line with the increasingly reduced focus on the importance of patient self-determination as a means for protecting patient well-being.

1.4.1 Objective Interests Given a well-founded and growing unease of treating respect for patient selfdetermination as a guarantor of patient well-being promotion, a patient’s best interests ought to be thought of independently of her self-determining choices. An independent approach for determining the best interests of the patient, and whether or not it is good for her to die, can be drawn from an objective theory of well-being. Such an account identifies a patient’s best interests in a way that is not solely dependent upon her preferences and values: The idea of the objective list is simply that what is intrinsically good for a person is fixed independently of that person’s attitudes or opinions; the items on the list for an individual are there independently of whether the individual has favourable attitudes toward them or himself judges that the items are valuable for him (Arneson 1999, 118–119).

Since the objective list account of well-being is independent of the individual’s preferences and desires it is possible for the account to determine that something will contribute to an individual’s well-being even if it directly conflicts with that individual’s preferences: 11

These are moral, rather than legal, conditions. Advocates for the legalization of assisted dying similarly agree on the importance of determining that the patient is suffering and that she consents to the intervention (Young 2014a, b). This paper focus on the moral framework underlying these legalistic conditions. Specifically, I argue that the determination of whether or not the person is suffering or whether her life is overly burdensome is not reducible to the self-determining patient’s conception of her best-interests.

8

E. C. Bullock What is essential is that these are theories according to which an assessment of a person’s well-being involves a substantive judgement about what things make life better, a judgement which may conflict with that of the person whose well-being is in question (Scanlon 1993, 188).

An objective list account identifies the substantive goods that contribute to the well-being of an individual and holds that they are good for the individual independently of that individual’s preferences. Objective lists generally contain more than one substantive good, such as: · · · · · · · · · · · · · · · · · ·

Life, consciousness, and activity Health and strength Pleasures and satisfactions of all or certain kinds Happiness, beatitude, contentment, etc. Truth Knowledge and true opinion of various kinds, understanding, wisdom Beauty, harmony, proportion in objects contemplated Aesthetic experience Morally good dispositions or virtues Mutual affection, love, friendship, cooperation Just distribution of goods and evils Harmony and proportion in one’s own life Power and experiences of achievement Self-expression Freedom Peace, security Adventure and novelty Good reputation, honour, esteem etc. (Frankena 1973, 87–8).12

An objective list theory of well-being provides a basis for balancing different substantive goods in different situations beyond relying upon patient selfdetermination. Accordingly, it is possible that in some instances the medical decision that will be in the patient’s best interests might not involve respecting her self-determination. A central suspicion regarding objective list accounts of well-being is that they are too rigid to account for the differences in individual values and interests: “…as if the same things must be valuable for everyone” (Scanlon 1993, 188). Specifically, it is argued that whatever is included on the list there are likely to be persons who do not want certain things to appear on it, and for others to want to add items that are not already included. In order to illustrate this concern Griffin uses the example of: “[a] group of scholars [who] may, with full understanding, prefer an extension to their library to exercise equipment for their health” (Griffin 1986, 45). In this example, the scholars regard their well-being to rest on increasing their knowledge, rather than improving their health. According to Griffin this is a problem for objective theories 12

The items on this list are not presented in any order of priority.

1 Assisted Dying and the Proper Role of Patient Autonomy

9

of well-being (assuming that health is always prior as a value to knowledge) as it is too rigid to account for what matters to the individual (Griffin 1986, 51). The supposed difficulty in responding to this objection is finding a way to allow for a variation in individual well-being with an account that identifies well-being independently of the individual’s experiences and desires. However, it is possible for an account of well-being to include variances in individual preferences and values without the account being reduced to them. Instead, a theory of well-being can be subject-related,13 assessing the objective well-being for a particular individual, as opposed to individuals in general. The possibility of developing a subject-related account of well-being that focuses on the well-being of each individual is not ruled out by the adoption of an objective account. Indeed, an objective account: […] does not deny that an individual’s attitudes may partly determine what is prudentially valuable for her. An individual’s attitudes do not determine what items properly belong on her objective list, but among the items that appear, some may include requirements concerning her attitudes and opinions. For example, an objective-list view might well hold that one good thing for an individual is that her important life aims be satisfied, with importance determined by her own subjective ranking of her aims (Arneson 1999, 117).

Thus, even though an objective list is not reducible to a subject’s preferences or desires, this does not mean that an objective list cannot accommodate individual aims, goals, preferences and values. One way in which this could be accommodated would be to include self-determination on the list: “[…] claiming that the informed and reflective living of one’s own life for oneself itself constitutes a good” (Crisp 2008). Significantly, just because self-determination features on the list this does not concede that self-determination should always be respected. Rather, exercising self-determination is only one element on the list, amongst others, that contributes to overall well-being. Instead of prioritising self-determination, self-determination becomes one good to consider among many. Furthermore, an objective list can also be treated as being related to circumstantial factors. Imagine, for instance, that Griffin’s scholar is in the midst of a heart attack. Given the scholar’s preference for furthering her knowledge over protecting her health, would it be right to hand her a rare academic text book to read in her last moments rather than taking her to hospital for treatment? The answer to this question is almost certainly no. The items on an objective list not only vary from individual to individual in terms of their personal preferences, but the ordering of items on the list can also vary from circumstance to circumstance. Such an account is not reducible to the subject’s preferences but remains related to an individual’s tastes and circumstances. The objection that objective list accounts are too rigid cannot be sustained.14 An advantage of a subject-related objective account of patient best interests is that it sits well with some intuitions regarding the permissibility of assisted dying. Take 13

Here, I use the term ‘subject-related’ in order to contrast my position with such subject-relative views that do not allow the ordering of items on the list to differ between individuals whilst remaining independent of their views on the matter (cf. Varelius 2003, 368 ff.). 14 The above discussion on subject-relatedness is adapted from Bullock (2012).

10

E. C. Bullock

for instance Scoccia’s example of an instance in which we are reluctant to assist in a suicide: Imagine that Frank suffered a shoulder injury a year ago that will prevent him from ever again playing competitive golf, his life’s passion. He is no longer depressed about his situation but feels certain that he has nothing to live for and would be better off dead. Legalization with limits rightly denies [assisted dying] eligibility to Frank. To extend eligibility to people like him would, as Daniel Callahan says, be ‘self-determination run amok’. [Callahan (1992)] [He] should be denied suicide assistance [because he is] better off alive than dead notwithstanding [his] belief to the contrary (Scoccia 2008, 367).

A benefit of a theory of well-being that determines whether or not it is objectively good or bad for patient to die is that it reflects standard practices and intuitions regarding the permissibility of assisted dying. We are, for instance, reluctant to permit assisted dying in cases where the patient is young or the disease is not terminal even if their decision is completely self-determined. This is because we rightly judge that it is not in the patient’s best interests. Likewise, it is often recognised that pain and suffering arising from a terminal disease towards the end of a natural life are good reasons for permitting assisted dying. It is certainly controversial to argue that sometimes it is objectively best for a patient to die, regardless of the patient’s wishes.15 Suspicions arise about how a doctor can possibly judge whether or not assisted dying is in the best interests of the patient. At this point I remind the reader that the aim of this paper is not to provide a justification for assisted dying. Certainly, it might turn out that such objective values cannot be determined and so we might not know whether or not assisted dying is ever justified. The purpose of this exposition is instead to determine the role that patient autonomy ought to play in matters of assisted dying, not to determine whether assisted dying is ever permissible.

1.4.2 Autonomy as a Side-Constraint Determining the permissibility of assisted dying cannot end with the claim that whether or not it is in the best interests of the patient to die can be determined objectively. Indeed, such a position could be easily objected to by those who are rightfully worried that the legalisation of assisted dying would lead to mandatory euthanasia. This is where the proper role of informed consent comes into play. Above I indicated that autonomy is often understood as self-determination. However, a second important way in which patient autonomy can be interpreted is as a side-constraint on the actions of others. Autonomy as a side-constraint prevents 15

Note that this position is different to the claim that there are objective standards for determining the permissibility of assisted dying, such as ‘the patient is suffering’ or ‘the patient has an incurable illness’. Whilst these conditions are relevant for determining the permissibility of assisted suicide, on my account, it could turn out that it is not in the objective best interests of a patient suffering from an incurable illness to die. It could equally turn out that it is in the best interests of a patient to die (when suffering from an incurable illness) even if she would prefer to continue living.

1 Assisted Dying and the Proper Role of Patient Autonomy

11

others from doing what they want with the individual, even when they correctly judge the interference to be objectively in the individual’s best interests. It can thus be invoked as a reason why a medical practitioner should not do something to her patient, specifically: “…that Y should not do something to X because: (a) X has a right to consent to things being done to her; and (b) the appropriate consent has not been given” (Foster 2009, 8–9). Given the defence of an objective-list account of well-being, it is an interesting question as to why we should care about patient autonomy at all. Those who doubt that there is any additional value to exercising autonomy beyond its contribution to individual well-being might be motivated to adopt a paternalistic approach to medical decision-making. However—without reaching any conclusions on medical paternalism in general—it seems to me that we have a special reason to respect patient autonomy as a side-constraint when it comes to life or death decision making. This is because whilst it is only morally controversial to interfere with an individual paternalistically in order to promote her well-being, assisted dying is both morally and legally serious. Arguments opposing the legalisation of assisted dying on the basis that it constitutes murder (see, e.g., Brock 1993, 208; Dworkin 1993, 21; Glover 1990, 45) are far more compelling than the comparatively mundane objection to paternalism that it constitutes unwarranted interference with an individual’s decision (see, e.g., Mill 2008, 92; Feinberg 1986). Thus, in cases of assisted-dying we should not only be concerned with patient well-being, but given the gravity of the decision, the patient’s consent or refusal. Justifications for autonomy as a side-constraint are both legal and moral in nature. In moral terms, Nozick argues that: “[s]ide constraints upon action reflect the underlying Kantian principle that individuals are ends and not merely means; they may not be sacrificed or used for the achieving of other ends without their consent” (Nozick 1974, 30–31), expressing the “inviolability of other persons” (Nozick 1974, 32). Others have suggested that autonomy as a constraint on action is justified on the basis that each individual has full ownership of her body and: “…no one else, not even the State, can interfere with the exercise of that ownership, in whatever way it manifests itself” (Macioce 2012, 102). In legal terms, autonomy as a side-constraint is protected by the laws of battery and assault (Maclean 2006, 323). Manson and O’Neill argue that informed consent procedures can only be successfully justified on the basis of laws of battery and assault. They argue that it is in the nature of medical practice to carry out procedures that would normally infringe upon an individual’s human rights. Informed consent permits a doctor to carry out actions that would usually be illegal or unethical and so: “…is a way of justifying action that would otherwise violate important norms, standards or expectations” (Manson and O’Neill 2007, 75). For example, an individual is protected in law from having his teeth removed by another individual. It is, however, sometimes necessary for dentists to remove rotten teeth in order to prevent infection and further pain in their patient. Without the patient’s consent the dentist will be breaching ethical norms and laws (Manson and O’Neill 2007, 76). In order to prevent such medical procedures being classified as battery or assault, the patient can thus consent to her rotten tooth being removed without the dentist being prosecuted.

12

E. C. Bullock

When we understand patient autonomy as a side-constraint on action instead of viewing patient self-determination as a means for protecting patient well-being we get a clearer picture of how assisted dying might be justified on the basis of an appeal to patient autonomy. I have argued that whether or not it is good or bad for a patient to be assisted in her death is to be determined objectively. But whether or not assisted dying is permissible will also depend on whether or not the patient has autonomously consented to the assistance, where autonomy is understood as having the role of a side-constraint on action. Take a case in which it is determined that it is objectively in the best interests of a patient to be assisted in their death.16 At this point the patient or her proxy can consent to the procedure or refuse to consent to the procedure as protected under the doctrine of informed consent. The role of patient autonomy as a side-constraint is to permit or refuse the procedure. Whether or not the procedure contributes to the patient’s overall well-being is not reducible to her self-determination.

1.4.3 Assisted Dying and the Proper Role of Informed Consent Given a growing reluctance towards the view that respect patient self-determination is the best means for promoting and protecting patient well-being, I have suggested a new model for determining the permissibility of assisted dying that incorporates a role for respecting patient autonomy. The two necessary conditions for permissible assisted dying are as follows: (a) assisted dying is in the objective best interests of the patient (b) assisted dying has been consented to by the patient. Determining (a) will involve communication with the patient and acknowledging her self-determination.17 However, given growing suspicions of the efficacy and accuracy of patient self-determination the patient’s values and views on the nature of her well-being should be used as one piece of information amongst others in the determination of her best interests. Self-determination thus still has a role to play—it is just not the only role, and it is certainly not the central role in determining whether or not it is in the patient’s best interests to die. Condition (b) provides an important restriction on permitting assisted dying and better captures the legal role of informed consent. Patient consent matters, but not because she has the ability to determine what is best for her, but because it protects her from unwanted interference.

16

I do not detail the conditions here but it will plausibly include things like, being near the end of life, having a terminal condition and/or the experience of great suffering. 17 As encouraged by models of shared decision making. See Sandman and Munthe (2009, 2010), Maclean (2006).

1 Assisted Dying and the Proper Role of Patient Autonomy

13

Keeping focus on the importance of consent as a side-constraint mutes the worry that overriding patient self-determination as a means for promoting patient wellbeing will lead us into full blown medical paternalism, and the legitimate worry that patients will be killed against their wishes. Autonomy as a side-constraint on action allows that even if assisted dying is judged to be in the patient’s best interests this does not mean it is permissible. The absence of the patient’s consent as a side-constraint (explicit, or hypothetical in the case of incompetent patients) renders assisted dying impermissible. Ultimately, this means that assisted dying is permissible not because it is the patient’s preference, but because it is in her objective best interests and she permitted it via her consent.

1.5 Conclusion The doctrine of informed consent is commonly justified by the claim that we ought to respect patient autonomy. This demand to respect patient autonomy is a central argument used in favour of permitting assisted dying. Specifically, it is claimed that respect for patient autonomy entails respect for a patient’s decisions about the nature of her death. An aim of this paper has been to square this argument from autonomy with the legitimate concern that medical ethics has overemphasised the role of patient self-determination in determining patient well-being. I have argued that respect for patient autonomy can be understood in at least two ways, firstly as patient self-determination and secondly as a side-constraint on action. Whilst there are many reasons to be suspicious of patient consent as selfdetermination I have shown that this does not mean that assisted dying is never justified, nor moreover that informed consent should be dispensed with. Instead, respect for patient autonomy as a side-constraint on action has a fundamental role to play in determining the permissibility of assisted dying. Specifically, I have argued that two necessary conditions for the moral permissibility of assisted dying are: (a) assisted dying is in the objective best interests of the patient (b) assisted dying has been consented to by the patient. Whilst these conditions are not sufficient for permitting assisted dying, the model importantly identifies the role that patient autonomy ought to play in end-of-life decision making in view of well-motivated arguments for attributing a lesser role to patient self-determination. The proper role of informed consent in relation to assisted dying is not to protect the patient’s self-determined views on the nature of her wellbeing, but to act as a side-constraint on objective assessments of whether or not it is good for the patient to die.

14

E. C. Bullock

References Arneson, Richard J. 1999. Human flourishing versus desire satisfaction. Social Philosophy and Policy 16: 113–142. Battin, Margaret Pabst. 2005. Ending life: Ethics and the way we die. Oxford: Oxford University Press. Beauchamp, Tom L. 2006. The right to die as the triumph of autonomy. Journal of Medicine and Philosophy 3: 643–654. Beauchamp, Tom L., and J.F. Childress. 1989. Principles of biomedical ethics, 3rd ed. Oxford: Oxford University Press. Biggs, Hazel. 2001. Euthanasia, death with dignity, and the law. Oxford: Hart Publishing. Birchley, Giles. 2014. Deciding together? Best interests and shared decision-making in paediatric intensive care. Health Care Analysis 22: 203–222. Blumenthal-Barby, J.S. 2013. Choice architecture: A mechanism for improving decisions while preserving liberty. In Paternalism: Theory and practice, eds. Christian Coons and Michael Weber, 178–196. Cambridge: Cambridge University Press. Botti, Simona, and Sheena S. Iyengar. 2006. The dark side of choice: When choice impairs social welfare. Journal of Public Policy and Marketing 25: 24–38. Brock, Dan W. 1993. Life and death: Philosophical essays in biomedical ethics. Cambridge: Cambridge University Press. Buchanan, Allen E., and Dan W. Brock. 1990. Deciding for others: The ethics of surrogate decision making. Cambridge: Cambridge University Press. Bullock, Emma C. 2012. Informed consent and justified hard paternalism. Doctoral thesis, University of Birmingham. Bullock, Emma C. 2014. Free choice and patient best interests. Health Care Analysis 1–19. https:// doi.org/10.1007/s10728-014-0281-8 Callahan, Daniel. 1992. When self-determination runs amok. Hastings Center Report 22: 52–55. Chetwynd, S.B. 2004. Right to life, right to die and assisted suicide. Journal of Applied Philosophy 21: 173–182. Conly, Sarah. 2013. Against autonomy: Justifying coercive paternalism. Cambridge: Cambridge University Press. Crisp, Roger. 2008. Well-being. Stanford encyclopedia of philosophy. http://plato.stanford.edu/arc hives/win2008/entries/well-being/. Accessed 2 Jan 2012. Declaration of Helsinki. 2008. Ethical principles for research involving human subjects. http://www. wma.net/en/30publications/10policies/b3/. Accessed 11 Oct 2008. Dworkin, Ronald. 1993. Life’s dominion: An argument about abortion and euthanasia. London: Harper Collins Publishers. Ende, Jack, Lewis Kazis, Arlene Ash, and Mark A. Moskowitz. 1989. Measuring patients’ desire for autonomy: Decision making and information seeking preferences among medical patients. Journal of General Internal Medicine 4: 23–30. English, Veronica, Gillian Romano-Critchley, and Ann Sommerville. 2004. Medical ethics today: The BMA’s handbook of ethics and law, 2nd ed. London: BMJ Books. Erbay, Hasan, Alan Sultan, and Selim Kadıo˘glu. 2010. A case study from the perspective of medical ethics: Refusal of treatment in an ambulance. Journal of Medical Ethics 36: 652–655. Faden, Ruth R., and Tom L. Beauchamp. 1986. A history and theory of informed consent. Oxford: Oxford University Press. Feinberg, Joel. 1986. Harm to self: The moral limits of the criminal law. Oxford: Oxford University Press. Foster, Charles. 2009. Choosing life, choosing death: The tyranny of autonomy in medical ethics and law. Portland: Hart Publishing. Frankena, William K. 1973. Ethics, 2nd ed. New Jersey: Prentice-Hall.

1 Assisted Dying and the Proper Role of Patient Autonomy

15

Glover, Jonathan. 1990. Causing death and saving lives: The moral problems of abortion, infanticide, suicide, euthanasia, capital punishment, war and other life-or-death choices. London: Penguin. Gray, William. 1999. Right to die or duty to live? The problem of euthanasia. Journal of Applied Philosophy 16: 19–32. Griffin, James. 1986. Well-being: Its meaning, measurement, and moral importance. Oxford: Clarendon Press. Loewy, Erich H. 2005. In defense of paternalism. Theoretical Medicine and Bioethics 26: 445–468. Macioce, Fabio. 2012. What can we do? A philosophical analysis of individual self-determination. Eidos 16: 100–129. Maclean, Alisdair. 2006. Autonomy, consent and persuasion. European Journal of Health Law 13: 321–338. Manson, Neill C., and Onora O’Neill. 2007. Rethinking informed consent in bioethics. Cambridge: Cambridge University Press. Mill, J.S. 2008. On liberty. Oxford: Oxford University Press. Nozick, Robert. 1974. Anarchy, state, and utopia. New York: Basic Books. Robinson, Angela, and Richard Thomson. 2001. Variability in patient preferences for participating in medical decision making: Implication for the use of decision support tools. Quality in Health Care 10: i34–i38. Sandman, Lars, and Christian Munthe. 2009. Shared decision-making and patient autonomy. Theoretical Medicine and Bioethics 30: 289–310. Sandman, Lars, and Christian Munthe. 2010. Shared decision making, paternalism and patient choice. Health Care Analysis 18: 60–84. Savulescu, Julian. 2003. Is the sale of body parts wrong? Journal of Medical Ethics 29: 138–139. Scanlon, Thomas. 1993. Value, desire and quality of life. In The quality of life, ed. Martha C. Nussbaum and Amartya Sen, 185–200. Oxford: Clarendon Press. Schermer, Maartje. 2002. The different faces of autonomy: Patient autonomy in ethical theory and hospital practice. London: Kluwer Academic Publishers. Schiavone, Giuseppe, Gabriele De Anna, Matteo Mameli, Vincenzo Rebba, and Giovanni Boniolo. 2014. Libertarian paternalism and health care policy: A deliberative proposal. Medicine, Health Care and Philosophy 17: 103–113. Schneider, Carl E. 1998. The practice of autonomy: Patients, doctors and medical decisions. New York: Oxford University Press. Scoccia, Danny. 2008. In defense of hard paternalism. Law and Philosophy 27: 351–381. Strull, William M., Bernard Lo, and Gerald Charles. 1984. Do patients want to participate in medical decision making? JAMA 252: 2990–2994. Tännsjö, Torbjörn. 1999. Coercive care: The ethics of choice in health and medicine. London: Routledge. Thaler, Richard H., and Cass R. Sunstein. 2006. Preferences, paternalism, and liberty. In Preferences and well-being, ed. Serena Olsaretti, 233–364. Cambridge: Cambridge University Press. Thaler, Richard H., and Cass R. Sunstein. 2008. Nudge: Improving decisions about health, wealth, and happiness. London: Yale University Press. Varelius, Jukka. 2003. Autonomy, subject-relativity, and subjective and objective theories of wellbeing in bioethics. Theoretical Medicine and Bioethics 24: 363–379. Varelius, Jukka. 2006. Voluntary euthanasia, physician-assisted suicide, and the goals of medicine. Journal of Medicine and Philosophy 31: 121–137. Veatch, Robert M. 2000. Doctor does not know best: Why in the new century physicians must stop trying to benefit patients. Journal of Medicine and Philosophy 25: 701–721. Young, Robert. 2014a. ‘Existential suffering’ and voluntary medically assisted dying. Journal of Medical Ethics 40: 108–109. Young, Robert. 2014b. Voluntary euthanasia. In Stanford encyclopedia of philosophy, ed. Edward N. Zalta. http://plato.stanford.edu/archives/sum2014b/entries/euthanasia-voluntary/. Accessed 29 Apr 2015.

16

E. C. Bullock

Emma C. Bullock was an assistant professor of philosophy and a research fellow at the Centre for Ethics and Law in Biomedicine at the Central European University, Budapest, between 2014 and 2018. She has published research on the philosophy of medicine and medical paternalism.

Chapter 2

Preventing Assistance to Die: Assessing Indirect Paternalism Regarding Voluntary Active Euthanasia and Assisted Suicide Thomas Schramme Abstract The chapter focuses on cases of assisted suicide and voluntary euthanasia in relation to the rarely discussed notion of indirect paternalism. Indirect paternalism involves not just a paternalistic intervener and a person whose welfare is supposed to be protected, but also another party, whom I call “assistant.” Indirect paternalism interferes with an assistant in order to prevent harm to another person. I will introduce a strategy that paternalists can pursue to justify indirect paternalism. It specifically targets an element of assistance cases, namely the fact that people do not necessarily have a justified claim or entitlement to demand such assistance. To prevent people from providing assistance seems normatively different from preventing a person to do something to herself by her own means. I critically discuss arguments from the goals of medicine and from the conscientious objection. These aspects are not deemed decisive when considering the case of indirect paternalistic intervention. Finally, I argue against the rationale of indirect paternalism by showing that there are at least two situations where it does not succeed. One such situation that undermines the justification of indirect paternalism is present when the offered service is itself harmless, another pertinent situation consists of a person necessarily requiring assistance to be really free. At least in some cases, these very conditions are given when contemplating assistance to die. Keywords Assisted suicide · Harm principle · Voluntary consent · Active euthanasia · Paternalistic intervention

2.1 Introduction Assisted suicide and voluntary active euthanasia have been discussed in bioethics for a number of years. Many arguments have been put forward, either supporting or rejecting such methods. Sometimes people reject the practices for principled reasons; T. Schramme (B) Department of Philosophy, University of Liverpool, Liverpool, UK e-mail: [email protected] © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 M. Cholbi and J. Varelius (eds.), New Directions in the Ethics of Assisted Suicide and Euthanasia, The International Library of Bioethics 103, https://doi.org/10.1007/978-3-031-25315-7_2

17

18

T. Schramme

sometimes they are worried about possible consequences of turning killing on request or helping people to die into a common procedure. In this chapter I will focus on a possible yet neglected rationale for preventing assistance to die. This is indirect paternalism: We might interfere with an assistant’s freedom to act in order to prevent harm or secure the good of another person who has asked for such assistance. I believe that indirect paternalism poses more complicated normative issues than “normal” cases of paternalism, which only involve one party interfered with. Direct paternalism briefly is an interference with the freedom of persons to choose or act in order to prevent harm to these persons themselves. A common case involving direct paternalism is suicide prevention. One possible strategy for defending the practice of assistance to die would be to draw an analogy between suicide and assisted killing on request. If we do not have sufficient reason, this argument goes, to prevent suicide on paternalistic grounds, we also do not have sufficient reason to prevent assisted suicide or voluntary active euthanasia. I want to show that this strategy of drawing a straightforward analogy between the direct form of paternalism and indirect paternalism fails. Yet I will nevertheless argue that indirect paternalism is not justified in certain circumstances.1 There might be a worry regarding the pertinence of paternalism as a rationale for intervening into assisted causes of death. It seems that it is not usually the good of the person willing to die that is decisive when considering interference, but for instance the immorality of killing itself. I agree that there might be other reasons than paternalistic ones that might justify a ban on assisted suicide or euthanasia, most importantly the already mentioned worries regarding the consequences of turning assisted killing into a regular practice. There might also be mixed motives when considering intervention. Yet it is important to acknowledge that even if intervening parties might be only concerned with the morality of killing, and not at all with the welfare of the people interfered with, it might nevertheless be a form of paternalism to interfere with a choice to die: Moral paternalism, which aims at securing the moral good of a person, can also take both a direct and an indirect form. Hence I believe there are actual cases of real or considered paternalistic interventions into assistance to die. For the purposes of the present chapter I will not distinguish between assisted suicide and voluntary active euthanasia. Though legally it is often important to know who has proximately caused the death of a person, for the normative problem I am pursuing this is not a major matter of concern. Both assisted suicide and voluntary active euthanasia are instances of what I call assistance cases. In one case the assistant provides the necessary means for a person who then kills herself, in the other case the assistants are themselves, as it were, the means to cause the death of a person willing to die. There might of course be reasons as to why a person cannot kill herself, most obviously when they are physically unable to do so, for instance when they are paralyzed. So active euthanasia might be the only possible means for a person willing to die to actually succeed in this plan. 1

This chapter draws on two other papers that focus on different problems but fall in the same area (Schramme 2013, 2015).

2 Preventing Assistance to Die: Assessing Indirect Paternalism Regarding …

19

The plan of the chapter is as follows: I will first introduce the notion of indirect paternalism. Indirect paternalism involves not just a paternalistic intervener and a person whose welfare is supposed to be protected, but also another party, whom I call “assistant”. Indirect paternalism interferes with an assistant in order to prevent supposed harm to another person. This sounds like established cases of preventing harm to others, but an important aspect of indirect paternalism is the fact that the assistance is sought by the person whose good is supposed to be secured by intervention. In the second section I will introduce a strategy that paternalists can pursue to justify indirect paternalism. It specifically targets an element of assistance cases, namely the fact that people do not necessarily have a justified claim or entitlement to demand such assistance. To prevent people from providing assistance seems normatively different from preventing a person to do something to herself by her own means. There are several aspects here that will require close inspection, especially when considering the situation of an assistant who is supposed to kill another person. Since I mainly focus on medical assistance to die, or physician-assisted suicide, which is after all the form of such assistance most often discussed in the relevant literature, I query an argument from the goals of medicine and another one from the conscientious objection. Yet I believe these aspects are not decisive when considering the case of indirect paternalistic intervention. Finally, I argue against the rationale of indirect paternalism by showing that there are at least two situations where it does not succeed. One such situation that undermines the justification of indirect paternalism is present when the offered service is itself harmless, another pertinent situation consists of a person necessarily requiring assistance to be really free.

2.2 Indirect Paternalism Cases where a person A requires the support or service of another person B to achieve a particular outcome or to perform an action can be called assistance cases. If the required assistance is forbidden, or by other means hindered or made impossible, for reasons of securing the good of person A, then we can deem these interventions instances of indirect paternalism. Indirect paternalism is therefore a form of multipleparty paternalism.2 Assistance to die might involve more than two parties, for instance when a psychiatrist evaluates the capacity of a patient to decide about their death and another party administers a lethal drug. Also, the paternalistic intent of a person or institution considering a ban on a particular service might not (only) aim at the person who requests a service but at the good of a potential assistant, for example when active euthanasia is prohibited for reasons of preventing psychological harm to the person who kills another person on request. This would transfer the same 2

Though at least theoretically there might be multiple-party cases of paternalism that are not forms of indirect paternalism, see Feinberg (1986, 9). Feinberg used the term “two-party cases”, but this might be confusing as there are more than two parties involved in the practice of indirect paternalism. He obviously meant that two parties are the target of a paternalistic interference, where one party is interfered with and the other benefits.

20

T. Schramme

case, which can be discussed under the rubric of indirect paternalism, (also) into a common case of paternalism where the intervenee, i.e. the person interfered with, and the supposed beneficiary are the same person. A central feature of many assistance cases is that a particular type of action, which is deemed an example of grave harm changes its normative status by a seemingly tiny bit of addition: the voluntary consent of a person. An action by B done towards A—say, to cut flesh from his body—would normally be a crime, but is a body modification (“scarification”) if requested. This ties in well with the legal principle volenti non fit injuria, which (roughly) translates to “no one is wronged willingly”. For anti-paternalists in the tradition of John Stuart Mill’s harm principle, only harm done to others (against their wishes) should be prevented, whereas “[o]ver himself, over his own body and mind, the individual is sovereign.”3 Mill’s general anti-paternalism has raised many doubts. Objections to the liberal harm principle might work in the following way: If harm is bad for a person, then it is always bad, whether it is wanted or chosen by the person herself or not. Indeed, harm, such as physical injury that involves pain, is intrinsically bad, so why should it matter for normative purposes whether the affected person desires it? There is, however, a convincing response to this objection. It stems from Joel Feinberg’s interpretation of the harm principle (Feinberg 1986, 10 ff.). Feinberg reads the harm principle as requiring the prevention of wrongfully inflicted harms. He also offers two interpretations of the notion “harm”, meaning firstly to injure or damage, and secondly to set back interests. We might want to call the first conceptualization “impersonal harm”, because it does not necessarily involve a point of view of a person; it is simply something undergone, for instance an alteration of the bodily structure. The second reading of the notion of harm might be called “personal harm”, as it involves the standpoint of a person.4 Only things that happen to a person, which are deemed a setback of her interests, are instances of personal harm. For Feinberg this second reading leads to a proper understanding of the harm principle: It requires prevention of any wrongfully inflicted setback of interests. This principle obviously does not prohibit voluntarily chosen injuries, disadvantages or other detriments; indeed, these are not even considered harms, or personal harms in my own terminology. In short, according to Feinberg’s account, we may stick to the general anti-paternalism implied by the harm principle and endorse the volenti maxim.5

3

Because B seems to harm A one might think that these cases were already banned by the Millian harm principle. Yet it should be obvious that the voluntary consent changes the normative status of the same action here and, as we will see, it is even slightly misleading to say that B harms A. 4 We could also say that impersonal harm is harm for a person, but only personal harm is harm to a person. 5 Note that it is even possible to accept that impersonal harm is intrinsically bad, and still allow for other considerations, which have to do with personal interests, to outweigh this kind of harm and to conclude that there is no personal harm present where a person has an interest in an impersonal harm. A person may reasonably choose what is intrinsically bad, as long at it is not only intrinsically bad.

2 Preventing Assistance to Die: Assessing Indirect Paternalism Regarding …

21

2.3 The Normative Difference Between Direct and Indirect Paternalism There are still plenty of strategies to defend paternalistic intervention. For instance, doubts might be raised regarding the voluntariness of particular choices. In cases of desired killing, interveners might want to quarrel with the reasonableness or, indeed, sanity of a desire to die; or one might want to put doubt on the voluntariness of choice by pointing out strong societal influences in many countries on people’s willingness to die or the terrible circumstances in which terminally patients often find themselves. Were these kinds of arguments to be successful, they would deem paternalistic intervention into these choices an instance of soft paternalism, as it would only account for a prevention of non-voluntary choices—something that is normatively less problematic than hard paternalism.6 I disagree with this strategy, mainly because I see voluntariness as a procedural feature that has to do with the way a choice has been reached. If no coercion or similar influences of will-formation are involved, a choice is voluntary.7 We cannot identify involuntariness by the content of a choice. Voluntariness and reasonableness are simply not the same—on whichever account of the reasonable we might come up with (Feinberg 1986, 104 ff.; cf. Möller 2005, 164 ff.). The wish to die is not always irrational, though it might be under certain circumstances (cf. Cholbi 2013a, b). Although it is true that cultural and similar influences on choices can be strong, this is not by itself sufficient warrant for deeming certain choices involuntary. Indeed one might wonder how we would otherwise be at all able to draw a distinction between voluntary and involuntary choices, as every choice is strongly influenced by our circumstances, our upbringing, our friends, etc. Roughly speaking, a choice is voluntary when a person is under no coercive influence and endorses, or identifies with, her choice.8 There is no principled argument that would exclude choices involving impersonal harm from the realm of voluntary choices.9 A more viable strategy for the paternalist against certain services is more closely related to a particular feature of indirect paternalism, namely that it involves more than one party, of which one is an assistant. As I said earlier, the services we are considering, such as voluntary active euthanasia, require assistance by other people. Now, there seems to be an important difference between preventing a person directly from doing something and preventing another person from offering a requested service 6

It is indeed arguable whether soft paternalism should be called “paternalism” at all (see Feinberg 1986, 12). But even if intervention into non-voluntary choices were not paternalism after all, this would of course still allow us to regard the cases we now refer to under the umbrella term “soft paternalism” as unjustified. 7 Obviously it is an important issue what kind of undue influences there might be, which consequently undermine consent to (impersonal) harm. I cannot discuss this question here, but see, for instance, Kleinig (2010, 13 ff.). 8 In these cases we might also want to use the notion “autonomous choice”. I disregard the relation between autonomy and voluntariness for the purposes of this essay. 9 There is an important debate regarding the possible coerciveness of inducements that I will ignore for the purposes of this chapter (see, for instance, Radcliffe 2010).

22

T. Schramme

to that person, even where the very same actions are involved. In one word, indirect paternalism—which implies preventing assistance—might well be normatively different from direct paternalism.10 This is mainly due to the fact that there does not seem to be an entitlement to be offered assistance, whereas a person usually is deemed to have the right of self-ownership. The latter allows persons to do lots of things to themselves.11 So what we need to look at now, when considering a possible justification of assistance to die, is whether the person seeking assistance has a justified claim to such a service. I believe there is indeed a normatively significant difference between direct and indirect paternalism, which would call for much more detailed reasoning than I can provide in this essay. There is very little that has been written so far on the topic of indirect paternalism [but see especially von Hirsch 2008; Simester and von Hirsch 2011, 166 ff.]. It seems that one can be an adamant anti-paternalist yet allow for indirect paternalism. Although in the following I will reject indirect paternalism in some cases, I do not deem this to amount to a rejection of the rationale per se.

2.4 When Indirect Paternalism is not Justified There seem to be at least two circumstances where it would be implausible to argue that the difference between intervention and preventing assistance has any normative significance: When a person has a justified claim, or entitlement, to a service,12 or when the service itself is not dubious for moral or other reasons, for instance reasons that deem a service imprudent. One aspect regards the person seeking a particular service, the other aspect is concerned with the nature of the service provided. The latter case applies to services such as selling sweets. Although we might have a paternalistic interest in banning it, because people tend to eat too many sweets, with well-known effects on their health, the service itself—offering a product in exchange

10

This is an important insight that is often ignored, for instance by Feinberg (see Hirsch 2008; du Bois-Pedain 2010). 11 Surely one may want to insist that the consent given by a person to the service of the assistant is normatively sufficient to justify providing assistance. Indeed, this seems to follow from the logic of the volenti principle. But one aspect of my chapter is to show why assuming a normatively different status of indirect paternalism is plausible and that the volenti principle cannot be an absolute principle. 12 I take ‘claim’ to be a moral notion here. It can be seen as a moral right, but I avoid the terminology to prevent confusion with legal rights. A person might have a legal entitlement to all kinds of morally dubious services, but these contracts are not my concern here, rather whether those contracts should be allowed. I also take ‘claim’ as to imply a duty of others to refrain from interference, so it is not just a ‘liberty’, in the Hohfeldian sense (Hohfeld 1923), where a person has permission to do something and hence is not doing something wrong. A justified claim, or entitlement, as such, does not include a duty of others to provide necessary means to pursue a goal, but I want to consider later how far such provision might indeed be morally required.

2 Preventing Assistance to Die: Assessing Indirect Paternalism Regarding …

23

for money—is neutral.13 The possible negative consequences are due to the service users. If the service is itself harmful, for instance because the offered good contains hazardous ingredients, a ban might either amount to avoiding third party harm and hence not be an issue of paternalism. Examples of these cases might be well-known health and safety measures we find in many legal requirements regarding production, sales, trades and services. A service or offered good might also be itself harmful, yet something a customer wants anyway. This differs from the situation just mentioned, where we can assume that people do not agree with certain harms, especially where they are not known or cannot serve as means to other purposes. If we remember the difference between impersonal and personal harm, we could say that a service such as killing is an impersonally harmful service, but it might not always be deemed a personal harm. Very often it is of course not easy to say whether a service is as such harmful in the way that is of significance for its normative assessment, namely in terms of posing personal harm. Obviously this makes many cases, where paternalistic intervention is considered, so difficult to assess. Regarding one of the mentioned aspects that would undermine the justification of indirect paternalism we can therefore conclude that services that are in themselves morally and prudentially neutral may not be banned. This seems straightforward enough in theory, as there is no harm involved—so nothing we can protect a person against—but there are complications in practice. Regarding services that are in themselves harmful in a certain respect, such as killing another person, we need to ask whether they are of a type people would normally try to avoid, hence could be deemed general personal harms. Again, this obviously poses many more questions that cannot be discussed here, such as whether a general ban that prevents all potential service users from gaining access to the service can ever be justified. After all, there might always be at least one person for which this impersonally harmful service is not personally harmful. It seem that this is a problem of the normative assessment of general rules, such as legal bans, as opposed to individual, single case interventions, hence they point at a possible normative difference between interpersonal and legal paternalism. The other condition undermining the rationale for indirect paternalism is fulfilled when a person has a moral claim or entitlement to a service, even where it could result in personal harm. Consider the case of parental education. Although we know that many parents raise their children in atrocious ways, we still respect children’s right to be raised by their parents. Hence even if we find a particular service dubious for moral or other reasons, we might still be convinced that it should be allowed, even judged from a paternalistic point of view. It would also be wrong to argue that any service failing on both criteria, i.e. that is deemed problematic for moral or other reasons and that does not involve entitlement to the service, should therefore be banned. After all, we need to balance the good of individual liberty against such a ban. I rather want to argue that there seems to be more scope for the paternalist in indirect paternalism than in direct paternalism. 13

Though we might want to introduce bans on, say, the aggressive marketing of sweets.

24

T. Schramme

Concerning direct paternalism, there is a kind of presumption of entitlement to do many things to oneself in virtue of self-ownership, but this does not automatically apply to the same actions performed by another person on request. This is probably best seen by the very example of suicide as opposed to assisted suicide. Although there might be a justification of the latter practice after all, the onus of justification is on the side of the defender of assisted suicide, whereas there seems to be presumption of the moral legitimacy of suicide (cf. Bergelson 2010).14 Obviously, there might be good reasons to oppose suicide and hence to overcome the presumption. Yet, my chapter is not concerned with the legitimacy of suicide but with the evaluation of assistant cases. I argue that even where suicide is not regarded as morally wrong all things considered, it does not automatically follow that people have a claim to be assisted in suicide. This is different from the common argument that if suicide is not morally wrong, assistance in suicide cannot be morally wrong either. I take it that the services considered for the purposes of this chapter—assisted suicide and voluntary active euthanasia—are indeed morally or prudentially problematic, hence fail on one (part) of the criteria: These services are not wholly neutral. They come along with at least impersonal harms. As I have just said, this would not alone justify a ban, as on balance a legislature might want liberty to prevail. It seems unlikely, though, that balancing alone would tip the scales in favor of the services under consideration. People who would like to argue against a ban on assistance to die need a more principled argument. One way would be to consider more closely the second criteria just mentioned. We therefore need to scrutinize whether people might have a justified claim to the services under consideration.15 If they do, the normative difference between indirect and direct paternalism breaks down and the paternalist strategy fails.

14

In contrast, Hill (2014, 277), in a recent contribution to a companion asserts: “Thus if suicide is not immoral in certain cases, then it will be permissible for others to assist unless there are further arguments against this.” (Emphasis in original.) It might be worth stressing that I do not want to argue that an assistant’s help to kill a person might be itself morally wrong, but that the case for an entitlement to such service needs to be made. 15 Some people might want to say that I am conceding far too much to the paternalist, as antipaternalists would maintain that service users always have a claim right, if not to the provision of services, but to purchase services on a free market. In addition they might want to say that service providers have the right to sell their services, as long as these are neither immoral nor illegal. But this argument relies (a) on the ideology of the free market, a topic I would like to avoid, and (b) on a liberal reading of what might be regarded as immoral—namely only services that cause personal harm to others. My aim here is to scrutinize the paternalist strategy in relation to indirect paternalism without begging the question in favor of a strongly liberal, or even libertarian, point of view, although I have of course already hinted at certain aspects of a liberal viewpoint that seem to me unavoidable.

2 Preventing Assistance to Die: Assessing Indirect Paternalism Regarding …

25

2.5 Some Arguments Specific to Health Care There are two special problems for a defense of entitlements to assistance to die as applied to medicine particularly. The first one is the idea of traditional goals of medicine, which might stand against those claims. The second one is the possibility of conscientious objections by medical personnel. Both problems are of some importance, as it is certainly important to also normatively assess the situation of an assistant and not just the person requesting assistance. Nevertheless, I want to quickly establish why these points should not cause too much trouble for the anti-paternalist before moving on to the main question, namely whether controversial services can be justifiably claimed at all. As far as the goals of medicine are concerned it is of course correct to state that health care’s primary task is to cure disease and to alleviate suffering. Arguably, assistance in dying therefore seems to be in line with the traditional goals of medicine, at least insofar as it relieves suffering. Yet, it might be argued that normally health care treatments are offered only when they are indicated, i.e. when a health issue, usually a disease, is present, and there might of course be cases where assistance to die is sought although no disease is (yet) present. Although not completely beside the point, this argument cannot by itself establish why medicine should stick to its traditional goals. As long as services are being paid for by customers themselves, there does not seem to be a general reason against offering medical skills and knowledge for the desired use of healthy people. In fact, medicine has always and traditionally offered at least a few services that were not treatments of disease, for instance abortion. Indeed, the whole profession for a long time was a paid service. So why should it not offer the whole range of its possible services, as long as clients choose them freely and other people do not suffer any harm because of these services? It is notoriously difficult to establish goals internal to the tradition or actual practice of medicine. Individual medical professionals must not be compelled to perform particular services, though. They might object due to conscientious reasons. Again, this is a valid point, and it seems all the more plausible regarding additional medical services than in the case of core treatments, where there is an extended debate about the justification of the conscientious objection. However, as long as there are some medical professionals who are willing and able to offer a service, such as killing a person on request, the possibility of conscientious objectors has no practical impact.

2.6 Moral Claims to Assistance to Die Why should clients have a claim to the services under scrutiny? Note first that to have a claim to have particular services offered is not the same as getting these services for free, or even as having a guarantee that they will be offered, for instance by introducing state services in case nobody wants to offer them on a private market. I am only interested in the option of particular services, i.e. whether people have

26

T. Schramme

a legitimate claim to demand that it might be performed by willing assistants, not whether everybody should be in a position to use the service.16 My argument here proceeds in two steps: Firstly, in the remainder of this section I will argue that respect for autonomy, a core principle that even paternalists agree with,17 needs to be seen as a demand to secure real or effective liberty,18 which, again, means to offer enabling conditions for important individual life choices. Secondly, I will explain, in the following section, why assistance to die is in congruence with established, widely accepted services. This is a kind of normalization argument, which proceeds by drawing analogies to uncontested practices. Its aim is to undermine the status of contestability of many services, such as voluntary active euthanasia. This latter point is not a strong argument, but I nevertheless see it as an important element in undermining the indirect paternalist approach. People regularly need assistance when leading their lives. This might be due to all sorts of reasons, for instance vulnerability, lack of capability, lack of time, and so on. These reasons might differ in their normative significance of course. People also differ widely in what they regard as valuable activities and pursuits. Everybody has his or her own individual and sometimes idiosyncratic life plan or idea of the good life. I have argued at the beginning that people ought to have the freedom to do what they want as long as they do not cause harm to others. This is the traditional liberal stance. Obviously, paternalists would disagree and maintain that people ought to have the freedom only to pursue what is really worthwhile. This is a very basic quarrel between paternalists and anti-paternalists that I will need to ignore. But the point we have reached in this chapter is a slightly different one: We want to consider whether indirect paternalism might be an option, even when direct paternalism regarding the very same outcome—desired death—is not justified. So the paternalist would agree that direct paternalism would not be justified in the cases under consideration, hence the very basic point about worthwhile options does not apply. The issue then really is whether the introduction of services leading to the same result might legitimately be prohibited or otherwise prevented after all. I stated earlier that the normative difference between indirect and direct paternalism hinges on the question whether service seekers have a claim to have a service introduced. This, again, is different from asking the question whether service providers have a claim to offer assistance. The latter question relates to the justification of a free market, the former is a question about the relevance of assistance for 16

This difference might not be significant in practice, after all, in case of assistance to die, as it is normally not discussed or in reality offered as a paid service. If indeed those services would only be affordable for rich people, this might lead to injustice. I disregard the issue of payments in this chapter. 17 Many paternalists support intervention into choices only where it enhances autonomy. Some paternalists have a particular, more demanding, reading of the concept of autonomy, which does not comply with the interpretation I endorse (but see Cholbi (2013a, b), for an important alternative). 18 I prefer the term “real liberty”, because it has been used in related discussions, especially in Philippe van Parijs’s book Real Freedom for All (1995). Occasionally “positive liberty” is also used in the debate, but it might cause some confusion with another notion of positive liberty that was discussed in a famous essay by Isaiah Berlin (2002).

2 Preventing Assistance to Die: Assessing Indirect Paternalism Regarding …

27

leading one’s own life. It is a question of what we mean when we say that we are free to do something, especially whether it requires the necessary means to be able to do it. Liberty, self-determination, and autonomy are terms that are often used interchangeably, and indeed they are surely closely related. Liberty to do what one wants to do,19 as long as one does not wrongfully harm other people, is a premise that is taken for granted at this stage of the argument. Where individual liberty has been granted, i.e. where we are allowed to do things to ourselves, respect for autonomy implies that we are not hindered by others to pursue our aims. But liberty is not effective where we rely on the assistance of other people to pursue these aims but they are hindered to offer their assistance. For example, to say a person is free to gain knowledge, where there are no teachers or books allowed, is making shambles of the notion of liberty to education. To be really free we constantly need the assistance of others. To respect autonomy therefore means to offer enabling conditions for services which support people in the pursuit of their individual lives (cf. Oshana 2003, 104; Möller 2009, 758). To be sure, this way of understanding autonomy as real liberty raises some problems. For instance, there seems to be a potential confusion between liberty itself and having the means to make use of one’s liberty. Indeed, a person who is not hindered from buying books might, in some important sense, be deemed free, even when she cannot afford these books. But note that here we are considering a different case, where in fact the assistance is banned from being provided at all. The analogy does not imply to only call those people free to educate themselves who have the necessary means, but the analogy is drawn to those who have access to the necessary means. So, in the analogical case, we cannot call someone free to gain knowledge where books and teachers are banned, in the same way as we cannot deem paralyzed persons free to kill themselves where assistance to die is prohibited. This might lead us to conclude that the option of being killed by an assistant should only be offered to people who are paralyzed and cannot physically kill themselves, because they are not really free—in contrast to physically able people who have the necessary means—to kill themselves. But we should not forget that there are also mental conditions that might prevent us from killing ourselves. People might be squeamish or in other respects unable to perform suicide though they indeed voluntarily wish to die. It would seem more adequate, therefore, to offer assistance to die not merely to certain groups of people, but generally to all people and then to introduce a procedure to consider their eligibility, where the criterion should be their level of individual freedom, hence whether they have the necessary means to kill themselves (see also Simester and von Hirsch 2011, 177). Some people might have a claim to such assistance, others not, but we cannot in general say who belongs to which group.

19

This formulation is less complicated than it should be. We might want to add that people should only be at liberty to do what they want to do when sufficiently informed, when no coercing influences are present, and so on.

28

T. Schramme

I have briefly used the notion of contestedness when referring to assisted suicide and voluntary active euthanasia. This has a certain empirical aspect: These services are in reality contested, due to value judgments by real people. Yet, we might also ask whether there are good reasons for these judgments. This is an issue of scrutinizing a feature of the practices, not an issue of finding out about the perception of these practices in real people. That is why I now talk about the contestability of these practices. So the second step in this part of my argument consists in pointing out that the ends that people pursue by using the services under consideration are decent and understandable: People who request active euthanasia or physician-assisted suicide want to end their suffering. Indeed, the sought services are merely extremes of widely accepted practices and they are often the only means available: In almost all societies, we offer services to alleviate suffering where we can and we usually allow people to die, even by their own hands. Hence, the services under consideration are in congruence with common practice in many countries, although admittedly we would have to say a bit more about the situations when the desire to die is really understandable and based on valid reasons (cf. Schramme 2013). Similarly, the intentions and goals of the assisting parties are generally morally valid. They want to help suffering people. Obviously this might not always be the case, for instance if the provided assistance is merely performed on grounds of financial reward. Here we might want to reconsider a ban on certain ways to provide services, but this concern does not apply to common practices of voluntary active euthanasia or physician-assisted suicide. In fact, one might see a special responsibility of medicine to provide the services sought because in many countries doctors have been installed with the power of control over effective and safe means to end one’s life.20 This license comes with an obligation to act in the public’s interest. Now, if my argument has been successful, either medicine should give access to such reliable means to suicide or provide these means themselves, i.e. take on the role of assistants.

2.7 Conclusion Our discussion of the possible normative difference between indirect and direct paternalism has brought us to the conclusion that, although it has normative significance, it is inconclusive as regards the services under consideration. If an individual is allowed to do certain things that are deemed morally or prudentially problematic—though she might not actually be able to perform it herself—then required services that offer the very same results should also be allowed. I therefore conclude that it does not matter, for cases of assistance to die, whether an action of a person or a related service by someone else is hindered or banned for paternalistic reasons. If we oppose direct paternalism, we should also oppose indirect paternalism in parallel cases. This might 20

Thanks to Michael Cholbi for raising this issue.

2 Preventing Assistance to Die: Assessing Indirect Paternalism Regarding …

29

still leave the paternalist with a strategy, but it is then a strategy that is not specific to indirect paternalism. It concerns whether the person who seeks assistance voluntarily agrees with it. This is similar to the question whether the self-harming person acts voluntarily. Yet, where there is a justified claim to assistance and no personal harm involved, voluntary consent is sufficient to justify the use of a service. Hence the volenti maxim is still in place, though in a slightly more complicated way, because not all assistance cases seem to be solved simply by applying this maxim.

References Bergelson, Vera. 2010. Consent to harm. In The ethics of consent: Theory and practice, ed. Franklin G. Miller and Alan Wertheimer, 163–192. New York: Oxford University Press. Berlin, Isaiah. 2002. Two concepts of liberty. In Liberty: Incorporating four essays on liberty, ed. Henry Hardy, 166–217. Oxford: Oxford University Press. Cholbi, Michael J. 2013a. Kantian paternalism and suicide intervention. In Paternalism: Theory and practice, ed. Christian Coons and Michael E. Weber, 115–133. Cambridge: Cambridge University Press. Cholbi, Michael J. 2013b. The terminal, the futile, and the psychiatrically disordered. International Journal of Law and Psychiatry 36: 498–505. Du Bois-Pedain, Antje. 2010. Die Beteiligung an fremder Selbstschädigung als eigenständiger Typus moralisch relevanten Verhaltens—Ein Beitrag zur Strukturanalyse des indirekten Paternalismus. In Paternalismus im Strafrecht: Die Kriminalisierung von selbstschädigendem Verhalten, eds. Andrew von Hirsch, Ulfrid Neumann, and Kurt Seelmann, 33–56. Baden-Baden: Nomos. Feinberg, Joel. 1986. Harm to self: The moral limits of the criminal law. Oxford: Oxford University Press. Hill, Thomas E. 2014. Killing ourselves. In Cambridge companion to life and death, ed. Steven Luper, 264–281. Cambridge: Cambridge University Press. Hohfeld, Wesley N. 1923. Fundamental legal conceptions as applied in judicial reasoning. New Haven: Yale University Press. Kleinig, John. 2010. The nature of consent. In The ethics of consent: Theory and practice, ed. Franklin G. Miller and Alan Wertheimer, 3–24. New York: Oxford University Press. Möller, Kai. 2005. Paternalismus und Persönlichkeitsrecht. Berlin: Duncker & Humblot. Möller, Kai. 2009. Two conceptions of positive liberty: Towards an autonomy-based theory of constitutional rights. Oxford Journal of Legal Studies 29: 757–786. Oshana, Marina A.L. 2003. How much should we value autonomy? Social Philosophy and Policy 20: 99–126. Radcliffe, Janet Richards. 2010. Consent with inducement: The case of body parts and services. In The ethics of consent: Theory and practice, eds. Franklin G. Miller and Alan Wertheimer. Schramme, Thomas. 2013. Rational suicide, assisted suicide, and indirect legal paternalism. International Journal of Law and Psychiatry 36: 477–484. Schramme, Thomas. 2015. Contested services, indirect paternalism and autonomy as real liberty. In New perspectives on paternalism and health care, ed. Thomas Schramme. Heidelberg: Springer. Simester, Andrew P., and Andreas Von Hirsch. 2011. Crimes, harms, and wrongs: On the principles of criminalisation. Oxford and Portland, Oregon: Hart Publishing. von Hirsch, Andreas. 2008. Direct paternalism: Criminalizing self-injurious conduct. Criminal Justice Ethics 27: 25–33.

30

T. Schramme

Thomas Schramme is professor of philosophy at the University of Liverpool. His main research interests are in political philosophy, ethics, and the philosophy of medicine. More particularly, he continues to work on problems of paternalism and freedom-limiting interventions in relation to health care and other contexts. His most recent book publication is Theories of Health Justice: Just Enough Health (Rowman & Littlefield Int., 2019).

Chapter 3

Autonomy, Interests, Justice and Active Medical Euthanasia Julian Savulescu

Abstract There are 4 main arguments for euthanasia: (1) arguments appealing to consistency (e.g., from passive to active euthanasia); (2) the argument from respect for autonomy; (3) appeals to justice; and (4) the argument from interests (mercy or relief of suffering). I will argue that only the last is directly relevant to active euthanasia as a medical intervention, though arguments together from autonomy and justice can in practice (through the backdoor) provide a ground for voluntary active medical euthanasia (AME). Keywords Passive and active euthanasia · Respect for autonomy · Distributive justice · Relief of suffering · Mercy in medicine

Should doctors perform active euthanasia? Theoretical background: definitions Euthanasia

X intentionally kills Y for Y’s benefit

Medical euthanasia

Euthanasia performed by a medical professional as a part of their job

Suicide

Y intentionally kills himself

Assisted suicide

X intentionally helps Y to kill himself

Active euthanasia

X performs an action which itself results in Y’s death

Passive euthanasia

X allows Y to die. X withholds life-saving treatment or withdraws life-saving treatment

Voluntary

Y requested death himself

Non-voluntary

Y is incapable of expressing a preference

Involuntary

Against Y’s wishes

J. Savulescu (B) The Oxford Uehiro Centre for Practical Ethics, Faculty of Philosophy, University of Oxford, Oxford, UK e-mail: [email protected] © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 M. Cholbi and J. Varelius (eds.), New Directions in the Ethics of Assisted Suicide and Euthanasia, The International Library of Bioethics 103, https://doi.org/10.1007/978-3-031-25315-7_3

31

32

J. Savulescu

3.1 Four Arguments for Medical Euthanasia There are 4 main arguments for euthanasia: (1) arguments appealing to consistency (e.g., from passive to active euthanasia); (2) the argument from respect for autonomy; (3) appeals to justice; and (4) the argument from interests (mercy or relief of suffering). I will argue that only the last is directly relevant to active euthanasia as a medical intervention, though arguments together from autonomy and justice can in practice (through the backdoor) provide a ground for voluntary active medical euthanasia (AME).

3.1.1 Consistency Arguments appealing to consistency have the following form: Premise 1. P is acceptable practice. Premise 2. Q is not different in a morally relevant way to P. So, Q is acceptable practice. Philosophers should be sceptical of this argument in the context of euthanasia. It can be an example of begging the question. The first premise assumes that current practice is justifiable. The argument, though valid, is not sound. Consider the most common examples. Morphine. Morphine is already used in doses in which it is foreseen that death will result (often invoking the doctrine of double effect). This is acceptable when the stated intention is to relieve pain. So, it is argued, it is acceptable to use other drugs which shorten life. However, this does not settle when, if ever, it is acceptable to use morphine (and other drugs) to shorten life. For example, morphine and sedatives have been used in doses which have been alleged to have shortened the lives of infants with Down syndrome and spina bifida (Gillon 1985; Kuhse 1992). This does not imply that euthanasia is justifiable in these cases. The same applies to quasieuthanasia practices involving sedatives.

The commonest area in which this argument from consistency is used is to move from passive to active euthanasia (Rachels 1975). Withdrawing life-prolonging treatment (passive euthanasia) is widely accepted and practised. One example is the case of Tony Bland (Hope et al. 2003). A person can be taken off a ventilator which provides support for his breathing and no crime is committed. Although a doctor turns off the switch, nature is said to have taken her course. But a doctor cannot administer an injection for the sole purpose of ending life. The slow death after treatment is withdrawn may cause great suffering for the patient and family. Given that a decision to end life has been taken, surely it is more humane to end life quickly, though actively. There are at least 2 ways in which this argument is problematic. Firstly, and again, whether passive euthanasia is justified turns on whether there is good normative reason for it. Secondly, whether active euthanasia is justifiable in the stated form of

3 Autonomy, Interests, Justice and Active Medical Euthanasia

33

the argument turns on whether it relieves suffering, or more specifically, produces more good than harm. While I do not believe there is an intrinsic moral difference between acts and omissions, there is one way in which withholding or withdrawing treatment can be relevantly different from administering a lethal injection. Treatment may legitimately be limited (withheld or withdrawn) for reasons of distributive justice (Wilkinson and Savulescu 2011) that would not apply directly to medical euthanasia. There is a great moral difference between distributing a scarce resource like intensive care or artificial feeding between this patient and another, and administering a lethal injection. Justice may require that, since the benefit, need or entitlement of A is greater than B, that A gets the treatment. Withholding the treatment from B may result in her death, but it is not the intentional killing of B for B’s own sake.1 This is brought out quite clearly by using a counterfactual: if more resources were available (or B could provide these herself) and B wanted treatment, would B be treated? I will return to distributive justice presently. It might be argued that passive euthanasia is justified enough—so active euthanasia is permissible in those circumstances.2 In circumstances in which passive euthanasia is justified, then those justifications will be on the grounds of either being in the best interests of the patient or being as a result of an autonomous refusal of medical treatment. I consider these grounds in the following section and whether they apply to active euthanasia.

3.1.2 Respect for Autonomy The commonest justification of active medical euthanasia is respect for individual autonomy. Battin summarises this argument in this way: “one ought to respect a competent person’s choices, where one can do so without undue costs to oneself, where doing so will not violate our moral obligations, and where these choices do not threaten harm to other persons or parties” (Battin 1994, 107). Respect for individual autonomy is historically a political doctrine about the State’s authority over the individual. In particular, according to Mill’s liberalism, two principles, or “maxims”, determine the limits of State interference in individual action:

1

It is in this sense that there is a relevant moral distinction between intended and foreseen effects. If I give the one magic life-saving bullet to A rather than B because A will live longer, or because A has had a rougher ride through life, or the coin fell in A’s favour, I am not saying that I intend B to die. That at least one dies is an inevitable consequence of whatever is done. I am intending to bring about the most just state of affairs. This is quite different from the usual way in which the distinction between intention and foresight is drawn: a doctor administers a lethal dose of morphine intending to relieve suffering but foreseeing that it will kill the patient. In this case, a death is not an inevitable consequence of whatever is done. 2 Thanks to the Editors for raising this objection.

34

J. Savulescu The maxims are, first, that the individual is not accountable to society for his actions, in so far as these concern the interests of no person but himself. Advice, instruction, persuasion, and avoidance by other people if thought necessary by them for their own good, are the only measures by which society can justifiably express its dislike or disapprobation of his conduct. Secondly, that for such actions as are prejudicial to the interests of others, the individual is accountable, and may be subjected either to social or legal punishment, if society is of opinion that the one or the other is requisite for its protection (Mill 1900, 150–151)

In relation to end of life, there are two distinct issues: · how far and when the State is entitled to interfere in an individual’s life · what the State should provide to individuals for them to live their lives. Respect for autonomy requires that we should not interfere in people’s choices, insofar as these choices affect only their own lives. This is true, as Mill emphasised, even when those choices are clearly prejudicial to that person’s interests. Respect for autonomy requires not interfering with an autonomous desire to commit suicide. How should the State respond when a person requests treatment or an intervention which he believes to be best for himself? Insofar as the treatment is a publicly funded intervention that is a part of a State sponsored medical service, the provision of that treatment necessarily affects others by consuming a public resource. The action passes outside the self-affecting realm. The principles which determine how these services are distributed, and indeed whether they should be provided at all, should be determined by the principles of distributive justice. The situation is different if the individual is requesting assistance in dying (either euthanasia or assisted suicide) which does not affect others through the consumption of public resources—either purely private medicine or other related private services. In countries with a national health service, the delivery of health services is a public service distributing public goods. In these countries, the aim of medicine is to offer medical treatment which there is good reason to believe (usually based on empirical evidence) provides more good than harm, and is better than the alternatives. The relevant outcomes for evaluating whether a treatment produces more good than harm are those based on community values and, I believe, those which themselves reflect ultimately some objective conception of human well-being. Let’s assume that we give up this conception of medicine and adopt a conception of medicine as respecting competent people’s choices. Consider the following cases: · Futile interventions. Ann has breast cancer. Her oncologist recommends surgery plus chemotherapy. She explains the reasons for this, based on existing evidence. Ann requests herbal therapy. · The money or the box. Mary has breast cancer. Her oncologist recommends surgery plus chemotherapy. She asks how much such treatment will cost. Her oncologist replies: “In the region of $100,000”. She says, “I’ve had enough of life as it is. I don’t want to live any longer. I’d prefer to take the money and have a good holiday.” · Harmful interventions. Jim asks a surgeon to remove his penis because he believes it is the source of his repeated sinning.

3 Autonomy, Interests, Justice and Active Medical Euthanasia

35

If medicine is about respecting the autonomous choices of people for their own health, then it should accede to each of these requests. Some libertarians might defend such a conception of medicine. Most people, however, would find acceding to such requests alien to the telos or goals of medicine. If respect for personal autonomy is the basis for providing euthanasia, then a person could request euthanasia for any reason, or no reason at all. Such a person need not be dying, or in pain, or even ill at all. One could autonomously desire to die simply because one was tired of life, even if that life was happy and fulfilling. But that is not a good reason for medicine to provide a euthanasia service. We do not believe that medicine should be offering any intervention which people request, even if it is related to their health, or their own conception of well-being. There is widespread agreement that doctors should not offer genital mutilating surgery, even if patients request it autonomously, and even if they believe it is in their interests to have such surgery. That is simply an intervention which there is no good reason to believe is of value to people. Should we give greater weight to personal autonomy? Is not medicine committed to giving some weight to personal autonomy, even at the expense of a person’s interests, such as when a person is told the truth about their medical diagnosis even when this will seriously affect their well-being? Here, I think there are two responses. Such arguments often employ a narrow conception of well-being. When we tell the truth, I think in many circumstances we are doing what is best for that patient, even if it makes the patient depressed or anxious. But even if such cases are really examples of promoting autonomy against a person’s interests, it still remains to be shown that they are relevant to the case where a person desires what is arguably the greatest harm of all—death. Finally, truth telling is not a question of the allocation of a scarce public good (apart from the physician’s time). The conclusion that AME cannot be justified in terms of respect for autonomy is rather startling as it flies in the face of much of the movement towards greater respect for patient autonomy. Such enthusiasm for autonomy, I have claimed, stems from a confusion between respect for autonomy as a political doctrine and respect for autonomy as a ground for distributive justice. One might object3 that there are some objective conceptions of the value of death that ground a right to euthanasia. Imagine a community that endorsed the idea of active ending of life as a component of medical care, and saw a particular type of death (peaceful, neat, free from suffering, in a manner and timing of the individual’s choosing) as intrinsically valuable—indeed objectively valuable. An analogy would be a good ending to a novel. In such a society, autonomy would provide a grounds for AME. There is such a society—the Netherlands. Accordingly, there is a strong autonomy argument for AME in the Netherlands. If this argument holds, then the question of whether or not there is an autonomy-based right to AME turns on the socially contingent issue of majority support for AME, and on the philosophical question of the objective value of death in a particular form.

3

This is Dominic Wilkinson’s objection.

36

J. Savulescu

This is in part an interests-based argument for AME—that certain kinds of dying are good for a person. As I will argue in the section on interests, it is hard to see how this kind of dying is superior to unconsciousness, that is, deep palliative sedation, from the perspective of the individual’s own interests.

3.1.3 Justice Active Medical Euthanasia to Bring About Justice. The relationship between distributive justice and euthanasia is a complex but important one. Battin puts one argument from justice in this way. When resources are limited, “it is better to deny [treatment] to those people who are … medically unsavalageable and will die soon anyway: the terminally ill, the extremely aged, and the seriously defective neonate” (Battin 1994, 114). Battin (1994, 115) extends the argument to active euthanasia. On this view, euthanasia should be available to bring about a just state of affairs. On the face of it, this justification for euthanasia appears invalid. Euthanasia to bring about a just state of affairs is not euthanasia at all. It is important to remember that euthanasia is X killing Y for Y’s benefit. Killing Y for Z’s benefit is not euthanasia, but murder. Similarly, the Nazi program of so-called “euthanasia” was involuntary killing and not for the person’s benefit at all, but to maintain the racial purity of the Volk. Battin admits that it is stretching the term euthanasia to use it to describe killings for the purpose of distributive justice. But this is not merely a terminological question. Distributive justice is about who gets a slice of some finite cake. It is about giving public goods to some, but not others. It is not typically about doing things to people actively, in particular killing them, to bring about a just state of affairs. Thus we do not think that justice requires that we take a kidney from those who are healthy to provide kidneys for those with renal failure. Kidneys are not a public good in this sense. They only become public goods when they are freely donated. In a similar way, a person’s life is not a public good, nor is it something that can be manipulated to bring about more just distribution of public goods. This is, to use the now hackneyed Kantian phrase, to use people as a means to some abstract concept, and not as an end in themselves. Battin argues that euthanasia may save money and thus justice may require it. Distributive justice may certainly require that we offer a cheaper alternative to a person, even though a more expensive but more effective alternative exists. However, it cannot require that we harm someone to save resources. Thus any argument from justice cannot rest solely on the dollar cost or saving of euthanasia, but also on whether it is in a person’s interests. There is one way in which bringing about a just state of affairs might require active euthanasia. Justice may require that we withhold antibiotics or another lifeprolonging treatment from one person, Y, because others would benefit more from scarce resources. The withholding of medical treatment may cause suffering. If so, it would be justifiable to kill Y if killing Y was the best way of relieving Y’s suffering.

3 Autonomy, Interests, Justice and Active Medical Euthanasia

37

This argument thus relies on the argument from interests for any plausibility. Given that Y is not entitled to a scarce public good, and will die as a result, it is more in Y’s interests to be killed than allowed to die. It is really an argument about what is in someone’s interests, given the constraints of scarce resources and the moral imperative to distribute these justly. Justice does not directly or necessarily require that we kill. Consider an example.4 Let’s say you have a country where, on resource grounds cardiac transplantation is not available. It is therefore inevitable that patients with end-stage cardiac failure will die. Would it be justifiable to provide AME to patients with end stage cardiac failure? The available options are: (1) provide no medical treatment, allow patient to suffer until they die naturally; (2) provide AME—ensuring that they do not suffer; or (3) provide palliative care, minimising patient symptoms and suffering until they die naturally. Of these three options, (2) would potentially lead to the least patient suffering, as well as the least cost. If the only options were (1) and (2), we should choose (2). However, we do have option (3), and other things being equal we should choose that as our policy, for two reasons. The first is that while the patient would have no further suffering at all with (2), they would also have no further potential conscious experience or possible benefit. The question is whether the patient’s suffering is able to be diminished sufficiently for them (and/or their family) to appreciate continued life (and without ablating conscious experience). If the only way of treating the patient’s suffering were by rendering them unconscious, the question then becomes, should we perform (2) rather than (3) because it will both guarantee zero suffering and cost less than terminal sedation. I will return to this point. Secondly, while AME would cost less and guarantee absent suffering for patients with terminal illness, the same argument would hold for any patient with a lifelimiting illness (or indeed any patient with an illness). It would almost always be cheaper and suffering could be reduced more effectively. However, that isn’t seen as the goal of medicine. The argument for AME as cheaper than existing treatment proves too much. One last note on euthanasia as justice. Preventing harm to others may sometimes require that we do certain things to individuals (Mill’s harm principle). For example, an individual may be detained and subjected against his will to medical examinations to determine whether he has Lassa fever or Ebola. This is not distributive justice. Moreover, it is not clear when, if ever, it would be justifiable to kill one person to prevent harm to others, short of times of war. The description of such a practice as euthanasia would be thoroughly misleading. Whether a person has a right to active euthanasia as a medical intervention turns on whether that use of public resources can be justified under the principles of distributive justice. Whether a patient should be offered a medical intervention turns on whether that intervention produces a net health benefit to the patient, that is, how much the good produced by the intervention outweighs the harms. That is a question

4

Thanks to Dominic Wilkinson for this example.

38

J. Savulescu

about what is in a person’s interests. The only justification for offering euthanasia as a medical intervention is that dying (sooner) is in a person’s best interests. Distributive Justice and the Entitlement to Active Medical Euthanasia. There is a much more important relationship between distributive justice and active medical euthanasia. If euthanasia can be shown to promote a state of affairs to which a person is entitled, then justice requires that people have access to euthanasia. Consider a parallel: cochlear implants. Deaf people are entitled to claim access to cochlear implants because this intervention is to be able to restore or establish hearing. Since deafness is a disability (Kahane and Savulescu 2009), and restoring hearing is one of the goals of medicine, it is appropriate that deaf people have access to such a treatment. How accessible we make cochlear implants depends on the magnitude of the benefits of such an intervention compared to the health benefits of other available interventions. If euthanasia promotes a state of affairs to which people have a strong entitlement, then justice requires that they be offered euthanasia. When a medical intervention promotes longer life, or less pain, or restores the ability to ambulate, hear or see, or improves the quality of life in other ways, we believe that people have a strong entitlement to it. Euthanasia causes the death of the person. Death is not usually thought to be a benefit for the person who dies. However, if death could be shown to be a benefit for the person who dies, then that person would have a strong claim to euthanasia. But that of course requires showing that euthanasia is in a person’s interests. The justice justification for access to euthanasia again requires the argument for interests. If what is good for people is determined by what people desire, and distributive justice is about providing what people desire, then respect for autonomy would bear upon whether active medical euthanasia should be provided. However, no matter how much Jim wants his penis removed to atone for his sins, we do not believe that this desire should be given any weight in determining the distribution of medical resources. Distributive justice concerns the distribution of primary goods which Rawls defines as goods which every rational man is presumed to want. Primary goods include social primary goods (rights, liberties, powers, opportunities, income and wealth) and natural primary goods (health, vigour, intelligence and imagination) (Rawls 1971, 62). However primary goods are defined, they cannot be defined just in terms of what people do actually now want.

3.1.4 Euthanasia in a Person’s Interests The crucial argument, I have argued, supporting active medical euthanasia is that from interests. According to this argument, euthanasia is justified in those cases

3 Autonomy, Interests, Justice and Active Medical Euthanasia

39

in which it promotes a person’s interests.5 The idea here is that the person is still suffering greatly after other medical treatment and euthanasia is the last resort. In essence, for euthanasia to be a beneficial intervention in these cases, death must be better for the person than continued life. Rachels relates a typical example of Jack, a patient with terminal melanoma. The story is related by a fellow patient. At the prescribed hour, a nurse would give Jack a shot of the synthetic analgesic, and this would control the pain for perhaps two hours or a bit more. Then he would begin to moan, or whimper, very low … Then he would begin to howl, like a dog. When this happened, he would ring for the nurse who would give him some codeine by mouth. …but it never did any real good—it affected him no more than half an aspirin might affect a man who had just broken his arm. Always the nurse would explain as encouragingly as she could that there was not long to go before the next intravenous shot—“Only about 50 min now.” And always poor Jack’s whimpers and howls would become more loud and frequent until the blessed relief came (Rachels 2007, 152–153).

Such cases are meant to show us that at least in some cases, euthanasia is in a person’s interests. Although emotionally very powerful, such examples have serious limitations in guiding any practical discussion of when euthanasia is justified. Firstly, there is the empirical issue of whether this person’s suffering really is unrelievable. In today’s light, the case of Jack would be shocking to many palliative care physicians: Jack was clearly underdosed on narcotic analgesic. The pain he was experiencing was effectively relieved by narcotics, but he was experiencing breakthrough pain. The very sad fact is that Jack need not have suffered any pain at all: he only required larger doses of narcotic, together with rescue doses for breakthrough pain. However, Rachels, writing originally in the 1980’s, remarked of this case: The NIH clinic is, of course, one of the most modern and best equipped hospitals we have. Jack’s suffering was not the result of poor treatment in some backward rural facility; it was the inevitable product of his disease, which medical science was powerless to prevent (Rachels 2007, 153).

Rachels was a top philosopher and one of the foremost writers on euthanasia. In my view, his article “Active and Passive Euthanasia” in 1975 is the best article in medical ethics (Rachels 1975). The fact that he made this remark shows just how difficult it is to determine whether a person’s suffering is relievable or not. The second problem is more fundamental and the one which I will focus on. It is very unclear when death is in a person’s interests. However, without some idea of when death is better than continued existence, it is hard to make an argument for medical euthanasia in terms of the goals of medicine. The usual way of proceeding is on a case by case approach, as Rachels does, pointing to “paradigm cases” in which euthanasia seems justified. There are at least two problems with such an approach. Firstly, some people often do not share the intuition that euthanasia is justified in the case stated. The argument which is often operating in cases like that of Jack is: 5

This is sometimes put as relieving suffering, as providing mercy (Rachels 1975). Battin (1994, 101) puts this as the principle of mercy: one ought to relieve a person’s suffering, when this does not preclude the sufferer attaining some overriding good.

40

J. Savulescu

X is suffering greatly Suffering can be so great as to make death in a person’s interests Therefore, death is in X’s interests. When the argument is spelt out in this way, it is clearly questionable. While X’s suffering is great, it remains open whether it is so great as to make life not worth living. To summarise the argument to this point: (a) Appeals to consistency between passive and active euthanasia may beg the question, in so far as the justification for passive euthanasia is that it is in a person’s interests. (b) The argument from respect for autonomy cannot justify active medical euthanasia, at least not within a public health system. The distribution of medical resources cannot be determined by an individual preference. Respect for autonomy may require that we withhold or withdraw a life-prolonging medical treatment. However, this does not imply that we must provide active euthanasia out of respect for autonomy. Arguments from passive “euthanasia” can only be used to justify active euthanasia when the basis for the passive euthanasia is best interests, and not respect for autonomy. (c) The argument from justice would justify a claim to euthanasia if it could be shown that euthanasia was in a person’s interests. (d) The crucial argument in justifying AME within a public health service is the argument from interests. This question cannot be settled without some conception of a life which is not worth living or a death worth having. This argument has other important implications. If the justification for euthanasia as a medical intervention is interests, non-voluntary euthanasia will be easiest to justify as those who are worst off will often not be competent in virtue of severe neurological impairment, as we shall see. However, the most important implication is that future progress requires explicit argument about which kinds of life are worse than death. In practice, most attempts at formulation of a policy of active euthanasia have required that a person be terminally ill, competent and wanting to die. For example, in the world’s first Act legalising euthanasia, medical euthanasia could be provided under the following conditions. Rights of the Terminally Ill Act 1995, Northern Territory, Australia: A patient can be killed if · over 18 · suffering from an illness which will kill the patient (without the application of extraordinary measures) · no medical measures acceptable to the patient which can cure it · A second medical practitioner confirms the medical diagnosis and prognosis · A psychiatrist has confirmed the patient is not suffering from treatable depression · Both have confirmed the patient:

3 Autonomy, Interests, Justice and Active Medical Euthanasia

41

(a) is suffering (b) has been informed of treatment options, including palliative care, counselling, psychiatric support and extraordinary measures (c) patient has considered the implications for family (d) patient is of sound mind and decision is made freely, voluntarily and after due consideration But many people’s lives with terminal illness are still worth living. Palliative care physicians and others are rightly sceptical that this is enough to provide a good reason to kill someone. This legislation was clearly based on respect for autonomy, not interests. If the debate on medical euthanasia is to go ahead, we need a more systematic, coherent approach on the value of life and death. When is a human being’s existence of equivalent value or less value to non-existence? One way of proceeding is to examine those cases in which a decision to withdraw or withhold treatment has been made on quality of life grounds. Are there any principles which might unify these decisions about actual cases? The possible range of states which seriously detract from life’s value are: 1. 2. 3. 4. 5. 6. 7. 8. 9. 10.

cognitive impairment motor impairment inability to communicate sensory deprivation affective or emotional disorder volitional disturbance memory disturbance disturbance of identity pain6 inability to control bodily functions (incontinence, poor swallowing, etc.).

Now clearly there is an issue of how much of each of these and how many are required to make life no longer worth living. Having surveyed the literature and worked on this for 10 years, I have not been able to identify or provide such an account. One thought might be that it will always be difficult to determine the exact point at which burdens outweigh benefits (the zero point). However, it might be possible to be certain further from zero (i.e. where burdens considerably/clearly outweigh benefits) (cf. Sorites paradox).7 The clearest cases are not the cases of permanent unconsciousness (since in those cases it appears that suffering is zero). The clearest cases are where there is profound

6

I am not sure whether pain is bad in itself, or bad because of the way it affects a life in the other ways. 7 Thanks to Dominic Wilkinson for this point.

42

J. Savulescu

brain injury with absent communication/purposeful movement, but evidence of ongoing suffering, such as patients with severe dementia, pressure sores, or fractures. Attempts have been made along these lines using reference to legal cases. For example, recently Willmott et al. concluded, To illustrate, in the four cases where treatment was withheld or withdrawn, the patients had profound brain injury with no prospect, or very little prospect, of neurological recovery (Slaveski, Melo, Herrington and Messiha). By contrast, the three remaining cases where lifesustaining treatment was commenced or continued involved patients in better neurological states (Northridge, JT and Astill). Although not couched in terms of quality of life, the capacity to engage meaningfully with the world seems to be relevant. It appears that, at least indirectly, judges consider how the proposed treatment will affect quality of life (Willmott et al. 2014).

One condition that was a ground for introducing euthanasia for infants in the Netherlands is epidermolysis bullosa, a genetic condition in which the skin peels off. There is no cure and death ensues a short time after birth after periods of severe pain and infection. In countries in which euthanasia is not allowed, such babies are often allowed to die by withholding or withdrawing life-prolonging medical treatment (however, we have seen that the ground for this might be distributive justice, not that life is not worth living.). Sometimes, there are attempts to apply the subjective standard to incompetent individuals like newborns. Some ask, what would the infant think about such a life, if he were competent. This is a hypothetical subjective standard for evaluating quality of life. Yet it cannot really be subjective because the projected evaluation must be according to certain objective criteria, like the degree of suffering, prospect of recovery, and the impairments listed above impairing in turn the capacity to have meaningful and rewarding social relationships, etc. So evaluation of quality of life and whether life is worth living must be partly objective. What makes life objectively worth living? Happiness and pleasure, and the absence of pain and suffering. Being able to develop talents and achieve worthwhile things, having a rich set of social relationships, understanding and appreciating the world, and being able to have some impact on it, raising children, and so on. Life becomes not worth living when it becomes devoid of most of these, and is dominated by negative mental states, like pain, suffering, isolation, fear and so on. I believe we need objective criteria to evaluate quality of life. But even if we can agree on these, it will remain a difficult question which to my knowledge has not been answered or even hardly addressed, as to when the threshold is crossed as to whether life is not worth living. One example is Trisomy 18. It is associated with severe intellectual disability (more severe than Down syndrome) and most infants die within the first year but a small number live until 10 years of age. Sometimes there are other complex congenital abnormalities. It is often misleadingly described as lethal—it is not universally lethal. Is it a life worth living? Perhaps in this grey zone where it is not clear whether objectively life is worth living or not worth living, we should defer to the judgements of carers such as parents.

3 Autonomy, Interests, Justice and Active Medical Euthanasia

43

There is thus a paradox. Those with the worst lives will have the greatest impairment of their minds but will be least able to articulate judgements of the worth of life. But in these cases it is hardest to draw a line on what actually makes life not worth living.

3.2 One Criterion for AME: Death and What Matters Many people believe that death is a misfortune and a very bad thing for the person who dies. Thus, killing is prima facie wrong for this reason. However, this is a mistake. There are many definitions of death. Whole brain death, brainstem death, death of the whole organism and so on. But nearly everyone is united in thinking that death matters a lot—it is often seen as the worst thing that can befall an organism. But this is mistaken. Here is the argument. It is based on Derek Parfit’s famous argument that personal identity is not what matters (Parfit 1984). Ceasing to exist is at least as bad as death. This may at first seem puzzling as it suggests that people could cease to exist without dying. But there are some ways of ceasing to exist that would not involve dying, in any ordinary sense, and which would have none of the badness of dying. That was Parfit’s claim. Consider an imaginary Star Trek teletransportation case. Imagine nanotechnology and synthetic biology progress. In 2050, organisms can be built precisely molecule by molecule, atom by atom. You can enter a teletransporter in London. Your whole body is scanned by a supercomputer and the atomic arrangement of your body is recorded, but the process of recording destroys the entire organism. Your entire organism is then reconstructed, atom by atom, in New York, milliseconds later. It appears that one minute you were in London, the next in New York. But of course, you ceased to exist when your whole organism was destroyed in London (this is most easily seen in the case of twinning, when the reconstructor mistakenly makes copies both in London and New York—this is the subject of the wonderful Arnold Schwarzenegger film, Sixth Day). Would you enter such a teletransporter, provided it perfectly copied you? I would. But importantly, nothing of moral significance would be lost when you ceased to exist. This example shows that what matters are mental states, not our original physical existence. More precisely, this example shows that what matters is that there will be someone in the future who will be psychologically just like us, even if that person won’t be us since we shall have ceased to exist. That is Parfit’s view, but the imagined case of Teletransportation isn’t enough to show this view to be true. Things are different with Parfit’s imagined case in which someone’s brain is successfully divided and transplanted into the empty skulls of two other similar bodies. The two resulting people here would not be merely psychologically exactly like the original person, but this psychological relation would have its normal cause: the continued existence of enough of the same brain. Of those who are persuaded by this example that personal identity isn’t what matters, many conclude that what matters is psychological continuity, even in a branching form, with its normal cause (Parfit 1984).

44

J. Savulescu

What matters is not biological life and death, but something to do with mental states (or perhaps some embodiments of them). Philosophers are criticised for using such thought experiments—they are said to be mere science fiction of no relevance to every day life. In fact, there is already real life analogue of teletransportation. Early in human development, until 14 days, the embryo can split into identical twins (or higher order identical multiples). Some people, such as leaders in the Catholic Church, claim the embryo is person with a right to life from the moment of conception. For such people, twinning involves the ceasing to exist of a human being and replacement by two clones, like when the teletransporter mistakenly makes two copies, one in London and the other in New York. Consider such a Catholic couple who naturally conceive an embryo, call him Danny. A few days after conception, Danny divides into identical twins: Patrick and David. Patrick and David are clones. Danny ceased to exist when he divided, just an amoeba ceases to exist when it divides. Danny could not be identical with Patrick and David as this would imply that Patrick and David are identical with each other, that is, they are the very same thing. This is clearly false. These examples show that what matters is not death, but loss of psychological continuity and connectedness. This has implications for ending life. It implies that when a human organism does not have mental states, it is not wrong to kill it. This lends support to the practice of withdrawing life prolonging interventions from people who are permanently unconscious (Sinnott-Armstrong and Miller 2013), early abortion and destruction of embryos. Thus active medical euthanasia would be justified in cases of severe cognitive impairment or permanent unconsciousness, where there is the absence of psychological continuity and connectedness. This would apply to advanced dementia. In such cases life would be of little or no value (indeed, it could be worse, if the organism experienced pain or other negative mental states).

3.3 Voluntary Palliated Starvation and AME Through the Ethical Backdoor We have seen that if euthanasia is to be practised by doctors as a part of medicine, it must be in the interests of patients. This requires determining that life is no longer objectively worth living. Even in those whose lives are worst, such as those with profound cognitive impairment, it is difficult to draw the line of when life is no longer worth living. However, when psychological continuity and connectedness are absent, it would not be bad to cease to exist. In such cases, euthanasia could be performed for reasons of relief of suffering, parental request, etc. However, there is another way in which AME could be ethically required of doctors. I have argued that respect for autonomy is insufficient to ground AME.

3 Autonomy, Interests, Justice and Active Medical Euthanasia

45

However, it is well established that people have the right to refuse life-prolonging medical treatment. Examples of respect for autonomy in health care are: 1. when we respect a competent person’s informed refusal of medical treatment judged to be in her interests. An example is a Jehovah’s Witness refusing a blood transfusion. 2. when we honour a living will of a now incompetent person refusing medical treatment. An example is an unconscious Jehovah’s Witness with a card refusing blood transfusions. 3. when a substituted judgement procedure is used to predict which medical treatments a previously competent but now incompetent person would refuse. An example is an unconscious bleeding Jehovah’s Witness whom it is decided would refuse a blood transfusion based on her prior beliefs, even though there has not been an explicit refusal. I have argued that those wishing to die also have the basic right to refuse to eat and drink. In addition, they have a right to palliative care as they die from dehydration and starvation. I called this Voluntary Palliated Starvation (VPS). This does not require any specific legislation and is likely to be legal in most jurisdictions (White et al. 2014; Wilkinson and Savulescu 2014). If people have a right to starve and dehydrate themselves to death based on respect for their autonomy, then this opens the backdoor to AME in the following way. Those availing themselves of VPS will certainly die, usually over a period of days to weeks. Given that they will die, in those cases in which they are heavily sedated and analgesed, there is an argument for AME being superior to palliative care. That argument is based on distributive justice—it saves resources for patients who will die, to die earlier rather than later. This applies to terminal illnesses. If a person will die and they request to die sooner rather than later, then distributive justice speaks in favour of AME rather than palliative care. This flies in the face of my previous argument that justice must be about providing treatments which are in the interests of patients. But in fact, that argument requires qualification. Patients need not be provided with medical care that is in their interests when that care would bring about greater benefits to others (or at least others in equal need). While I have said that judgements about when life is worth living are difficult to make, it is clear that relative judgements of value of life between patients must be made and are made. For example, a longer life is better than a shorter life. It is better to be without pain than in pain (Wilkinson and Savulescu 2014). Thus relative value of life judgements are more tractable than absolute judgements of when a person’s life is no longer worth living. Thus a person might be denied medical care because it is very expensive, would produce only a very small benefit or only has a very small chance of working. These kinds of judgements are in fact the basis of determinations of “futility” (Wilkinson and Savulescu 2011). We need not claim such interventions are not in the interests of patients but only that they are cost-ineffective. Thus we need not show that AME is in the interests of patients, but only that it is more cost-effective than other interventions. Given that resources could be saved

46

J. Savulescu

in the dying phase, and the patient wishes autonomously to die, both respect for autonomy and distributive justice speak in favour of AME. Since respect for autonomy applies to decisions about future health states when the person will be incompetent—advance directives or living wills—a person could refuse food and fluids in advance if, for example, she had advanced dementia. Since the person has that legal right, then she could equally request AME if she ever developed advanced dementia. This is not to imply that AME must be performed when a patient will die but only that the patient has the right to it as an alternative, when he or she wants it and when he or she will die, regardless of whether that is by disease, limitation of treatment for justice reasons or because of a valid refusal of treatment. Palliative care should be an option, provided it is sanctioned by principles of justice. Imagine that I am diagnosed with a progressive neurological disorder that will lead to death in 6 months, but which is currently minimally symptomatic. If I choose VPS, I wouldn’t need analgesia. I might become hungry and thirsty, but I could be prescribed appetite suppressants and medication to diminish thirst. If I am depressed by my condition, palliative care would demand that I am given anti-depressants. Perhaps I am given some combination of opiates, amphetamines and mood-altering agents, to minimise sensation of thirst and engender a feeling of general euphoria.8 It isn’t clear in such a case that AME would be superior to drug-induced wellbeing. Nonetheless, both should be options open to the patient.

3.4 Conclusion Respect for autonomy does not ground a right to AME in the simple way many people have argued. However, respect for autonomy does ground a right, now and in advance, to refuse to eat and drink. And given that a person will die (whether by disease, limitation of treatment on distributive justice grounds or legal refusal of food and fluid), justice requires that AME be an option for that person. The combination of respect for autonomy together with distributive justice provides a justification for AME. Thus, we can share the scepticism of opponents of AME about making quality of life judgements and decisions about when life is not worth living. We can in fact eschew such judgements. And we can share their belief that the role of medicine should be to serve patients’ interests. But that principle should extend to all patients, including those competing for limited resources. Because AME saves resources that can be used by patients who need them, distributive justice requires AME, at least for those who will die and request it.

8

Thanks to Dominic Wilkinson for this example.

3 Autonomy, Interests, Justice and Active Medical Euthanasia

47

Respect for autonomy, together with justice, speak together in favour of voluntary AME. What of nonvoluntary AME? This is the most controversial practice. Considerations of autonomy do not apply. For example, neonatal euthanasia is practised in the Netherlands (Verhagen 2013) for conditions like epidermolysis bullosa. I have argued that it is difficult to argue that such practises are straightforwardly in the interests of patients, that is, that their lives are not worth living. However, I have also argued that justice can speak in favour of more cost-effective alternatives, when a patient will die. Elsewhere, I have argued with Dominic Wilkinson for Organ Donation Euthanasia (Wilkinson and Savulescu 2012). We argued that organs could be extracted to save the lives of others if it was certain a person would die, for example, through the limitation of life prolonging medical treatment. This would be a case of altruistic euthanasia. In a similar way, euthanasia could be provided on justice grounds in those incompetent patients who would certainly die. Their deaths could be caused either by their disease or by limitation of life prolonging medical treatment, including the withholding or withdrawing of artificial nutrition. In such cases, nonvoluntary AME would save resources over a slower death. In addition, parental interests might support euthanasia. Moreover, considerations of lack of psychological continuity and connectedness in cases of unconsciousness or near-unconsciousness would reduce or eliminate the wrongness of killing in such incompetent patients. This is not to say that AME should be provided whenever palliative care is an option. There need to be good reasons for AME. Many cases will be grey and perhaps in this case it should not be used. But some cases will be clear. So if a patient is going to die soon (say in days to a couple of weeks), and the patient is unconscious, or has minimal consciousness, the case for AME is strongest. In short, in cases in which a human being will certainly die, whether by disease, injury, their decision or the legitimate decisions of others, euthanasia offers a more just way of dying when that human being would otherwise require palliative or other care during dying that imposes costs indirectly on others. Distributive justice may in this way justify both voluntary and nonvoluntary AME. But of course whether AME should be offered, or is the best option, will depend on weighing all the reasons in particular contexts.

References Battin, Margaret Pabst. 1994. The least worst death. New York: Oxford University Press. Gillon, Raanan. 1985. Philosophical medical ethics. Rights. British Medical Journal 290: 1890– 1891. Hope, Tony, Julian Savulescu, and Judith Hendrick. 2003. Medical ethics and law: The core curriculum. London: Churchill Livingstone. Kahane, Guy, and Julian Savulescu. 2009. The welfarist account of disability. In Disability and disadvantage, ed. Adam Cureton and Kimberley Brownlee, 14–53. Oxford: Oxford University Press.

48

J. Savulescu

Kuhse, Helga. 1992. Quality of life and the death of “Baby M”: A report from Australia. Bioethics 6: 233–250. Mill, J.S. 1900. Principles of political economy. New York: P. F. Collier and Sons. Parfit, Derek. 1984. Reasons and persons. Oxford: Clarendon Press. Rachels, James. 1975. Active and passive euthanasia. New England Journal of Medicine 292: 78–80. Rachels, James. 2007. The morality of euthanasia. In The right thing to do, ed. James Rachels and Stuart Rachels, 151–155. New York: McGraw Hill. Rawls, John. 1971. A theory of justice. Cambridge, MA: Harvard University Press. Savulescu, Julian. 2014. A simple solution to the puzzles of end of life? Voluntary palliated starvation. Journal of Medical Ethics 40: 110–113. Sinnott-Armstrong, Walter F., and Franklin G. Miller. 2013. What makes killing wrong? Journal of Medical Ethics 39: 3–7. Verhagen, A.A. Eduard. 2013. The groningen protocol for newborn euthanasia: Which way did the slippery slope tilt? Journal of Medical Ethics 39: 293–295. White, Ben, Willmott Lindy, and Julian Savulescu. 2014. Voluntary palliated starvation: A lawful and ethical way to die? Journal of Law and Medicine 22: 376–386. Wilkinson, Dominic, and Julian Savulescu. 2011. Knowing when to stop: Futility in the ICU. Current Opinion in Anaesthesiology 24: 160–165. Wilkinson, Dominic J.C., and Julian Savulescu. 2012. Should we allow organ donation euthanasia? Alternatives for maximizing the number and quality of organs for transplantation. Bioethics 26: 32–48. Wilkinson, Dominic J.C., and Julian Savulescu. 2014. Disability, discrimination and death: Is it justified to ration life saving treatment for disabled newborn infants? Monash Bioethics Review 32: 43–62. Willmott, Lindy, Ben White, Malcolm K. Smith, and Dominic J.C. Wilkinson. 2014. Withholding and withdrawing life-sustaining treatment in a patient’s best interests: Australian judicial deliberations. Medical Journal of Australia 201: 545–547.

Julian Savulescu is Uehiro Professor of Practical Ethics at the University of Oxford, Fellow of St Cross College, Oxford, Director of the Oxford Uehiro Centre for Practical Ethics, Sir Louis Matheson Distinguished Visiting Professor at Monash University, and Head of the Melbourne– Oxford Stem Cell Collaboration. He is the editor of the Journal of Medical Ethics. In addition to his background in applied ethics and philosophy, he also has a background in medicine. His areas of research include: the ethics of genetics, especially predictive genetic testing, preimplantation genetic diagnosis, prenatal testing, behavioral genetics, genetic enhancement, gene therapy, and research ethics (especially ethics of embryo research, including embryonic stem cell research), new forms of reproduction (including cloning and assisted reproduction), medical ethics (including end of life decision-making, resource allocation, consent, confidentiality, decisionmaking involving incompetent people, and other areas), sports ethics, and the analytic philosophical basis of practical ethics.

Chapter 4

Mental Illness, Lack of Autonomy, and Physician-Assisted Death Jukka Varelius

Abstract In this chapter, I consider the idea that physician-assisted death might come into question in the cases of psychiatric patients who are incapable of making autonomous choices about ending their lives. I maintain that the main arguments for physician-assisted death found in recent medical ethical literature support physicianassisted death in some of those cases. After assessing several possible criticisms of what I have argued, I conclude that the idea that physician-assisted death can be acceptable in some cases of psychiatric patients who lack autonomy ought to be taken into account in assessing the moral and legal acceptability of physician-assisted death. Keywords Severe mental illness · Assisted suicide and euthanasia · Physical illness · Mental distress · Autonomous choice

4.1 Introduction Charlie spends most of his time desperately fighting and escaping tormentors that others cannot see. This has been going on for years. Because of his predicament, Charlie is suffering severely and has repeatedly attempted to kill himself. As he is convinced that the personnel of the mental institution in which he now resides are also conspiring against him, they fail to have meaningful contact with him. Yet, following the common psychiatric goal of suicide prevention, the staff have done their best to stop Charlie from killing himself. Especially as the suicide methods to which he has resorted have often, but not always, been as distorted as his conception of reality, this has not been difficult. However, since his distress is clearly very intense and his condition is deemed incurable, some of the mental health care providers treating Charlie have recently started to wonder whether they should help him to end his life rather than try to prevent him from killing himself. J. Varelius (B) Department of Philosophy, Contemporary History, and Political Science, University of Turku, Turku, Finland e-mail: [email protected] © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 M. Cholbi and J. Varelius (eds.), New Directions in the Ethics of Assisted Suicide and Euthanasia, The International Library of Bioethics 103, https://doi.org/10.1007/978-3-031-25315-7_4

49

50

J. Varelius

The question whether a health care provider sometimes ought to end her patient’s life or to assist him to end it himself has recently attracted increasing public attention worldwide. The academic discussion on the topic has mostly focused on patients who suffer from physical illnesses or injuries but recently some ethicists have proposed that mental illness could sometimes provide grounds for physician-assisted death (see, e.g., Appel 2007; Cholbi 2013a, b; Hewitt 2010a, b, 2013; Parker 2013).1 Yet proponents of the idea typically stress that physician-assisted death could be morally acceptable only in the cases of competent patients, patients who are able to make autonomous decisions about ending their lives. Given the seriousness of Charlie’s condition, this requirement would rule him out as a candidate for physician-assisted death (see below). However, as just suggested (see also Burgess and Hawton 1998, 121; Hardcastle and Stewart 2002, 432–433), it could be taken that physician-assisted death might sometimes come into question even when a suicidal mentally ill person lacks decision-making ability. In this chapter, I consider the idea that psychiatric patients lacking autonomy could be candidates for physician-assisted death in light of the main arguments for physician-assisted death presented in recent medical ethical literature. First I briefly characterize the starting points I have in doing that. On the basis of assessing the role of patient autonomy in making end-of-life decisions and the moral importance of non-autonomous mental suffering in the end-of-life context, I propose that the central arguments in favor of physician-assisted death advocate allowing the pertinent practices in cases of patients such as Charlie. Then I consider several possible objections to this suggestion. I conclude by briefly clarifying what I have and what I have not argued.

4.2 The Main Points of Departure In today’s medical ethics, it is commonly accepted that an autonomous patient can refuse and withdraw from (even) vital treatment. Accordingly, the debate on the moral acceptability of physician-assisted death focuses on cases in which a health care provider not merely refrains from or stops providing life-sustaining treatment to her patient, but takes what is considered a more active role in bringing about her patient’s death. In what is called active voluntary euthanasia, this role typically consists of administering a lethal medication to a patient. In the procedure named physicianassisted suicide, a health care provider supplies her patient with appropriate means of

1

According to The Royal Dutch Association of Medicine—which allows that physical illnesses and injuries can provide grounds for physician-assisted death—suicide based on psychiatric reasons should not be treated differently from assisted suicide in medicine generally. In Belgium, also, suffering unrelated to physical illness or injury is acknowledged in law as a valid basis for physicianassisted death (see The Royal Dutch Medical Association 2011, 21–22 and, e.g., Naudts et al. 2006). On the legal status of physician-assisted death for patients suffering from physical illnesses or injuries see, e.g., the chapter by Jocelyn Downie and Georgia Lloyd-Smith in this volume.

4 Mental Illness, Lack of Autonomy, and Physician-Assisted Death

51

ending his life by himself. I refer here to both of these procedures with the expression ‘physician-assisted death.’ The main arguments for physician-assisted death found in recent medical ethical literature refer to respect for patient autonomy and the relief of suffering. It is maintained, roughly, that, within the limits resulting from granting a similar right to others, an autonomous person has the right to live his life in accordance with his own view of how it should proceed. And this right is seen to entail that, when he autonomously wants that, a person is allowed to end his life and entitled to receive assistance from health care providers in doing that (cf., e.g., Beauchamp 1993, 101, 2006). Autonomy is thus here typically perceived in procedural terms, as, again roughly, deciding and acting freely, intentionally, and with sufficient understanding about what one is doing (see, e.g., Beauchamp and Childress 2009, 99 ff.). This allows individuals to have quite differing conceptions about how their lives are to go and to end. Below I call this argument for physician-assisted death the argument from autonomy. In terms of suffering, proponents of physician-assisted death (but not only them) maintain that a person should not have to experience more distress than he can bear. Typically, only distress caused by severe physical illness or injury is seen to provide grounds for physician-assisted death. Yet, as was also already stated, some authors accept that mental illness can suffice, and sometimes also suffering unrelated to any illness or injury is deemed a legitimate ground for physician-assisted death (see, e.g., Wijsbek 2012 and also footnote 1). Whatever the nature of the suffering deemed relevant, it is commonly required that the distress that can warrant physicianassisted death must be enduring and unavoidable in the sense that there is no way of adequately alleviating it other than by ending the suffering patient’s life. Sometimes it is also maintained that only distress caused by a terminal illness or injury can provide grounds for physician-assisted death. Below I refer to the position that the alleviation of intolerable distress provides grounds for physician-assisted death as the argument from relief of suffering. As already mentioned, authors writing on the moral acceptability of physicianassisted death usually emphasize that only a competent person can present a morally authoritative request to die. According to what has been called the standard analysis of the notion of competence, a competent patient has the capacity to understand, is able to reason and deliberate, possesses a set of interests and concerns relevant to the decision she is to make, and is able to communicate her decision (see, e.g., Appelbaum 2007, 1836; Beauchamp and Childress 2009, 114; Buchanan and Brock 1989, 23; Stewart et al. 2011). When a patient is unable to make autonomous choices about her treatment, the health care providers attending her must resort to surrogate decision-making methods. The main types of these procedures refer to the will that the patient possibly expressed when she was competent, the surrogates’ approximation of what she would now want if she had decision-making ability, and to what is seen to be in the patient’s best interests (see, e.g., Buchanan and Brock 1989). Charlie lacks insight into his illness, he is unable to understand what is really happening around and to him, and his reasoning is often quite illogical. Therefore, I

52

J. Varelius

take it that he lacks decision-making ability in the sense described above.2 In what has been referred to as the orthodox psychiatric view on suicide, because the desire to kill oneself nearly always results from mental illness, “it is always necessary to intervene in suicide attempts whenever possible,” with the aim of preventing patients from killing themselves (see, e.g., Hewitt 2013, 361; Fairbairn 1995, 28–29; cf., e.g., Szasz 2011; Callaghan et al. 2013, see also, e.g., Appel 2012). Accordingly, as in Charlie’s case so far, the surrogate decision-makers deciding on the behalf of a suicidal psychiatric patient most likely aim to stop the patient from ending her life. Below, I turn to assessing whether the argument from autonomy and the argument from relief of suffering provide reason for questioning this policy in the cases of patients such as Charlie.

4.3 The Relevance of Charlie’s Lack of Autonomy Do Charlie’s lack of competence and his consequent inability to make an autonomous decision about ending his life mean that physician-assisted death cannot be morally justified in his case? Asking this leads to the question: Why is patient autonomy valued in health care? To assess that question, let us consider the case of Ben. Ben falls seriously physically ill and, consequently, seeks medical care. The health care providers treating him are obligated to request his autonomous consent for the therapies they offer him. If the way in which Ben then exercises his autonomy does not enable the health care providers to determine whether or not giving him the therapies would accord with how he wishes his life to proceed, Ben has failed to engage successfully in the main procedure protecting patient autonomy in current health care, medical informed consent. The central reason for a patient’s autonomy being given a prominent role in health care is thus to ensure that the treatments which the patient receives do not conflict with her own view of how her life should go. Consider that Ben’s health care providers are to determine whether he ought to have a certain surgical procedure. The operation is a quite complicated one. If it is successful, Ben will regain some of the abilities he now lacks because of his illness. But the procedure cannot restore Ben to full health. The probability that the procedure goes well in Ben’s case is approximately 65%. If it does not go well, Ben may suffer from complications that could significantly worsen his condition. Now, as it depends on his evaluation of the life he could have if the operation is successful as compared 2

Though it has often been considered clear that mentally ill patients lack decision-making ability, it has also recently been emphasized that even a severe mental disorder need not preclude at least periodical competence (see, e.g., Hewitt 2010a). Accordingly, it may be that psychiatric patients who are altogether unable to autonomously decide about their treatment are rarer than has commonly been supposed. Yet that does not preclude the possibility that patients such as Charlie exist (cf. also, e.g., Hardcastle and Stewart 2002, 431–432) nor makes them morally unimportant. Also, the possibility that a patient such as Charlie may be able to make some choices autonomously—such as choosing between tea and coffee at breakfast—does not entail that he is autonomous in the sense that he can make an autonomous choice about ending his life.

4 Mental Illness, Lack of Autonomy, and Physician-Assisted Death

53

to his current existence and on his assessment of whether the risk is worth taking, determining whether Ben ought to have the procedure appears quite difficult without recourse to his autonomous decision. Yet the case of Charlie appears significantly different. According to the procedural understanding of autonomy referred to here, even the decision to spend an indeterminable period in intolerable and pointless agony can, in principle, be autonomous. But it would seem clear that, rather than agreeing to live such a life, an autonomous person would typically do her best to avoid it. Indeed, a person’s willingness to lead such an existence very strongly suggests that she lacks the kind of understanding about its nature and consequences that autonomy presupposes. Accordingly, there is also reason to think that, were he autonomous, Charlie would be very much against living the kind of life he now must endure. This suggests that the fact that Charlie is unable to make an autonomous decision about ending his life does not disqualify him as a candidate for physician-assisted death. But does the fact that Charlie is not autonomous mean that his suffering is ultimately not of real moral significance? The view that non-autonomous distress— which besides the suffering of people such as Charlie includes the anguish of small children, the demented elderly, and non-human animals—would not really matter morally appears quite counter-intuitive. But could there still be something to it? Consider the case of Mary. After her husband went on a cruise with their children, some evil-minded people told her that the ship they were on sank and all of its passengers drowned. Consequently, Mary is devastated. As her feelings are not based on adequate understanding about what has actually happened, they are not autonomous. If Mary was informed in the way autonomy as it is here conceived presupposes, she would not be suffering. This, someone might take it, demonstrates that a person’s suffering has real moral significance only if it is autonomous. However, if she is not provided with adequate information, Mary’s subjective evaluation of her existence does not differ from what it would be if her family really had drowned. In other words, as long as she remains in her uninformed state, Mary’s distress is quite real to her. Charlie differs from Mary, among other things, in that he cannot grasp that he lacks adequate understanding about his situation. Therefore, even if he were provided with all possible evidence about what he really does and what actually happens to and around him, it would not affect how he experiences his life. He would just consider the provision of the information as a further stage in the plots against him. Hence, the possibility of successfully informing Charlie is not open. But, just as Mary—in her uninformed state—would continue to be very distressed, in Charlie’s view his suffering is quite real. Indeed, that he is unable to acquire insight into his condition means that his experiential world is more significant from his own viewpoint than it would be if he had the ability to understand his situation. If he understood that the tormentors he believes to be persistently pursuing him do not exist in the external reality, his suffering would not be as intense as it now is. And that it is unbearable from his point of view is the basis of the moral significance of his distress. Even though the viewpoint is uninformed, it is the only perspective available to Charlie, the one he most plausibly must make do with from day to day, as long as his life continues (cf. Sects. 4.6.3 and 4.6.4). Accordingly, though Mary’s wellbeing

54

J. Varelius

should credibly be ultimately assessed in terms of her autonomous views on her life— the ones she would have after being informed,—the fact that Charlie lacks autonomy does not undermine the moral relevance of his distress. Indeed, a significant moral reason for Mary to be provided with adequate information about what has actually happened is that she is suffering without it. That it is non-autonomous does not undermine the moral significance of suffering in her case either. Of course, there are philosophers who deny that what is good and bad for a person is to be determined by reference to her subjective states. According to proponents of the so-called objective theories of human good, whether an individual fares well or ill is to be defined in terms of, for instance, the extent to which she exemplifies ideals such as rationality and virtue. Some authors maintain that things that affect a person’s interests need not enter her experience at all (cf., e.g., Ferkany 2012; Lauinger 2013; Sarch 2011; Tiberius 2007). Yet these kinds of considerations clearly do not undermine the moral relevance of Charlie’s suffering. Because of the severity of his illness, Charlie is not capable of exemplifying ideals such as rationality or moral virtue to any significant extent. And even if he had interests that could be affected by events that never enter his experience, it would still be quite counter-intuitive to maintain that his life is going well for him. Indeed, if a theory of wellbeing entails that a person such as Charlie is doing fine, that is a reason to reject the theory rather than to accept the implication. If (1) patients’ autonomous choices are valued in health care because of their role in determining what patients would really want, (2) there is good reason to believe that Charlie would reject the kind of life he now leads were he autonomous, and (3) his lack of autonomy does not compromise the moral relevance of his suffering, it would seem that the fact that Charlie is not autonomous does not entail that physician-assisted death cannot come into question in his case. It appears clear that he is suffering in a morally relevant sense and that continuing to live the life that he is leading at present is not what he really wants or would want were he autonomous. But does the fact that Charlie’s suffering is of the mental kind disqualify him as a candidate for physician-assisted death? The above remarks on his distress suggest that it does not but let us consider the question in more detail.

4.4 Mental Suffering in the End-of-Life Context In the current medical ethical literature, the position that mental distress—suffering that is not directly caused by physical illness or injury (see also below)—can justify assisting a patient to end her life is a marginal one. Among those who do not reject physician-assisted death altogether, the main reason against the position would appear to relate to the perceived subjectivity of mental distress (see, e.g., Gill 2009, 31). It is taken that, because of this subjectivity, mental suffering is less real than distress directly caused by physical illness or injury—henceforth physical suffering—and, therefore, mental suffering does not have the same moral relevance as physical distress in the end-of-life context.

4 Mental Illness, Lack of Autonomy, and Physician-Assisted Death

55

However, the realization in the 1950s that some mental illnesses can be treated with medication gave rise to a new research paradigm within which mental disorders came to be understood as distinctive neurobiological entities. The position that mental illnesses have a neurobiological basis has recently derived further support from pertinent genetic research and studies using novel neuroimaging techniques (see, e.g., Tsou 2012; Cross-Disorder Group of the Psychiatric Genomics Consortium 2013). Though the research into the neurobiological grounds of mental disorders is still at a fairly early stage, the results of the studies conducted so far suggest that, in terms of their basis, many mental illnesses can be as objective as physical trauma. And those who think that mental suffering is subjective and not as real as physical distress can, and would indeed often appear to do, accept that the mental distress resulting directly from severe mental illness is not as subjective as the mental suffering experienced by mentally healthy people. Accordingly, the view that mental suffering is subjective in a way that physical distress is not would rather appear to relate to the mental distress experienced by mentally healthy persons than to the mental suffering of patients with severe psychiatric disorders. Moreover, as has been pointed out, people have different pain thresholds. Accordingly, individuals can react differently to a similar physical trauma. Their response can vary with factors such as their personal histories and their mental state while undergoing the trauma. It also seems that individuals are not always similarly affected by what—as far as is known—are similar experiences of physical pain. Even severe physical pain need not cause suffering: at least those who consider it a part of something they see worthwhile may not suffer as a result of undergoing it. And what is called chronic pain is known to sometimes outlast the physical trauma that is seen as its initial cause. (cf., e.g., Cholbi 2013b; Corns 2014; Niv and Marshall 2004; Hardy 2005; Saariaho et al. 2012) This suggests that physical suffering too is subjective, not only in the sense that it is experienced from the first person viewpoint, but also in that its causes partly depend on factors that can vary from one person to another.3 In light of empirical studies, mental suffering also plays an important role in connection with the requests for assistance in dying made by physically ill or injured patients. A major reason for their wish to end their existence, when that is what they desire, is often the mental distress related to the effects that their illness or injury has on their lives, effects such as loss of autonomy, inability to engage in what they see as meaningful activities, indignity, hopelessness, and pointlessness. The studies also suggest that, because it usually is more continuous than physical suffering, mental suffering is often considered worse than physical distress (see, e.g., The Oregon Department of Human Services 2013; Dees et al. 2011). And, as demonstrated by physically healthy suicides, mental distress alone can be severe enough to make a person to (want to) end her life. Finally, the argument from autonomy allows a patient to end her life when that is what she autonomously wants. And, as has been 3

Pain is among the topics of the burgeoning neurosciences but at least so far the prospect of completely objective assessment of pain and suffering remains rather distant (see, e.g., Giordano 2010; and also Button et al. 2013). If it became possible, objective pain assessment could imply that mental pain and suffering is as objective as physical pain and suffering.

56

J. Varelius

explained, according to the argument from relief of suffering, a physician-assisted death is morally permissible when the patient’s suffering is unbearable to her. Hence, both of the main arguments for physician-assisted death focus on the desires and experiences of the patient, on how her life feels for her, on what it is like from her own viewpoint. In view of the considerations of this section, there would not seem to be adequate reason to consider mental suffering morally less significant than physical distress in the end-of-life context. Accordingly, if it is accepted that her physical suffering can provide grounds for ending a patient’s life, there does not appear to be sufficient reason to rule mental suffering out as grounds for physician-assisted death. The relevance of mental suffering in the end-of-life context is indeed already acknowledged, at least implicitly, in that even patients who are not suffering physically are allowed to refuse and to withdraw from vital treatment (when they do it autonomously).4

4.5 Countervailing Considerations Pertaining to Charlie’s Distress? That it is permissible to punish criminal offenders for their crimes by inflicting mental suffering on them is quite commonly accepted. Furthermore, many people apparently think that, whether or not it is connected to punishment, mentally distressing experiences can develop one’s character: suffering refines and purifies and, at best, makes one merciful, compassionate, and noble. As with, for instance, artistic creation, mental suffering can also be seen as an unfortunate but essential experience that is considered highly valuable. And, for example, grief, as distressing as it can be, is widely believed to be an appropriate emotion in particular circumstances, when, say, one has lost a close friend or relative. Finally, some of those with masochistic inclinations may be able to enjoy even mental suffering for its own sake.5 Perhaps then, there is also at least some reason to endorse the mental distress that Charlie is experiencing, something that speaks against physician-assisted death in his case? The kind of suffering that Charlie is undergoing would not seem to qualify as an acceptable punishment for any crime (cf. also, e.g., Adams 2014). And, as things are, Charlie has done nothing to deserve to be harshly punished. It also seems clear that the mental suffering that a person experiences can be helpful in building her character—as distinguished from, for instance, just making her numb—only if she 4

It might be objected that if mental suffering is deemed relevant in the end-of-life context, then the request for euthanasia made by a heartbroken teenager who has just lost the person she believes to be the love of her life must be obeyed (cf., e.g., Young 2014). However, as has already been pointed out (see, e.g., Varelius 2014), accepting the above conception of morally acceptable physician-assisted death, the teenager does not qualify as a candidate. For, in cases like hers, the suffering typically is not enduring and unavoidable in that there is no way of adequately alleviating it other than that of ending her life. 5 As it does not affect the main argument of this chapter, I will not now go into the question whether a masochist actually enjoys her ability to endure suffering rather than the suffering itself (or both).

4 Mental Illness, Lack of Autonomy, and Physician-Assisted Death

57

can learn something from the distress. That, in its turn, most plausibly requires mental abilities that Charlie, because of the severity of his mental illness, lacks. Charlie might learn something valuable from his distress if he recovered from his condition. But, then again, he might not (see also, e.g., Olsen 2006). And, at any rate, the probability of his recuperating from his illness is very low. Neither is Charlie’s distress something that he has decided to accept because he sees it as an inevitable part of some whole that he deems valuable. His distress might be seen as an appropriate response to his circumstances in the sense that other people are also known to have become similarly mentally ill after undergoing as gravely deprived a childhood as he did. Yet falling seriously mentally ill is clearly not a reaction that is morally required from people with a similar—or any—background. Mental suffering is arguably to be endorsed, if at all, only in situations in which avoiding it would result in, or sustain, an unfortunate loss of contact with reality. But after Charlie fell ill, his situation has been the opposite. Charlie is also not a masochist. Consequently, I take it that there is no good reason to endorse the suffering he undergoes.

4.6 Possible Objections As far as the above considerations are plausible, the main arguments for physicianassisted death found in recent medical ethical literature support physician-assisted death also in cases of psychiatric patients such as Charlie. Yet several possible objections for allowing a physician to assist a patient such as Charlie to end his life suggest themselves. Below I briefly assess eight criticisms that seem to be the most central ones.

4.6.1 Charlie’s Condition is not Terminal As already noted, it has been maintained that only suffering caused by a terminal illness or injury can provide grounds for physician-assisted death. An advocate of the view might argue that in Charlie’s case the fact that his condition is not terminal rules out the possibility of physician-assisted death. Therefore, even if the above considerations were acceptable, the conclusion of this possible objection could be that the mental health care providers treating Charlie ought not to help him to end his life. The view that Charlie’s condition is not terminal could be questioned (see Cholbi 2013b, 501–503). But, for the sake of argument, let us assume that his illness is not fatal. Would the non-terminality of Charlie’s condition then suffice to disqualify him as a candidate for physician-assisted death? Allowing physician-assisted death only in the case of terminal patients limits the end-of-life choices open to nonterminal patients and, hence, restricts the scope of their autonomy (see also, e.g.,

58

J. Varelius

Plaisted 2013). Accordingly, considerations motivating one of the two main arguments for physician-assisted death—the argument from autonomy—would also appear to support permitting physician-assisted death in the case of non-terminal patients. In terms of relieving the patients’ suffering, other things being equal, the only difference between a terminal patient who suffers unbearably and a non-terminal patient in intolerable distress is that the agony of the latter is likely to last longer.6 When avoidance of unbearable suffering is seen as a central consideration in favor of physician-assisted death, as it now is, it would therefore seem that helping a nonterminal patient in ending his life is more justified than assisting a terminal patient to die (see also, e.g., Beauchamp and Davidson 1979). Accordingly, seen from the viewpoints of the argument from autonomy and the argument from relief of suffering, physician-assisted death would appear to be at least as justified in the case of nonterminal patients as it is in the case of terminal patients. And, in light of the above considerations, both of the arguments are also relevant in Charlie’s case.

4.6.2 Is Charlie’s Condition Certainly Incurable? It might be argued, however, that mental disorders differ from physical disorders and injuries in that the incurability of the former is never certain (cf., e.g., Cowley 2013). Accordingly, a critic could argue, even if it were very improbable, it is still possible that Charlie will recover from his illness. In view of this prospect, the critic could conclude that the mental health care providers treating Charlie ought not to help him to end his life. However, people are known to have recovered from physical illnesses thought to be incurable.7 New findings in medical science may also suddenly alter existing conceptions of which physical illnesses and injuries can be healed and which cannot. On the other hand, though the prognoses of patients suffering from severe mental illnesses such as schizophrenia are currently often less pessimistic than they used to be, not all patients with severe mental illness return to a normal existence (see, e.g., Frese et al. 2009; cf., e.g., Tamminga and Lahti 2001). Accordingly, it is not at all clear that there really is the kind of difference between physical illnesses and injuries and mental disorders that this possible objection presupposes. Someone might now object that, instead of allowing physician-assisted death, it would still be more reasonable to apply a version of Pascal’s Wager here. In this view, as the costs of staying alive for the patient are small as compared to the benefits he 6

It might be objected that ending the life of a terminal patient is not as bad as terminating the life of a non-terminal patient because the former would soon die anyhow. However, this presupposes that ending the life of a patient in unbearable suffering is a bad thing. It is deeply regrettable that there are cases in which people suffer so horribly that they want to end their existence. But when the situation is as severe as that, ending the patient’s life is arguably not a bad thing. Therefore, this possible objection is unconvincing. 7 That some of such cases involve misdiagnoses does not lessen their relevance here.

4 Mental Illness, Lack of Autonomy, and Physician-Assisted Death

59

would gain if he recovered from his illness, the best bet is against dying. However, considering that even many healthy people can have great difficulties with finding accommodation and employment and maintaining relationships, the life of a person who recovers after years or decades of severe mental illness can unfortunately be rather unappealing to several people, possibly including Charlie. Accordingly, even if he recovered from his illness, he might also find the life he would then have to be overly burdensome. Moreover, given the severity of Charlie’s suffering, it would seem that the costs of his staying alive, while he remains ill, could be insignificant to him only if he were sedated into unconsciousness during the possibly quite futile wait for a cure for his condition. Though someone might be willing to choose such an existence, it would not—to say the least—seem to be the only reasonable choice to be made in Charlie’s situation (see also Sect. 4.6.4).

4.6.3 What if Charlie Changes His Mind? A critic could still insist that even if Charlie does not recover from his illness, his mind as regards the desirability of death might change. If a competent patient holds on for an extended period to her wish to die, we can be reasonably convinced that she really does want to die. But precisely because he is severely mentally ill, the critic could continue, the case of Charlie is different (also) in this respect. Charlie’s thinking follows its own peculiar logic and, hence, it may be that tomorrow he will be quite happy with his life. Because of this possibility, the critic could conclude that assisting Charlie to end his life would be morally unacceptable. This possible critic is quite correct in that patients should not be helped to end their lives if that is not what they want. It is also true that the possibility of Charlie’s changing his mind about the desirability of dying cannot be ruled out with absolute certainty. However, again, that degree of certitude is not available in the cases of physically ill or injured patients who request assistance in ending their lives either. The facts that Charlie has been suicidal for years and that the mental health care providers treating him foresee no dramatic alteration in his state of mind also argue against this possible objection. Indeed, to be plausible, the view that Charlie should be made to continue his life because of the possibility that he suddenly gains the desire to live should be supported with good reasons for thinking that he really would change his mind.

4.6.4 Ending Charlie’s Life is Undignified and Inhumane When suffering patients autonomously choose to die, they exercise their human capacities and control the ending of their lives. But in the cases of incompetent persons such an expression of agency is not possible. Consequently, a critic might maintain that terminating the life of an incompetent person, or even just assisting

60

J. Varelius

in it, amounts to treating the person like an animal that is put out of its misery. Behaving like that toward Charlie would be both undignified and inhumane. In the case of human beings, the criticism could continue, more considerate and respectful ways of acting should be found. Therefore, the critic could conclude that physicianassisted death does not come into question in Charlie’s case, the main argument of this chapter notwithstanding. Let us assume that physician-assisted death would be undignified and inhumane in Charlie’s case. Would this possible objection then be plausible? That depends on how physician-assisted death compares with its alternatives. Given that recovery and change of mind are quite improbable in Charlie’s case, there are two options to consider. First, Charlie continues his life as it now is. Second, he is sedated heavily enough for his persecutors to leave him alone. Now, putting up with the situation can, at least sometimes, be deemed a dignified response to one’s suffering. But if one is able to tolerate it, one’s distress is not unendurable. And, as already mentioned, physician-assisted death is now assumed to be acceptable only in cases of intolerable suffering. Coming to terms with one’s distress would also appear to presuppose mental capacities—the ability to step back from it and to put it into some perspective—which Charlie lacks. And having to experience further meaningless suffering for years or even decades is arguably, if not evidently, less humane than its alternatives, including the option of ending such existence. In terms of the second alternative, if it were possible to medicate Charlie so that he would regain contact with reality but just cease to have the hallucinations and delusions he now has, his condition would not be as severe as it is. As things are, it would seem that sedation could alleviate Charlie’s agony only if the medication were strong enough to significantly affect the level of his consciousness. Perhaps persons whose level of awareness is somewhat lower than normal could lead a dignified or at least a humane existence? But, unfortunately, that it would be possible for Charlie to lose his persecutors without being sedated to, or at least very near to, unconsciousness seems like an unrealistically sanguine prospect. And spending years, or even decades, sedated into (near) unconsciousness hardly qualifies as leading a dignified or a humane existence. Accordingly, even if physician-assisted death were undignified and inhumane, it would seem that the most probable alternatives available in Charlie’s case are at least as, if not more, undignified and inhumane.8

8

Of course, the notion of human dignity can be understood in significantly different ways. Accordingly, someone might argue that the common understanding of human dignity employed above should be replaced with a different one, one that would also imply that assisting Charlie to end his life would be more undignified than its alternatives. However, as I am unable to here assess whether there could be a justifiable notion of human dignity of the kind it refers to, I must now put this possible objection aside.

4 Mental Illness, Lack of Autonomy, and Physician-Assisted Death

61

4.6.5 Can Physician-Assisted Death Really Be Acceptable in Cases of Non-autonomous Patients? The view that even non-autonomous patients could qualify as candidates for physician-assisted death could be deemed counterintuitive despite the reasons for it presented above. If the view is accepted, a critic could maintain, then we must also endorse physician-assisted death in the cases of infants. But advocating that health care providers ought to help small children in ending their lives is clearly implausible. Therefore, the critic could conclude, the above considerations do not show that physician-assisted death could come to question in Charlie’s case either. However, withholding and withdrawing even life-sustaining treatment from incurably physically ill or injured severely suffering infants is widely accepted (cf., e.g., Porta and Frader 2007). This demonstrates that the distress of non-autonomous children is usually already deemed significant in the end-of-life context. But, of course, if physician-assisted death comes into question in the case of infants, it can be acceptable only if they are incurably ill, their suffering is unbearable, and ending their lives is the only adequate way of avoiding it. When these criteria are satisfied, that a health care provider—not the patient herself—ends the life of an infant is arguably not morally unacceptable (see also, e.g., Vanden Eijnden and Martinovici 2013; cf., e.g., Kon 2007). Accordingly, this possible objection is implausible.

4.6.6 Charlie is not the Only One Whose Interests Count Above, the focus has been on what is good from Charlie’s viewpoint. Yet, obviously, Charlie is not the only party in the case. At least the interests of the mental health care providers treating him and those of the members of his family are also involved (see also Sect. 4.6.7). It could thus be argued that even if assisting Charlie to end his life were best from his viewpoint, it does not necessarily follow that physician-assisted death would be morally warranted in his case. The interests of the other parties may argue strongly enough against helping Charlie to end his life to make it morally unacceptable. However, the main purpose of the institution of mental health care is arguably, if not evidently, that of serving the good of those who suffer from psychiatric problems. Accordingly, whether assisting Charlie to end his life accords with the relevant interests of mental health care providers—primarily their interests qua the professionals they are—should mainly depend on whether doing that is what is best from Charlie’s viewpoint (cf. also, e.g., Kantymir and McLeod 2014). Given that Charlie is especially vulnerable (cf., e.g., Tavaglione et al. 2015), it is also not unreasonable to require his family members to give extra weight to his interests. At least apart from very exceptional circumstances, it is difficult to imagine that a member of his family could have an interest that would justify continuing, to the indeterminable future, the agony that Charlie’s life is for him. And, for reasons presented above, sedating

62

J. Varelius

him into unconsciousness could be deemed undignified and inhumane. Accordingly, if ending his life were best for a patient like Charlie, it would seem that giving due weight to the interests of mental health care providers and to those of the patient’s family members would, at least normally, not suffice to make physician-assisted death impermissible in the case.

4.6.7 Accepting Physician-Assisted Death in Charlie’s Case Would Compromise the Value of Human Life in Society Maintaining that physician-assisted death could be acceptable in cases like that of Charlie, a critic might still argue, would be to say that the lives of the severely mentally ill have no worth. Besides being bad in itself, sending such a message would, the criticism could continue, surely compromise the value put on human life in a society. Therefore, the conclusion of the objection could be, physician-assisted death is not acceptable in cases of patients like Charlie, irrespective of what has been argued above. However, the main argument of this chapter does not entail that the lives of the severely mentally ill have no value or that human life has no significant worth. The argument concerns only the cases of the severely mentally ill who have a persistent wish to die because of their continuing unbearable and incurable suffering. And the argument does not entail that their lives have no value, but that the value of their lives can be outweighed by the worth of relieving their distress and enabling them to avoid the kind of existence they would most plausibly autonomously eschew. Interpreting the main argument of this chapter as saying that the lives of the mentally ill have no value—or that human life in general has no worth—would simply be to make a mistake. Accordingly, it would arguably be better to inform people who might make the error about the proper implications of allowing physician-assisted death in cases of persons such as Charlie than to force persons like him to reluctantly continue lives of unbearable distress for years, or even decades.

4.6.8 Incompetent Patients Are More Likely to Be Abused Than Competent Patients Finally, a critic might maintain that competent patients are usually able to stand up for themselves whereas incompetent patients often are not. Therefore, the possible objection could continue that abuses of the laws permitting physician-assisted death would be much more likely in the cases of patients who are incapable of autonomously deciding about ending their lives than in cases of competent patients. And that, the critic could conclude, is why physician-assisted death should not be allowed in the cases of patients such as Charlie.

4 Mental Illness, Lack of Autonomy, and Physician-Assisted Death

63

Incompetent patients are, in general, more vulnerable to abuse than competent patients. Yet it does not necessarily follow that there would actually be more maltreatment in the cases of incompetent patients. To a significant extent, whether there would be would appear to depend on how physician-assisted death would be arranged. When physician-assisted death is allowed in the cases of competent patients, the pertinent rules include requirements to the effect that a request for death must be assessed by independent parties, for instance, by at least one medical expert besides the one responsible for treating the patient in question (see, e.g., The Oregon Death with Dignity Act). A corresponding requirement should evidently also be used in the case of incompetent patients. Whether formulating additional rules so as to adequately account for the possible remaining threats of abuse would be impossible is an empirical question that I am unfortunately unable to answer. But, in the absence of good evidence for the claim that abuses of the laws permitting physician-assisted death would really be significantly more frequent in the cases of incompetent patients than in those of competent patients, this possible objection appears unconvincing.

4.7 Conclusion In this chapter, I have considered the question whether physician-assisted death could be morally acceptable in the cases of persistently suicidal unbearably and incurably suffering psychiatric patients who are unable to make autonomous end-of-life choices. I focused on one case of that kind, the case of Charlie, from the viewpoint of the main arguments for physician-assisted death found in recent medical ethical literature, the argument from autonomy and the argument from relief of suffering. First I argued that patients’ autonomous choices are valued in health care because of their role in determining what patients would really want; that there is adequate reason to believe that an autonomous person would not choose a life of meaningless unbearable suffering; and that a patient’s lack of autonomy does not make his suffering morally unimportant. Then I maintained that mental suffering is not relevantly different from physical distress in the end-of-life context and argued that there is no good reason to endorse the suffering of a patient like Charlie. On these grounds, I proposed that that the main arguments for physician-assisted death found in recent medical ethical literature also support physician-assisted death in the cases of patients such as Charlie.9 After this, I briefly assessed eight possible objections to the suggestion and proposed that they are not sufficiently convincing to disprove it. As already proposed, I did not argue that it is permissible for health care providers to end the lives of their patients against the patients’ will. Neither did I maintain that health care providers should urge their patients to commit suicide or to have 9

An affective disorder, such as depression, would appear to influence agency differently than a delusional disorder from which Charlie suffers. Yet it would seem that sometimes an affective disorder can also make a patient incompetent to make autonomous end-of-life choices (cf., e.g., Meynen 2011). The above considerations support physician-assisted death in such cases too, provided, again, that the patient is suffering incurably and unbearably and wants to end her existence.

64

J. Varelius

euthanasia. Nor did I argue that physician-assisted death is acceptable when there is a cure for the condition that makes life intolerable or adequate ways of alleviating a patient’s distress other than ending her life. Though I touched upon some considerations related to the moral acceptability of physician-assisted death other than the arguments from autonomy and from relief of suffering, I did not show that physician-assisted death is morally acceptable in the cases of incompetent patients such as Charlie: perhaps there are stronger objections to the idea than the ones discussed here (cf., e.g., footnote 8) and someone might also reject my responses to the latter. Hence, the conclusion of this chapter is modest: the main arguments for physicianassisted death found in recent medical ethical literature support physician-assisted death also in cases of incompetent psychiatric patients whose illness is incurable and who persistently want to end the existence that they find unbearable. How many patients like this there are, is an empirical question I am unable to answer. But even if they were rare—as they hopefully are—the above considerations support the view that such patients should not be ignored in connection with assessing the moral and legal acceptability of physician-assisted death. As already suggested above, the idea is not novel (see Burgess and Hawton 1998, 121; Hardcastle and Stewart 2002, 432– 433), but the explication of the support the main arguments for physician-assisted death provide for it and the assessment of the possible criticisms of it presented above, I believe, are.10

References Adams, David M. 2014. Belief and death: Capital punishment and the competence-for-execution requirement. Criminal Law and Philosophy. https://doi.org/10.1007/s11572-014-9293-6. Appel, Jacob M. 2007. A suicide right for the mentally ill: A Swiss case opens new debate. Hastings Center Report 3: 21–23. Appel, Jacob M. 2012. “How hard it is that we have to die”: Rethinking suicide liability for psychiatrists. Cambridge Quarterly of Healthcare Ethics 21: 527–536. Appelbaum, Paul S. 2007. Assessment of patients’ competence to consent to treatment. The New England Journal of Medicine 357: 1834–1840. Beauchamp, Tom L. 1993. Suicide. In Matters of life and death: New introductory essays in moral philosophy, 3rd ed., ed. Tom Regan, 69–120. New York: McGraw-Hill Inc. Beauchamp, Tom L. 2006. The right to die as the triumph of autonomy. Journal of Medicine and Philosophy 3: 643–654. Beauchamp, Tom L., and James F. Childress. 2009. Principles of biomedical ethics, 6th ed. New York: Oxford University Press. Beauchamp, Tom L., and Arnold I. Davidson. 1979. The definition of euthanasia. Journal of Medicine and Philosophy 4: 294–312. Buchanan, Allen E., and Dan W. Brock. 1989. Deciding for others: The ethics of surrogate decisionmaking. Cambridge: Cambridge University Press.

10

I thank Michael Cholbi for valuable comments on an earlier version of this chapter and Marion Lupu for revising my English. All remaining errors are mine.

4 Mental Illness, Lack of Autonomy, and Physician-Assisted Death

65

Burgess, Sally, and Keith Hawton. 1998. Suicide, euthanasia, and the psychiatrist. Philosophy, Psychiatry, & Psychology 5: 113–126. Button, Katherine S., John P.A.. Ioannidis, Claire Mokrysz, Brian A. Nosek, Jonathan Flint, Emma S.J.. Robinson, and Marcus R. Munafò. 2013. Power failure: Why small sample size undermines the reliability of neuroscience. Nature Reviews Neuroscience 14: 365–376. Callaghan, Sascha, Christopher Ryan, and Ian Kerridge. 2013. Risk of suicide is insufficient warrant for coercive treatment for mental illness. International Journal of Law and Psychiatry 36: 374– 385. Cholbi, Michael J. 2013a. Kantian paternalism and suicide intervention. In Paternalism: Theory and practice, ed. Christian Coons and Michael E. Weber, 115–133. Cambridge: Cambridge University Press. Cholbi, Michael J. 2013b. The terminal, the futile, and the psychiatrically disordered. International Journal of Law and Psychiatry 36: 498–505. Corns, Jennifer. 2014. The inadequacy of unitary characterizations of pain. Philosophical Studies 169: 355–378. Cowley, Christopher. 2013. Euthanasia in psychiatry can never be justified. A reply to Wijsbek. Theoretical Medicine and Bioethics 34: 227–238. Cross-Disorder Group of the Psychiatric Genomics Consortium. 2013. Identification of risk loci with shared effects on five major psychiatric disorders: A genome-wide analysis. The Lancet 381: 1371–1379. Dees, Marianne, Myrra Vernooij-Dassen, Wim Dekkers, Kris C. Vissers, and Chris van Wee. 2011. ‘Unbearable suffering’: A qualitative study on the perspectives of patients who request assistance in dying. Journal of Medical Ethics 37: 727–734. Fairbairn, Gavin J. 1995. Contemplating suicide: The language and ethics of self-harm. Florence: Routledge. Ferkany, Matt. 2012. The objectivity of wellbeing. Pacific Philosophical Quarterly 93: 472–492. Frese, I.I.I., J. Frederick, Edward L. Knight, and Elyn Saks. 2009. Recovery from schizophrenia: With views of psychiatrists, psychologists, and others diagnosed with this disorder. Schizophrenia Bulletin 35: 370–380. Gill, Michael B. 2009. Is the legalization of physician-assisted suicide compatible with good endof-life care? Journal of Applied Philosophy 26: 27–45. Giordano, James. 2010. The neuroscience of pain, and a neuroethics of pain care. Neuroethics 3: 89–94. Hardcastle, Valerie G., and Rosalyn W. Stewart. 2002. Supporting irrational suicide. Bioethics 16: 425–438. Hardy, Rona. 2005. Re-weaving the self: Approaches to chronic pain. Healthcare Counselling & Psychotherapy Journal 5: 14–17. Hewitt, Jeanette. 2010a. Rational suicide: Philosophical perspectives on schizophrenia. Medicine, Health Care and Philosophy 13: 25–31. Hewitt, Jeanette. 2010b. Schizophrenia, mental capacity, and rational suicide. Theoretical Medicine and Bioethics 31: 63–77. Hewitt, Jeanette. 2013. Why are people with mental illness excluded from the rational suicide debate? International Journal of Law and Psychiatry 36: 358–365. Kantymir, Lori, and Carolyn McLeod. 2014. Justification for conscience exemptions in health care. Bioethics 28: 16–23. Kon, Alexander A. 2007. Neonatal euthanasia is unsupportable: The Groningen protocol should be abandoned. Theoretical Medicine and Bioethics 28: 453–463. Lauinger, William A. 2013. The strong-tie requirement and objective-list theories of well-being. Ethical Theory and Moral Practice 16: 953–968. Meynen, Gerben. 2011. Depression, possibilities, and competence: A phenomenological perspective. Theoretical Medicine and Bioethics 32: 181–193.

66

J. Varelius

Naudts, Kris, Caroline Ducatelle, Jozsef Kovacs, Kristin Laurens, Frederique Van Den Eynde, and Cornelis Van Heeringen. 2006. Euthanasia: The role of the psychiatrist. British Journal of Psychiatry 188: 405–409. Niv, David, and Marshall Devor. 2004. Chronic pain as a disease in its own right. Pain Practice 4: 179–181. Olsen, J. Mark. 2006. Depression, SSRIs, and the supposed obligation to suffer mentally. Kennedy Institute of Ethics Journal 16: 283–303. Parker, Malcolm. 2013. Defending the indefensible? Psychiatry, assisted suicide and human freedom. International Journal of Law and Psychiatry 36: 485–497. Plaisted, Dennis. 2013. An undignified side of death with dignity legislation. Kennedy Institute of Ethics Journal 23: 201–227. Porta, Nicolas, and Joel Frader. 2007. Withholding hydration and nutrition in newborns. Theoretical Medicine and Bioethics 28: 443–451. Royal Dutch Medical Association. 2011. The role of the physician in the voluntary termination of life. http://knmg.artsennet.nl/Publicaties/KNMGpublicatie/Position-paper-The-role-ofthe-physicianin-the-voluntary-termination-of-life-2011.htm. Accessed 29 Oct 2013. Saariaho, Tom, Anita Saariaho, Irma Karila, and Matti Joukamaa. 2012. Early maladaptive schema factors, chronic pain and depressiveness: A study with 271 chronic pain patients and 331 control participants. Clinical Psychology & Psychotherapy 19: 214–223. Sarch, Alexander. 2011. Internalism about a person’s good: Don’t believe it. Philosophical Studies 154: 161–184. Stewart, Cameron, Carmelle Peisah, and Brian Draper. 2011. A test for mental capacity to request assisted suicide. Journal of Medical Ethics 37: 34–39. Szasz, Thomas. 2011. Suicide prohibition: The shame of medicine. Syracuse: Syracuse University Press. Tamminga, Carol A., and Adrienne C. Lahti. 2001. Treatments for chronic psychosis. Dialogues in Clinical Neuroscience 3: 281–292. Tavaglione, Nicolas, Angela K. Martin, Nathalie Mezger, Sophie Durieux-Paillard, Anne François, Yves Jackson, and Samia A. Hurst. 2015. Fleshing out vulnerability. Bioethics 29: 98–107. The Oregon Department of Human Services. 2013. Annual Oregon’s death with dignity act report year 15. http://public.health.oregon.gov/ProviderPartnerResources/EvaluationResearch/Deathw ithDignityAct/Pages/arindex.aspx http://public.health.oregon.gov/ProviderPartnerResources/ EvaluationResearch/DeathwithDignityAct/Pages/arindex.aspx. Accessed 2 December 2013. Tiberius, Valerie. 2007. Substance and procedure in theories of prudential value. Australasian Journal of Philosophy 85: 373–391. Tsou, Jonathan Y. 2012. Intervention, causal reasoning, and the neurobiology of mental disorders: Pharmacological drugs as experimental instruments. Studies in History and Philosophy of Biological and Biomedical Sciences 43: 542–551. Vanden Eijnden, Serge, and Martinovici Dana. 2013. Neonatal euthanasia: A claim for an immoral law. Clinical Ethics 8: 75–84. Varelius, Jukka. 2014. On the relevance of an argument as regards the role of existential suffering in the end-of-life context. Journal of Medical Ethics 40: 114–116. Young, Robert. 2014. ‘Existential suffering’ and voluntary medically assisted dying. Journal of Medical Ethics 40: 108–109. Wijsbek, Henri. 2012. ‘To thine own self be true’: On the loss of integrity as a kind of suffering. Bioethics 26: 1–7.

Jukka Varelius is a Research Fellow at the Department of Philosophy, Contemporary History, and Political Science at the University of Turku, Finland. His work focuses on questions of applied ethics.

Chapter 5

Assisted Dying for Individuals with Dementia: Challenges for Translating Ethical Positions into Law Jocelyn Downie and Georgia Lloyd-Smith

Abstract In this chapter, we explore the issue of assisted dying for individuals with dementia at the nexus of ethics and law. We set out the basic medical realities of dementia and the available data about the desire for the option of assisted dying in the face of dementia. We then describe law and practice with respect to voluntary euthanasia and assisted suicide in jurisdictions that permit at least some assisted dying. We conclude that, because of the peculiar ways in which some of the features of dementia interact with specific legislative provisions, less access to assisted dying for persons with dementia can be realized through the legislation than might have been intended or expected. We then describe draft assisted dying legislation that is before the federal Parliament of Canada. We conclude that, because of the peculiar ways in which some of the features of dementia interact with specific legislative provisions, more access to assisted dying for persons with dementia would be realized, if the legislation was passed as drafted, than might be intended or expected. This exercise reveals that very careful attention needs to be paid to the features of dementia if drafting assisted dying legislation in order to actually achieve the desired law reform effect (whatever that might be). The interplay of common ethical rationales for permitting assisted dying (e.g., autonomy and alleviation of suffering) with common features of dementia (e.g., loss of competence before the terminal phase of the illness, loss of the capacity to communicate before the loss of the capacity to suffer) raises challenges for translating one’s ethical position on assisted dying for individuals with dementia into law. NB: This chapter is current up to 2015. Keywords Advance directive · Palliative sedation · Voluntary euthanasia · Nuffield Council · Dementia

J. Downie (B) Faculty of Medicine, Schulich School of Law, Dalhousie University, Halifax, Canada e-mail: [email protected] G. Lloyd-Smith West Coast Environmental Law, Vancouver, Canada © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 M. Cholbi and J. Varelius (eds.), New Directions in the Ethics of Assisted Suicide and Euthanasia, The International Library of Bioethics 103, https://doi.org/10.1007/978-3-031-25315-7_5

67

68

J. Downie and G. Lloyd-Smith

5.1 Introduction Can individuals with dementia access assisted dying in jurisdictions where assisted dying is at least sometimes permitted? Should they be able to do so? These questions were raised publicly and powerfully in Canada through the recent case of Gillian Bennett (CBC News 2014), an 84-year-old woman who took her life three years after being diagnosed with dementia; her open letter (published on the purposebuilt website www.deadatnoon.com (Bennett (2014)) called for law reform to allow assisted dying through a living will. These questions are also being asked in many other countries around the world as they too grapple with the issue of assisted dying for individuals with dementia. In this chapter, we discuss the assisted dying laws in various countries in a way that we hope will illuminate the status quo and illustrate the effects of particular ways of approaching law reform in relation to these questions. The goal is not to argue for or against the decriminalization of assisted dying for persons with dementia1 but rather to explore the implications of positions taken on various key legislative variables for access to assisted dying for individuals with dementia. We first define our terms and then set out the medical realities that lie behind the questions asked above. Next we provide data (such as there is) about the desire for the option of assisted dying in the face of dementia. We then describe law and practice with respect to voluntary euthanasia and assisted suicide in jurisdictions that permit at least some assisted dying (“permissive jurisdictions”). We conclude that, because of the peculiar ways in which some of the features of dementia interact with specific legislative provisions, less access to assisted dying for persons with dementia can be realized through the legislation than might have been intended or expected. We then describe draft assisted dying legislation that is before the federal Parliament of Canada. We conclude that, because of the peculiar ways in which some of the features of dementia interact with specific legislative provisions, more access to assisted dying for persons with dementia would be realized, if the legislation was passed as drafted, than might be intended or expected. This exercise reveals that very careful attention needs to be paid to the features of dementia if drafting assisted dying legislation in order to actually achieve the desired law reform effect (whatever that might be). The interplay of common ethical rationales for permitting assisted dying (e.g., autonomy and alleviation of suffering) with common features of dementia (e.g., loss of competence before the terminal phase of the illness, loss of the capacity to communicate before the loss of the capacity to suffer) raises challenges for translating one’s ethical position on assisted dying for individuals with dementia into law.

1

Recent philosophical explorations of assisted dying and dementia include Cholbi (2014), Menzel and Steinbock (2013), and Gastmans and De Lepeleire (2010).

5 Assisted Dying for Individuals with Dementia: Challenges …

69

5.2 Terminology It is essential to be clear at the outset precisely what is meant by the terms being used. For the purposes of this paper, we have adopted most of the definitions provided by the Royal Society of Canada Expert Panel on End of Life Decision-Making (“RSC Expert Panel”)—a multi-disciplinary, international panel of experts (Schüklenk et al. 2011, 6–8): “Advance directives” are directions given by a competent individual concerning what and/or how and/or by whom decisions should be made in the event that, at some time in the future, the individual becomes incompetent to make health care decisions. An example is a woman who has signed a document that states that, should she fall into a persistent vegetative state, she does not wish to receive artificial hydration or nutrition. Or, as another example, a man who has signed a document that states that, when he is incompetent, he wishes his wife to make all health care decisions on his behalf. There are two kinds of advance directives: instruction directives, which establish what and/or how health care decisions are to be made; and proxy directives, which establish who is to make health care decisions. “Withholding of potentially life-sustaining treatment” is the failure to start treatment that has the potential to sustain a person’s life. An example is not providing cardiopulmonary resuscitation to a person having a cardiac arrest. “Withdrawal of potentially life-sustaining treatment” is stopping treatment that has the potential to sustain a person’s life. An example is the removal of a ventilator from a patient with a devastatingly severe head injury after a motorcycle accident with no prospect of improvement. “Potentially life-shortening symptom relief” is a suffering control medication given in amounts that may - but are not certain to - shorten a person’s life. An example is giving everincreasing levels of morphine necessary to control an individual’s suffering from terminal cancer when the morphine is known to potentially depress respiration even to the point of causing death (but it is not known precisely how much is too much as the levels are slowly increased). “Palliative sedation” is an umbrella term used to explain intermittent and continuous as well as superficial and deep sedation. The most contested subtype of palliative sedation is known as “terminal sedation.” “Terminal sedation” is potentially life-shortening deep and continuous sedation intentionally combined with the cessation of nutrition and hydration. “Assisted suicide” is the act of intentionally killing oneself with the assistance of another. An example is a woman with advanced ALS who gets a prescription from her physician for barbiturates and uses the drug to kill herself. “Voluntary euthanasia” is an act undertaken by one person to kill another person whose life is no longer worth living to them in accordance with the wishes of that person. An example is a man bedridden with many of the consequences of a massive stroke whose physician, at his request, gives him a lethal injection of barbiturates and muscle relaxants. “Voluntary” means in accordance with the wishes expressed by a competent person or through a valid advance directive. “Non-voluntary” means without the knowledge of the wishes expressed by a competent person or through a valid advance directive. “Involuntary” means against the wishes expressed by a competent person or through a valid advance directive.

70

J. Downie and G. Lloyd-Smith

We also use two umbrella terms. First, “potentially life-shortening palliative interventions.” Palliative interventions are a range of medical practices that provide some relief from pain or suffering at the end of life. There are two categories of palliative interventions that can potentially cause death: potentially life-shortening opioid use (captured under what the RSC Report calls “potentially life-shortening symptom relief”); and terminal sedation. Second, “assisted dying” which, for the purposes of this paper, we use to capture assisted suicide and voluntary euthanasia. An additional term central to this paper but not defined by the RSC Expert Panel is “dementia.” For a definition of this term, we rely on the Nuffield Council on Bioethics (2009, xvii): The term “dementia” describes a collection of signs and symptoms such as memory and communication problems, changes in mood and behavior, and the gradual loss of control of physical functions which, taken together, are an indication of damage to the brain as a result of the progressive degeneration of nerve cells. This can be caused by a variety of different diseases, of which Alzheimer’s disease is the most common. Others include vascular dementia, Lewy body dementia, dementia related to Parkinson’s disease, frontotemporal dementia, alcohol-related dementias and prion diseases.

5.3 Background Medical Realities Before considering the issue of the legal status of assisted dying for individuals with dementia, it is important to first understand the background medical realities of dementia.

5.3.1 Incidence, Prevalence, and Demographics of Dementia According to a 2013 systematic review of studies on the prevalence of dementia around the world, in 2010, 35.6 million people were living with dementia. It was predicted that the incidence would nearly double every twenty years resulting in 65.7 million people living with dementia in 2030 and 115.4 million in 2050 (Prince et al. 2013, 68). Dementia is clearly more common in older populations—with most regions ranging from 5 to 7% prevalence for individuals 60 years old and above and doubling every 5.5–6.7 years (Ibid., 67). It is also more common for women than men with a 19–29% lower prevalence for men than women in most regions of the world (Ibid., 67). It can also be revealing to look at one country as an example of these numbers on a national versus global scale (given the considerable variability across regions of the world this allows a bit more precision) (Ibid., 66). The incidence in Canada has been predicted to rise from 103,728 new cases per year in individuals over the age of 65 in 2008 to 257,811 new cases per year in 2038. The prevalence of dementia for individuals of all ages has been predicted to rise from 1.5% of the population to 2.8%. Dementia is more common in women than men with a ratio of 1.36:1. It

5 Assisted Dying for Individuals with Dementia: Challenges …

71

is also more common in older individuals: 7% of individuals over the age of 65 in 2008 to 9% in 2038; 40% of individuals over the age of 90 in 2008 to 50% in 2038; and 55% of individuals with dementia being over the age of 80 in 2008 and 68% in 2038 (Alzheimer Society 2010, 16–18). Dementia is also more common in First Nations populations than non-First Nations (7.5 per 1000 compared to 5.6 per 1000) and it “disproportionately affects younger age groups and males… in First Nations populations compared to non-First Nations” (Jacklin et al. 2013, e39).2

5.3.2 Symptoms of Dementia Dementia is commonly broken down into three stages: early, middle, and late.3 The Nuffield Council on Bioethics (2009, 9) report on dementia sets out the symptoms associated with each stage in the context of Alzheimer’s disease. In the early (or mild) stage, symptoms include: · · · · · · ·

memory loss difficulty learning new things difficulty making decisions disorientation and bewilderment social withdrawal losing track of the time becoming lost in familiar places.

In the middle (or moderate) stage, symptoms include: · · · ·

more serious disorientation (e.g. difficulty distinguishing between day and night) becoming lost at home increasing difficulty with communication problems with visual perception which may have a very significant effect on their ability to function independently · behaviour changes (including wandering and repeated questioning). 2

The increased prevalence may be a result of a number of different factors: “higher rates of associated conditions, such as hypertension, heart disease, stroke, and diabetes, and higher smoking and obesity rates, all of which increase the risk of dementia. First Nations populations also have an increased vulnerability to the social determinants of health; most notable for dementia are lower incomes and lower levels of educational attainment. Recent studies suggest that sufferers of PostTraumatic Stress Disorder (PTSD) are at an increased risk of dementia. As a result of the historical trauma attributed to residential school experience and the ongoing intergenerational effects, PTSD may also contribute to the overall prevalence of dementia for First Nations, Inuit and Metis people” (Jacklin et al. 2013, e39–e40). 3 Other, more complex, scales are used by health-care professionals to identify the progression of dementia with more precision. These include, for example, the Clinical Dementia Rating (CDR) which rates dementia on a five step scale. The Global Deterioration Scale for Assessment of Primary Degenerative Dementia (or Reisberg scale) is the measure used for the progression of Alzheimer’s disease and divides Alzheimer’s disease into seven stages (Olde Rikkert et al. 2011).

72

J. Downie and G. Lloyd-Smith

In late (or severe) stage dementia, symptoms include: (a) (b) (c) (d) (e) (f) (g) (h)

becoming unaware of the time and place having difficulty recognizing relatives and friends having an increasing need for assisted self-care difficulty walking behaviour changes (including aggression) difficulties swallowing and eating loss of control over bodily functions loss of speech (including loss of capacity to report pain and suffering).

Other forms of dementia may not follow the same slow and steady exacerbation of symptoms. Rather, for example, someone with vascular dementia may experience a significant loss of function, then plateau, then some months or years later, experience another significant loss.4 Dementia is also associated with “unpredictable anger and aggression, depression and apathy” (The Nuffield Council on Bioethics 2009, 8). There is obviously a range of scenarios that people may have in mind when considering the issue of assisted dying for patients with dementia. The vision you have of dementia can greatly impact your analysis of the issue. In order to better understand the range of scenarios, consider the contrasting descriptions of two women with dementia, Margo and Margot. Margo was described in a short piece in JAMA in 1991. A medical student described a 55-year-old woman with Alzheimer’s living in New York (Firlik 1991). She would wander away from her apartment, sometimes to be found a couple of days later in a nightgown “roaming Central Park.” The author notes opaquely, “[b]ad things have happened during these excursions.” Margo is described as enjoying reading, listening to music, and eating peanut butter and jelly sandwiches. She paints— although she has painted the same four concentric circles “every day, exactly the same way, for the last five years.” The author claims “Margo is undeniably one of the happiest people I have known. There is something graceful about the degeneration her mind is undergoing, leaving her carefree, always cheerful.” Margot, by contrast, is an 82-year-old in the final stages of dementia living in British Columbia. In his description of Margot’s condition in the context of a legal contest over the continuation of spoon-feeding, Justice Greyell writes: Mrs. Bentley has advanced Alzheimer’s disease. It was clear from the petitioners’ and respondents’ evidence that Mrs. Bentley makes very few physical movements. She occasionally rubs the back of her hand, arm, or face. She is routinely transferred from her bed to a wheelchair. Her eyes are closed much of the time. She has not spoken since 2010. She does not indicate through her behaviour that she recognizes her family members or any other person.5

4 5

For a review of the progression of various forms of dementia, see Alzheimer’s Society (2011). Bentley v Maplewood Seniors Care Society (2014) at para 18.

5 Assisted Dying for Individuals with Dementia: Challenges …

73

Mrs. Bentley grasps the hands of people who speak to her, but she does not make eye contact or appear to respond in other ways when people try to interact with her. The staff reported to Dr. O’Connor that Mrs. Bentley conveys when she is in pain by moaning and tightening her facial muscles. She is being given a small amount of hydromorphine regularly to address any pain she may be in.6

5.3.3 Decision-Making Capacity and Dementia Decision-making capacity is often compromised in patients with dementia. Berghmans et al. distinguish four standards or criteria for capacity: “(i) the capacity to make and express a choice; (ii) the capacity to understand relevant information; (iii) the capacity to evaluate the character of the situation and possible consequences; and (iv) the capacity to handle information rationally” (Berghmans et al. 2004, 255). Each of these abilities can be compromised by dementia. The assessment of capacity can be particularly difficult in the context of dementia. First, “cognitive fluctuations” occur in patients with dementia, e.g., in 12% of patients with Alzheimer’s disease (Trachsel et al. 2014). The specific timing of the assessment can therefore have a significant impact on the findings. Second, the results of the assessment can be dependent upon whether the assessor is familiar with the particular ways in which asking the questions and supporting the patient’s decision-making can impact upon the ability of the patient to perform (Kim et al. 2002). Difficulties aside, there is some data re: the incidence of incapacity in persons with dementia. For example, one pooled study found that 54% of 1425 patients with Alzheimer disease were incapable (95% CI, 28–79%) (Sessums et al. 2011, 422). One study of 48 patients with very mild to moderate Alzheimer disease found 40% to be competent to make an Alzheimer treatment decision. This study also suggested the following relationship between the stage of the disease and incapacity: very mild very likely to be competent; moderate and severe very likely to be incompetent; and mild and early moderate in a grey zone (Karlawish et al. 2005, 1516). Of course, great care needs to be taken with these statistics as they report on small numbers and, more significantly, are not specific to end of life decisions. Some relate to the capacity to make decisions to enter a research trial, some to respond to hypothetical treatment scenarios, and some to actual disease treatment decisions. Nonetheless, the conclusion relevant to the purposes of this paper can be drawn with confidence and that is that there are some individuals with dementia who clearly have the decision-making capacity needed for all decisions about assisted dying, some who have the capacity for some but not other decisions about assisted dying, some whose capacity is in flux for some or all such decisions (i.e., coming and going at different times of the day or across spans of time), and some who have no capacity for any such decision-making.

6

Ibid. at para 27.

74

J. Downie and G. Lloyd-Smith

5.4 The Desire for the Option of Assisted Dying in the Face of Dementia Not surprisingly, against these background medical realities, some individuals with dementia wish to have the option to access assisted dying at some point in the progression of their disease and some individuals who do not have dementia would like the option of access to assisted dying to be available should they develop dementia. A 2014 “systematic review of the international literature data on attitudes of health professionals, patients, carers and the public” (Tomlinson and Stott 2015) found studies revealing the following attitudes among the public, for example: (a) 10% of 447 American adults found physician-assisted suicide to be an acceptable option for people with mild dementia (citing Cicirelli 1998) (b) 24% of 1960 Dutch adults were in favour of allowing physician-assisted suicide for people with mild dementia (citing Kouwenhoven 2013) (c) Over half of a 725-person sample from the United Kingdom indicated that they would want physician-assisted suicide to be an option for themselves (59.5%) or their partner (57.4%) if they had severe dementia (citing Williams et al. 2007) (d) 50% of the public sampled in the Netherlands (1998), Finland (2002), and the UK (2007) were in favour of allowing euthanasia in cases of severe dementia (citing Ryynänen et al. 2002; van Holsteyn and Trappenburg 1998; Williams et al. 2007) (e) 77% of the public sampled in the Netherlands in 2013 were in favour of allowing euthanasia in cases of severe dementia (citing Kouwenhoven et al. 2013). Obviously, there is a wide range of support revealed here and few conclusions can be drawn on the basis of them. However, it does seem reasonable to at least conclude that some individuals in some countries want to have the option of legal access to assisted dying for themselves in the face of dementia and some want that option to be available to others. So three important questions are whether individuals with dementia legally have access to assisted dying anywhere in the world, whether people with dementia living in permissive jurisdictions actually take advantage of access to assisted dying, and what lessons can be taken from permissive (or potentially permissive) jurisdictions with respect to how to make assisted dying legally accessible to individuals with dementia or how to ensure that it is not (depending on the desired result).7

7

A distinct, but critically important, question is whether individuals with dementia are adequately protected at the end of life and are not having potentially life-sustaining treatment withheld or withdrawn or having potentially-life-shortening palliative interventions administered without their consent (or authorization from a substitute decision-maker) or without their knowledge (or without the knowledge of their substitute decision-maker). This must remain the topic for another paper. Although not specific to dementia, readers interested in unilateral decision-making are directed to Downie et al. (2014).

5 Assisted Dying for Individuals with Dementia: Challenges …

75

5.5 Legal Status of Assisted Dying for Individuals with Dementia in Permissive Jurisdictions 5.5.1 Netherlands In the Netherlands, the Termination of Life on Request and Assisted Suicide (Review Procedures) Act (2002) (The Dutch Act) establishes criteria for access to legally permissible assisted suicide and euthanasia. Under s. 2 of the Dutch Act: 1. In order to comply with the due care criteria … the attending physician must: · be satisfied that the patient has made a voluntary and carefully considered request; · be satisfied that the patient’s suffering was unbearable, and that there was no prospect of improvement; · have informed the patient about his situation and his prospects; · have come to the conclusion, together with the patient, that there is no reasonable alternative in the light of the patient’s situation; · have consulted at least one other, independent physician, who must have seen the patient and given a written opinion on the due care criteria referred to in (a)–(d) above; and · have terminated the patient’s life or provided assistance with suicide with due medical care and attention.8 Under Article 2.2 of the Dutch Act, If a patient aged sixteen or over who is no longer capable of expressing his will, but before reaching this state was deemed capable of making a reasonable appraisal of his own interests, has made a written declaration requesting that his life be terminated, the attending physician may comply with this request. The due care criteria referred to in subsection 1 apply mutatis mutandis.

Therefore, physicians are legally allowed to perform euthanasia on, or provide for assisted suicide by, patients capable of making a “voluntary and carefully considered request” and to perform euthanasia on incapable patients based on an advance directive from the patient written at the time when they were still capable as long as the other requirements for due care are met. Two main issues have been raised with the law as it plays out in the context of dementia. First, in the context of requests for euthanasia in early stages of dementia, questions have been raised about whether suffering of capable individuals in anticipation of increasingly severe dementia can meet the unbearable suffering criterion. Second, it has been suggested that it is difficult for physicians to ascertain whether a patient’s suffering is “unbearable” without being able to communicate with the 8

Netherlands, Termination of Life on Request and Assisted Suicide (Review Procedures Act) (2002).

76

J. Downie and G. Lloyd-Smith

patient (Rietjens et al. 2009). Patients with advanced dementia lose their ability to communicate the extent of their suffering to a physician. This makes it difficult for physicians to assess whether the suffering is unbearable (and it is the physician who must make this assessment under the Dutch law). Regional Review Committee decisions are illuminating with respect to how the law is being interpreted in relation to these concerns. In relation to what kinds of suffering count, the Committee noted: What makes their suffering unbearable is often their awareness of the deterioration in their personality, functions and skills that is already taking place, coupled with the realisation that this will get worse and worse and will eventually lead to utter dependence and total loss of self. Already being aware of their disease and the prognosis may cause patients great and immediate suffering. In that sense, ‘fear of future suffering’ is a realistic assessment of the prospect of further deterioration. (Regional Euthanasia Review Committees 2010, 15)

In relation to the issue of how to determine suffering when the patient can no longer communicate, the Committee noted that the nature of a patient with dementia’s suffering can be determined not only from statements written when competent but also from body language and contemporaneous oral statements (Legemaate and Bolt 2013, 456).9 In a case in 2011, physicians performed euthanasia on a woman with advanced dementia. The Regional Review Committee noted: It is true that just before the termination of her life the patient was no longer able to properly express the unbearable nature of her suffering in words; however the physician and the consultants had received the very strong impression from her body language and verbal responses that the patient felt that her suffering was unbearable, having previously described it as such. The physician considered that impression was reinforced by the fact that just before the euthanasia was carried out the patient, even if imperfectly, had on a number of occasions expressed the wish to die. (Regional Euthanasia Review Committees 2011 in Legemaate and Bolt 2013, 456)

This case caused considerable controversy and has been the subject of a vigorous debate. The Royal Dutch Medical Association responded with guidelines cautioning physicians about the difficulty of dementia cases and asking them to “act with extreme caution and restraint (KNMG 2011, 7).” The Regional Review Committee (2010) has said that “in the case of people suffering from dementia, ‘the response must in general be one of extra caution’” (Regional Review Committee 2010 in Legemaate and Bolt 2013, 455). In sum, euthanasia or assisted suicide for individuals with dementia is permissible (under strict conditions) for early through to end stage dementia. However, because of the way the Dutch Act is drafted, it is thought to be difficult to meet the conditions in the context of dementia and the practice is uncommon for early dementia and very rare for end stage dementia. 9

Methods are being developed to assess pain in persons with dementia. See for example, Lichtner et al. (2014). However, “there are a considerable number of pain assessment tools available for use with the elderly cognitive impaired population. However there is limited evidence about their reliability, validity and clinical utility. On the basis of this review no one tool can be recommended giving the existing evidence” (Ibid., 138).

5 Assisted Dying for Individuals with Dementia: Challenges …

77

5.5.2 Belgium The Belgian legislation on euthanasia (2002) (The Belgian Act) states that euthanasia is not illegal if a physician follows the specific conditions and procedures set out in the statute.10 As noted by Justice Smith in Carter v Canada (Attorney General), “Section 3(1) of the Belgian Act provides that a physician who performs euthanasia does not commit a criminal offence when he or she ensures that [“il s’est assuré que”]: (b) the request is voluntary, well-considered, repeated and not the result of external pressure [the request must also be in writing (s. 4)]; and. (c) the patient is in a medically futile condition of constant and unbearable physical or mental suffering that cannot be alleviated, resulting from a serious and incurable disorder caused by illness or accident.”11 While the Belgian Act does not expressly permit physician-assisted suicide, the Federal Control and Evaluation Commission (the body with oversight of assisted dying in Belgium) has accepted that assisted suicide is also legal if a doctor follows the conditions and procedures established under the Act.12 Assisted suicide is subject to the same requirements as euthanasia. In cases of individuals with dementia, if they are still capable of a voluntary, well-considered and repeated request and their constant and unbearable physical or mental suffering cannot be alleviated, euthanasia and assisted suicide can legally be provided if the other criteria are also met. However, as with the Dutch Act, the scope of this provision is unclear as it is the physician’s assessment that is determinative and it is not clear what will be considered to constitute sufficient suffering in the context of dementia. In cases in which a patient is no longer capable, a physician may legally perform euthanasia based on an advance directive if: 1. the patient suffers from a serious and incurable disorder, caused by illness or accident; 2. the patient is no longer conscious; 3. and this condition is irreversible given the current state of medical science.13 The legislation also requires that the advance directive was drafted or confirmed no more than five years before the start of the incapacity.14 Euthanasia on patients with dementia on the basis of an advance directive is therefore illegal in Belgium if the patient is still conscious. Unconsciousness is obviously far later in the disease process than the point at which many people with 10

Belgium, The Belgian Act on Euthanasia of May, 28th (2002), s 4, § 2, 184. Carter v Canada (Attorney General) (2012) at para 509. 12 Ibid. at para 508. 13 Belgium, The Belgian Act on Euthanasia of May, 28th (2002), s 4, § 2, 184. 14 “La déclaration ne peut être prise en compte que si elle a été établie ou confirmée moins de cinq ans avant le début de l’impossibilité de manifester sa volonté.” Belgium, The Belgian Act on Euthanasia of May, 28th (2002), s 4, § 1, 183. 11

78

J. Downie and G. Lloyd-Smith

dementia would be requesting euthanasia even through an advance directive. That said, euthanasia is available to individuals with dementia through an advance directive at the point of unconsciousness where the patient is seen by the physician to suffer from a serious and incurable disorder (which dementia is) and the condition is irreversible (which dementia is) and the person completed or confirmed the advance directive no more than five years before the start of his or her incapacity.

5.5.3 Luxembourg In Luxembourg, a physician will not be prosecuted for euthanasia or assisted suicide when: 1. the patient is a capable and conscious adult at the time of their request; 2. the request is made voluntarily, after reflection and, if necessary, repeated, and does not result from external pressure; 3. the patient is in a terminal medical situation and shows constant and unbearable physical or mental suffering without prospects of improvement, resulting from an accidental or pathological disorder; 4. the patient’s request for euthanasia or assisted suicide is made in writing.15 In addition, a physician will not be prosecuted for euthanasia in accordance with an advance directive where the physician observes: · that they are afflicted by a severe and incurable accidental or pathological disorder, · that they are unconscious, and · that the situation is irreversible according to science at the time.16 While not a condition for access under the law, the National Commission for Control and Assessment (NCCA) is required to request confirmation of the wishes expressed in the advance directive at least once every five years from the date of registration.17 These requirements probably prevent many patients with dementia from accessing assisted dying. First, for those in early stages of dementia, they will fail to meet the requirement of a “terminal medical situation.” For those in late stages of dementia, they will not be competent. They will then need to rely on an advance directive to achieve their goals but this will only be possible when they are unconscious (again, as in Belgium, much later than many people with dementia may want their advance directive to be acted upon). In addition, if they have been incompetent for more 15

Luxembourg Euthanasia and Assisted Suicide: Law of 16 March (2009), art 2.1, 37. Ibid., art 4(1). 17 Luxembourg Ministries of Health and Social Security 2010, 20. Note the difference here between the Belgian and Luxembourgian requirements. In Belgium, the advance directive must have been made at least five years before the start of the incapacity. In Luxembourg, that restriction is not present. As long as the patient is competent, she can create a valid advance directive. 16

5 Assisted Dying for Individuals with Dementia: Challenges …

79

than five years, it is possible that a healthcare provider might (mis)interpret the requirement of NCCA confirmation of the patient’s wishes every five years to mean that their advance directive will no longer be valid for the purposes of requesting euthanasia.

5.5.4 Switzerland In Switzerland, euthanasia is always illegal while assisting suicide is illegal if was done for selfish motives.18 Assisted suicide is not restricted to individuals with a terminal illness or a specific kind or level of suffering. However, it is restricted to individuals who voluntarily administer the drugs themselves and are competent. Under Article 111f of the Swiss Penal Code, assisting the death of an individual who lacks mental capacity constitutes homicide because, in that case, suicide is not considered to be the result of a free decision. Therefore, assisted suicide is legal for competent patients with dementia (hence largely in the early stages of dementia but not later).

5.5.5 Oregon Oregon was the first American state to pass legislation that legalized physician assisted dying in certain circumstances. The Death with Dignity Act (the Oregon Act) allows a physician to prescribe medication for the purpose of ending the patient’s life only when a patient is capable (defined as having the ability to make and communicate health decisions) and is suffering from a terminal disease (death within six months).19 This precludes individuals with dementia from accessing assisted suicide (unless they have a separate illness that will be in its terminal phase before the dementia has caused a loss of capacity). In early stages of dementia when a patient may still be deemed capable, they will not satisfy the requirement of being terminally ill. In a more advanced stage of dementia when their condition might be considered terminal, they will no longer be capable of making and communicating health decisions.

18 19

Switzerland, Swiss Criminal Code, SR 311.0, art 115 (1937, amended 2015). Oregon, Death with Dignity Act (1997).

80

J. Downie and G. Lloyd-Smith

5.5.6 Washington State/Vermont The legislation passed in Washington State20 and Vermont21 is based on the Oregon Act. Both Acts include the criteria of competence and terminal illness that, taken together, prevent patients with dementia from accessing assisted suicide (unless they have a separate illness that will be in its terminal phase before the dementia has caused a loss of capacity).

5.5.7 Montana In Montana, assisted suicide has been made legal by a decision of the Supreme Court that recognized that the consent of a mentally competent, terminally ill patient can constitute a statutory defence to a charge of homicide against an aiding physician.22 As in the other states, this decision only makes assisted suicide legal for mentally competent, terminally ill patients and therefore assisted suicide is not available for patients with dementia (unless they have a separate illness that will be in its terminal phase before the dementia has caused a loss of capacity).

5.5.8 New Mexico In the 2014 case of Morris v New Mexico,23 an Albuquerque trial court recognized a patient’s fundamental right to receive their physician’s assistance in death only if the patient is mentally competent and terminally ill. As in Oregon, these criteria would preclude individuals with dementia from accessing assistance (unless they have a separate illness that will be in its terminal phase before the dementia has caused a loss of capacity). This decision has been appealed by New Mexico Attorney General Gary King but the appeal has yet to be heard.

5.5.9 Colombia In 1997, the Colombian Constitutional Court considered a challenge to the criminal law provisions that prohibit euthanasia but reduce the sentence when certain conditions are met: 20

Washington State, Death with Dignity Act (2009). Vermont, An Act Relating to Patient Choice and Control at End of Life (2013). 22 Baxter v Montana (2009). 23 Morris v New Mexico (2014). 21

5 Assisted Dying for Individuals with Dementia: Challenges …

81

He who kills another person out of compassion, to put an end to intense suffering caused by physical injuries or grave or incurable illness, will be punished with imprisonment of 1 to 3 years.24

This provision was challenged on the grounds that, by providing for a lesser penalty for euthanasia, the Criminal Code violated the Constitution (specifically, the rights to life and equality). In May 1997, the Court found, first, that the lesser penalty attaching to euthanasia does not violate the Constitution.25 The Court also found that “whenever a medical doctor or physician performed such an act [euthanasia under the conditions in the Code], in the presence of informed consent, no criminal sentence, conviction or penalty could result, in observance that it is a lawfully justified conduct.”26 The Court grounded its decision in the values of autonomy, human dignity, and solidarity. The Court called on the Parliament to establish a regulatory regime to guide the practice of euthanasia. Despite a number of attempts (as recently as 2013), Parliament has been unable to pass such legislation (Colombia Reports 2013). As a consequence of this decision, it appears that euthanasia in the face of intense suffering caused by “physical injuries or grave or incurable illness” (either dementia itself or coincident with dementia) without consent from the individual (but not against the wishes of the individual) would be a crime but could be subject to a lesser penalty than that applied to other forms of homicide. It also appears that euthanasia performed by a physician in the face of intense suffering caused by “physical injuries or grave or incurable illness” (either dementia itself or coincident with dementia) would have a complete defence (Michlowski 2009, 184).27

5.5.10 England and Wales In England and Wales, while the act of assisting suicide remains illegal, prosecutors make decisions about whether or not to prosecute depending on the circumstances of each case under a specific set of prosecutorial charging guidelines. In 2010, the Director of Public Prosecutions issued guidelines that set out 16 criteria that would be taken into consideration when deciding whether or not to prosecute an individual 24

Exhibit D referred to in the Affidavit of Sabine Michalowski sworn (or affirmed) 20 September 2011 in Carter v Canada (2012). 25 The majority of the Court did not address the constitutionality of the assisted suicide provision. (“Incitement to suicide or assisted suicide. He who effectively incites another person to commit suicide, or provides effective assistance for its commission, will be punished with imprisonment of two to six years. The incitement or assistance are directed towards putting an end to intense suffering caused by physical injuries or grave or incurable illness, it will be punished with imprisonment of one to two years.”) Michlowski (2009, 197). 26 Statement made by Carlos Gaviria Diaz (author of the majority decision), May 20, 1997, Exhibit F, Affidavit of Sabine Michalowski, supra note 24. 27 For a rich discussion of the law in Colombia, see Michlowski (2009).

82

J. Downie and G. Lloyd-Smith

for assisting suicide.28 The list of 16 factors in favour of prosecution includes “the victim did not have the capacity to reach an informed decision to commit suicide.” When the individual had reached a “voluntary, clear, settled and informed decision” and the individual was acting with compassion, these factors will weigh against prosecuting the individual for assisting suicide. Assisted suicide in cases of late dementia would clearly be limited by the competency requirement. Another factor in favour of prosecution is when “the victim was physically able to undertake the act that constituted the assistance him or herself.” This might preclude assisted suicide for individuals with early dementia able to perform the suicide themselves.

5.5.11 Quebec In Canada, both assisted suicide and euthanasia are criminal offences under the Criminal Code.29 However, in Quebec, “An Act respecting end of life care” establishes a right to “medical aid in dying”30 for patients who: are of full age and capable of giving consent to care; are at the end of life; suffer from a serious and incurable illness; are in an advanced state of irreversible decline in capability; and experience constant and unbearable physical or psychological pain. While the possibility of allowing requests for medical aid in dying to be made through advance directives was explicitly considered by the Quebec National Assembly, it was not included in the final version of the Act. Indeed, the Act states that “in such directives the person may not request medical aid in dying.”31 Therefore, individuals with dementia are not likely to be able to access medical aid in dying under the Quebec legislation— they will usually either be competent but not meet the “end of life” requirement or meet the “end of life” requirement for access but no longer be competent. Of course, there may be circumstances in which the individual has dementia, is still competent, and is “at the end of life” due to some condition other than the dementia and, in those circumstances, they could qualify for access.

28

Director of Public Prosecutions, England and Wales (2010). Criminal Code (1985). 30 Section 3: “(3) ‘End-of-life care’ means palliative care provided to end-of-life patients and medical aid in dying.” “(6) ‘Medical aid in dying’ means care consisting in the administration by a physician of medications or substances to an end-of-life patient, at the patient’s request, in order to relive their suffering by hastening death.” Section 4: “Every person whose condition requires it has the right to receive end-of-life care.” An Act Respecting End-of-Life Care (2014), ss 3–4. 31 Ibid., s 51. 29

5 Assisted Dying for Individuals with Dementia: Challenges …

83

5.6 Access in Permissive Jurisdictions One might reasonably wonder what the rates of assisted dying for people with dementia are in countries in which there is some level of permission or tolerance. There is some data available on this: · In the Netherlands, euthanasia or assisted suicide in patients with dementia has been steadily increasing since 2009 (12 in 2009, 25 in 2010, 49 in 2011, 42 in 2012, 97 in 2013).32 In their Annual Report 2010, the Regional Review Committees noted that all of the 25 patients with dementia who received euthanasia were at an early stage of the disease and were competent to make autonomous decisions at the time of their death (Regional Euthanasia Committees 2010). In 2011, the Regional Review Committees reviewed the first case of a patient with advanced stage dementia. This case involved a 64-year-old woman who had written an advance directive in which she expressed her desire to die rather than go to a nursing home. The Regional Review Committees found that the physician met all due care requirements because she was experiencing unbearable suffering and continued to express her desire to die (Legemaate and Bolt 2013, 456). Post2011, the majority of patients with dementia who received euthanasia or assisted suicide were in the early stage of dementia and still competent. · A recent study of end-of-life decisions in individuals dying with dementia in Belgium revealed no cases of euthanasia being provided but in 1.3% of 361 cases, there was an ungranted request for euthanasia (Chambaere et al. 2015, 293). · In Switzerland, the Federal Statistical Office reported that 0.3% of the cases of assisted suicide between 1998 and 2009 involved individuals with dementia (Swiss Federal Statistical Office 2012). “Exit Deutsche Schweiz,” a Swiss right to die organization, provided suicide assistance to four individuals with early-stage dementia in 2010 (Gather and Vollmann 2013). It has been reported that in 2013, a British man with early stage dementia chose to end his life at the Swiss Dignitas clinic (Stevens 2013). · There are no official reports of physician-assisted suicide deaths of patients with dementia in Oregon (Oregon Public Health Division 2011). However, Gather and Vollmann note that “some neuropsychiatric diseases placed in the category of ‘other illnesses’, are definitely, or at least potentially, accompanied by dementia during the course of the disease (including Huntington’s disease, corticobasal degeneration)” (Gather and Vollmann 2013, 445). Therefore it is possible that someone could be terminally ill (as a result of something other than dementia) and be at an early stage of dementia (and therefore still competent) and access physician-assisted suicide. · There have been no reported cases of physician-assisted suicide deaths of patients with dementia in Washington.

32

See collection of Regional Euthanasia Review Committees Annual Reports available online at http://www.euthanasiecommissie.nl/archiefjaarverslagen.asp.

84

J. Downie and G. Lloyd-Smith

· It has been reported that, in practice, some Colombian physicians will euthanize unconscious patients when families state that the patient had expressed desires not to be kept alive after all hope of recovery was gone (Ceaser 2008). · Clive Seale’s study in the UK involving 2869 doctors treating an estimated 4106 patients with dementia reported no “ending of life without explicit request of the patient” or voluntary euthanasia deaths in patients with dementia (Seale 2009, 198).

5.7 Lessons to Be Learned from Permissive Jurisdictions Thus it can be seen that assisted dying is permitted for individuals with dementia to varying (from zero to limited) degrees in the various countries in the world that have somewhat permissive regimes with respect to assisted dying. Jurisdictions that at first glance might seem to allow for access to assisted dying for individuals with dementia are often less permissive than they at first appear. For those seeking to have assisted dying available to persons with dementia, reflecting on the experiences in permissive jurisdictions can reveal what positions should be taken (or avoided) on key legislative variables in order to ensure that the access desired is the access achieved.

5.7.1 Suffering Two aspects of suffering are relevant here. First, what kind of suffering is required for access? Physical only? Or also psychological? Or also existential? Second, who determines whether the patient’s suffering is of the nature and to the level required for access? The patient contemporaneously? The patient through an advance directive? The patient’s substitute decision-maker? A physician? Requiring that suffering be physical only will restrict access for many individuals with dementia. Frequently the suffering of individuals with dementia is caused not by physical pain but rather by such emotions as fear, frustration, and grief over loss of self and relationships. Requiring that physicians determine that the patient is suffering to the level required by the law restricts access for individuals with severe dementia as they are unlikely to be able to communicate about their suffering. For individuals at the late stages, only those who are demonstrating suffering through non-verbal means will have access. Yet, individuals can be in unbearable pain and especially other forms of suffering without manifesting that through nonvoluntary or involuntary bodily indicators.

5 Assisted Dying for Individuals with Dementia: Challenges …

85

To be responsive to the specifics of dementia re: kinds of suffering, permissive legislation would need to explicitly acknowledge that suffering can be physical or psychological and that either of these kinds of suffering can be experienced to the level required. Suffering doesn’t need to be physical to be agonizing. To be responsive to the specifics of dementia re: detection of suffering, permissive legislation would need to not require a finding of contemporaneous suffering where the individual is incompetent. Once the patient is incompetent, it is possible that a physician could be able to determine that patient is suffering (e.g., through involuntary physical manifestations of pain).33 However, it is not possible for a physician to determine that a patient is not suffering. We simply cannot know whether someone is not suffering exquisitely when they are no longer capable of competent communication—reflecting on anaesthetic awareness and locked-in syndrome and individuals who have been misdiagnosed as being in a persistent vegetative state (rather than minimally conscious) should give us pause in presuming we can reliably observe suffering (Deschepper et al. 2013, 1505–1506). We can know with some confidence that a person is in physical pain (there will sometimes be very reliable evidence of physical pain). But we cannot know with any confidence that a person is not suffering (absence of evidence of suffering is not evidence of absence of suffering).

5.7.2 Consciousness The requirement that patients be unconscious delays access for individuals with dementia long past the time when most would have wanted to have assisted death. Dementia alone does not cause unconsciousness. Rather, unconsciousness in the context of dementia is the result of some complication (e.g., urinary tract infections, pneumonia, dehydration), and patients usually die within a matter of days of onset of unconsciousness. Yet individuals with dementia who want an assisted death are likely to want it well before they are imminently dying. They may want it when, for example, they have lost the capacity to recognize their loved ones or to communicate and yet they can live for months or years having lost those capacities. It is also not at all clear what the moral significance of unconsciousness is. It isn’t linked to competence which correlates with autonomy and so is linked directly to the moral principle of respect for autonomy. It isn’t linked to the presence of suffering which correlates with the moral justification of alleviating suffering. So consciousness as a requirement precludes some individuals with dementia from accessing assisted dying without any of the values underlying permissive legislation being protected or promoted.

33

For a review of pain assessment tools for persons with dementia, see Lichtner et al. (2014).

86

J. Downie and G. Lloyd-Smith

5.7.3 Five Yearly Confirmation Requiring confirmation at least once every five years restricts access for individuals who are incompetent for more than five years. Individuals can live for more than five years after dementia has robbed them of competence for end of life decisionmaking. The goal that presumably motivates the five-year confirmation requirement can be accomplished by insisting on confirmation every five years while competent. This prevents an individual writing a directive when very young, forgetting about it, changing her mind but not documenting the change. It makes sense to take steps to ensure that a person hasn’t changed her mind while competent but not to insist on such non-ambivalence through incompetence.

5.7.4 Competence and Terminal Illness Restricting access to assisted dying to those who are terminally ill and competent creates a Catch-22—the closer to death one needs to be, the less likely one is to be competent. This is particularly acute for individuals with dementia—while still competent, they are not yet terminally ill yet once terminally ill, they are no longer competent. To be responsive to the specifics of dementia re: the inverse relationship between competence and the terminal phase of the disease, the permissive legislation would need to either allow individuals to request assisted dying through advance directives (to allow it at end stage) or allow individuals to access assisted dying while not yet terminally ill (to allow it at early stage).

5.8 Lessons to Be Learned from a Potentially Permissive Jurisdiction In contrast to the legislation in Europe that may have prevented more access than intended or expected with respect to persons with dementia, the Bill that is presently before the federal Parliament in Canada, may allow more access than intended or expected. Arguably, because of the way it has been worded, it may allow access to persons with dementia through contemporaneous consent (early stage) or through advance directives (all stages). Consider Bill C-581 in relation to each of the elements shown earlier to have created barriers to access for persons with dementia in currently permissive jurisdictions.34

34

Bill C-581 (2014).

5 Assisted Dying for Individuals with Dementia: Challenges …

87

5.8.1 Consciousness There is no discussion of consciousness in the Bill.

5.8.2 Five Years Before Onset/Every Five Years The Bill’s only temporal restriction is that “at least fourteen days must elapse between the time the assisting physician, the person making the request for assisted death or the witnesses, as the case may be, signed the request and the time the request is carried out.”35 This limit would not pose any significant barrier in the context of dementia.

5.8.3 Terminal Illness The Bill does not limit access to assisted dying to those who are terminally ill. Rather, as noted above, a person must only “have been diagnosed by a physician as having an illness, a disease or a disability (including disability arising from traumatic injury) that causes physical or psychological suffering that is intolerable to that person and that cannot be alleviated by any medical treatment acceptable to that person or the person must be in a state of weakening capacities with no chance of improvement.”36

5.8.4 Suffering Suffering under Bill C-581 is determined by the person seeking assisted death (not the physician as, e.g., in the Netherlands) as the person must “have been diagnosed by a physician as having an illness, a disease or a disability (including disability arising from traumatic injury) that causes physical or psychological suffering that is intolerable to that person and that cannot be alleviated by any medical treatment acceptable to that person.” Note, also, that suffering is not limited to physical suffering and so could include the suffering of early stage dementia. Furthermore, and most significantly, access to assisted death is not limited to those who are suffering as the conditions in subsection (c) (only one of which must be met because of the use of the term “or” instead of “and”) include “or the person must be in a state of weakening capacities with no chance of improvement.” 35 36

Ibid., s 3 [creating s. 241.1(9)]. Ibid., s 3 [creating s. 241.1(3)(c)].

88

J. Downie and G. Lloyd-Smith

Therefore, a person could, in principle meet the suffering condition in Bill C-581 in early, mid, or late stages of dementia.

5.8.5 Competence Unlike other permissive legislation, the Bill does not require competence to access assisted dying. Rather, it only requires competence to make a request. The only competence requirement is that the individual is “of sound mind and capable of fully understanding the information provided to him or her under subsection (6)”37 at the time of the request. This is extremely significant. If the Bill required individuals to be competent for access to assisted dying, then assisted dying would (obviously) not be available to any incompetent individuals (whether they had valid advance directives or not). However, since the Bill only requires that individuals be competent when the request is made, then assisted dying could be available to some incompetent individuals: i.e., those who requested assisted dying through a valid advance directive completed while competent. The logic of this argument is as follows: provincial/territorial advance directives legislation allows individuals to request in advance of incompetence that which would be legal for them to request when competent; Bill C-581 would make assisted dying legal; therefore, individuals could request assisted dying through advance directives. This conclusion could, of course, be avoided by the inclusion of a further premise— that assisted dying is legal only for individuals who are competent at the time of access. But that premise is not included in the Bill. In sum, under Bill C-581, individuals with dementia could access assisted dying in at least two ways. First, at an early stage, they may “have been diagnosed by a physician as having an illness, a disease or a disability (including disability arising from traumatic injury) that causes physical or psychological suffering that is intolerable to that person and that cannot be alleviated by any medical treatment acceptable to that person.” Requests for assisted dying by individuals qualifying on these grounds would have to be made contemporaneously (i.e., not through an advance directive). Second, even more remarkably, at early, mid, or late stages, they may “be in a state of weakening capacities with no chance of improvement.” Individuals qualifying on this ground can meet the conditions for eligibility to make a request by being “of sound mind and capable of fully understanding the information provided to him or her under subsection (6)” at the time of either a contemporaneous request or a request made through an advance directive. It is not clear whether this level of access was intended—that will come clear during any debate on the Bill. Bill C-581 will then come to be seen as either a Bill that learned the lessons taught by the legislation in the permissive jurisdictions and achieved the goal of allowing for some access for people with dementia or as yet another example of inadvertent consequences of drafting without working 37

Ibid., s 3 [creating s. 241.1(3)(d)].

5 Assisted Dying for Individuals with Dementia: Challenges …

89

through all of the implications for persons with dementia (this time, unlike the existing permissive jurisdictions reviewed earlier, being more permissive than less).

5.9 Conclusion In sum, it can be seen that dementia creates its own special challenges for legislation on assisted dying. It can be the source of great fear, sadness, frustration, and loss and a consequential desire for access to assisted dying. It involves symptoms and the progression of disease that makes it unusual with respect to the lack of correlation between meeting the common legal conditions for access and meeting the actual conditions for desire for assisted death. Countries that have moved toward permissive regimes with respect to assisted dying have ended up permitting access to fewer people than perhaps intended or expected. As Canada considers legislation that would take it into the realm of the permissive, it may end up permitting access to more people than intended or expected. The goal of this paper has been to demonstrate the peculiar ways in which some of the features of dementia interact with specific legislative provisions so that legislators in the future will draft their legislation such that it permits access to exactly (all and only) the people intended.

References Alzheimer’s Society. 2010. Rising tide: The impact of dementia on Canadian Society. http://www.alz heimer.ca/media/Files/national/Advocacy/ASC_Rising_Tide_Full_Report_e.pdf. Accessed 28 April 2015. Alzheimer’s Society. 2011. The progression of Alzheimer’s disease and other dementias. http:// www.alzheimers.org.uk/site/scripts/documents_info.php?documentID=133. Accessed 28 April 2015. Bennett, Gillian. 2014. Dead at noon. http://www.deadatnoon.com/. Accessed 18 August 2014. Berghmans, Ron, Donna Dickenson, and Ruud Ter Meulen. 2004. Mental capacity: In search of alternative perspectives. Health Care Analysis 12: 251–263. CBC News. 2014. Suffering with dementia, dies leaving right-to-die plea, 21 August 2014. Online: http://www.cbc.ca/m/touch/news/story/1.2742440 Ceaser, Mike. 2008. Euthanasia in legal limbo in Colombia. The Lancet 371: 290–291. Chambaere, Kenneth, S. Joachim Cohen, Lenzo Robijn, Kathleen Bailey, and Luc Deliens. 2015. End-of-life decisions in individuals dying with dementia in Belgium. Journal of the American Geriatric Society 63: 290–296. Cholbi, Michael J. 2014. Kant on euthanasia and the duty to die: Clearing the air. Journal of Medical Ethics. https://doi.org/10.1136/medethics-2013-101781. Cicirelli, Victor G. 1998. Views of elderly people concerning end-of-life decisions. Journal of Applied Gerontology 17: 186–203. Colombia Reports. 2013. Euthanasia bill runs out of time in Colombia’s senate. http://colombiar eports.co/colombia-euthanasia-bill-runs-out-of-time-in-senate Deschepper, Reginald, Steven Laureys, Said Hachimi Idrissi, Jan Poelaert, and Johan Bilsen. 2013. Palliative sedation: Why we should be more concerned about the risks that patients experience an uncomfortable death. Pain 154: 1505–1508.

90

J. Downie and G. Lloyd-Smith

Director of Public Prosecutions, England and Wales. 2010. Policy for prosecutors in respect of cases of encouraging or assisting suicide. London: Crown Prosecution Service, DPP. http://www.cps. gov.uk/publications/prosecution/assisted_suicide_policy.pdf. Accessed 28 April 2015. Downie, Jocelyn, Lindy Willmott, and Benjamin P. White. 2014. Cutting the Gordian knot of futility: A case for law reform on unilateral withholding and withdrawal of potentially life-sustaining treatment. New Zealand Universities Law Review 26: 24–59. Firlik, Andrew D. 1991. A piece of my mind: Margo’s logo. JAMA 265: 201. Gastmans, Chris, and Jan De Lepeleire. 2010. Living to the bitter end? A personalist approach to euthanasia in persons with severe dementia. Bioethics 24: 78–86. Gather, Jakov, and Jochen Vollmann. 2013. Physician-assisted suicide of patients with dementia: A medical ethical analysis with a special focus on patient autonomy. International Journal of Law and Psychiatry 36: 444–452. Jacklin, Kristen M., Jennifer D. Walker, and Marjory Shawande. 2013. The emergence of dementia as a health concern among First Nations populations in Alberta, Canada. Canadian Journal of Public Health e39–e44. Karlawish, Jason H., David J. Casarett, Bryan D. James, Sharon X. Xie, and Scott Y.H. Kim. 2005. The ability of persons with Alzheimer disease (AD) to make a decision about taking an AD treatment. Neurology 64: 1514–1519. Kim, Scott Y., Jason H. Karlawish, and Eric D. Caine. 2002. Current state of research on decisionmaking competence of cognitively impaired elderly persons. American Journal of Geriatric Psychiatry 10: 151–165. KNMG (Koninklijke Nederlandsche Maatschappij tot bevordering der Geneeskunst). 2011. Position paper: The role of the physician in the voluntary termination of life. http://knmg.artsen net.nl/Publicaties/KNMGpublicatie/Position-paper-The-roleof-the-physician-in-the-voluntarytermination-of-life-2011.htm. Accessed 29 April 2015. Kouwenhoven, Pauline S.C., et al. 2013. Opinions of health care professionals and the public after eight years of euthanasia legislation in the Netherlands: A mixed methods approach. Palliative Medicine 27 (3): 273–280 at 279. Legemaate, Johan, and Ineke Bolt. 2013. The Dutch Euthanasia Act: Recent legal developments. European Journal of Health Law 20: 451–469. Lichtner, Valentina, Dawn Dowding, Philip Esterhuizen, S. Closs, Andrew F. Long, Anne Corbett, and Michelle Briggs. 2014. Pain assessment for people with dementia: A systematic review of systematic reviews of pain assessment tools. BMC Geriatrics 138. Luxembourg. 2010. Ministries of Health and Social Security, Euthanasia and Assisted Suicide: Law of 16 March 2009: 25 Questions 25 Answers. http://www.sante.public.lu/publications/ sante-fil-vie/fin-vie/euthanasie-assistance-suicide-25-questionsreponses/euthanasie-assistancesuicide-25-questions-reponses-en.pdf Menzel, Paul T., and Bonnie Steinbock. 2013. Advance directives, dementia, and physician-assisted death. Journal of Law, Medicine, and Ethics 41: 484–500. Michlowski, Sabine. 2009. Legalising active voluntary euthanasia through the courts: Some lessons from Colombia. Medical Law Review 17: 183–218. Nuffield Council on Bioethics. 2009. Dementia: Ethical issues. http://nuffieldbioethics.org/wp-con tent/uploads/2014/07/Dementia-report-Oct-09.pdf. Accessed 29 April 2015. Olde Rikkert, Marcel G.M., et al. 2011. Validity, reliability, and feasibility of clinical staging scales in dementia: A systematic review. American Journal of Alzheimer’s Disease & Other Dementias 26 (5): 357–365. Oregon Public Health Division. 2011. Oregon’s Death with Dignity Act. Annual reports. http:// public.health.oregon.gov/ProviderPartnerResources/EvaluationResearch/DeathwithDignityAct/ Pages/ar-index.aspx. Accessed 29 April 2015. Prince, Martin, Renata Bryce, Emiliano Albanese, Anders Wimo, Wagner Ribeiro, and Cleusa P. Ferri. 2013. The global prevalence of dementia: A systematic review and metaanalysis. Alzheimer’s and Dementia 9: 63–75.

5 Assisted Dying for Individuals with Dementia: Challenges …

91

Regional Euthanasia Review Committees. Annual reports. Online: http://www.euthanasiecommi ssie.nl/archiefjaarverslagen.asp Regional Euthanasia Review Committees. 2010. Annual report 2009. http://www.rtde.eu/node/19. Accessed 29 April 2015. Rietjens, Judith A.C., Paul J. van der Maas, Bregje D. Onwuteaka-Philipsen, Johannes J.M. van Delden, and Agnes van der Heide. 2009. Two decades of research on euthanasia from the Netherlands. What have we learnt and what questions remain? Journal of Bioethical Inquiry 6: 271–283. Ryynänen, Olli-Pekka, Markku Myllykangas, Markku Viren, and Harri Heino. 2002. Attitudes towards euthanasia among physicians, nurses and the general public in Finland. Public Health 116: 322–331. Schüklenk, Udo, Johannes J.M. van Delden, Jocelyn Downie, Sheila McLean, Ross Upshur, and Daniel Weinstock. 2011. The royal society of Canada expert panel: End-of-life decision making. Ottawa: Royal Society of Canada. https://rsc-src.ca/en/expert-panels/rsc-reports/endlife-decision-making. Accessed 29 April 2015. Seale, Clive. 2009. End-of-life decisions in the UK involving medical practitioners. Palliative Medicine 23: 198–204. Sessums, Laura L., Hanna Zembrzuska, and Jeffrey L. Jackson. 2011. Does this patient have medical decision-making capacity? JAMA 306: 420–427. Stevens, John. 2013. Man, 83, Becomes first Briton to choose Dignitas assisted suicide because he had dementia. Mail Online, 31 May 2013. http://www.dailymail.co.uk/news/article-2333133/ Man-83-Briton-Dignitas-assisted-suicide-dementia.html Swiss Federal Statistical Office. 2012. Cause of death statistics 2009: Assisted suicide and suicide in Switzerland. Online: http://www.bfs.admin.ch/bfs/portal/en/index/themen/14/22/publ.html?pub licationID=4732 Tomlinson, Emile, and Joshua Stott. 2015. Assisted dying in dementia: A systematic review of the international literature data on the attitudes of health professionals, patients, carers and the public, and the factors associated with these. International Journal of Geriatric Psychiatry 30: 10–20. Trachsel, Manuel, Helena Hermann, and Nikola Biller-Andorno. 2014. Cognitive fluctuations as a challenge for the assessment of decision-making capacity in patients with dementia. American Journal of Alzheimer’s Disease and Other Dementias ePub June 18: 1–4. van Holsteyn, Joop, and Margo Trappenburg. 1998. Citizens’ opinions on new forms of euthanasia: A report from the Netherlands. Patient Education and Counseling 35: 63–73. Williams, Nia, Charlotte Dunford, Alice Knowles, and James Warner. 2007. Public attitudes to lifesustaining treatments and euthanasia in dementia. International Journal of Geriatric Psychiatry 22: 1229–1234.

Court Cases and Rulings Baxter v Montana. 2009 MT 449, 224 P (3d) 1211 (Mont Sup Ct 2009). Bentley v Maplewood Seniors Care Society. 2014 BCSC 165. Carter v Canada (Attorney General). 2012 BCSC 886. Morris v New Mexico. No. D-202-CV 2012-02909 (N Mex Dist Ct 2014).

Laws and Legislation Belgium, Loi Relative à l’Euthanasie, BS. 2002. http://eol.law.dal.ca/wp-content/uploads/2015/06/ Belgian-Euthanasia-Act.pdf. Translated in Kidd, Dale. 2002. The Belgian Act on Euthanasia of May, 28th 2002. Ethical Perspectives 9 (2–3): 182–188. http://eol.law.dal.ca/wp-content/uploads/ 2015/06/Euthanasia-Act.pdf

92

J. Downie and G. Lloyd-Smith

Bill 52 (Quebec). An Act Respecting End-of -Life Care, SQ 2014, c2. Bill C-581 (Canada). 2014. An Act to Amend the Criminal Code (Physician-Assisted Death). First Reading, March 27, 2014, Second Session, Forty-first Parliament. Criminal Code. RSC 1985, c C-46. Luxembourg. 2010. Euthanasia and Assisted Suicide: Law of 16 March 2009 (Luxembourg 2010). http://www.sante.public.lu/publications/sante-fil-vie/fin-vie/euthanasie-assistancesuicide-25-questions-reponses/euthanasie-assistance-suicide-25-questions-reponses-en.pdf Netherlands. 2002. Termination of Life on Request and Assisted Suicide (Review Procedures) Act. http://www.euthanasiecommissie.nl/Images/Wet%20toetsing%20levensbeeindiging%20op% 20verzoek%20en%20hulp%20bij%20zelfdoding%20Engels_tcm52-36287.pdf Oregon. 1997. Death with Dignity Act. Or Rev Stat, §§ 127.800–127.995. http://public.health.ore gon.gov/ProviderPartnerResources/EvaluationResearch/DeathwithDignityAct/Documents/sta tute.pdf Switzerland. 2015. Swiss Criminal Code. SR 1937 (amended 2015), 311.0. http://eol.law.dal.ca/ wp-content/uploads/2015/06/Penal-Code.pdf Vermont. 2013. An Act Relating to Patient Choice and Control at End of Life. 18 Vt Stat Ann § 77. http://legislature.vermont.gov/assets/Documents/2014/Docs/ACTS/ACT039/ ACT039%20As%20Enacted.pdf Washington State. 2009. Death with Dignity Act. Wash Rev Code tit 70 § 70.245. http://app.leg.wa. gov/rcw/default.aspx?cite=70.245

Jocelyn Downie is a Member of the Order of Canada and a Fellow of the Royal Society of Canada and the Canadian Academy of Health Sciences. She is a Full Professor in the Faculties of Law and Medicine at Dalhousie University and a Faculty Associate of the Dalhousie Health Law Institute. Her work on end-of-life law and policy goes back many years and includes: Special Advisor to the Canadian Senate Committee on Euthanasia and Assisted Suicide; author of Dying Justice: A Case for the Decriminalizing Euthanasia and Assisted Suicide in Canada (winner of the AbbyAnn D. Lynch Medal in Bioethics from the Royal Society of Canada); member of the Royal Society of Canada Expert Panel on End-of-Life Decision-Making; member of the plaintiffs’ legal team in Carter v. Canada (Attorney General), the Provincial-Territorial Expert Advisory Group on Physician-Assisted Dying, and the Canadian Council of Academies Expert Panel on Medical Assistance in Dying.

Chapter 6

Clinical Ethics Consultation and Physician Assisted Suicide David M. Adams

Abstract In this paper I attempt to address what appears to be a novel theoretical and practical problem concerning physician-assisted suicide (PAS). This problem arises out of a newly created set of circumstances in which persons are hospitalized in jurisdictions where PAS, though now legally available to patients, remains morally contentious. When moral disagreements over PAS come to divide physicians, patients, and family members, it is quite likely they will today find their way to the hospital’s consulting ethicist, a member of an emerging group of professionals charged with the responsibility (so we are told) of resolving moral conflict in the clinic. What can or should an ethics consultant do to fulfill this mandate in such circumstances? I argue that the now predominant conception of clinical ethics consultation is importantly deficient in that it provides no final guidance to ethicists on how they should respond in such cases. The ethicist may only recommend—or assist others to reach—solutions to moral disagreements falling within the limits of established norms. Since the normative landscape of PAS appears to be changing, whether it falls within these limits is unsettled. I conclude that the accepted goals of ethics consultation require a revised and expanded conception of the ethicist’s role in cases involving PAS. I close by outlining what that revised understanding would look like and the further questions it raises. Keywords Clinical ethics · Ethics consultation · Moral inquiry · Ethical disagreement · Physician-assisted suicide

6.1 Introduction In this paper I attempt to address what appears to be a novel theoretical and practical problem concerning physician-assisted suicide (PAS). That problem arises out of a newly created set of circumstances in which persons are hospitalized in jurisdictions where PAS, though now legally available to patients, remains morally contentious. D. M. Adams (B) Department of Philosophy, California State Polytechnic University, Pomona, USA e-mail: [email protected] © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 M. Cholbi and J. Varelius (eds.), New Directions in the Ethics of Assisted Suicide and Euthanasia, The International Library of Bioethics 103, https://doi.org/10.1007/978-3-031-25315-7_6

93

94

D. M. Adams

When moral disagreements over PAS come to divide physicians, patients, and family members, it is quite likely they will today find their way to the hospital’s consulting ethicist, a member of an emerging group of professionals charged with the responsibility (so we are told) of resolving moral conflict in the clinic. What can or should an ethics consultant do to fulfill this mandate in such circumstances? The burden of my argument will be that the now predominant conception of the ethicist’s role is incapable of satisfactorily answering this question and that therefore a fresh understanding of the ethicist’s competence and commission needs to be developed. I turn first to the nature of the problem, illustrated by the following case.

6.2 The Problem Mr. Smith is a 79-year-old man suffering from end-stage COPD. He has dealt with recurrent pneumonia and has been hospitalized several times in the past eight years with respiratory failure. During his most recent hospitalization Mr. Smith required endotracheal intubation and ventilatory support, though he was subsequently successfully weaned off the respirator. The patient completed an advance healthcare directive after discharge; the directive requests no attempt at resuscitation in the event of cardiac arrest. Mr. Smith has become increasingly frail in the last three months, though cognitively he is still intact. Five days ago his wife found him down on the floor of their home with labored breathing. Brought to the hospital by ambulance, he was intubated emergently upon arrival, diagnosed with a fresh case of pneumonia, and placed on a regimen of IV antibiotics. The pulmonologist and critical care specialist caring for Mr. Smith believe he can again be weaned and returned home, though they admit his prognosis is extremely poor. He is entering the terminal phase of his illness, they believe, and will almost certainly be dead in less than six months. Mr. Smith has been in the ICU since his admission; he is tired but awake and alert. He communicates in writing with his wife, adult daughter, and treating physicians. Two days ago the hospital’s clinical ethics consultation service was contacted with a request to meet with the patient, family and treatment team. Mr. Smith has conveyed in writing to all concerned that he detests being hospitalized and wishes to go home to die. He insists he “has had it living with COPD” and can’t bear the thought of enduring even one more intubation. Weak yet restless, Smith hates his growing dependence upon machines—and other people—for the maintenance of his failing body. Mr. Smith lives in a state that permits PAS by law and he has indicated his desire to leave the hospital—against medical advice, if necessary—to seek PAS. He has researched the law and concluded he satisfies the requisites it specifies (for example, that the patient have a terminal illness). Though the patient’s daughter and his longtime primary care physician are reluctantly supportive of his resolve to end his life in this way, his wife and other treating physicians are strongly opposed, believing it would be wrong both for the patient to cause his death and for them to support such an outcome. Mr. Smith responds angrily: “Don’t you have any compassion? It’s pointless to go on.” His wife doubts this entreaty expresses Smith’s real reasons

6 Clinical Ethics Consultation and Physician Assisted Suicide

95

for requesting assisted death, however, suspecting he actually wants to rid his family of a burden that will only grow more onerous with time. Smith’s doctors refuse to credit either of the forgoing reasons for pursuing death, believing the promotion of health, not the appeasement of existential distress or alleviation of hardship, to be the true and only good of medicine. The pursuit of that good is incompatible with intending death, even a death that is undeniably foreseeable. Better management of his symptoms is the answer, they say, urging him to re-consider. Mr. Smith’s wife and daughter page the hospital’s consulting ethicist, posing a stark question: Is it ethical for her father to obtain PAS? “None of us can agree on whether it is right for him to do this,” the daughter confides. “We need your help. Can you tell us if PAS is ethical in a case like this?”. Two recent developments in the practice of healthcare are converging in a way that makes cases of the foregoing sort inevitable. The emergence of clinical ethics consultation, evolving out of previously established hospital ethics committees, coincides with renewed debate over—and incremental legal expansion of—physicianassisted death. In addition to their contemporaneity, the practice of ethics consultation and continuing disputes over PAS share in the considerable uncertainty each occasions. The forms such uncertainty takes, upon which I shall elaborate in this paper, come together in cases like that of Mr. Smith, and in ways making a straightforward answer to the daughter’s questions extraordinarily difficult to obtain. Yet some kind of response from the ethicist is surely called for and expected, both by the members of Mr. Smith’s family and also by the involved physicians, the patient himself, and (presumably) the hospital for which the ethicist works. What is the ethicist to say? It is important to be clear at the outset what the involved parties here are (and are not) asking. They are not asking about the legality of PAS: all parties to the Smith case understand that assisted suicide is lawful in their state. Moreover, and as they realize, if the law were all that mattered the disputing parties could clear up their disagreement by consulting an attorney. There would be no need of an ethicist. But they quite sensibly realize the legality of a practice does not settle its moral status. Nor are the parties in the Smith case asking for the ethicist to make a decision for Mr. Smith on whether he should seek PAS. All concede that the patient (so long as decisional capacity is retained) has the final say in this. The issue is not who has the right to decide on the clinical course of action to be taken. The issue is what decisions may rightfully be made. The question is about the moral status of the choices open to Mr. Smith, and about the choice for PAS in particular. Finally, Mr. Smith’s family members are not inquiring about general public attitudes toward assisted suicide. The patient and his family members can reasonably be assumed to appreciate that opinions on this issue are divided. They want to know what is, in fact, the moral status of PAS. Is it allowable in this case or not? What then should an ethics consultant say to the family of Mr. Smith? This is the question I shall tackle. My answer will unfold in two stages. At the first stage, I argue that the now predominant conception of clinical ethics consultation—what it should aim to achieve and how it should do so—is importantly deficient, in that it provides no final guidance to ethicists on what they should say to Mr. Smith’s family. To support this claim, I begin in Sect. 6.3 by summarizing briefly the conception of clinical

96

D. M. Adams

ethics consultation widely accepted and practiced in America and (increasingly) in some other parts of the world. Since as we shall see this predominant theory makes the ethicist’s advice a function of how law, professional standards, scholarly literature, and prevailing practice currently view assisted suicide, I set out in Sect. 6.4 a detailed summary of the current state of affairs regarding PAS. I then seek to show (in the remaining sections of the paper) that this predominant conception of ethics consultation necessarily fails to satisfy its own stated goals in cases turning on contentious and unsettled areas of bioethical concern such as PAS. I conclude that the accepted goals of ethics consultation require a revised and expanded conception of the ethicist’s role in cases involving PAS. I close by outlining what that revised understanding would look like and the further questions it raises.

6.3 Ethics Consultation Many hospitals in the U.S. now have some form of ethics consultation service, developed to address ethical concerns regarding patient care (Fox et al. 2007).1 Put most generally, clinical ethics consultants deal with moral conflicts arising in the care of the sick. The kinds of moral conflict addressed by clinical ethicists typically involve difficult decisions. Would it be best to attempt resuscitation of a critically ill patient with a terminal disease? What should be done for an individual with fast-spreading cancer that cannot be stopped? Should a feeding tube be withheld in a patient with end-stage disease who can no longer swallow? Naturally, it is of preeminent importance to know how hospital ethicists ought to respond when approached about these and related moral concerns. The most influential conception of the goals and aims of ethics consultation, emerging in the last twenty years, is set forth in a series of documents developed and disseminated by the American Society for Bioethics and Humanities (ASBH), most particularly the Core Competencies for Health Care Ethics Consultations (ASBH 1998).2 The first edition of the Core Competencies report endorsed a methodology for ethics consultation it called ethics facilitation, the overarching aim of which is to resolve moral disagreement by “facilitating the building of consensus among the involved parties” within “a range of morally acceptable options” (ASBH 1998, 8). This theory of ethics consultation was an outgrowth of work published by clinical ethicists going back to the 1980s.3 The theories and ideas expressed in this early

1

While European countries for some time lagged behind the U.S. in this regard, things appear to be changing, particularly in the nations of Eastern Europe. See the contributions from many European clinical ethicists in Schildmann et al. (2010). See also McLean (2009), Reiter-Theil (2001), and Nagao et al. (2008). 2 Summaries of the report also appeared in Aulisio and Arnold (2000), Aulisio et al. (2003). This document was subsequently revised. See ASBH (2011). 3 See, e.g., Ackerman (1987), Thomasma (1987), Glover et al. (1986), and Veatch (1987).

6 Clinical Ethics Consultation and Physician Assisted Suicide

97

work have now coalesced into a widely-accepted view.4 On this understanding, ethics consultation is indicated in situations of moral “uncertainty or conflict” (ASBH 2011, 4) where the goal is to “identify, analyze, and resolve ethical problems” (Dubler et al. 2009, 25) by assisting disputing parties to find consensus within a range of “morally acceptable options” (ASBH 1998, 6).5 Ethics consultants are to assist in “the building of morally acceptable shared commitments” bounded by “societal values, law, and institutional policy” (ASBH 1998, 7); no acceptable consensus resolution of a conflict in the clinic may “fall outside the boundaries of widely accepted ethical and legal norms” (ASBH 2011, 7). Imagine, for example, that Mr. Smith specifically refuses in writing any further mechanical ventilation, yet his family and physicians entirely ignore this instruction and orally intubate him. Such a course of action would be an obvious “violation of the patient’s right of medical self-determination” (ASBH 1998, 6). Clearly acknowledged rights and widely accepted standards and principles set “boundaries that must be respected in ethics consultation” (ASBH 1998, 4). The ASBH Clinical Ethics Credentialing Project summarized, by way of reaffirming, the basic features of this predominant understanding of ethics consultation: The task of the [clinical ethics consultant] is to identify, clarify, and analyze the ethical issues in the case and the interests and rights of patients, family members, providers, and administrators, and—where they clash—to facilitate or mediate a ‘principled resolution’: a plan that falls clearly within accepted ethical principles, legal stipulations, and moral rules defined by ethical discourse, legislatures, and courts and that facilitates a clear plan for future interventions. (Dubler et al. 2009, 28)6

For convenience, I will refer to this as the received view of ethics consultation. As I have elsewhere argued, the distinctive feature of ethics consultation on the received view is the notion that whatever consensus plan is finally adopted is necessarily constrained: the collective reflection and deliberation that constitute the ethics consultation process are to be guided along particular dimensions and conducted in 4 There is “general consensus in the field,” according to the revised Core Competencies Report, “that ‘ethics facilitation’ is the best model for HCEC [health care ethics consultation]” (American Society for Bioethics and Humanities 2011, 6). The model of ethics consultation defended in the Core Competencies has been adopted by any number of hospitals, including the Veterans Health Administration (VHA). See Veterans Health Administration Integrated Ethics. It is also arguably embraced by the American Medical Association, which advises that “where there is a dispute, the [ethics] consultant’s role is one of negotiation and resolution.” See American Medical Association (2012, 2). 5 See also Aulisio and Arnold (2008, 422): The role of the ethics consultant “is to help the clinicians, patients, and families arrive at a decision that optimally respects and promotes the values of the appropriate decision-makers—most often the values of the once competent but now incompetent patient—within the bounds of medically acceptable options.” The ethicist facilitates the building of a “principled ethical resolution” of a moral problem by “helping involved parties identify a range of ethically acceptable options, and providing an ethical justification for each option” (Tarzian and ASBH Core Competencies Update Task Force 2013, 5). 6 This model is the centerpiece of a nascent movement to standardize the training of clinical ethicists by, for example, specifying minimum qualifications for consultants and establishing uniform curricula for credentialing programs. See Dubler et al. (2009), and Tarzian and Wocial (2015). For further support of the facilitated consensus model see Aulisio et al. (1999), Aulisio and Arnold (2000, 2008), DeRenzo (1994), and (Dubler 2011).

98

D. M. Adams

a way that rules out appeal to certain kinds of considerations and reasons (Adams 2009). And it is the ethics consultant’s responsibility to ensure that the consensus is in this way ethically justified, guiding deliberations among a circumscribed set of allowable medical choices. I use the expression ‘allowable’ (and its converse, ‘disallowable’) throughout—as opposed to ‘permissible’ and ‘impermissible’—as the latter terms, suggesting moral permissibility, appear too restrictive: for on the received view morality is not the only source of normativity from which the constraints upon consensus formation are to be drawn. Rather, ethicists must “ensure that identified options comport with relevant bioethics, medical, and scholarly literature, and with laws, pertinent institutional policies, and current [ethics consultation] practice standards” (Tarzian and ASBH Core Competencies Update Task Force 2013, 5). In deciding whether a given clinical course of action is allowable, the ethics consultant is ascertaining what I will call the normative status of a general practice or of a particular decision. On the received view then the ethicist is to seek resolution of bioethical disputes among parties who presumably hold widely divergent moral and non-moral beliefs (for example, concerning the worth of human life, the meaning of suffering, the finality of death) by specifying as initial conditions normative boundaries demanding the disputants select from among a limited range of allowable resolutions. The boundaries of the allowable are drawn by standards deriving from several sources of normativity: statutes and regulations, widespread institutional policies, prevailing standards of professional practice, and positions adopted in the bioethics literature. Clearly, this theory of ethics consultation will work only if it is possible in each case reliably to identify the allowable choices; and answering the question posed by Mr. Smith’s daughter requires that we discern what these designated sources of normativity say about the allowability of assisted suicide. As we shall see, this is not as straightforward a matter as the theory assumes.

6.4 The ‘Allowability’ of PAS My contentions in this section will be two. First, I will try to show that despite careful examination of the available evidence, the normative status of PAS remains indeterminate with respect to the sources of normativity the clinical ethicist is supposed to consider. Second, I will claim that the (unstated) assumption of the received view— that determinate answers to questions about the normative status of various clinical choices will be forthcoming—is false in just those cases, like PAS, where ethical guidance on deeply controversial matters is most needed. I begin by looking at what each of the normative sources singled out by the received view says about PAS.

6 Clinical Ethics Consultation and Physician Assisted Suicide

99

6.4.1 Law The first source of norms meant to constrain deliberations about the allowability of a given clinical course of action is law. The most salient legal fact in the case of Mr. Smith is of course that he resides in a state that allows PAS and has legislation in place detailing the procedures for obtaining access to it. It is also important, though, to situate Mr. Smith’s legal rights within a larger jurisprudential context. The legal landscape of PAS, both in the U.S. and elsewhere, is changing. While as recently as the mid-1990s, no American jurisdictions allowed PAS by law, five (as of the time of this writing) now do so.7 Three of these states (Oregon, Washington, and Vermont) have legal protocols in place specifying criteria and procedures for obtaining access to PAS, approved either by ballot initiative (Oregon and Washington) or legislative enactment (Vermont). Courts in both Montana and New Mexico have ruled that PAS is not illegal in those states, but currently no set of enabling procedures is in place. Since the early 1990s (and in addition to the three states noted above), ballot initiatives undertaking to legalize PAS have been voted on in California, Maine, Massachusetts, and Michigan. Though all were defeated, the margins in the first three states were close.8 Support among state legislators for the legalization of PAS, while not yet sufficiently widespread to win large-scale passage of proposed bills, appears nonetheless to be growing. According to the Death with Dignity National Center, bills proposing legislation similar to that currently in place in Oregon, Washington, and Vermont have since 2013 been introduced in Connecticut, Hawaii, Kansas, New Hampshire, New Jersey, and Pennsylvania.9 It is reasonable to suppose that the forgoing numbers reflect changing attitudes apparent in public opinion surveys. Three polling organizations recently surveyed Americans on the subject of PAS and reported similar results. A survey published in late 2013 by the Pew Research Center found that 62% of Americans now say a person suffering great pain with no real hope of improvement should have a right to commit suicide; a majority (56%) say the same of those with an incurable illness.10 According to the same survey, nearly half of Americans polled (47%) approve of laws 7

See data from Pro/Con.org. Retrieved July 3, 2014 from (http://euthanasia.procon.org/view.res ource.php?resourceID=000132) and (http://www.procon.org/headline.php?headlineID=005198). For the text of the recent court ruling in New Mexico, see http://agoodgoodbye.com/wp-content/ uploads/2014/01/199446010-Physician-aid-in-dying-Ruling.pdf. 8 See information at Patients Rights Council. Retrieved July 3, 2014 from http://www.patientsrigh tscouncil.org/site/failed-attempts-usa/. 9 Retrieved July 3, 2014 from http://www.deathwithdignity.org/advocates/national. See also, “‘AidIn-Dying’ Movement Takes Hold in Some States,” New York Times, Feb. 7, 2014. Retrieved July 10, 2014 from http://www.nytimes.com/2014/02/08/us/easing-terminal-patients-path-to-death-leg ally.html?_r=0. The text of the Connecticut bill is representative. See State of Connecticut General Assembly Bill 1138, January, 2009 Session. Retrieved July 3, 2014 from http://www.cga.ct.gov/ 2009/TOB/S/2009SB-01138-R00-SB.htm. 10 See “Views on End-Of-Life Medical Treatments,” Pew Research Religion and Public Life Project, November 21, 2013. Retrieved July 3, 2014 from http://www.pewforum.org/2013/11/21/views-onend-of-life-medical-treatments/.

100

D. M. Adams

permitting physicians to prescribe a lethal dose of drugs to a terminally ill patient. A BBC/Harris Poll indicated 70% of Americans believe terminally ill patients in great pain and with no chance of recovery should have the right to choose to end their lives.11 A 2013 Gallup poll found that 70% of Americans responded favorably when asked if doctors should be allowed to “end the patient’s life by some painless means”; 51% approved of doctors acting to “assist the patient to commit suicide.”12 Majorities supporting PAS (though by smaller percentages) were reported by two other polls.13 A published study of shifts in American attitudes regarding PAS over a number of decades concluded that “Americans who reported that physician-assisted suicide should be legal rose from 37% in 1947 to 61% in 1999” and “during these years, those who agreed that physician-assisted suicide should be allowed were invariably in the majority” (Allen et al. 2006). Increasing legislative and popular support for PAS is evident as well in other countries. A bill to allow “voluntary assisted dying” was introduced in Tasmania in 2013; a bill modeled after the Oregon Death With Dignity Act was proposed in Scotland in the same year.14 A French ethics advisory body in late 2013 recommended legalization of “medical assisted suicide,”15 and a poll from 2011 showed that a majority of Canadians support the legalization of PAS16 ; at least some reports show similar numbers among Europeans in twelve nations.17 Luxembourg legalized PAS in

11

“Large Majorities Support Doctor Assisted Suicide for Terminally Ill Patients in Great Pain,” January 25, 2011. Retrieved July 3, 2014 from http://www.harrisinteractive.com/NewsRoom/Har risPolls/tabid/447/mid/1508/articleId/677/ctl/ReadCustomDefault/Default.aspx. 12 “U.S. Support for Euthanasia Hinges on How It’s Described.” Retrieved July 24, 2014 from http://www.gallup.com/poll/162815/support-euthanasia-hinges-described.aspx. 13 A HuffingtonPost/YouGov poll from 2013 reported that 50% of Americans say it should be legal for a doctor to assist the terminally ill in suicide. “Assisted Suicide Legalization Supported by Half of Americans,” June 23, 2013. Retrieved July 10, 2014 from http://www.huffingtonpost. com/2013/05/22/assisted-suicide-legalization_n_3314849.html. An NPR-Truven Health Analytics survey pegged the number of Americans in favor of assisted suicide at 55%. “Americans Support Physician-Assisted Suicide for Terminally Ill,” December 28, 2012. Retrieved July 10, 2014 from http://www.npr.org/blogs/health/2012/12/27/168150886/americans-support-physicianassisted-suicide-for-terminally-ill. 14 For the text of the bill, see http://www.scottish.parliament.uk/parliamentarybusiness/Bills/69604. aspx. 15 See “French Panel Recommends Legalizing Assisted Suicide for Terminally Ill,” Deutsche Welle, December 16, 2013. Retrieved July 1, 2014 from http://www.dw.de/french-panel-recommends-leg alizing-assisted-suicide-for-terminally-ill/a-17300805. 16 See “67% of Canadians Support Legalizing Assisted Suicide: Poll,” National Post, December 29, 2011. Retrieved July 1, 2014 from http://news.nationalpost.com/2011/12/29/67-of-canadianssupport-legalizing-assisted-suicide-poll/. 17 See “Large Europe Majorities for Assisted Suicide,” Reuters, Nov. 30, 2012. Retrieved July 1, 2014 from http://www.reuters.com/article/2012/11/30/us-euthanasia-europe-survey-idUSBRE8A T0EG20121130.

6 Clinical Ethics Consultation and Physician Assisted Suicide

101

2009 (joining Belgium, the Netherlands, and Switzerland). In early 2015 the Supreme Court of Canada invalidated laws in the country banning assisted suicide.18 These changes in law and attitudes suggest if nothing else that Mr. Smith, his family members and physicians should not regard the lawfulness of PAS in their state as anomalous, despite representing at least for now a minority legal view.

6.4.2 Institutional Policy The received view directs the ethicist’s attention to institutional or organizational policy as a further source of guidance concerning the range of allowable medical options within which parties to an ethical disagreement must choose. Not surprisingly, shifting legal conditions in respect of PAS are bringing with them changes in institutional arrangements. Naturally, most hospitals and healthcare facilities are located in jurisdictions that do not (as yet, at any rate) recognize PAS; these institutions may thus be presumed to lack policies supporting the practice. Organizational policy on end-of-life care is of course common, and it is not unusual to find facilities with protocols concerning withdrawal of treatment or use of palliative care that incorporate boilerplate language of the following sort: “nothing in this Policy and Procedure shall be construed to condone, authorize, or approve mercy killing, assisted suicide, or euthanasia.”19 Nonetheless, hospitals in states sanctioning PAS are beginning to formulate protocols and plans in anticipation of patient requests for assistance in dying. Laws in the three states with a regulatory framework for assisted suicide contain similar provisions regarding the participation of healthcare providers. None impose a duty upon a doctor, nurse, or hospital to provide access to PAS, though each insists that a facility electing not to provide assisted suicide must promulgate policies making this stance clear.20 What data is available from Oregon 18

See Pope (2009). See also Lewy (2011). See “Canada Court Strikes Down Ban on Aiding Patient Suicide,” New York Times, Feb. 6, 2015. Retrieved July 23, 2015 from http://www.nytimes.com/ 2015/02/07/world/americas/supreme-court-of-canada-overturns-bans-on-doctor-assisted-suicide. html?_r=0. 19 Taken from policy statement on “End of Life Care,” San Antonio Regional Hospital, Upland, CA. Copies available from author. 20 Relevant provisions of the Washington Death with Dignity Act are illustrative. See Revised Code of Washington, Sec. 70.245.190: “Only willing health care providers shall participate in the provision to a qualified patient of medication to end his or her life in a humane and dignified manner… A health care provider may prohibit another health care provider from participating under [the law] on the premises of the prohibiting provider if the prohibiting provider has given notice to all health care providers with privileges to practice on the premises and to the general public of the prohibiting provider’s policy regarding participating…” Retrieved July 3, 2014 from http://apps.leg.wa.gov/RCW/default.aspx?cite=70.245. See also Oregon Death with Dignity Act, Oregon Revised Statutes, Sec. 127.885 §4.01, (4) and (5a); and Vermont Statutes Annotated, Title 18 chapter 113, sec. 1.

102

D. M. Adams

and Washington—states with the longest record of implementation with respect to PAS—show that some hospitals and medical centers are supporting access to PAS, while others (notably, Catholic facilities) are not.21 The Washington State Hospital Association responded to the change in that state’s law by formulating model policies both for facilities electing to participate and those opting not to do so.22 In a much discussed piece in the New England Journal of Medicine a group of physicians at the Seattle Cancer Care Alliance—a consortium of the Fred Hutchinson Cancer Research Center, the University of Washington School of Medicine, and Seattle Children’s Hospital—described the implementation of their policy for PAS (Loggers et al. 2013). This protocol was apparently modeled on guidelines developed by Oregon Health & Science University Hospitals & Clinics in Portland.23 A study of hospice programs in Oregon found that while most refuse to participate “directly” or “actively” in PAS—by, for example, procuring or administering medications to end life—a majority do provide information about the Death With Dignity law, refer patients for further assistance, and furnish bereavement support (Campbell and Cox 2010). Lastly, it is worth noting that research in countries where PAS is legally tolerated shows a broad and growing number of institutions formulating guidelines and policies on assisted suicide. One Dutch study revealed that 70% of all healthcare facilities in the Netherlands had a written policy statement on assisted suicide, with the percentage of hospitals having such guidelines rising to 80% (Passman et al. 2009). Mr. Smith resides in one of the states where PAS is regulated by statute. Thus, while almost certainly some hospitals or clinics in his state do not allow PAS to be conducted under their auspices, it is likely that with some effort Mr. Smith would be able to locate a facility whose policies permit its physicians and nurses to assist him in obtaining the death he seeks.

21

See “Healthcare Facilities Scrambling to Deal with Dignity Act,” Seattle Times, Feb. 2, 2009. Retrieved July 3, 2014 from http://seattletimes.com/html/health/2008696811_death02m.html. See also “Island Hospital Readies Plan for Death with Dignity Act,” Skagit Valley Herald, Feb. 26, 2009. Retrieved July 3, 2014 from http://www.goskagit.com/news/island-hospital-readies-plansfor-death-with-dignity-act/article_70db5389-2bd9-52d0-91ca-e6420b12bfb6.html?mode=jqm; “Death with Dignity at WGH,” Whidbey News Times, March 3, 2009. Retrieved July 3, 2014 from http://www.whidbeynewstimes.com/news/40682408.html; “Grace Cottage Opts Out of Assisted Suicide Law—For Now,” Battleboro Reformer, June 22, 2013. Retrieved July 3, 2014 from http://www.reformer.com/localnews/ci_23515442/grace-cottage-opts-out-assisted-suicidelaw-now. “Cancer Center Goes Public with Assisted Suicide Protocol,” Amednews.com, April 22, 2013. Retrieved July 3, 2014 from http://www.amednews.com/article/20130422/profession/130 429973/2/. 22 See “Death With Dignity Act,” Washington State Hospital Association. Retrieved July 3, 2014 from http://www.wsha.org/0302.cfm. 23 “Cancer Center Goes Public with Assisted Suicide Protocol,” Amednews.com, April 22, 2013. Retrieved July 3, 2014 from http://www.amednews.com/article/20130422/profession/130429 973/2/.

6 Clinical Ethics Consultation and Physician Assisted Suicide

103

6.4.3 Professional Standards of Practice Professional standards of practice constitute a further source of guidance concerning the range of allowable options, according to the received model of ethics consultation. Such norms are evident from an inspection of generally promulgated practice guidelines and organizational codes of ethics, and are largely reflected in widespread clinical practice. The clear majority of American medical organizations and societies of healthcare professionals are on record as being opposed to the practice of PAS, though this stance is by no means unanimous; and some such bodies have explicitly refused to take a position on the allowability of PAS. The most prominent organizations, such as the American Medical Association, the College of Physicians, and the American Nurses Association have voiced long-standing opposition to assisted suicide.24 Other associations including the American Society of Pain Management Nursing, the American Medical Directors Association, and the American College of Medical Quality join them.25 At odds with these views is the position of the American Public Health Association, which “supports allowing a mentally competent, terminally ill adult to obtain a prescription for medication that the person could self-administer to control the time, place, and manner of his or her impending death…” (APHA 2008). More nuanced is the view of the Hospice and Palliative Nurses Association. While maintaining that it “opposes the legalization of assisted suicide,” the organization recognizes that “nurses have important roles in supporting patients requesting [PAS]” and, when working in states where PAS is legal, must “guard against communicating a negative judgment” on patients who elect to pursue it (HPNA 2011). Perhaps most notable is the growing number of societies adopting what amounts to a neutral stance on the moral status of PAS. Emblematic of this stand is the position staked out by the American Academy of Hospice and Palliative Medicine (AAHPM). Speaking of “physician assisted death” (PAD) the AAHPM “takes a position of ‘studied neutrality’ on the subject of whether PAD should be legally regulated or prohibited,” and cautions its physician members to “carefully scrutinize” requests for assisted suicide and proceed with “great caution before instituting PAD…” (AAHPM 2007). The American Academy of Physician Assistants argues that though “liberalization of assisted suicide could have unintended consequences, so too could absolute restriction on assisted suicide” (AAPA 1997). While the AAPA does not advocate PAS, it offers “decriminalization with well-defined guidelines” as one solution and insists that “the ethical, compassionate, well-intentioned provider who discusses voluntary self-termination” with a capable patient must not be subject to legal liability (AAPA 1997). The American Pharmacists Association “supports 24

See American Medical Association (2012), Opinion 2.211. Available at http://www.ama-assn. org/ama/pub/physician-resources/medical-ethics/code-medical-ethics/opinion2211.page? See also American College of Physicians (2012), “Care of Patients Near the End of Life: Physician-Assisted Suicide and Euthanasia”; American Nurses Association (2013). 25 See American Society for Pain Management Nursing (2003), American Medical Director’s Association (1997), American College of Medical Quality (2000).

104

D. M. Adams

informed decision-making based upon the professional judgment of pharmacists, rather than endorsing a particular moral stance on the issue of physician-assisted suicide,” but “opposes laws and regulations which mandate or prohibit the participation of pharmacists in physician-assisted suicide” (APA 2004). The American Society of Health-System Pharmacists affirms the “right of a pharmacist to participate or not in morally, religiously, or ethically troubling therapies,” while insisting pharmacists must apprise patients of “all legally available treatment options” (ASHSP 1999). Concerning PAS, the National League for Nursing adheres to the “belief that a single position on the part of this organization would not be helpful at this time” (NLN 1999). Among hospice nurses and social workers surveyed in Oregon, 95% indicated that their hospice agency should either support a patient’s choice for PAS, or at least remain neutral (see Miller et al. 2004). Some prominent ethicists, such as Timothy Quill, encourage the adoption of such a neutral position (Quill and Cassell 2003). State medical societies in jurisdictions sanctioning PAS by law appear as well to be opting for this neutral view, neither opposing nor supporting PAS. The Oregon Medical Association states that its members “may choose whether to participate” in PAS, and it has made available checklists and forms to be used by doctors helping their patients under the state’s Death with Dignity legislation (OMA 2011). The Vermont Medical Society refers the public and its members to a policy adopted in 2003. According to the Society, that policy “does not support laws for or against physician assisted suicide.”26 The Washington State Medical Association has posted a “Physician’s Guide” to that state’s “Death with Dignity Law” containing information and links to documents, but with neither condemnation nor endorsement of assisted suicide (WSMA 2009). In May 2015, the California Medical Association dropped its opposition to PAS in the face of widespread support for proposed legislation that would legalize PAS in the state. The CMA is the largest state-wide medical association to adopt a position of neutrality.27 Though he is undertaking to obtain PAS in a state recognizing it by law, it is reasonable to suppose Mr. Smith will encounter resistance to his plan by practitioners whose attitudes have been shaped in part by professional opposition to assisted suicide, though he and his family may well find local organizations of healthcare providers unopposed to his intentions.

26

See Vermont Medical Society (2003). The text of the Society’s policy states “there should be no laws concerning physician-assisted suicide.”. 27 See “CMA changes stance on physician aid in dying, takes neutral position on end of life option act.” Retrieved July 23, 2015 from http://www.cmanet.org/news/detail/?article=cma-changes-sta nce-on-physician-aid-in-dying.

6 Clinical Ethics Consultation and Physician Assisted Suicide

105

6.4.4 The Bioethics Literature The received view insists that agreements facilitated with or recommendations given by a clinical ethicist must “comport with the bioethics literature” and “other relevant scholarly literature,” in addition to statutes, judicial opinions, and pertinent institutional policies (ASBH 2011, 6). This presumably means looking for positions broadly supported in published bioethics research. It may appear surprising to assume any body of literature as voluminous as that in bioethics to speak as if with one voice on any subject, though it is in fact possible to identify some widely endorsed views. For example, bioethics scholars now overwhelmingly support the proposition (defended extensively by the 1983 President’s Commission) that decisionally capable patients have a right to refuse unwanted medical treatment of whatever kind (President’s Commission for the Study of Ethical Problems in Medicine and Biomedical and Behavioral Research 1983). But predictably, matters are by no mean so clear respecting PAS. Since it looks for positions generally supported in the scholarly literature, the received view assigns some weight to the distribution of professional opinion on matters like assisted suicide. It is therefore relevant to make two observations about that literature in this connection. It is large. And it is deeply split, as a brief (but far from exhaustive) list of prominent bioethicists on opposing sides of the PAS question attests. Among those arguing in support of PAS are Angell (1982), Battin (2005), Brock (2004), Caplan (2012), Dworkin and Frey (1998), Dworkin et al. (1997) (along with Thomas Nagel, Robert Nozick, John Rawls, Thomas Scanlon, and Judith Jarvis Thomson), Kamm (1998), Miller and Truog (2012), Quill (2008), Singer (2005) and Sumner (2011). The ranks of those opposing PAS include Arras (1998), Bok (1998), Callahan (2002), Cohn and Lynn (2002), Dorff (1998), Kass (2002), Marquis (1998), Meilaender (2005), and Pellegrino (2002). Unsurprisingly, this “bioethics literature” enunciates no overall position respecting PAS. Pursuant to it, assisted suicide is both inconsistent and fully compatible with the aims of medicine; both absolutely distinct from and decidedly no different than death following removal of life-sustaining interventions; both too dangerous to the vulnerable ever to permit and acceptably safe when properly regulated; both a dignified way to alleviate misery and a dehumanizing ploy to silence suffering by eliminating the sufferer; both a necessary means to address intolerable pain and an unnecessary over-reaction to inadequate palliative treatment. A recent editorial in the New England Journal of Medicine is emblematic of the current state of the PAS literature, with the two contributors taking entirely opposing positions on whether assisted suicide should be permitted (Boudreau and Sumerville 2013; Biller-Andorno 2013). We can therefore be confident that Mr. Smith, his family, and physicians could consult the scholarship and there find a wealth of arguments, objections, and counterarguments respecting PAS, though without any prospect of discovering what the “bioethics literature,” as a corporate body, says about the allowability of PAS.

106

D. M. Adams

6.5 The Unsettled Status of PAS Let me recap my argument thus far. Confronted with a man determined to forgo further treatment for his disease and leave the care of his doctors to seek assisted suicide, his family and care providers are troubled and divided over the ethics of PAS. Having been contacted by the patient and his family, the clinical ethicist prepares to follow the prevailing conception of the ethicist’s role, endeavoring to facilitate a dialogue with the disputing parties over the options open to them. The ethicist recognizes this discussion must be confined to courses of action allowed by the combined operation of several sources of normativity. The ethicist in our case thus undertakes an investigation, the fruits of which we can assume are summarized in the analysis of Sect. 6.4. But what now is the ethicist to say to the question he or she was originally asked: Is PAS ethical in Mr. Smith’s case? On the received view this question must be reformulated: Does PAS fall within the range of allowable options open to Mr. Smith? The question plainly assumes a clear answer will be forthcoming. Just as obviously, however, this assumption is untenable. The only reasonable conclusion to draw from the foregoing investigation is that current laws, policies, professional standards, and relevant literature, when taken together, say nothing definitive about the normative status of PAS. There simply is no settled answer reflecting the aggregate input of the sources of normativity upon which the allowability of a course of action is supposed to depend. As we saw, some jurisdictions (including Smith’s) unequivocally permit PAS by law; and while laws in the great majority of jurisdictions forbid it, the number that do is slowly decreasing, perhaps a reflection of the fact that majority public opinion appears now supportive of PAS. Proposals to permit assisted suicide are regularly being introduced in legislatures both in the U.S. and elsewhere. Institutional policies generally track the law, but as observed above there is evidence of such policies being modified to accommodate PAS in jurisdictions where it is legal. Many professional bodies still prohibit healthcare providers from participating in PAS, though some organizations have adopted positions of “studied neutrality,” while others encourage their members to act as their conscience dictates. The bioethics literature is quite simply all over the place, reflecting deep divisions of opinion among scholars about the permissibility of PAS and the wisdom of its legalization. It may be tempting to suppose that this state of irresolution—the lack of a settled normative status—is a singular result, a consequence of the uniquely divisive nature of the debate over assisted death. But to suppose this would be a mistake. Feeding tubes for infants with “incompatible with life” anomalies; use of palliative sedation; organ donation “after circulatory determination of death”; hemodialysis for octogenarians suffering from end-stage conditions; deactivation of implanted pacemakers—there are many disputed practices and procedures in clinical biomedicine. As William Winslade and I have argued, PAS is far from the only controversial matter currently dividing bioethicists, members of the medical profession, and the wider public in a way that leaves its normative status unsettled (Adams and Winslade 2011). Naturally, some of these contentious issues may become normatively settled

6 Clinical Ethics Consultation and Physician Assisted Suicide

107

over time; and it is equally possible that some now settled areas of moral, professional, and legal agreement may become destabilized and eventually break down. (We may be witness to the beginnings of such destabilization with recent challenges to the notion of “death by neurological criteria” or brain death.)28 There are at least two general reasons for thinking that proper identification of what I am calling the normative status of difficult medical choices is not likely to be the straightforward matter it is presumed to be on the received view. First, and as we have seen, whether a clinical choice is allowable is supposed to be a function of several sources of normativity acting together. Yet the received view says nothing about how this overall sum is supposed to be arrived at or how the results vouchsafed by these sources are to be combined when they pull in different directions. Nor, second, does it assign comparative weights to the various sources of normativity to which it directs the ethicist’s attention. When, for example, two or more sources of normativity give conflicting answers, which is to take priority? These seem to be serious deficiencies, and it is not immediately clear how proponents of the received view could easily rectify them.

6.6 The Role of the Clinical Ethicist We arrive then at this question: How is the ethicist to respond when the normative status of a clinical course of action he or she has been called upon to clarify is unsettled? What ought to be expected of a hospital ethicist when questioned about the normative status of a contentious practice such as PAS? I have argued that the conceptual resources of the received view neither isolate the normative status of PAS nor enable the ethicist in Smith’s case correctly to discover it. (Indeed, if I am right there is nothing definitive to discover.) Yet at the same time we are told that clinical ethics consultants are ethics experts equipped with knowledge allowing them to offer “expert opinions, at least ones consistent with established ethical standards” (Tarzian and ASBH Core Competencies Update Task Force 2013, 5). Such avowals surely give rise to an expectation that ethicists will have something to say about puzzlement over the normative status of particular clinical decisions. But what could it mean for someone to give expert ethics input regarding a disputable moral matter such as PAS? We seem to have reached something of an impasse. There are at least two ways to address this impasse. I will reject the first and endorse the second. To begin with, we could imagine an ethicist responding to questions about the allowability of PAS by first informing the patient and other involved parties that their disagreement cannot be resolved because the normative status of assisted suicide remains indeterminate. The ethicist then indicates that by so informing them she has 28

See, e.g., “Jahi McMath Case Renews Moral Debate Over Brain Death Diagnoses,” National Catholic Register, January 14, 2014. Retrieved August 21, 2014 from http://www.ncregister. com/daily-news/jahi-mcmath-case-renews-moral-debate-over-brain-death-diagnoses/-ixzz3B4Ma 64GQ. See also Miller and Truog (2014).

108

D. M. Adams

discharged her responsibilities and can be of no further service. Now, in what sense hospital ethicists are experts is an issue far too large to take up here. But in whatever such expertise exactly consists, it is surely abandoned by an ethicist acting in the manner just described. We can see this by contrasting the ethicist’s response with a similar one by a different kind of specialist. Suppose Mr. Smith has a brain tumor. A neurosurgeon is consulted and talks with the patient and his family. “My job,” says the specialist, “is to present you with a range of potential treatments, including possible surgical options based upon my neurosurgical expertise and training. Having studied the situation carefully, however, I believe there is no clear answer to which surgical option is best. Therefore, I can be of no further help to you.” Surely something is badly amiss here, and the patient had better find another surgeon. For plainly the expertise of the neurosurgeon must consist in at least this much: instead of signingoff the case, he should engage further with the patient, laying out the considerations for and against possible treatment modalities with the aim of arriving along with the patient at agreement upon a treatment that seems best supported in his or her particular case. I want to suggest that something similar must be true where an ethics expert or specialist confronts uncertainty over the allowability of practices like assisted suicide. The basic idea would be that in situations where the range of allowable choices is not clear the ethicist should engage the disputing parties in a process of moral inquiry and deliberation with the aim of help them determine for themselves what ought to be regarded as allowable. The ethicist does this by inviting them to participate in a deeper moral inquiry—evaluating arguments, probing distinctions, questioning assumptions—thereby working toward the goal of articulating reasons and arguments that could function as a shared basis for agreement. In this way the ethicist can assist the parties to resolve their moral disagreement on the basis of ethically wellinformed deliberations. This suggestion seeks to take seriously the alleged expertise of the consultant—his or her status as an ethics specialist. A medical specialist is called into a case with the expectation he or she will provide the requesting parties with a more detailed account of the medical options open to them, permitting the attending physician and patient (or surrogate) to make a better, more informed and carefully reasoned choice. In the same way, when disagreement occurs concerning the normative status of a moral choice, the allowability of which is unsettled, the consulting ethicist should help those involved arrive at a better understanding of the contested moral territory and of the reasons that can be advanced for and against treating a given choice as allowable. Thinking about ethics consultation in unsettled cases as an engagement in moral inquiry and deliberation underscores the need to reconsider the basic goal of clinical ethics consultation and, along with that, the nature of the ethicist’s expertise. It also points up questions for further investigation. The received view instructs the ethicist in Mr. Smith’s case to resolve value uncertainty and conflict. On the face of it this is indeed a tall order—for given the normatively unsettled state of affairs regarding assisted suicide it is not entirely clear what could count as success in fulfilling it. It hardly makes sense to expect Mr. Smith’s ethicist to settle definitively the long-running debate over the morality of assisted suicide (or, indeed, of any

6 Clinical Ethics Consultation and Physician Assisted Suicide

109

other normatively unsettled bioethical issue). The immediate need is of course to clear up the conflict between Mr. Smith, members of his family, and the treating team. If the source of their antagonism were poor communication or interpersonal friction, then the obvious goal would be to address that tension with the aid of a social worker, ombudsperson, or perhaps a therapist; should they be of a particular religious background and their differences were over the interpretation of scripture as it relates to end-of-life decisions, the intervention of a chaplain would be called for. But none of these matters appear to be what is chiefly at stake in the case of Mr. Smith. The root of dissension in Smith’s case, we are to suppose, is moral disagreement. Since the normative status of PAS is unsettled and the contending parties have appealed to deeply contested ideas—for example, about causing death, the aims of medicine, the duty to alleviate suffering—the only meaningful goal of an ethics consultation can be to ensure that whatever decision is arrived at—either by the group as a whole or by Mr. Smith acting on his own—is the best ethically informed decision possible under the circumstances. The ethicist should help the parties achieve a deeper understanding of what their respective moral beliefs and commitments entail, how they conflict, and the reasons that can be given to support or contest them. For the work of the ethicist to be genuinely a form of ethics consultation (as opposed to pastoral counseling, psychotherapy, or dispute resolution) it must include an examination of the parties’ moral beliefs and convictions. Mr. Smith’s physicians are generally opposed to his plans to seek PAS. Do they believe all cases of intentionally assisting to bring about the death are wrong? If so, how do they defend intentionally removing critically ill people from life-sustaining interventions? Does discontinuing such interventions count as taking a life? Mr. Smith thinks others are failing to honor a duty of compassion by refusing to assist him. But is this a duty to relieve suffering irrespective of the means involved? Would compliance with such a duty bring Smith’s doctors into conflict with another duty, to refrain from causing harm? Mr. Smith’s wife seemingly distinguishes amongst reasons for pursuing assisted suicide. Yet which reasons are licit and which not? The patient’s plea underscores his suffering. But if suffering is an “afflicted state of being” (Cassell 2004), are all of the ways in which an individual suffers equally the business of medicine to address? Most of the doctors insist that healing the sick is their primary moral duty. What does it mean, though, to heal an obviously dying man? Smith himself evidently sees PAS as a release from a dehumanizing ordeal. Does assisted suicide put an end to such debasement or serve as its tragic culmination? Naturally, many more equally weighty and challenging questions lie just beneath the surface of the confusion caused by Smith’s declared intention to end his life. But perhaps enough has been said to demonstrate the salience of such serious moral questions and the need straightforwardly to confront them—or at least some of them—as indispensible to disciplined reflection on a factious issue. Such reflection affords at least some assurance that whatever decision is reached is informed by the best moral reasons and arguments the parties can identify. The role I have just sketched for ethics consultation in unsettled cases is a richer and more convincing conception of the consultant’s ethics expertise than is accepted

110

D. M. Adams

by the received view; for on that view such expertise is confined to knowledge of what is and isn’t generally accepted or allowed, and the ethicist’s role limited to relating the implications of well-established options for the case at hand. This attenuated understanding of ethics and its relevance to controversies in the clinic is problematic for two reasons. First, it is unhelpful in just those cases where ethical reflection is most needed. When what is generally accepted cannot with confidence be discovered, attention needs to turn more fundamentally to reflection on what should be regarded as allowed—and this requires careful reconstruction and evaluation of competing arguments, analysis of critical distinctions, and interrogation of pivotal assumptions. Second, to insist that only those medical options the acceptability of which is broadly recognized should be supported by an ethicist—in other words, that principles and distinctions embedded in law, policy, and practice must be respected—is to embrace a conservatism inconsistent with serious moral reflection and careful moral reasoning. Return once more to Smith’s case. His doctors assert that actions taken when death is intended are distinct from conduct foreseeably leading to a patient’s demise. Is this distinction truly tenable? Unquestionably, it is frequently cited in codes of ethics and hospital policies. But moral reasoning that is rigorous and discerning cannot begin by assuming that because certain judgments have gained traction in professional codes and bioethical discussion they are supportable by the best arguments. It is likely the case that the best ethically informed decision about PAS—or about any other practice whose normative status remains unsettled—will be the product of a wide-ranging discussion, one that not only examines but challenges claims or ideas largely accepted.

6.7 Conclusion The received view of ethics consultation fails to accomplish its stated goal of resolving the conflict dividing Mr. Smith, his family, and doctors. This is because genuine moral disagreements, like that over PAS, lack the settled normative status indispensable on the received view to the attenuated role assigned to the ethicist. In addition, there appears to be no accepted procedure for deriving clear conclusions about the normative status of a given clinical practice or decision from a set of disparate and imprecise normative sources. I have tried to illustrate each of these concerns in the forgoing, though it should be clear that many questions remain. What, for instance, should an ethicist report about the normative status of assisted suicide in a jurisdiction not currently extending it legal recognition? Is the illegality of a practice a decisive reason for concluding it is disallowed? Suppose that despite the law, the state’s medical societies and major medical centers come out in support of PAS. What then is its normative status? Could a clinical decision be widely viewed as (morally) impermissible yet remain allowable because it is both legal and supported by a significant body of bioethical scholarship? (Might certain forms of abortion be an example?).

6 Clinical Ethics Consultation and Physician Assisted Suicide

111

I have argued that moral disagreement over particular medical decisions occurring in contexts where legal, professional, and institutional practices are at odds over the acceptability of those decisions, requires a new understanding of the aims of ethics consultation—a conception of the consulting ethicist’s role differing markedly from how that role is constructed under what I have called the received view. The revised understanding of ethics consultation I have sketched is by no means complete, however, either in its theoretical underpinnings or in its practical applications. A number of further questions await investigation, among which are these. Is an ethicist a “moral authority,” in the way (for example) a neurosurgeon is an authority in operating on the brain? May an ethicist give moral advice reflecting his or her own moral judgment, as opposed simply to reporting what is generally regarded as acceptable or unacceptable? What responsibility does the ethicist have to share minority positions—views dissenting from established standards or scholarly consensus? If as I suggest ethics expertise consists in helping others to reason better about the moral questions confronting them, what does this imply about the training of clinical ethics consultants? Must they, for example, be moral philosophers? If the parties to a clinical moral disagreement aim to resolve their differences by reaching a consensus to which they are collectively bound, how must they justify their decision? Ought they, for example, reach agreement (assuming they can) only by appeal to reasons that all can accept? These questions need to be taken up and carefully discussed by bioethicists before a revised conception of ethics consultation along the lines I have proposed can be developed, and before more effective ways of responding to unsettled bioethical disputes can be implemented. I have not undertaken that work in this paper, but I hope here at least to have begun the conversation.

References Ackerman, Terrence F. 1987. The role of an ethicist in health care. In Health care ethics: A guide for decision makers, ed. Gary R. Anderson, and Valerie A. Glesnes-Anderson, 309–320. Gaithersburg: Aspen Publications. Adams, David M. 2009. Ethics consultation and ‘facilitated’ consensus. Journal of Clinical Ethics 20: 44–55. Adams, David M., and William J. Winslade. 2011. Consensus, clinical decision making, and unsettled cases. Journal of Clinical Ethics 22: 310–327. Allen, Jen, Sonia Chavez, Sara DeSimone, Debbie Howard, Keadron Johnson, Lucinda LaPierre, Darrel Montero, and Jerry Sanders. 2006. Americans’ attitudes toward euthanasia and physicianassisted suicide, 1936–2002. Journal of Sociology and Social Welfare 33: 5–23. American Academy of Hospice and Palliative Medicine (AAHPM). 2007. Statement on physicianassisted death. http://aahpm.org/positions/pad. Accessed 7 July 2014. American Academy of Physician Assistants (AAPA). 1997/2009. End-of-life decision making. http://www.aapa.org/workarea/downloadasset.aspx?id=812. Accessed 2 August 2014. American College of Medical Quality (ACMQ). 2000. Policy 34: Physician-assisted suicide and end of life care. http://www.acmq.org/policies/policy34.pdf. Accessed 7 July 2014.

112

D. M. Adams

American College of Physicians (ACP). 2012. Care of patients near the end of life: Physicianassisted suicide and euthanasia. In Ethics manual, 6th ed. http://www.acponline.org/runningpr actice/ethics/manual/manual6th.htm-euthanasia. Accessed 7 July 2014. American Medical Association (AMA), Council on ethical and judicial affairs. 2012. Code of medical ethics. Opinion 2.211. http://www.ama-assn.org/ama/pub/physician-resources/medicalethics/code-medical-ethics/opinion2211.page? Accessed 2 July 2014. American Medical Director’s Association (AMDA). 1997. Position statement on care at the end of life. http://www.amda.com/governance/resolutions/p97.cfm. Accessed 7 July 2014. American Nurses Association (ANA). 2013. Position statement: Euthanasia, assisted suicide, and aid in dying. http://www.nursingworld.org/euthanasiaanddying. Accessed 7 July 2014. American Pharmacists Association (APA). 2004. Policy: Physician-assisted suicide. http://www. pharmacist.com/policy/physician-assisted-suicide-13. Accessed 7 July 2014. American Public Health Association (APHA). 2008. Patients’ rights to self-determination at the endof-life. Policy 20086. http://www.apha.org/advocacy/policy/policysearch/default.htm?id=1372. Accessed 7 July 2014. American Society for Bioethics and Humanities (ASBH). 1998. Core competencies for health care ethics consultation. American Society for Bioethics and Humanities (ASBH). 2011. Core competencies for health care ethics consultation, 2nd ed. American Society of Health-System Pharmacists (ASHSP). 1999. Statement on pharmacist’s decision-making on assisted suicide. http://www.ashp.org/DocLibrary/BestPractices/EthicsStS uicide.aspx. Accessed 7 July 2014. American Society for Pain Management Nursing (ASPMN). 2003. Position statement on assisted suicide. http://www.aspmn.org/pdfs/assistedsuicide.pdf. Accessed 7 July 2014. Angell, Marcia. 1982. The quality of mercy. The New England Journal of Medicine 306: 98–99. Arras, John D. 1998. Physician assisted suicide: A tragic view. In Physician assisted suicide: Expanding the debate, ed. Margaret Pabst Battin, Rosamond Rhodes, and Anita Silvers, 279–300. New York: Routledge. Aulisio, Mark P., and Robert M. Arnold. 2000. Health care ethics consultation: Nature, goals, and competencies. Annals of Internal Medicine 133: 59–70. Aulisio, Mark P., and Robert M. Arnold. 2008. Role of the ethics committee: Helping to address value conflicts or uncertainties. Chest 134: 417–424. Aulisio, Mark P., Robert M. Arnold, and Stuart J. Younger. 1999. Moving the conversation forward. Journal of Clinical Ethics 10: 49–56. Aulisio, Mark P., Robert M. Arnold, and Stuart J. Younger. 2003. Ethics consultation: From theory to practice. Baltimore: The Johns Hopkins University Press. Battin, Margaret Pabst. 2005. Ending life: Ethics and the way we die. Oxford: Oxford University Press. Biller-Andorno, Nikola. 2013. Physician-assisted suicide should be permitted. New England Journal of Medicine 368: 1451–1452. Bok, Sissela. 1998. Physician-assisted suicide. In Euthanasia & physician assisted suicide: For and against, ed. Gerald Dworkin, Raymond G. Frey, and Sissela Bok, 128–139. Cambridge: Cambridge University Press. Boudreau, Donald J., and Margaret A. Somerville. 2013. Physician-assisted suicide should not be permitted. New England Journal of Medicine 368: 1450–1451. Brock, Dan W. 2004. Physician assisted suicide as a last resort option. In Physician-assisted dying: The case for palliative care and patient choice, ed. Timothy E. Quill, and Margaret Pabst Battin, 130–149. Baltimore: The Johns Hopkins Press. Callahan, Daniel. 2002. Reason, self-determination, and physician-assisted suicide. In The case against assisted suicide, ed. Kathleen Foley, and Herbert Hendin, 52–68. Baltimore: The Johns Hopkins University Press. Campbell, Courtney S., and Jessica C. Cox. 2010. Hospice and physician-assisted death. Hastings Center Report 40: 26–35.

6 Clinical Ethics Consultation and Physician Assisted Suicide

113

Caplan, Arthur. 2012. Massachusetts should legalize physician-assisted suicide. Harvard Law Petrie-Flom Center. http://blogs.law.harvard.edu/billofhealth/2012/11/02/art-caplan-ma-shouldlegalize-physician-assisted-suicide/. Accessed 25 July 2014. Cassell, Eric T. 2004. When suffering patients seek death. In Physician-assisted dying: The case for palliative care and patient choice, ed. Timothy E. Quill, and Margaret Pabst Battin, 75–89. Baltimore: The Johns Hopkins Press. Cohn, Felicia, and Joanne Lynn. 2002. Vulnerable people: Practical rejoinders to claims in favor of assisted suicide. In The case against assisted suicide, ed. Kathleen M. Foley, and Herbert Hendin, 238–260. Baltimore: The Johns Hopkins University Press. DeRenzo, Evan G. 1994. Providing clinical ethics consultation. HEC Forum 6: 384–389. Dorff, Elliott N. 1998. Matters of life and death: A Jewish approach to modern medical ethics. Philadelphia: Jewish Publication Society. Dubler, Nancy N. 2011. A ‘Principled Resolution’: The fulcrum for bioethics mediation. Law and Contemporary Problems, Summer 2011: 177–200. Dubler, Nancy N., Mayris P. Webber, and Deborah M. Swiderski. 2009. Charting the future: Credentialing, privileging, quality, and evaluation in clinical ethics consultation. Hastings Center Report 39: 23–33. Dworkin, Gerald, and Raymond G. Frey. 1998. The nature of medicine. In Euthanasia and physician assisted suicide: For and against, ed. Gerald Dworkin, Raymond G. Frey, and Sissela Bok Dworkin, 6–16. Cambridge: Cambridge University Press. Dworkin, Ronald, Thomas Nagel, Robert Nozick, John Rawls, Thomas Scanlon, and Judith Jarvis Thomson. 1997. Assisted suicide: The philosophers’ brief. The New York Review of Books 27: 41–47. Fox, Ellen, Sarah Myers, and Robert A. Pearlman. 2007. Ethics consultation in United States hospitals: A national survey. American Journal of Bioethics 7: 13–25. Glover, Jacqueline J., David T. Ozar, and David C. Thomasma. 1986. Teaching ethics on rounds: The ethicist as teacher, consultant, and decision-maker. Theoretical Medicine 7: 13–32. Hospice and Palliative Nurses Association (HPNA). 2011. Position statement: Legalization of assisted suicide. https://www.hpna.org/DisplayPage.aspx?Title1=PositionStatements. Accessed 7 July 2014. Kamm, Frances M. 1998. Physician-assisted suicide, euthanasia, and intending death. In Physician assisted suicide: Expanding the debate, ed. Margaret Pabst Battin, Rosamond Rhodes, and Anita Silvers, 28–62. New York: Routledge. Kass, Leonard R. 2002. ‘I will give no deadly drug’: Why doctors must not kill. In The case against assisted suicide, ed. Kathleen Foley and Herbert Hendin, 17–40. Baltimore: The Johns Hopkins University Press. Lewy, Guenter. 2011. Assisted death in Europe and America. Oxford: Oxford University Press. Loggers, Elizabeth Trice, Helene Starks, Moreen Shannon-Dudley, Anthony L. Back, Frederick R. Appelbaum, and F. Marc Stewart. 2013. Implementing a death with dignity program at a comprehensive cancer center. New England Journal of Medicine 368: 1417–1424. Marquis, Don. 1998. The weakness of the case for legalizing physician assisted suicide. In Physician assisted suicide: Expanding the debate, ed. Margaret Pabst Battin, Rosamond Rhodes, and Anita Silvers, 267–278. New York: Routledge. McLean, Sheila A.M. 2009. Clinical ethics consultation in the United Kingdom. Diametros 22: 76–89. Meilaender, Gilbert. 2005. Bioethics: A primer for Christians, 2nd ed. Grand Rapids: William B. Eerdmans Publishing Co. Miller, Franklin G., and Robert D. Truog. 2012. Death, dying, and organ transplantation. Oxford: Oxford University Press. Miller, Franklin G., and Robert D. Truog. 2014. Bioethics and the dogma of brain death. Bioethics Forum, 3 February 2014. http://www.thehastingscenter.org/Bioethicsforum/Post.aspx?id=6769& blogid=140. Accessed 21 August 2014.

114

D. M. Adams

Miller, Lois L., Theresa A. Harvath, Linda Ganzini, Elizabeth R. Goy, Molly A. Delorit, and Ann Jackson. 2004. Attitudes and experiences of Oregon hospice nurses and social workers regarding assisted suicide. Palliative Medicine 18: 685–691. Nagao, Noriko, Mark P. Aulisio, Yoshio Nukaga, Misao Fujita, Shinji Kosugi, Stuart Youngner, and Akira Akabayashi. 2008. Clinical ethics consultation: Examining how American and Japanese experts analyze an Alzheimer’s case. BMC Medical Ethics 9. http://www.biomedcentral.com/con tent/pdf/1472-6939-9-2.pdf. Accessed 16 July 2014. National League for Nursing (NLN). 1999. Life-terminating choices: A framework for nursing decision-making. http://www.nln.org/aboutnln/info-suicide.htm. Accessed 7 July 2014. Oregon Medical Association (OMA). 2011. ‘PAS patient form’ and ‘PAS compliance checklist.’ http://www.theoma.org/. Accessed 7 July 2014. Pasman, H. Roeline W., Johanna E. Hanssen-de Wolf, Berniek A.M. Hesselink, Agnes van der Heide, Gerrit van der Wal, Paul J. van der Maas, and Bregje D. Onwuteaka-Philipsen. 2009. Policy statements and practice guidelines for medical end-of-life decisions in Dutch health care institutions: Development in the past decade. Health Policy 92: 79–88. Pellegrino, Edmund D. 2002. Compassion is not enough. In The case against assisted suicide, ed. Kathleen M. Foley, and Herbert Hendin, 41–51. Baltimore: The Johns Hopkins University Press. Pope, Thaddeus M. 2009. Legal briefing: Medical futility and assisted suicide. Journal of Clinical Ethics 29: 274–286. President’s Commission for the Study of Ethical Problems in Medicine and Biomedical and Behavioral Research. 1983. Deciding to forego life-sustaining treatment. https://repository.library.geo rgetown.edu/handle/10822/559344. Accessed 29 April 2015. Quill, Timothy E. 2008. Physician-assisted death in the United States: Are the existing ‘last resorts’ enough? Hastings Center Report 38: 1–22. Quill, Timothy E., and Christine K. Cassel. 2003. Professional organizations’ position statements on physician-assisted suicide: A case for studied neutrality. Annals of Internal Medicine 138: 208–211. Reiter-Theil, Stella. 2001. Ethics consultation in Germany: The present situation. HEC Forum 13: 265–280. Schildmann, Jan, John-Stewart Gordon, and Jochen Vollmann, eds. 2010. Clinical ethics consultation. Surrey: Ashgate Publishing. Singer, Peter. 2005. Decisions about death. Free Inquiry 25. https://www.secularhumanism.org/ index.php/cont_index_25. Accessed 17 July 2014. Sumner, L.W. 2011. Assisted death: A study in ethics and law. Oxford: Oxford University Press. Tarzian, Anita J., and ASBH Core Competencies Update Task Force. 2013. Health care ethics consultation: An update on core competencies and emerging standards from the American Society for Bioethics and Humanities’ Core Competencies Update Task Force. American Journal of Bioethics 13: 3–13. Tarzian, Anita J., and Lucia D. Wocial. 2015. A code of ethics for health care ethics consultants. The American Journal of Bioethics 15: 38–51. Thomasma, David C. 1987. Legitimate and illegitimate roles for the medical ethicist. In Clinical medical ethics: Exploration and assessment, ed. Terrence F. Ackerman, Glenn C. Graber, Charles H. Reynolds, and David C. Thomasma. Lanham: University Press of America. Veatch, Robert M. 1987. The medical ethicist as agent for the patient. In Clinical medical ethics: Exploration and assessment, ed. Terrence F. Ackerman, Glenn C. Graber, Charles H. Reynolds, and David C. Thomasma. Lanham: University Press of America. Vermont Medical Society (VMS). 2003. Physician assisted suicide. http://www.vtmd.org/physic ian-assisted-suicide-likely-top-priority-2011. Accessed 7 July 2014. Washington State Medical Association (WSMA). 2009. Physicians’ guide to health law: The Death with Dignity Act. https://www.wsma.org/doc_library/LegalResourceCenter/RulesRegu latoryInformation/Rules_Washington_Physicians_Guide_DWDA_20120511-1.pdf. Accessed 7 July 2014.

6 Clinical Ethics Consultation and Physician Assisted Suicide

115

David M. Adams is Emeritus Professor of Philosophy at California State Polytechnic University, Pomona and Clinical Ethicist at Pomona Valley Hospital Medical Center. His recent research has focused on problems in clinical medical ethics.

Chapter 7

License to Kill: A New Model for Excusing Medically Assisted Dying? Richard Huxtable and Jonathan Ives

Abstract In this chapter, we seek to offer a fresh perspective on whether or not doctors should be “licensed to kill”. As that phrase indicates, we metaphorically refer to the adventures of fictional spy James Bond, although we hope, in doing so, that readers will not think that we are belittling the serious topic with which the chapter is concerned. Having surveyed some of the familiar arguments for and against allowing medically-assisted dying, we advance a new proposal, which seeks to strike a compromise between such arguments. Our exploratory model builds on an analogy with armed response police units as they operate in England, and specifically with situations in which such officers are found to have killed someone in the course of their duties. Rather than confer immunity prospectively (as proponents of assisted dying often seek) or deny immunity completely (as opponents insist), our policy would provide a potential retrospective excuse. Keywords Slippery slope · Criminal sanction · Assisted suicide · Autonomous choice · Excuses

7.1 Introduction: A View to a Kill Gaylin et al. (1998, 2140) have suggested that physician-assisted dying: touches medicine at its very moral center; if this moral center collapses, if physicians become killers or are even merely licensed to kill, the profession – and, therewith each physician – will never again be worthy of trust and respect as healer and protector of life in all its frailty.

R. Huxtable (B) · J. Ives Centre for Ethics in Medicine, University of Bristol, Bristol, UK e-mail: [email protected] J. Ives e-mail: [email protected] © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 M. Cholbi and J. Varelius (eds.), New Directions in the Ethics of Assisted Suicide and Euthanasia, The International Library of Bioethics 103, https://doi.org/10.1007/978-3-031-25315-7_7

117

118

R. Huxtable and J. Ives

Are Gaylin et al. right to be concerned, or should doctors be “licensed to kill”? The idea of a “licence to kill” conjures the image of James Bond, the fictional British spy, whose adventures are featured in an internationally-renowned series of books and films. We deploy a Bondian metaphor throughout this paper, which is linked to the analogy we use to moot a new direction, and we hope, in doing so, that readers will not think that we are belittling the serious topic with which the paper is concerned. Opinions understandably differ on this complex, sensitive phenomenon and specifically about the appropriateness of such killing. All too often the disputants reach a familiar crossroads: one fork points towards permission, in view of the alleged justifiability of the practice, the other towards prohibition, in view of its alleged unjustifiability. As this volume suggests, new directions of travel are needed, which are capable of leading us away from these well-trodden and entrenched paths. In this chapter, therefore, we seek to forge and explore a different path. Our exploratory model builds on an analogy with armed response police units as they operate in England, and specifically with situations in which such officers are found to have killed someone in the course of their duties. Rather than confer immunity prospectively (as proponents of assisted dying often seek) or deny immunity completely (as opponents insist), our policy would provide a potential retrospective excuse. We start by surveying the main features of the debates to date. We first look to the advocates of assisted dying, who argue that there are obligations to respect autonomous choices and to relieve suffering, which suggest that assistance in dying is justifiable. We then, however, encounter the opponents. Opponents will typically point to the intrinsic value of human life, arguing that life should never intentionally be shortened. They also argue that, in principle and in practice, a policy of permission is dangerous, since it will lead us onto slippery slopes. Finally, as Gaylin et al. suggested, opponents will argue that doctors should not be involved in any such policy because it will erode trust in the profession. In many of the permissive laws that are proposed, as well as in many of those already in operation, doctors are typically positioned as the providers of assistance in dying. Opponents fear the affront to the integrity of medicine that such a development poses. Each set of arguments has a long history and continues to command substantial support; indeed, the to-and-fro of argument, counter-argument and counter-counterargument appears to be unrelenting. Yet, despite this history, we still appear to lack a definitive, principled answer to the euthanasia question. In the absence of such an answer, we look to alternative ways forward. One such route would lead us to assess the prospects for consensus. Whilst we detect some areas of convergence, we acknowledge that the disputants are unlikely ever to agree on the fundamental question. Perhaps, instead, a compromise might indicate the way forward. The options here include the possibility of creating an excusable (but not justifiable) offence.

7 License to Kill: A New Model for Excusing Medically Assisted Dying?

119

Our proposed model takes a different line, albeit explicitly a compromise position, which acknowledges that there are good justifications both for being permissive and prohibitive, and seeks a middle way that allows us to make progress in the debate.1

7.2 Never Say Never Again In current parlance, acting to end the life of a suffering individual tends to be described as “assisted dying.” A multiplicity of concepts and phrases litter debates about the (ethical and legal) rights and wrongs of such a practice. References are therefore made to “euthanasia” and “assisted suicide”, as well as to slogans like “death with dignity” and the “right to die”. It is not always obvious that the terms capture a single practice. Indeed, sometimes neighbouring, but potentially distinguishable, practices feature in the discussions, such as the withdrawal of life-sustaining treatment or the use of drugs to relieve symptoms, which might be said to “ease the passing”. Without a common referent, there is a risk that disputants will talk past one another. Unfortunately, the task of defining the practice with which we are concerned has been rather neglected.2 Although we will occasionally return to questions of definition, we will focus on what we term “assistance in dying”. For the label to be applied, the assistance will have been intentionally achieved by some positive means (such as supplying or administering a lethal medication) and it will have been provided by a particular person, specifically a medical doctor. Furthermore, this assistant— and, where there is one, the request—will have been motivated by the current or anticipated suffering of the individual who is to receive the assistance. Perhaps “It’s Over, Debbie”, the article to which Gaylin et al. were responding (along with numerous others), provides an illustration of such assistance in action. The piece outlined an anonymous gynaecology resident’s decision to administer morphine to a 20-year-old patient who was dying of ovarian cancer (Anonnymous 1988). Debbie, who had not slept or eaten for two days, was struggling to breathe. When the resident appeared, Debbie said only: “Let’s get this over with”. Debbie appeared to die minutes after the injection. Maybe this was “double effect” in action: the drug brought Debbie relief from her symptoms, but it also had the unintended secondary effect of suppressing her breathing (see, e.g., Huxtable 2007, 11–12). Or perhaps this was assistance in dying—and perhaps, in any event, that was what Debbie had sought. Gaylin et al. propose that we should never countenance assistance in dying, at least by doctors, but others argue that we should never say “never”. Proponents will usually point to one or both of two ethical arguments (see De Haan 2002). The first argument is grounded in respect for autonomy and thus in the view that life has a self-determined value (Huxtable 2007, 13–15). Many would maintain that we should allow scope for 1

See, e.g., Huxtable (2012) and Ives (2014) for an exposition of the theoretical basis of the need for, and value of, compromise in genuinely dilemmatic situations. 2 But see, e.g., Beauchamp and Davidson (1979), Wreen (1998), and Huxtable (2007).

120

R. Huxtable and J. Ives

self-rule (e.g. Gillon 2003). If this is morally worthwhile, the argument runs, then we should allow autonomous individuals to decide for themselves what should, and should not, happen to them in relation not only to their lives, but also to their deaths. Proponents of the autonomy argument would therefore say that a patient like Debbie should have a wish to die granted, provided (of course) that death is autonomously sought. “Making someone die in a way that others approve, but he [himself] believes a horrifying contradiction of his life, is a devastating, odious form of tyranny”, suggests Dworkin (1993, 217). Many would agree, and they do so from a variety of ethical perspectives, including deontology, consequentialism, feminism(s), and virtue ethics (e.g. Beyleveld and Brownsword 2001; Singer 1993; Biggs 1998; van Zyl 2000). The support is not merely theoretical or abstract. Some patients appear to want the option of ending their lives early, and they will occasionally approach their clinicians or the courts with their requests.3 There is also public approval, at least in some jurisdictions: in England, for example, opinion polls typically reveal a high proportion of support for a reform to the law in order to allow some form of assistance in dying.4 Such support is usually offered for voluntary forms of assistance in dying, such as voluntary euthanasia (in which the assistant takes the final fatal step) or assisted suicide (in which the patient does so). These latter forms of assisted dying appear most obviously premised on respect for autonomy, but the idea can also offer principled support for some of its other forms. Where the patient in question is no longer autonomous, but has made a prior request for assistance whilst autonomous, then this might count as not only nonvoluntary (since the patient now lacks autonomy), but also voluntary (as the patient made an autonomous advance request) assistance in dying. Perhaps the same idea can also be stretched to encompass proxy requests, and therefore to cases in which a loved one authorises assistance in dying for the incapacitated, suffering patient, on the premise that death would be in their best interests (Singer 1993, 182–3). Yet, there might be no need to stretch the logic of autonomy to cover these sorts of cases, since an alternative argument might provide a better moral basis. This latter argument is variously articulated in terms of beneficence, mercy, and the relief of suffering. The central idea here is that life has only an instrumental value: it is good only insofar as it can be put to use in pursuing that which makes life worthwhile (Huxtable 2007, 15–17). Where the life is sufficiently afflicted, and the quality of that life is sufficiently poor, its value may be said to diminish. In such circumstances the patient, or perhaps others acting in the patient’s interest, might come to judge it better—kinder, more merciful—to end a life of suffering, rather than allow (or force) its continuation. These sorts of arguments, which tend to be consequentialist, are captured by Singer, who refers to the example of critically ill infants: “When the life of an infant will be so miserable as not to be worth living [and] there are no ‘extrinsic’ reasons for keeping the infant alive—like the feelings of the parents—it is better that the child should be helped to die without further suffering” (Singer 1993, 182–183; see also Doyal 2006). 3 4

E.g. the recent English ruling in R (on the application of Nicklinson and another) (2014). House of Lords Select Committee (2005), Commission on Assisted Dying (2011).

7 License to Kill: A New Model for Excusing Medically Assisted Dying?

121

As Singer implies, sometimes the autonomy and mercy arguments will work together to justify an apparent instance of non-voluntary euthanasia. Indeed, this might even be the case with voluntary euthanasia and assisted suicide. Although autonomy might be capable of doing the relevant moral work unaided, autonomy and mercy can work together to form a “joint view” on the appropriateness of consensual killing (De Haan 2002). Indeed, both grounds typically feature in those laws which permit voluntary forms of assistance in dying. As is well known, the Netherlands has the longest history, but its Benelux neighbours have recently taken similar steps, as have certain states in the USA, not least Oregon, which allows physician assisted suicide.5 Respect for autonomy is a central feature of these permissive laws, and Belgium has recently extended its provision of assistance in dying to encompass competent minors (Siegel et al. 2014). Mercy, nevertheless, also plays a part, as the individuals who can avail themselves of such laws usually need to be patients (in some sense) i.e. they need to have particular medical conditions that lead to unbearable suffering. In some legal systems, the condition in question is narrowly defined: in Oregon, for example, the patient must be in the last six months of terminal illness. Elsewhere the suffering criterion is more expansive: the Dutch policy, for example, originated in somatic disease, but thereafter extended to psychiatric suffering and there has even been discussion of its ability to accommodate assistance in dying for cases of so-called “life fatigue” (Huxtable and Möller 2007; Varelius 2007). As we have seen, the argument from mercy seems applicable not only to voluntary, but also to non-voluntary assistance in dying. Presumably the argument could even extend to involuntary instances of assistance i.e. to assistance in dying which is provided contrary to the wishes of the autonomous individual or without them even being consulted. No one seems to support such an extension, presumably given the affront to autonomy. Yet, it is possible to find support for non-voluntary assistance in dying, including in practice. The Dutch policy only explicitly allows for voluntary forms of assistance in dying, but a protocol openly operates in Groningen, by which euthanasia is provided to critically ill infants (Manninen 2006).

7.3 Dr – No! These arguments in favour of assisted dying, and their deployment in practice, may leave opponents of the phenomenon shaken, but not stirred. First, the opponents are unstirred by the proclaimed moral force of the case for allowing assistance in dying, as they dispute the meaning, scope and appeal of the proponents’ appeals to autonomy and mercy (see Huxtable 2007, 24–27). The proper meaning of respect for autonomy gives some critics reason to pause. The idea is enduring something of a backlash in various contexts, from various quarters (Donchin 2001). Some critics complain that respect for autonomy should not be reduced to “I want”, with the expectation being that this translates unequivocally to “I should get”: we still need to be convinced 5

For an overview of some of these laws, see Lewis and Black (2012).

122

R. Huxtable and J. Ives

that what “I want” is in some sense “good” and that “I” am entitled to the provision of that good. Operating alone, the logic of autonomy “is ostentatiously silent about whether death is desirable or undesirable, but insists only that each individual should make this value choice for himself” (Burt 2005, s13).6 If that is the case, then the logic is also difficult to constrain. Why restrict a policy to only assisted suicide, and not also allow voluntary euthanasia, and why not also extend that logic to being able to choose the manner of death? Furthermore, why insist that there be any underlying medical condition—surely, if the autonomous choice is central, it matters not why that choice was formed, but only that it was formed autonomously (see e.g. Huxtable and Moller 2007)? Critics even suspect that the logic of autonomy will prove self-defeating. If autonomy is a crucial value for proponents, then they will certainly need to attend to any risks that assistance in dying poses to that value. Perhaps there are such risks: maybe the very existence of an option to die will be sufficient to place pressure on people to have the option exercised in their case, which will therefore serve to threaten their autonomy (Velleman 1992). Finally, say the opponents, there is not necessarily the public appetite for assisted dying that the proponents claim there is. The surveys that reveal public support are sometimes badly phrased and badly timed, coming amidst difficult, high-profile cases (Hagelin et al. 2004). Indeed, the suggestion has been made that, when the surveys are amended to include counter-arguments, public support appears to drop (Saunders 2014). Similar charges are levelled against the proponents’ appeals to mercy. Singer spoke of ending “miserable” lives, but the qualifying criteria need to be spelled out, especially if charges of “disablism” are to be avoided.7 Here, too, we might well ask whether something like “life fatigue” should (or should not) be considered a good ground for euthanasia? The individuals who should be empowered to make such life-ending judgments also need to be identified. We might then wonder whether and how the brakes can be put on any merciful policy: to condemn involuntary euthanasia as “murder” is simplistic (e.g. Harris 1997), since every form thereof will be murder in many jurisdictions, but the condemnation of this specific practice does at least require proponents to spell out clearly what is (and is not) to be allowed. In addition to criticising proponents of assisted dying, the opponents also provide standalone arguments against affording doctors a licence to kill. There are three main clusters of arguments offered, which respectively refer to the intrinsic value of human life, the avoidance of slippery slopes, and the need to preserve the integrity of the medical profession. First, as will be apparent from some of the preceding points, opponents of assisted dying do not judge the value of life in self-determined or instrumental terms. Rather, they argue, life has an intrinsic value, in and of itself: life is inviolable and it should not be ended intentionally, whether by action or by omission (e.g. Keown 2002). This essentially deontological position is traditionally associated with theistic (and often Judaeo-Christian) thinking, but the same idea 6

Sometimes proponents will point to dignity as a basis for respecting individual choice, but critics suspect that these efforts are equally misconceived (Foster 2011). 7 For (critical) discussion of the disability rights perspective, see Ouellette (2006).

7 License to Kill: A New Model for Excusing Medically Assisted Dying?

123

recurs in more secular forms, such as in the right to life. For some, respect for autonomy can only be understood in such terms: a truly autonomous choice will be one which recognises the value of life, whether that life is one’s own or another’s (Keown 2002, 53). The second argument is more consequentialist in nature, since it points to the disastrous effects that could—or logically should—follow on from endorsing assistance in dying in some form or another. Variously expressed in terms of the “thin end of the wedge” or “opening the floodgates”, there are usually two types of “slippery slopes” which are mentioned, the first logical, the second empirical or psychological. One logical objection makes a familiar claim: involuntary euthanasia follows the same logic as voluntary or non-voluntary euthanasia, insofar as they can all be premised on the poor quality of the patient’s life (e.g. Keown 2002, 70). The empirical objections, meanwhile, refer to probabilities: we are more likely in fact to be led into dangerous territory, if we make the first step of allowing some form of assisted dying. Opponents fear that life will be de-valued, and lives will be imperilled, and they draw various analogies, including with the Nazi atrocities (which are said to rest on the same judgment of worthless lives) and with terminations of pregnancy (which are said to have expanded considerably from apparently modest beginnings) (e.g. Keown 2002, 73). The move in the Netherlands, from allowing euthanasia only in cases of somatic disease, to mental health disorders, and possibly now to “life fatigue”, is an example of such a slope. As Bond is told by Dryden in the opening sequence of Casino Royale,8 the first kill is always the hardest, because you feel it. The second, Bond laconically agrees as he pulls his trigger, is considerably easier. Most apposite, perhaps, are the references to data emerging from jurisdictions that have passed measures to allow assistance in dying. As we have seen, the Dutch policy widened its scope beyond the unbearable suffering associated with terminal illness. Notably, not every instance of euthanasia in that jurisdiction appears to be entirely voluntary: not only is non-voluntary euthanasia now openly practised in that jurisdiction, but early reports suggested that 1000 (0.8%) of the annual occurrences of “voluntary” euthanasia took place without any explicit request from the patient (see Keown 2002, 104–106). The assistants involved appear not to have been censured. More recently, as we also saw, Belgium has amended its law: the original law was only passed in 2002, but in 2014 the provisions were extended to encompass minors (Siegel et al. 2014). Citing such examples, opponents of assistance in dying argue that lines can neither be drawn in principle nor defended in practice, so the practise is best left unlawful. Finally, the opponents point to the integrity of medicine, which would be threatened by any moves to equip doctors with a license to kill. This was a central objection from Gaylin et al., whose opposition can be linked to efforts by Kass and Pellegrino to spell out the distinctive “internal morality of medicine”. Kass (1975) sees medicine as a moral enterprise, which is orientated towards the end of health. Pellegrino (2001) takes a similarly Aristotelean approach, in which the correct function

8

Campbell (2006).

124

R. Huxtable and J. Ives

of medicine is to serve the good of the patient, which he sees as having four dimensions. Doctors should therefore seek to promote the medical good (which requires technical competence), the good as perceived by the patient (which requires respect for their autonomous wishes), the good for the patient as a human being, and the spiritual good. These goods ascend in order of importance: ultimately, the patient’s apparent choices or perceived needs are subordinate to the spiritual good, which, on Pellegrino’s account, is incompatible with assistance in dying. Such incompatibility, in turn, renders the practice incompatible with medicine’s moral mission. In short, assistance in dying is contrary to the integrity of medicine, so the final answer to the euthanasia question should be: “Dr – No!”.

7.4 Live and Let Die Proponents of assisted dying take issue with each of the opponents’ objections, and whilst they may agree that it is generally better to live, we also ought to let (people) die when certain conditions are met. First, they question the view that life has an intrinsic value, citing problems with its meaning, scope and appeal, similar to those the opponents had levelled against rival accounts of the value of life. Proponents of assisted dying initially criticise the concepts and distinctions on which the ethic apparently relies. The intentional ending of life is prohibited according to adherents to the intrinsic value of life, but critics suspect that “intention” is being defined too narrowly, in ways which serve the moral judgments these adherents seek to make (e.g. Rachels 1986, 92). Even if we think that Debbie’s doctor did not intend to end her life, this doctor did act voluntarily and did appear to cause her death—so why not hold this doctor to moral account? There are said to be similar problems with opponents’ efforts to justify some fatal omissions (e.g. Rachels 1986, 106–150). Those opponents who cleave to the intrinsic value of life believe that they can differentiate situations of “live and let die”: they therefore distinguish between acts and omissions, and then between different sorts of omissions (see, e.g., Keown 2002, 42–43). According to such arguments, lifesupporting interventions can be withdrawn or withheld (i.e. the doctors can omit to treat), provided that the interventions are judged to be “futile” or more burdensome than beneficial. The proponents of assisted dying remain unconvinced, and they again argue that key terms like “futility” are being defined in ways that beg the crucial moral questions (e.g. Halliday 1997). These problems lead some proponents to suspect that the inviolability ethic is dishonest and self-defeating. Since the approach allows doctors to withdraw allegedly futile life-support or administer symptom-relieving drugs that might incidentally (but not intentionally) shorten life, proponents insist that it tacitly embraces consequentialist (instrumentalist) judgments that some poor quality lives are best ended (e.g. Singer 1993, 210, 211). Perhaps the ethic can only be rescued by insisting on a more vitalistic stance, which would require every effort to be made to protect life (Kuhse 1987). However, supporters of the sanctity of life find this extreme just as

7 License to Kill: A New Model for Excusing Medically Assisted Dying?

125

unappealing as the opposite, consequentialist extreme (Keown 2002). Yet, even their alleged middle ground position is found to be unappealing: although the central idea commands some secular support, the sanctity of life looks like a distinctively theological notion, which will scarcely convince those who have no, or alternative, faith-based commitments (e.g. Dworkin 1993, 217). Proponents secondly dispute the logic and likelihood of claims that, by allowing assisting dying in some cases, we will be led down a slippery slope. Some simply dismiss the alleged logic of the slippery slope argument as fallacious (e.g. Smith 2005). Others suspect that the claim merely camouflages the opponents’ real objection, which is premised on the intrinsic value of life (e.g. Doyal 2006). The empirical objections and analogies are also rejected. The Nazi eugenic policies are said to be some considerable moral distance from policies advocated by those in favour of assistance in dying (e.g. Rachels 1986, 177–178). Any slide in relation to termination of pregnancy is also disputed: for example, the fact that doctors can abort disabled foetuses has not generally inclined them towards killing disabled children or adults (e.g. Singer 1993, 217 and Rachels 1993, 62–63). Furthermore, the Dutch data, including the problematic 1000 deaths, might not be all that they appear; maybe, indeed, these deaths were not non-consensual (Downie 2000). In short, advocates insist that ways can be found to pinpoint and police the necessary boundaries (e.g. Rachels 1993, 61–62). Finally, the integrity of medicine is said to present no barrier to allowing assistance in dying. For one thing, it might be possible to create a new specialty—“thanatology”, for example—whose practitioners need not be drawn from medicine’s ranks (Brazier 1996). Of course, this might not be a complete response, since clinicians seem likely to be involved in some way in any state-endorsed system.9 Yet, proponents still perceive no problem in principle with affording doctors a licence to kill, or with otherwise involving them in such a licensed system, because robust measures can always be put in place to ensure that abuses of trust are identified and appropriately dealt with, if not eliminated entirely. In contrast to Kass et al., Miller and Brody suggest that medicine is not only about healing and promoting health, but also concerned with enabling patients to achieve dignified and peaceful deaths (Miller and Brody 1995). They deny that this latter goal begs the question about medicine’s compatibility with assistance in dying (it might only refer to the need for palliative care), although they do still conclude that medicine can encompass the practice. On their account, doctors cannot be obliged to assist in dying, but they have the discretion to do so, provided that the patient has chosen this autonomously and the doctor is satisfied that this is best in the circumstances. Momeyer argues to similar effect (Momeyer 1995). Although he agrees with Kass that medicine is about healing, Momeyer rejects Kass’s Hippocratic beneficence and charges him with paternalism. For Momeyer, medicine’s ends are chosen by humans, not stipulated by nature. Such ends should allow room for patient autonomy. There 9

However, this is not the position in Switzerland (see Lewis and Black 2012). There is also some scholarly support for less medicalised models, e.g., Ost (2010).

126

R. Huxtable and J. Ives

may (but only may) be extrinsic reasons for resisting assisted dying, but neither Momeyer nor Miller and Brody believe that there is anything intrinsic to medicine which requires such opposition. One of the most significant difficulties faced by proponents, however, is to propose a legislative system that permits assisted dying and also provides adequate protections to the potentially vulnerable, so that assistance to die is only provided to those who want it and where it is the only reasonable option available. Any legislation that allows assisted dying admits the possibility—perhaps the probability—that some unwilling patients will slip through the net. This possibility alone, opponents will argue, is enough to support prohibition.

7.5 Die Another Day? We seem to be left with an assisted dying equation: although people will tend to prefer the arguments on one particular side, neither side appears conclusively to have greater value than the other. Each side seems to offer important insights: proponents understandably proclaim the importance of self-determination and the avoidance of suffering; opponents plausibly point to the value of human life and the risks of unintended consequences; and both sides seem to make viable suggestions about the core business of medicine. Yet, the arguments on each side are also afflicted by considerable problems of meaning, scope and appeal. The ethical battles will undoubtedly continue to be waged, and we should not be beguiled into thinking that better evidence can instigate a ceasefire (Parker 2005). Of course, the arguments that are raised continue to become ever more nuanced, and there is certainly more to these debates than our summary can convey. At the same time, however, the essential points are well-established, but the process of argument and counter-argument is unceasing. As far back as 1958, the debate about assisted dying was deemed “jaded” (Williams 1969). Perhaps this is a debate that must wait to die another day. Prior to its demise, and as this collection attests, we need to consider whether there are new directions in which assisted dying can travel. One possibility might be to look for areas of consensus and to construct our laws and policies on such bases. Maybe this is not an unrealistic ambition, since proponents and opponents will agree about the appropriateness of many practices, although they will cast their respective ethical justifications in different terms. For example, both sides support the cessation of life-supporting treatment in particular cases, and also the use of strong painkillers and sedatives for some patients, even in the (admittedly extremely rare) case where the drugs might shorten life. Opponents of assisted dying will justify such cases by reference to acts, omissions and futility, on the one hand, and double effect, on the other; proponents, meanwhile, might reject such convoluted labels, but they will nevertheless agree with the course proposed in many such cases. Proponents and opponents might also be united in other ways, ways which are too seldom noticed. Writing about those who support assistance in dying, Gillett

7 License to Kill: A New Model for Excusing Medically Assisted Dying?

127

referred to the moral significance of “the pause”: “a crucial element in the moral competence of a doctor is a tendency to hesitate, have misgivings or feel a ‘pause’ about certain principled medical decisions involving life and death” (Gillett 1988, 61).10 Perhaps this pause is already present, even in proponents’ arguments: notice the care with which qualifying conditions and criteria tend to be formulated, and the fact that assistance in dying is typically conceived as an exception to any general rule against killing. In short, proponents do not seek a wholesale dismantling of the ethic that prohibits the ending of life. Equally, opponents do not wholly insist that life must be preserved, no matter what. As such, even those opponents of assisted dying who support the intrinsic value of life appear mindful of the need to pause, since they allow for situations in which not every effort must be taken to prolong or protect life. However, we should not be too optimistic about the prospects for convergence and consensus. Each side might exhibit a degree of caution, and be inclined to support some practices that are also endorsed by their opponents, but on the crucial question they still fail to reach consensus. In short, proponents still insist that assisted dying can be justified (at least in some cases), while opponents still insist the opposite. Determining a victor continues to prove difficult: jurisdictions might adopt more or less permissive policies in practice, but the principled disputes rage on. Absolutists and universalists on either side might insist that they are in the right, but their arguments will scarcely convince everyone. Closure on the justifiability—or unjustifiability—of assisted dying seems highly unlikely; as van Willigenburg commented, albeit in another context, either “there is no superior way of mixing values or we are unable to rationally trace that superior mix” (Van Willigenburg 2000, 400). But maybe both sides are deploying the wrong conceptual category: perhaps, if we move away from notions of justification, and towards the concept of excuse, we can begin to attend to concerns on both sides. An excuse can convey a compromise, since it signals a degree of wrong-doing, albeit in the presence of factors which suggest a degree of right-doing (e.g. Austin 1956). Assisted dying might be amenable to such a policy solution: rather than judge this as justified or unjustified, it might be possible to treat it as criminal (in line with opponents) but also to ensure a suitably humane disposal by the criminal courts (in line with proponents) (e.g. Huxtable 2007). It remains to be seen whether such a proposal is defensible or, indeed, offers enough to either side (see, e.g., Holm 2010). We will leave this proposal to one side, although there is one element thereof which merits further consideration here. In contrast to many (envisaged or operative) legal models, this sort of compromise proposal does not provide a template for conferring immunity prospectively. Rather, the excuse operates retrospectively: the alleged assistant’s actions are judged against pre-ordained criteria, to ascertain whether the conditions of the excuse are satisfied, such that it should be made available in this case. This sort of approach merits further exploration, since it promises to take the assisted dying debate in a new direction. In the next section we therefore outline an alternative model, by drawing an analogy with UK armed response units. 10

See similarly Burt (2005) on ambivalence.

128

R. Huxtable and J. Ives

7.6 The Man with the Golden Gun This section outlines a model of “retrospective excuse”, which has the potential to accommodate permissiveness towards assisted dying, whilst maintaining both appropriate legislative safeguards and an appropriately prohibitive attitude towards killing in general. We will do this by drawing an analogy with the way in which killing is dealt with in the context of UK police armed response units, which are permitted to use lethal force in the carrying out of their duties. As we shall see, the analogy is not perfect, but it does provide food for thought, in offering a new direction to our thinking about assisted dying. We first outline the armed response model, and then use this to sketch an analogous model for doctors and assisted dying.

7.6.1 UK Armed Response In the UK,11 police officers do not routinely carry firearms. They are authorised to use physical force as necessary, but are subject to the same laws regarding physical assault as all UK citizens. There are, however, a special subset of police officers who are specially selected and trained in the use of, and therefore authorised (or “licensed”) to carry and use, firearms. These officers are deployed when the use of lethal force may be necessary in the discharge of the police force’s duty to protect the public. Whilst the firearms officer is authorised to carry a weapon, trained in its use and educated in the circumstances in which it may be lawful to discharge it, the responsibility of any decision to shoot a person belongs to that officer alone. There is no such thing as prior authorisation or immunity, and the Nuremberg defence (“I was only following orders”) cannot be relied upon. The officer who pulls the trigger must never do so on the order of someone else, but must be satisfied that it is necessary and correct to do so. Furthermore, the trigger is pulled in the knowledge that the resulting death will be investigated thoroughly, and the officer must be prepared to defend and justify his or her actions and to accept criminal sanctions if the death is consequently judged unlawful. Importantly, because no shooting, and therefore no resultant killing, is ever pre-approved, any death is automatically treated as potentially criminal and investigated as such. Following a shooting, an investigator from the Independent Police Complaints Commission (IPCC) is appointed and called to the scene. A solicitor is called out to represent any officer who has discharged a firearm. Officers involved may make 11

Our primary sources for this section are: Association of Chief Police Officers, Association of Chief Police Officers in Scotland, and National Policing Improvement Agency (2011), Accessed 15th Aug 2014, 1001 GMT and UK Police Firearms Officers Association (see PFOA. Post Shooting Procedures (2015), Accessed 14th Aug 2014, 10:57 GMT). We would also like to acknowledge and thank M, otherwise known as Martin Cooper, who gave us invaluable advice on this section. Any mistakes are, of course, our own.

7 License to Kill: A New Model for Excusing Medically Assisted Dying?

129

written notes of the event, subject to legal advice, and witnesses to the shooting (but not participants in it) will be asked to write a statement giving their account of events. All weapons discharged will be handed over to a forensics officer, alongside anything else of forensic value. In the following few days, officers involved must meet with their solicitor and make a detailed statement about the event for the investigators. Officers may confer about times, locations, routes, but may not confer about their honestly held beliefs at the time they discharged their weapons. At this time, officers are given the opportunity to speak to a welfare counsellor. Following the completion of individual statements, and any debriefing meetings (which would be attended by the investigator and recorded), all officers who discharged their weapon are removed (with pay) from active duty. If it was a fatal shooting, an inquest will be held. If the inquest determines that the officers acted appropriately and their use of firearms was justified, they will return to active duty. The Association of Chief Police Officers states that “[a] critical shot should only be fired when absolutely necessary in defence of a person when there is an imminent and extreme risk to life from unlawful violence. A critical shot is a shot or shots aimed to the head, if possible, or otherwise to the central nervous system or major organs”.12 If, in terms of guidelines quoted above, the inquest determines that the use of firearms was not justified, then the officer may be subject to a criminal investigation and, if found guilty, criminal sanctions.

7.6.2 Medically Assisted Dying: The Analogous Model In order to develop an analogous model, we would first have to note that, in the UK, doctors are not routinely expected to carry the equipment to assist death, nor be mindful that this may be part of their job. That would not rule out, however, the creation of a subset of the profession who could be specially selected and trained in the use of lethal medication (or other methods of assisting death), who are therefore authorised (or “licensed”) to assist death, and who are called upon when the use of lethal medications (for example) may be necessary in the discharge of the medical profession’s duty to respect patient autonomy and alleviate suffering. In keeping with the terminology we introduced earlier, let us call this group “thanatologists”. We will presume that such professionals will have voluntarily chosen to assume these duties (and, therefore, that no professional will be required to do so contrary to their conscience) (see Huxtable and Mullock 2015). Whilst the thanatologist would be authorised to carry death-assisting equipment, would be trained in its use and educated in the circumstances in which it may be lawful to use it, the responsibility for any decision to assist a patient to die belongs to that practitioner alone. There would be no such thing as prior authorisation or immunity, and the Nuremberg defence could not be relied upon. Furthermore, the 12

Association of Chief Police Officers, Association of Chief Police Officers in Scotland, and National Policing Improvement Agency (2011), Accessed 15th Aug 2014, 1001 GMT. S2.43, p. 35.

130

R. Huxtable and J. Ives

assistance would be given in the full knowledge that the resulting death will be investigated thoroughly, and the thanatologist must be prepared to defend and justify his or her actions and to accept criminal sanctions if the death is consequently judged unlawful. Importantly, because no assistance to die could ever be pre-approved, any death will be automatically treated as potentially criminal, and must be investigated as such. Following an assisted death, an investigator would be appointed and called to the scene. Those involved would have access to a solicitor, and make written notes of the events. All “witnesses”, including members of the multi-disciplinary team, family members and/or friends of the patient would be asked to make statement giving their account of events. Any relevant forensic material will be handed over to forensics officers, and a post-mortem examination will be held to determine cause of death. In the following few days, the thanatologist(s) must meet with their solicitor and make a detailed statement about the event for the investigators. At this time, the thanatologist(s) involved would be given the opportunity to speak to a welfare counsellor. Following the completion of (an) individual statement(s), the thanotologist(s) involved would be suspended from practice (with pay) whilst an inquest is held into the death. If the inquest determines that the thanatologist(s) acted appropriately and their role in assisting a death was justified, they will return to active duty. If, in terms of whatever guidelines are used, the inquest determines that the assistance was not justified, then the thanatologist may be subject to a criminal investigation and, if found guilty, criminal sanctions. We might tentatively suggest, based on the arguments from proponents, and on the assumption that it may represent some common ground, that the following guidelines might suffice: “assistance to die should only be provided when absolutely necessary, and must be performed in the sincere and honest pursuit of humanitarian ends—those ends being to relieve intolerable suffering and act according to the demonstrable interests and/or autonomous decisions of the patient”.

7.6.3 The Tightrope of Compromise? This model represents a significant divergence from the well-trodden permissive path, because it proposes that an act of assisted dying can never be approved or justified in advance, and it also avoids the equally familiar prohibitive path, since it does allow for retrospective authorisation in some circumstances. As such, assisted in dying would only be judged lawful after the act. An act of killing is so serious, so final, and so prima facie reprehensible, that an individual who assists a death must do so according to their own conscience, taking full responsibility, and be prepared to be judged retrospectively. Under such a model, the state, and the law, would not sanction the act of killing before the act. The state and law would, however, outline circumstances under which an act of killing might be excusable. Just as the criminal law accepts that actions that result in death can be lawful, for example, when acting

7 License to Kill: A New Model for Excusing Medically Assisted Dying?

131

under an honestly held belief that they were necessary for self-defence, it could accept that an action that results in death may be lawful, if it is performed in the sincere and honest pursuit of humanitarian ends—those ends being to relieve suffering and act according to the demonstrable interests and/or autonomous decisions of the patient. We suggest that this model ought to satisfy both opponents and proponents of assisted dying on a number of contested points, which we will briefly outline. First, because it would treat all assisted dying as potentially criminal, which must then be proven to be excusable (lawful), the model retains the message that life is to be valued. Certainly, in-roads are made into the notion that life must never be ended, but these would appear to be exceptional and would therefore capture something of Gillett’s “pause”. Secondly, because the thanatologist can never be given prospective permission to end life, or immunity from criminal sanctions, the decision to assist a death is never “safe” and could never be taken lightly. The question will never be “can I do this without being prosecuted or investigated?”, as would be the case if assisted dying were to be legalised in a prospective fashion. Rather, the question would be “can I excuse my action when I am investigated, and prove beyond reasonable doubt that I acted in the interests of my patient?”. This ought to serve to put the brakes on any potential slippery slope, as it does not make assisted dying easy, and will ensure that the thanatologist only assists a death when s/he is certain that it is right thing to do, and can demonstrate this. It may, in fact, make assisting death so unappealing, because it is so personally risky, that it will fail to satisfy proponents of assisted dying.13 We feel, however, that this is the correct point at which to compromise as it will ensure, as far as possible, that assisting dying never becomes comfortable. It will always be done at personal risk (just as is the case with firearms officers who make the decision to pull the trigger). Thirdly, this model accepts that the medical profession in general should not be involved in assisted dying, but acknowledges that assisted dying is not incompatible with medicine per se, and may be in accordance with it on occasion (as has been discussed above). By creating a specialised subset, which is trained and “licensed”, a clear message will be sent to the public, and to the profession, that there will remain a clear distinction between those professionals who will be involved in assisting death, and those who will not and cannot. Just as there is no good reason to mistrust all police officers simply because a small minority carry a gun and may use it, there is similarly no good reason to mistrust all doctors just because a small minority are permitted to assist death (see, e.g., Miller and Brody 1995, 14–15). That specialist subset must be rigorously selected and monitored, including psychological assessment—not just anyone should be given a “license to kill”.

13

The practice might also remain rare because, as Stevens’ (2006) research reveals, many doctors who have assisted in dying “are adversely affected emotionally and psychologically by their experiences.”.

132

R. Huxtable and J. Ives

7.7 Conclusion: Quantum of Solace? We accept that not all parties will find solace in our proposed compromise. There are problems with our proposal, and a great deal more thought needs to go into it before it could be ever adopted as a policy. The first such problem concerns the criteria against which an instance of assistance in dying should be judged, retrospectively, to be appropriate or inappropriate. We were (maybe noticeably) cautious about the qualifying criteria when we outlined our model. Perhaps assistance in dying should only be excused if it was initiated by an express, autonomous wish from the patient; or perhaps the model should accommodate proxy requests, such as those from the parents of critically ill infants or from other loved ones of incapacitated adult patients. Or maybe, more boldly still, the model should allow for these decisions to be made for incapacitated patients in the absence of such a proxy or prior request, on the basis that the worth (of the life) is not enough. We leave it to others to judge the exact qualifying criteria that might be adopted, if such a model is considered to have any merit. Of course, if the model is endorsed, and it were to be adopted, then it will entail that some patients will be helped to die, albeit in exceptional circumstances. This would appear to offer more to proponents of assisted dying than to opponents. It effectively does allow assisted dying, and that may be too much for those who believe that life is sacrosanct and we should strive to protect and prolong life. For the less fundamentalist opponents, what we offer in our proposal is a safeguard that ought to prevent empirical/psychological slippery slopes, which maintains a high value on life, and which acknowledges the need to separate assisted dying from mainstream medical practice. We note, also, that the personal risk involved to the thanatologist is potentially so great, that it might dissuade many from ever assisting death, and this may present too large a barrier to be acceptable to proponents. We fall back, however, on the nature of compromise and the task we have been set in this volume to explore new directions. A new direction should admit the possibility of moving forward and making progress and, we would argue, making progress in this debate requires compromise. A compromise, by its very nature, means that each side gets something they want, but no side gets everything they want (see, e.g., Huxtable 2012). Our proposal fits the bill, and our hope is that it gives each side just enough of what they need, in order for them to accept it. The latter point about preventing the slippery slope may fail, nonetheless, to accommodate legitimate concerns about “Bondian mavericks”.14 Momeyer (1995, 13), for example, refers to: the crusading, self-righteous Jack Kevorkian with his Volkswagen van, intravenous lines, face masks and tanks of carbon monoxide ‘servicing’ desperate strangers seeking to escape lives they no longer find endurable.

This kind of “maverick”, dedicated to the “cause” and willing to court controversy may be encouraged by our proposal, and see it as a licence to push the boundaries, 14

We are grateful to Genevieve Liveley, whose golden eye spotted this potential slope.

7 License to Kill: A New Model for Excusing Medically Assisted Dying?

133

which would certainly threaten the essence of our model.15 Similarly, “villains” may be encouraged to become thanatologists and to use it to camouflage more questionable killings—one can easily imagine a character such as Harold Shipman taking such an opportunity.16 One response may be to invoke the old adages about one bad apple and not throwing out the baby with the bathwater. There will always be rogue agents, but this is no reason to dispense with espionage altogether. A more useful response may be to note that it is precisely the idea of selection, training, assessment and ultimately “licencing” that would aim to filter out any putative thanatologist who is “suspect”. Anyone who is too keen, too quick, or just does not display the appropriate attitude would not get a license. The Platonic adage about not affording power to those who most want it springs to mind here, and we suggest that mavericks ought not be licenced in the first place, or would have their licence revoked if they display maverick tendencies. It is important, nonetheless, that this model incorporate some notion of the conscientious professional, who is correctly motivated, appropriately reflective, and will assist death as a last resort only when s/he is convinced that it is necessary and appropriate. Such a professional will also have to accept the possibility of criminal sanction, be prepared to be open and transparent, and be prepared to document and defend his or her actions at every turn.

7.8 Final Remarks: For Your Eyes Only? The proposal we have outlined is deliberately controversial, and is offered as a thought experiment to encourage a stale debate to explore new directions and new possibilities for fruitful and defensible compromise. The proposal for compromise offered here is broadly permissive towards assisted dying, but it does not let the sky fall in on the value of life. Furthermore, it is not for your eyes only—we now turn this proposal over to the reader, to consider, discuss and respond to, in the hope that, even if the model ultimately proves unacceptable or unworkable, we have presented a new direction that is worthy of consideration.

References Anonymous. 1988. A piece of my mind. JAMA 259: 272. 15

Kevorkian is not the only ‘maverick’ of this kind; other examples include the Australian Dr. Philip Nitschke, and, in the UK, Michael Irwin, as well as Nicholas Reed and (self-appointed) “Dr” Lyons. In the 1980s, the latter pair had collaborated in a clandestine assistance-in-dying service, before they were discovered and convicted: see e.g. Huxtable (2007: 55–57, 78). 16 Harold Shipman was a UK family doctor who, in 2000, was found guilty of murdering 15 of his patients. The subsequent judicial inquiry, examining nearly 500 of his patient’s deaths, found that he had killed 215 of his patients in total, the majority of whom were elderly women. See Smith (2002), 0917 GMT.

134

R. Huxtable and J. Ives

Association of Chief Police Officers, Association of Chief Police Officers in Scotland, and National Policing Improvement Agency. 2011. Manual of guidance on the management, command, and deployment of armed officers, 3rd ed. http://www.acpo.police.uk/documents/uniformed/2011/201 111MCDofAO3.pdf Austin, J.L. 1956. A plea for excuses. Proceedings of the Aristotelian Society 57: 1–30. Beauchamp, Tom L., and Arnold I. Davidson. 1979. The definition of euthanasia. Journal of Medicine and Philosophy 4: 294–312. Beyleveld, Deryck, and Roger Brownsword. 2001. Human dignity in bioethics and biolaw. Oxford: Oxford University Press. Biggs, Hazel. 1998. ‘I don’t want to be a burden!’ A feminist reflects on women’s experiences of death and dying. In Feminist perspectives on health care law, ed. Sally Sheldon, and Michael Thomson, 279–295. London: Cavendish. Brazier, Margaret. 1996. Euthanasia and the law. British Medical Bulletin 52: 317–325. Burt, Robert A. 2005. The end of autonomy. Hastings Center Report, Special Report 35: s9–s13. Campbell, Martin (dir.). 2006. Casino royale [film]. UK: EON Productions. Commission on Assisted Dying. 2011. The current legal status of assisted dying is inadequate and incoherent. London: Demos. De Haan, Jurriaan. 2002. The ethics of euthanasia: Advocates’ perspectives. Bioethics 16: 154–172. Donchin, Anne. 2001. Understanding autonomy relationally: Toward a reconfiguration of bioethical principles. Journal of Medicine and Philosophy 26: 365–386. Downie, Jocelyn. 2000. The contested lessons of euthanasia in the Netherlands. Health Law Journal 8: 119–139. Doyal, Len. 2006. Dignity in dying should include the legalisation of non-voluntary euthanasia. Clinical Ethics 1: 65–67. Dworkin, Ronald. 1993. Life’s dominion: An argument about abortion and euthanasia. London: Harper Collins Publishers. Foster, Charles. 2011. Human dignity in bioethics and law. Oxford: Hart. Gaylin, Willard, Leon R. Kass, Edmund D. Pellegrino, and Mark Siegler. 1998. Doctors must not kill. JAMA 259: 2139–2140. Gillett, Grant. 1988. Euthanasia, letting die and the pause. Journal of Medical Ethics 14: 61–67. Gillon, Raanan. 2003. Ethics needs principles—Four can encompass the rest—And respect for autonomy should be “first among equals.” Journal of Medical Ethics 29: 307–312. Hagelin, Joakim, Tore Nilstun, Jann Hau, and Hans-Erik Carlsson. 2004. Surveys on attitudes towards legalisation of euthanasia: Importance of question phrasing. Journal of Medical Ethics 30: 521–523. Halliday, Robert. 1997. Medical futility and the social context. Journal of Medical Ethics 23: 148–153. Harris, John. 1997. Euthanasia and the value of life. In Euthanasia examined: Ethical, clinical and legal perspectives, ed. John Keown, 6–22. Cambridge: Cambridge University Press. Holm, Sören. 2010. Euthanasia: Agreeing to disagree? Medicine, Health Care and Philosophy 13: 399–402. Huxtable, Richard. 2007. Euthanasia, ethics and the law: From conflict to compromise. London: Routledge-Cavendish. Huxtable, Richard. 2012. Law, ethics and compromise at the limits of life: To treat or not to treat? Abingdon: Routledge. Huxtable, Richard, and Maaike Möller. 2007. “Setting a principled boundary”? Euthanasia as a response to ‘life fatigue.’ Bioethics 21: 117–126. Huxtable, Richard, and Alexandra Mullock. 2015. Voices of discontent? Conscience, compromise, and assisted dying. Medical Law Review 23: 242–262. Ives, Jonathan. 2014. A method of reflexive balancing in a pragmatic, interdisciplinary and reflexive bioethics. Bioethics 28: 302–312. Kass, Leonard R. 1975. Regarding the end of medicine and the pursuit of health. The Public Interest 40: 11.

7 License to Kill: A New Model for Excusing Medically Assisted Dying?

135

Keown, John. 2002. Euthanasia, ethics and public policy: An argument against legalization. Cambridge: Cambridge University Press. Kuhse, Helga. 1987. The sanctity-of-life doctrine in medicine: A critique. Oxford: Clarendon Press. Lewis, Penney, and Isra Black. 2012. The effectiveness of legal safeguards in jurisdictions that allow assisted dying. London: Demos. Manninen, Bertha A. 2006. A case for justified non-voluntary active euthanasia: Exploring the ethics of the Groningen protocol. Journal of Medical Ethics 32: 643–651. Miller, Franklin G., and Howard Brody. 1995. Professional integrity and physician-assisted death. Hastings Center Report 25: 8–17. Momeyer, Richard. 1995. Does physician assisted suicide violate the integrity of medicine? Journal of Medicine and Philosophy 20: 13–24. Ost, Suzanne. 2010. The de-medicalisation of assisted dying: Is a less medicalised model the way forward? Medical Law Review 18: 497–540. Ouellette, Alicia. 2006. Disability and the end of life. Oregon Law Review 85: 123–182. Parker, Malcolm. 2005. End games: Euthanasia under interminable scrutiny. Bioethics 19: 523–536. Pellegrino, Edmund D. 2001. The internal morality of clinical medicine: A paradigm for the ethics of the helping and healing professions. Journal of Medicine and Philosophy 26: 559–579. Rachels, James. 1986. The end of life: Euthanasia and morality. Oxford: Oxford University Press. Rachels, James. 1993. Euthanasia. In Matters of life and death: New introductory essays in moral philosophy, ed. T. Regan, 30–68. New York: McGraw Hill. Saunders, Peter. 2014. Support for UK assisted dying bill plummets to 43 % after hearing opposing arguments. http://www.lifesitenews.com/pulse/support-for-uk-assisted-dying-bill-plummets-to43-after-hearing-opposing-ar. Accessed 29 April 2015. Siegel, Andrew M., Dominic A. Sisti, and Arthur L. Caplan. 2014. Pediatric euthanasia in Belgium: Disturbing developments. JAMA 311: 1963–1964. Singer, Peter. 1993. Practical ethics, 2nd ed. Cambridge: Cambridge University Press. Smith, Dame J. (2002). The shipman enquiry—First report: Death disguised. The National Archives. http://webarchive.nationalarchives.gov.uk/20090808154959/; http://www.the-shipman-inquiry. org.uk/home.asp. Accessed 15 August 2014 Smith, Stephen W. 2005. Fallacies of the logical slippery slope in the debate on physician-assisted suicide and euthanasia. Medical Law Review 13: 224–243. Stevens Jr., Kenneth R. 2006. Emotional and psychological effects of physician-assisted suicide and euthanasia on participating physicians. Issues in Law and Medicine 21: 187–200. UK Police Firearms Officers Association (PFOA), n.d. Post-shooting procedures. 2015. https:// www.pfoa.co.uk/211/post-shooting-procedures. Accessed 29 April 2015. Varelius, Jukka. 2007. Illness, suffering and voluntary euthanasia. Bioethics 21: 75–83. Van Willigenburg, Theo. 2000. Moral compromises, moral integrity and the indeterminacy of value rankings. Ethical Theory and Moral Practice 3: 385–404. van Zyl, Liezl. 2000. Death and compassion: A virtue-based approach to euthanasia. Aldershot: Ashgate. Velleman, David J. 1992. Against the right to die. Journal of Medicine and Philosophy 17: 665–681. Williams, Glanville. 1969. Euthanasia legislation: A rejoinder to the non-religious objections. In Euthanasia and the right to death: The case for voluntary euthanasia, ed. A.B. Downing, 134–147. London: Peter Owen. Wreen, Michael. 1998. The definition of euthanasia. Philosophy and Phenomenological Research 48: 637–653.

Court Cases and Rulings R (on the application of Nicklinson and another) v Ministry of Justice; R (on the application of AM) v The Director of Public Prosecutions [2014] UKSC 38.

136

R. Huxtable and J. Ives

Laws and Legislation UK, House of Lords Select Committee. 2005. Assisted dying for the terminally ill bill [HL], vol. I—Report, HL Paper 86-I, 125–127. London: Her Majesty’s Stationery Office.

Richard Huxtable is Professor of Medical Ethics & Law and Director of the Centre for Ethics in Medicine, in the Medical School at the University of Bristol. His research primarily concerns legal and ethical issues in end-of-life decision-making, surgical ethics and clinical ethics. He has published widely in the field of medical ethics and law and his eight books include Healthcare Ethics, Law and Professionalism (Routledge, 2018), Law, Ethics and Compromise at the Limits of Life: To Treat or Not to Treat? (Routledge, 2012), and Euthanasia, Ethics and the Law: From Conflict to Compromise (Routledge-Cavendish, 2007). Richard’s research projects include a major Wellcome Trust collaborative project, Balancing Best Interests in Healthcare, Ethics and Law (BABEL). He has given advice and expert evidence to a range of commissions on assisted dying internationally. He has also served on various ethics committees, including those of the British Medical Association and the Royal College of General Practitioners, and he is Chair of the UK Clinical Ethics Network. Jonathan Ives is Professor of Empirical Bioethics and Deputy Director of the Centre for Ethics in Medicine, in the Medical School at the University of Bristol. He has published on a wide range of topics across clinical ethics, research ethics, and parenting/fatherhood ethics, and is lead editor of ‘Empirical Bioethics: Theoretical and Practical Perspectives’ (Cambridge University Press). Jon sits on the UK’s National Institute for Clinical Excellence (NICE) Highly Specialised Technology Evaluation Committee, and is Chair of the Bristol, North Somerset and South Gloucestershire Risk and Ethics Advisory Forum. At the University of Bristol, Jon is Head of Section for Health Care Evaluation in the Medical School, and Deputy Director of the Centre for Ethics in Medicine. He teaches on a wide variety of undergraduate and postgraduate programmes.

Chapter 8

Saving Lives with Assisted Suicide and Euthanasia: Organ Donation After Assisted Dying David M. Shaw

Abstract In this chapter I consider the narrow and wider benefits of permitting assisted dying in the specific context of organ donation and transplantation. In addition to the commonly used arguments, there are two other neglected reasons for permitting assisted suicide and/or euthanasia: assisted dying enables those who do not wish to remain alive to prolong the lives of those who do, and also allows many more people to fulfill their wish to donate organs after death. In the first part of this chapter I explore the possibility of allowing those who die with assistance to donate their organs and the potential benefits of doing so in countries where some form of assisted dying is legal; in the second part I consider the added force that organ donation considerations bring to the argument in favour of legalizing assisted dying in countries where such practices remain forbidden. Keywords Organ donation · Assisted suicide · Transplant recipients · Wishes of the dying · Organ retrieval

8.1 Introduction The ongoing debate concerning the ethics of assisted suicide and euthanasia continues to focus on the interests of the person who wants to die and the perceived risk to vulnerable patients posed by legalizing the practice. On the one hand, proponents of assisted dying argue that it is inhuman to deny assistance in dying to terminally ill or severely handicapped people when they are experiencing immense suffering or indignity and wish to die; on the other, campaigners for the disabled warn that assisted dying legislation implies that their lives are not worth living, and other opponents warn that vulnerable groups might feel themselves a burden and that legislating for assisted dying is the beginning of a slippery slope. These are all old arguments. Although this war appears to be slowly being won by advocates of assisted dying, this is largely due not to any novel reasons but rather to modern society’s acceptance D. M. Shaw (B) Institute for Biomedical Ethics, University of Basel, Basel, Switzerland e-mail: [email protected] © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 M. Cholbi and J. Varelius (eds.), New Directions in the Ethics of Assisted Suicide and Euthanasia, The International Library of Bioethics 103, https://doi.org/10.1007/978-3-031-25315-7_8

137

138

D. M. Shaw

of individual autonomy, particularly in the face of an ugly death. But there are other arguments in favour of assisted dying that also have substantial force. In this chapter I consider the narrow and wider benefits of permitting assisted dying in the specific context of organ donation and transplantation. There are two other additional and neglected reasons for permitting assisted suicide and/or euthanasia: assisted dying enables those who do not wish to remain alive to prolong the lives of those who do, and also allows many more people to fulfill their wish to donate organs after death. In the first part of this chapter I explore the possibility of allowing those who die with assistance to donate their organs and the potential benefits of doing so in countries where some form of assisted dying is legal; in the second part I consider the added force that organ donation considerations bring to the argument in favour of legalizing assisted dying in countries where such practices remain forbidden.

8.2 Part I Despite the seeming turning of the tide in the debate about assisted dying, only a few jurisdictions currently permit assisted suicide or euthanasia. Switzerland, Germany, Columbia, Albania and Japan all permit assisted suicide under certain circumstances, as do five American states (Vermont, Washington, New Mexico, Montana and Oregon). Euthanasia is only permitted in Belgium, the Netherlands and Luxembourg. In the Netherlands, euthanasia accounts for almost 3% of all deaths (Steck et al. 2013). While some countries have terminal illness as a requirement for accessing assisted dying services, many do not. Several thousand people use assisted dying services globally every year, but they are typically not able to donate their organs, for a variety of practical reasons. It might be assumed that organs would not be suitable for transplantation following assisted suicide both because of the manner of death and because of the fact that most people who opt for assisted suicide have terminal illnesses that would tend to rule out donation. In fact, it is not medical contraindication that is the main barrier. Most assisted suicide and euthanasia techniques are not toxic to organs (Wood et al. 2003), and as many as 50% of those using assisted dying do not have conditions that would rule out donation, such as cancer (and even cancer is not a total contraindication) (Fischer et al. 2008, 810). Advanced age is also no barrier to donation, and some of those using assisted suicide services are neither old nor terminally ill, meaning that their organs might be in prime condition (Intensive Care Society (UK) 2005). The real barriers to organ donation after assisted dying concern location and attitude. Patients often choose to die at home, meaning organs would no longer be viable by the time they reached a hospital. Furthermore, donation after assisted dying would involve donation after cardiac death or DCD (rather than donation after brain death or DBD), a procedure which is still relatively underdeveloped and unsupported in many countries. Furthermore, many doctors and hospitals are resistant to the idea of assisted dying and are therefore reluctant to use organs from this source.

8 Saving Lives with Assisted Suicide and Euthanasia: Organ Donation …

139

However, facilitating organ donation after assisted dying in these countries would be relatively easy given sufficient investment of time and resources (Shaw 2014). Most countries are slowly improving and expanding their DCD capabilities in order to improve donation rates, which will make donation after assisted suicide more viable. If hospitals provided space for assisted suicide or euthanasia on their premises, organs could be transplanted speedily without compromising their function or risking harm to recipients. With regard to attitude, it is well-known that many doctors regard assisting suicide and euthanasia as contrary to their professional duties, but extending this view to the point that it prevents organs reaching recipients who could die without them appears to be taking things rather too far. In fact, transplant teams in Belgium have already cooperated with assisted dying organizations to transplant organs from patients who were euthanized, indicating that donation after assisted dying is possible given collaboration (Ysebaert et al. 2009). It might be asked why we should bother investing time and resources simply to allow a few more people to donate their organs. The answer is that the wider benefits of facilitating organ donation after assisted suicide are potentially immense. If we take the example of Switzerland, around 100 people donate organs after death every year, with around 400 recipients benefiting. However, at least 1000 people remain in need on that waiting list in an average year. Some of these people will die soon if they do not receive an organ; many of them are not at immediate risk of death but will have to continue on dialysis until a suitable organ is found. Now consider that 500 people die via assisted suicide every year in Switzerland. If half of them (250) agreed to donate their organs after death and were medically suitable, 1000 more recipients could benefit; Switzerland could clear its waiting list within one year, and would soon have a surplus of organs. Although Switzerland also allows foreigners to access assisted suicide services, similar arithmetic would apply in other countries. Essentially, any jurisdiction where assisted dying is legal could solve its organ scarcity problem by investing in facilitation of organ donation after assisted suicide and/or euthanasia. In addition to the having the potential to vastly improve the supply of organs, donation after assisted dying will also allow better testing and targeting of organs, reducing the chances of immune rejection. Most people cannot donate their organs unless they have a spontaneous catastrophic brain event or die violently, which means that donation rates are largely dependent on the rate of car crashes and other accidents in a given country. (Indeed, it has been suggested that improved road safety leads to greater scarcity of organs.) In contrast, assisted suicide and euthanasia are planned procedures. This means that potential recipients of organs could be preidentified and matched to the organs of the donor days or even weeks in advance of the planned death. In this sense, donation after assisted death is actually better than some more traditional forms of donation. Another advantage of organ donation after assisted dying is that it does not cause any additional harm. While the dying patient might experience some inconvenience due to the necessity of dying near a hospital rather than at home, this does not amount

140

D. M. Shaw

to harm and any patient could die at home and not donate his or her organs if he wished. It is possible that some premortem organ preservation interventions (such as use of heparin) might make the patient uncomfortable, but again, this would only take place with the patient’s consent. We should also consider donation after assisted dying from the perspective of justice. In terms of resource allocation, it makes no sense whatsoever to have a system that permits assisted dying yet allows the immensely precious resource of solid organs to go to waste. Research is continuing into lab- and animal-based generation of organs for transplantation into humans, but it may be decades before any such biotechnology becomes clinically useful. Many terminally ill people want to die but still have perfectly functioning organs, and it is irresponsible of modern healthcare systems to neglect this most invaluable resource. The benefits of permitting both assisted dying and donation after assisted dying are potentially substantial. But a case can be made for enabling donation after assisted suicide or euthanasia even without considering the wider benefits. Suicide and euthanasia are intended to provide an easier death for patients. Knowing that one’s organs will be used to prolong other people’s lives and to alleviate their suffering could also make one’s own death easier to bear. [I have argued elsewhere than euthanasia and assisted suicide can aid eudaimonia, and organ donation after assisted dying adds further force to this argument (Shaw 2009).] Furthermore, in the normal context of organ donation after sudden death, bereaved families are often comforted by the idea that the death of their loved one has the silver lining of helping other patients. While assisted dying is planned rather than accidental, families may nonetheless derive similar consolation from the thought that others will benefit from their relative’s death. Finally, most people who want to donate their organs after death never get a chance to, because they die in the ‘wrong’ manner. In contrast, people who use assisted dying services die in a controlled manner, making donation much more feasible than for most citizens. Given that many of those using assisted dying services around the world are registered organ donors, their wish to donate should be respected like any other registered donor.

8.3 Part II Given the aforementioned considerations, it appears sensible for countries that currently permit assisted suicide or euthanasia to invest resources in facilitating organ donation after assisted dying. However, in countries where all forms of assisted dying are illegal, these factors also provide additional ammunition to those who support assisted dying. While the main reasons to permit assisted suicide and euthanasia will always remain to prevent dying patients’ suffering and grant them death at a time of their choosing, the fact that legalizing assisted dying could save or improve thousands of lives every year is also a highly persuasive one. To take an example, if assisted suicide were legalized in all of the United States, tens of thousands of

8 Saving Lives with Assisted Suicide and Euthanasia: Organ Donation …

141

patients would no longer have to wait for an organ—an immense benefit. Given efficient organ donation and transplantation infrastructure, it is possible that any country that legalized assisted dying could solve its organ shortage at a stroke. (However, this is not to suggest that assisted dying should be legalized merely for this reason—see below.) Legalising assisted dying and facilitating organ donation from those who use the service also makes sense in terms of justice. It is illogical and unethical to have a system where people who want to stop using their organs are forced to carry on doing so, while people who are in desperate need of organs are denied them. It has been estimated that 6000 people per year experience pain and suffering that cannot be managed by palliative care in the last few months of their life in the UK alone; many of these people would probably want to access assisted dying services. At the same time, there are several thousand people who are suffering and/or dying because they need a new organ. It is a sad irony that tens of thousands of healthy organs are prolonging the lives of those who want to die, while failing organs cause the deaths of those who want to carry on living. [Indeed, forcing people to persist in life amounts to their organs providing unwarranted life support (Shaw 2007).] Of course, it is already controversial merely to suggest that organ donation from those using services in countries where assisted dying is legal should be permitted. Going further and suggesting that increasing organ supply is actually a good reason to legalize assisted dying is an even more provocative proposition. Opponents of assisted dying are likely to distort the arguments and insist that the idea is to kill people against their will in order to procure organs for other people. Indeed, I hesitated before writing this chapter given the relative progress in England and Wales represented by Lord Falconer’s Assisted Dying Bill, which is currently progressing through the House of Lords. Even though it is true that legalizing assisted suicide could provide England with a surplus of organs, many people are likely to misunderstand (and misconstrue) the argument, meaning that a good additional reason to legalise assisted dying might be used as a reason against doing so. Controversy itself can be a powerful persuader, even if arguments based solely upon it tend to be very weak. It seems appropriate to anticipate and discuss some of the objections that might be raised in opposition to organ donation after assisted dying. As mentioned above, a key criticism is likely to be that the suggestion is legalizing assisted suicide in order to procure organs for people. This is clearly not the idea at all. There are two separate arguments: · If someone is suffering and wants to die, she should be able to obtain assistance in doing so. · Anyone who wishes to be an organ donor should have those wishes respected if at all possible. It is a happy coincidence that any country that legislates in support of these two assertions will also have an abundance of organs. Currently, most countries support 2 but not 1; somewhat ironically, those that support 1 also support 2 but tend to abandon 2 for those people who put 1 into practice.

142

D. M. Shaw

Despite the weakness of this first objection, there are stronger potential arguments against donation after assisted dying, all of which also apply to countries where assisted dying is already legal. One of these is a variant of the familiar “burden” argument against assisted dying: people might feel obliged to die because they are a burden on their families and friends. This argument has been largely discredited, and opponents of assisted dying seemingly remain oblivious to the fact that people who ask for life-prolonging treatment to be withdrawn might themselves do so because they feel themselves a burden. (And as has been remarked by several commentators, feeling yourself to be a burden can be a perfectly reasonable reason to want to die.) The modified version of the burden argument for this context is that people might feel obliged to end their lives in order to save the lives of others. A similar rebuttal also applies here: it seems very unlikely that the possibility of donating one’s organs after death would be decisive in any choice regarding assisted dying. As I argue elsewhere, “The decision about suitability for assisted suicide must be kept separate from the decision to donate one’s organs” (Shaw 2014): many people will already be registered organ donors before even considering assisted suicide, and careful use of protocols should ensure that those who are not yet decided should only be asked about donation after the decision to end one’s life has been made. (Furthermore, assisted dying legislation in most countries is limited to the terminally ill and severely disabled.) Even if this factor were decisive in a decision to end one’s life, some would argue that would be fair enough; if anything, dying so that several other people might live would be an astonishingly altruistic way to end one’s life. However, this is not to concede that people might regard themselves as burdens on others because they refuse to donate their organs. Only if we subscribed to Hardwig’s concept of a “duty to die” would we be concerned about people feeling that they had an obligation to die in order to help others (Hardwig 1997). It is conceivable that an elderly family member might be tempted to consider assisted suicide if his daughter or granddaughter required a heart transplant, but the same is true of unassisted suicide. Although it is unlikely that anyone would choose to end his life simply because he could donate their organs, it is possible that some people who choose to use assisted suicide might end their lives a little earlier in order to improve viability of any donated organs. However, most people who use assisted suicide die at a time of their choosing and could have gone on living at least a little longer if they wished; if someone wants to die a little sooner in order to donate his organs, that would seem to be compatible with most current criteria for assisted dying. A third objection is that some people who do want assisted dying might not want their organs to be taken after obtaining assistance. But this is a straw man: any such people could simply opt out (or not opt in) to organ donation. It is true that some people who do want to donate their organs might nonetheless prefer to die at home, making organ donation impractical, but that would be their choice. As donation after assisted dying involves DCD, organ viability could be improved by use of certain pre-mortem interventions such as cannulation, which some patients might not want; again, this would be their choice.

8 Saving Lives with Assisted Suicide and Euthanasia: Organ Donation …

143

It might also be objected that people in need of an organ might not want to receive one that has been obtained from someone who received assistance in dying. The NHS does not normally provide detailed information about the circumstances of a donor’s death, so this is unlikely to be a problem. If assisted dying were legalized in the UK it might be decided to make an exception to this rule, but this too seems unlikely. If the system gave recipients the chance to refuse organs on these grounds, it would lead to organ wastage and indeed to the potential death of the recipient, assuming that any potential recipient was so opposed to the idea of assisted suicide that he or she would effectively commit suicide by refusing an organ. Therefore, it seems probable that recipients would not be told about organs’ origin. It might be argued that all potential recipients could at least be made aware of the possibility that they could receive an organ from someone who committed suicide. However, organ donation after violent non-assisted suicide is a quite routine practice in both Switzerland and the UK, provided that the local coroner gives permission. Patients are not usually informed of this possibility, so it is not obvious that the possibility of donation after assisted suicide should be brought to their attention either. Ironically, some of the ethical issues raised by traditional DCD do not occur in the context of donation after assisted dying. Concerns are sometimes voiced about pre-transplantation procedures such as cannulation: in normal DCD cases, this raises issues as the family often cannot be contacted for consent, but the preplanned nature of donation after assisted dying avoids this problem. Another concern with normal DCD is that treatment might be withdrawn before it is futile, but this is also not a problem in donation after assisted dying, because the patient wants to die. Normal DCD also involves a “cooling-off” period after the heart stops to ensure that the patient is dead before organ retrieval begins. However, given that the patient wants to die and is unconscious, it is not obvious that this precaution would be required in donation after assisted dying. (In any case, no heart has ever spontaneously restarted after 60 s and organ retrieval has been initiated after as little as 75 s in some cases.) It has already been suggested that a limited form of euthanasia should be legalized specifically for organ donors to avoid this practical problem with DCD (Wilkinson and Savulescu 2012). While “organ donation euthanasia” would certainly increase the number of viable organs for transplantation, it remains a limited proposal that is unlikely to be realized without more general assisted dying legislation.

8.4 Conclusion I have argued that countries where assisted suicide or euthanasia are legal should endeavour to enable all patients to donate their organs after death. More importantly, countries where assisted dying is not (yet) legal should also consider the potential benefits to other patients in need of adopting a system that allows donation after assisted death. Legalising assisted dying and facilitating organ donation after it would

144

D. M. Shaw

ease the suffering of both the dying and the living. Doing so would grant death to those who want to die and save the lives of those who want to live, while also enabling those dying to fulfil their wish to donate. It is important for everyone involved in the assisted dying debate to remember that each assisted death could also save several lives.

References Fischer, Susanne, et al. 2008. Suicide assisted by two Swiss right-to-die organizations. Journal of Medical Ethics 34: 810–814. Hardwig, John. 1997. Is there a duty to die? Hastings Center Report 27: 34–42. Intensive Care Society (UK). 2005. Guidelines for adult organ and tissue donation. http://www. ics.ac.uk/professional/standards_safety_quality/standards_and_guidelines/organ_and_tissue_ donation_2005. Accessed 15 August 2013. Shaw, David M. 2007. The body as unwarranted life support: A new perspective on euthanasia. Journal of Medical Ethics 33: 519–521. Shaw, David M. 2009. Euthanasia and eudaimonia. Journal of Medical Ethics 35: 530–533. Shaw, David M. 2014. Organ donation after assisted suicide. Transplantation 98: 247–251. Steck, Nicole, Matthias Egger, Maud Maessen, Thomas Reisch, and Marcel Zwahlen. 2013. Euthanasia and assisted suicide in selected European countries and US states: Systematic literature review. Medical Care 51: 938–944. Wilkinson, Dominic J.C., and Julian Savulescu. 2012. Should we allow organ donation euthanasia? Alternatives for maximizing the number and quality of organs for transplantation. Bioethics 26: 32–48. Wood, David, Paul Dargan, and Alison Jones. 2003. Poisoned patients as potential organ donors: Postal survey of transplant centres and intensive care units. Critical Care 7: 147–154. Ysebaert, Dirk, G. Van Beeumen, Kathleen De Greef, Jean-Paul Squifflet, Olivier Detry, Arnaud de Roover, W. van Marie-Héléne Delbouille, Geert Roeyen Donink, Thiery Chapelle, D. van Jean Louis Bosmans, Marie-Elisabeth Faymonville Raemdonck, Steven Laureys, Maurice Lamy, and Patrick Cras. 2009. Organ procurement after euthanasia: Belgian experience. Transplantation Proceedings 41: 585–586.

David M. Shaw is Associate Professor of Health Ethics & Law at the Care and Public Health Research Institute at Maastricht University, and Senior Research Fellow at the Institute for Biomedical Ethics at the University of Basel. He previously worked in the philosophy department at the University of St Andrews and the medical school at the University of Glasgow. He is interested in all areas of bioethics, but particularly public health ethics, research ethics and shared decision-making.

Chapter 9

Everyday Attitudes About Euthanasia and the Slippery Slope Argument Adam Feltz

Abstract This chapter provides empirical evidence about everyday attitudes concerning euthanasia. These attitudes have important implications for some ethical arguments about euthanasia. Two experiments suggested that some different descriptions of euthanasia have modest effects on people’s moral permissibility judgments regarding euthanasia. Experiment 1 (N = 422) used two different types of materials (scenarios and scales) and found that describing euthanasia differently (‘euthanasia’, ‘aid in dying’, and ‘physician assisted suicide’) had modest effects (≈ 3% of the total variance) on permissibility judgments. These effects were largely replicated in Experiment 2 (N = 409). However, in Experiment 2, judgments about euthanasia’s moral permissibility were best predicted by the voluntariness of the treatment. Voluntariness was a stronger predictor than some demographic factors and some domain general elements of moral judgments. These results help inform some debates about the moral permissibility of euthanasia (e.g., the slippery slope argument) suggesting that some of the key premises of those arguments are unwarranted. Keywords Moral judgment · Political orientation · Physician assisted suicide · Life decision · Active euthanasia

9.1 Introduction In the United States, voluntary passive euthanasia is often thought to be both legally and morally permissible. One reason for the permissibility of voluntary passive euthanasia is that it promotes the two main goals of contemporary medical decision making in the United States—protecting patient autonomy and promoting patient well-being. Allowing the patient to die can respect the patient’s wishes and could result in promoting patient well-being by preventing unavoidable future suffering. However, other types of euthanasia are often thought to be immoral and are illegal in many parts of the world. For instance, the American Medical Association does not A. Feltz (B) Department of Psychology, University of Oklahoma, Norman, OK, USA e-mail: [email protected] © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 M. Cholbi and J. Varelius (eds.), New Directions in the Ethics of Assisted Suicide and Euthanasia, The International Library of Bioethics 103, https://doi.org/10.1007/978-3-031-25315-7_9

145

146

A. Feltz

condone actively taking steps to end a patient’s life (i.e., active euthanasia) (American Medical Association (AMA), Council on Ethical and Judicial Affairs 1992). The conflict concerning the ethical, legal, and procedural permissibility of some types of euthanasia is also reflected in the philosophical literature. Some argue that some kinds of euthanasia are sometimes morally permissible, others argue that those same kinds of euthanasia are not morally permissible (Battin 2005; Beauchamp 2006; Brock 1992; Jackson and Keown 2012; McLachlan 2010; Raz 2013; Velleman 1992). While the correct ethical, legal, and procedural views about euthanasia are important, weighing in on those debates is not the primary goal of this paper. Rather, there is a more modest goal. Parties to the debate often give detailed and nuanced arguments about the permissibility of different kinds of euthanasia (see, for example, some of the chapters in this volume). While these arguments are philosophically rich, they often reference everyday thought, attitudes, or other empirically discoverable facts about whether some kinds of euthanasia are permissible. For example, some have argued that allowing voluntary active euthanasia would result in a slippery slope toward other, less morally permissible forms of euthanasia (e.g., involuntary active euthanasia). In its empirical form, the slippery slope is most efficiently and perhaps best addressed by using empirical methods. Does allowing some kinds of euthanasia actually lead to an acceptance of other less ethically desirable kinds of euthanasia? The overarching aim of this chapter is to provide evidence about everyday attitudes concerning euthanasia by addressing two main issues. The first issue involved measuring the extent to which different ways of characterizing euthanasia (e.g., ‘assisted suicide’ versus ‘aid in dying’) influence everyday attitudes about the morality of those practices. Results from the two experiments suggested that the effect of description is real but small. The second issue involved predicting everyday attitudes about euthanasia. One of the major factors predicting judgments about the moral permissibility of euthanasia was the voluntariness of the decision. Voluntariness predicted attitudes independently of other demographic variables and some domain general components of moral cognition. These results not only provide additional evidence about everyday attitudes about euthanasia’s moral permissibility, they also help inform some philosophical arguments about the ethics of euthanasia’s (e.g., the slippery slope argument). These results suggest that a key empirical premise in the slippery slope argument against euthanasia is false. People who accept some forms of euthanasia simply are not led to accept other, more morally objectionable forms of euthanasia.

9.2 Euthanasia: Philosophical and Empirical Work Generally, it is agreed that there are at least six conceptually distinct kinds of euthanasia. Euthanasia can be passive (allowing a patient do die) or active (actively taking steps to end a patient’s life). Euthanasia can also be voluntary (the procedure is requested), non-voluntary (the patient is unable to request the procedure),

9 Everyday Attitudes About Euthanasia and the Slippery Slope Argument

147

or involuntary (the patient does not want the procedure). Crossing these two categories of euthanasia produces six distinct kinds of euthanasia (e.g., voluntary active euthanasia) (Brock 1992). There is a growing body of empirical research about attitudes toward euthanasia (Achille and Ogloff 1997; DeCesare 2000; Domino 2002; Emanuel 2002; Gamliel 2013; Genuis et al. 1994; Ho and Penney 1992; Jorgenson and Neubecker 1981; MacDonald 1998; Ostheimer 1980; Parkinson et al. 2005; Rogers 1996; Singh 1979; Verbakel et al. 2009; Wolfe et al. 1999). Unfortunately there are some common conceptual problems and ambiguities that make interpreting the ethical significance of these empirical data difficult (Rogers 1996; Wasserman et al. 2005). First, there are definitional confusions about euthanasia. For example, the AMA defines ‘euthanasia’ as “the administering of a lethal agent by another person to a patient for the purpose of relieving the patient’s intolerable and incurable suffering” (American Medical Association (AMA), Council on Ethical and Judicial Affairs 1992). This definition clearly involves active euthanasia only. On this definition, all passive ways to end life are not part of the definition of ‘euthanasia’. More than that, the AMA’s definition does not reference the voluntariness of the euthanasia. These ambiguities also occur in many experimental explorations of attitudes toward euthanasia. For example, in one study, participants were instructed to rate whether practices indicated by a word on a card were justified. One of the words was ‘euthanasia’ (Verbakel et al. 2009). It is unclear which, if any, of the six general notions of euthanasia participants thought ‘euthanasia’ referred to. Second, terms may be appropriately disambiguated yet impermissible inferences are made to euthanasia in general. For example, The National Opinion Research Center has one prominent question that has been analyzed a number of times (DeCesare 2000; Jorgenson and Neubecker 1981; Ostheimer 1980; Singh 1979): “When a person has a disease that cannot be cured, do you think doctors should be allowed to end the patient’s life by some painless means if the patient and his family request it?”. This question clearly focuses on voluntary euthanasia and is naturally (although not necessarily) interpreted as actively ending a patient’s life as opposed to allowing the patient’s life to end. The other five types of euthanasia are left unexplored by this question. Since this procedure constitutes just one kind of euthanasia, it is impermissible (or at least risky) to infer that answers to this question reflect attitudes about euthanasia in general or to make inferences about the moral permissibility of some other types of euthanasia. Finally, kinds of euthanasia can be nested, yet the nested nature is not noticed or is glossed. For example, questions of euthanasia are also discussed under the rubric of “physician assisted death.” Though this is not always acknowledged or made clear, physician assisted death divides into two “species.” In physician assisted suicide, the patient is the one who actually administers the deadly treatment whereas in voluntary active euthanasia the agent who initiates the lethal treatment is typically a doctor (Brock 1992, 10). If there is this conceptual and practical distinction, one might think that physician assisted suicide is an acceptable form of physician assisted

148

A. Feltz

death but voluntary active euthanasia is not. Or one might think that killing one’s self is not permissible but having a professional do it might be. So, one could think that one type of physician assisted death is permissible, but not both types of physician assisted death. Theorists can avoid many of these problems by stipulating definitions of euthanasia. But, not paying attention to these conceptual distinctions is risky. These conceptual confusions raise the possibility that attitudes toward euthanasia are confounded by terminology rather than assessing core issues about euthanasia—a phenomenon that is similar to the psychological effect known as framing. Typically, framing occurs when apparently logically identical, but different, descriptions of a choice elicit different decisions (for a review, see Levin et al. 1998). The classic example of framing is Tversky and Kahneman’s (1981) Asian Flu case. In this case, participants were asked to decide between two programs to combat a new Asian Flu that will affect up to 600 people. Participants could choose program A that would save 200 lives for sure, or program B that has a 1/3 chance of saving everybody and a 2/3 chance of saving nobody. A different group of participants received a similar description but their choices were between program C where 400 people will die for sure and program D where there is a 1/3 chance nobody dies and a 2/3 chance everybody dies. On the surface, these two descriptions are logically identical. However, in the “save” condition, 72% took that less risky program A whereas in the “die” condition 78% took the more risky program D. One explanation for this phenomenon is that people become risk averse in the “gain frame” to lock in the desirable outcome, whereas people become risk seeking in the “loss frame” to have a chance of avoiding the negative outcome. Similar framing may happen when using different terminology to refer to euthanasia. To illustrate, “physician assisted suicide” may focus people’s attention on a specific type of goal—suicide. One may be inclined under this description to avoid that undesirable goal thereby increasing the odds that one judges it morally impermissible. Another description that may be logically identical to physician assisted suicide is “aid in dying.” ‘Aid’ may focus attention on a very different goal that is evaluated as more favorable. Given that favorable evaluation, one may be more inclined to obtain that positive goal and thus judge the action morally permissible. If physician assisted suicide and aid in dying refer to the very same thing, then perhaps describing the type of euthanasia one way may generate a very different reaction compared to describing euthanasia in a logically identical, alternative way.1 Some data suggest that different descriptions of euthanasia can influence judgments about the legality of euthanasia (Barry 2007). A 1997 public opinion poll conducted by Princeton University found that 45% of people responded ‘yes’ to the following question “Do you think that it should be legal for a doctor to help a terminally ill patient commit suicide, or not?”. However, a poll conducted by Louis Harris and Associates found that 69% of people responded ‘yes’ to the following question 1

It is unclear whether these two descriptions really are logically identical. Even if they are not logically identical, it is an open question whether attitudes about them vary sufficiently for there to be an empirical distinction between the two. See discussion.

9 Everyday Attitudes About Euthanasia and the Slippery Slope Argument

149

“Do you think that the law should allow doctors to comply with the wishes of a dying patient in severe distress who ask to have his or her life ended, or not?” (Barry 2007). On the face of it, the only substantive difference between the two questions is whether it should be legal for doctors to help patients commit ‘suicide’ or respect patients’ wishes to end their life. This small difference saw the majority of people disagree that the former should be legal while the majority thought the latter should be legal. Others have found a similar difference comparing different measures of euthanasia and physician assisted suicide (Hains and Hulbert-Williams 2013). There are subtle but possibly important differences in the wording of the questions in the two polls, making interpretation of direct comparisons difficult. Huber et al. (1992) provide more systematic and direct evidence. Their studies suggest that there are important differences between end of life decisions described as ‘euthanasia’, ‘mercy killing’, ‘physician assisted suicide’, and ‘some form of control over death’. They asked participants “If adequate safeguards could be developed, would you like to see (one of the four terms) legalized?” (Huber et al. 1992, 7). Averaging across all four descriptions, 64% of people thought that these treatments should be legalized. However, there was variability associated with different descriptions. More people thought that euthanasia should be legalized (about 78%) compared to physician assisted suicide (about 40%). These results suggest that the description can have an impact on judgments about whether euthanasia should be legalized. These studies highlight some difficulties in assessing everyday attitudes about the moral permissibility of euthanasia. First, there are conceptual problems. Terms used in existing studies are often not sufficiently clear to measure the relevant attitudes. Second, studies often measure the legality and not morality of euthanasia. It is sometimes difficult to infer moral permissibility from legality. For example, one could think that euthanasia should be legal while at the same time think it is morally impermissible. Even if there is likely to be some correlation between many legal and moral judgments, the strength of that relation remains unknown. Third, most of the research about wording does not directly compare responses in the same studies or samples. The one study that does relies on one question that may have questionable reliability and that requires replication. Finally, given that there are ambiguous and varied descriptions of end of life decisions involving death, framing effects may influence some judgments about the moral permissibility of those decisions. These conceptual and empirical issues are important for assessing and interpreting some arguments about euthanasia. The slippery slope argument will serve as an illustrative example. The slippery slope argument is often presented in a logical or an empirical form (see Lewis 2007 for an overview). On both versions, accepting some, perhaps morally permissible, version of euthanasia would lead one either conceptually or empirically to accept less morally permissible versions. For example, on the conceptual version, people may not be able to fully appreciate the conceptual difference between non-voluntary and involuntary euthanasia once they have already accepted non-voluntary euthanasia. Or, on the empirical version, accepting some forms of euthanasia would cause one to accept other definitions of euthanasia—or at

150

A. Feltz

least endorse practices that are consistent with those morally objectionable types of euthanasia. So the slippery slope arguments have key premises that, in fact, people (a) do not appreciate conceptual distinctions among types of euthanasia once they accept some types of euthanasia, or (b) accepting some types of euthanasia causes people to accept other types of euthanasia. Unfortunately, the current state of the science does not help much to address either (a) or (b). Conceptual problems make it difficult to interpret whether those who endorse some acceptable forms of euthanasia see no conceptual distinction between less acceptable forms of euthanasia. Relatedly, given the conceptual problems in the currently existing empirical data, it is difficult to understand any of the causal relations among those definitions. Finally, different ways to frame euthanasia could give divergent evidence for (a) and (b). Theoretically, if one focuses on positive aspects (e.g., “aid in dying”) one may find fuller endorsement of all types of euthanasia compared to negative frames for euthanasia (e.g., “physician assisted suicide”). The former may support slippery slope arguments while the latter may not. Without knowing the extent of the influence of framing, it will be difficult to interpret people’s core attitudes about euthanasia. To fully address (a) and (b), new data are required. Experiments 1 and 2 were designed to help provide some of these data.

9.3 Experiment 1 Experiment 1 had three different goals. The first was to measure the effect of different descriptions of end of life decisions on the moral permissibility of those decisions. This was done using two different kinds of materials. The first set of materials was scenarios that systematically altered the description of the end of life decision. The second set of materials involved scales that systematically altered the description of the end of life decision. Based on previous research, it was predicted that the most morally permissible action would be described as “aid in dying.” The least morally permissible treatment would be described as “physician assisted suicide.” Treatments described as ‘euthanasia’ were predicted to be morally permissible, but not as acceptable as aid in dying since “aid in dying” is a proper subset of euthanasia in general (euthanasia could be interpreted in one of its less acceptable forms, i.e., involuntary active euthanasia).

9 Everyday Attitudes About Euthanasia and the Slippery Slope Argument

151

9.3.1 Participants Four hundred and twenty-two participants were recruited from Amazon’s Mechanical Turk.2 Twenty-five participants were excluded for not completing the survey. One participant was excluded for reporting an age less than 18. The mean age was 35.59, SD = 12.9 ranging from 18 to 79. Fifty-six percent (N = 223) were women.

9.3.2 Materials The scenarios were inspired by those developed by Frileux et al. (2003). Their scenarios focused on physician assisted suicide and euthanasia. Their data suggested that generally, physician assisted suicide is less preferred than euthanasia. In addition, their data suggested that requests for euthanasia were one of the primary factors in whether the treatment was acceptable (along with age of patient, mental health, and prognosis). Their scenarios were modified in this experiment to include a description of “aid in dying” in addition to descriptions of physician assisted suicide and euthanasia. Finally, the scenarios were modified to make the non-voluntary versus voluntary nature of the decision clear (see Appendix for the actual text of all six scenarios).3 Participants responded to the moral permissibility of the procedure on a 6-point scale (1 = strongly disagree, 6 = strongly agree). The second set of materials involved scales composed of 11 items concerning euthanasia, physician assisted suicide, and aid in dying (see Appendix for full scales). Participants responded to each prompt on a 6-point scale (1 = strongly disagree, 6 = strongly agree). These scales were based on Roger’s (1996) scale that measured attitudes about euthanasia. The basic methodology was adopted from Kemmelmeier et al. (1999) who systematically changed ‘euthanasia’ to ‘physician assisted suicide’ in each of Roger’s scale items that had ‘euthanasia’ in it.4 In addition to systematically altering ‘euthanasia’ to ‘physician assisted suicide’, one scale also used the phrase ‘aid in dying’. Scales were used in addition to scenarios because one-item measures can be of limited validity. Many extraneous factors, question wordings, or other features idiosyncratic to the scenario or question may influence participants’ responses. Instruments with more than one question allow assessing the degree 2

For an overview of the quality of Amazon Mechanical Turk’s participants, see Buhrmester et al. (2011), Paollacci et al. (2010). 3 It may seem somewhat forced to include the category “non-voluntary physician assisted suicide” since physician assisted suicide is typically taken to be a kind of voluntary, active euthanasia. In the non-voluntary scenario, the wishes of the patient are left unspecified so one cannot be sure if the patient volunteers for the treatment. Alternatively, the patient may be understood to be functioning, yet incompetent (hence, not able to give adequate consent). 4 Kemmelmeier et al. (1999) did not gather data on the ‘euthanasia’ scale, so direct comparisons between the two scales was not possible.

152

A. Feltz

of internal reliability of responses. If items measure roughly the same underlying construct, then the internal reliability of the scale should be relatively high. In this way, the scales provide an additional source of evidence that can converge with evidence from the scenarios. Participants first answered each of the three scales for euthanasia, physician assisted suicide, and aid in dying (counter balanced for order). Participants then were randomly assigned to only one of the six scenarios. Next, participants completed the Ten Item Personality Inventory (TIPI) (Gosling et al. 2003). The TIPI is a brief, 10-item measure of the Big Five personality traits extraversion, openness to experience, emotional stability, agreeableness, and conscientiousness. Next, participants completed the Berlin Numeracy Test (BNT) (Cokely et al. 2012). The BNT is a brief measure of the ability to understand and use statistical information and has been related to increased focus, attention, and some normatively correct choices. Finally, basic demographic information was collected including a brief measure of political orientation: “Here is a seven point scale on which political views people might hold are arranged from extremely liberal to extremely conservative. Where would you place yourself on this scale” (1 = extremely liberal, 4 = moderate, 7 = extremely conservative). This measure of political orientation is an efficient and reliable way to measure general political orientations (Kroh 2007).

9.3.3 Results 9.3.3.1

Scenarios

Scenarios were analyzed first. Means and standard deviations for the 6 scenarios are reported in Table 9.1. An Analysis of Variance (ANOVA) with the different scenarios as the independent variable and responses to the permissibility question as the dependent variable showed an overall difference between scenarios F (5, 390) = 20.43, p < 0.001, ηp 2 = 0.21. There was no main effect of sex F < 1, and sex did not reliably interact with judgments F (5, 384) = 1.67, p = 0.14, ηp 2 = 0.02. Because sex was not reliably related to judgments and for ease of analyses, sex was excluded as an independent variable for all subsequent analyses. Planned comparisons with voluntariness as the independent variable and responses to the permissibility question as the dependent variable revealed an overall Table 9.1 Means and standard deviations from scenarios in Experiment 1 Non-voluntary

Voluntary

Euthanasia

N = 57, M = 3, SD = 1.91

N = 61, M = 4.67, SD = 1.42

PAS

N = 62, M = 2.95, SD = 1.83

N = 74, M = 4.5, SD = 1.8

Aid in dying

N = 73, M = 3.14, SD = 1.86

N = 69, M = 5.04, SD = 1.33

9 Everyday Attitudes About Euthanasia and the Slippery Slope Argument

153

difference between non-voluntary (M = 3.04, SD = 1.86) and voluntary (M = 4.74, SD = 1.55) conditions, F (1, 394) = 96.93, p < 0.001, ηp 2 = 0.2. Pairwise comparisons were next performed for each type of end of life decision (euthanasia, PAS, and aid in dying) to determine the effect of voluntariness on judgments of permissibility. These analyses revealed large overall differences in judgments as a function of voluntariness: Euthanasia, F (1, 116) = 29.37, p < 0.001, ηp 2 = 0.2, physician assisted suicide F (1, 134) = 24.56, p < 0.001, ηp 2 = 0.16, aid in dying F (1, 140) = 48.85, p < 0.001, ηp 2 = 0.26. Finally, analyses were conducted to determine differences in judgments of permissibility as a function of the description. There were no detectable differences in permissibility judgments for non-voluntary descriptions of euthanasia, Fs < 1. There was a significant difference between voluntary physician assisted suicide and voluntary aid in dying F (1, 141) = 4.16, p = 0.04, ηp 2 = 0.03. There was no reliable difference between voluntary euthanasia and voluntary physician assisted suicide, F < 1. There was no reliable difference between voluntary euthanasia and voluntary aid in dying F (1, 128) = 2.36, p = 0.13, ηp 2 = 0.02. Correlations among the dependent variables for the scenarios are reported in Table 9.2. There were no systematic relations between the permissibility question and these demographic factors.

9.3.3.2

Scales

The mean responses and internal reliabilities were similar for the euthanasia scale (M = 4.11, SD = 1.23, α = 0.92), physician assisted suicide scale (M = 4.11, SD = 1.28, α = 0.92) and aid in dying scale (M = 4.22, SD = 1.18, α = 0.91). A mixed-model ANOVA with responses to the three scales as within subjects factors and order of presentation as between subjects factors revealed an overall small effect of description F (2, 392) = 7.33, p = 0.001, ηp 2 = 0.02. However, this effect was qualified by an interaction of order of presentation F (2, 393) = 6.04, p < 0.001, ηp 2 = 0.03. To control for the order effect, only first responses were analyzed. An ANOVA revealed no overall difference among first responses F < 1. Correlations among the dependent variables are reported in Table 9.3. Political orientation predicted permissibility to all three scales. No other reliable relations to the three scales were found.

9.4 Experiment 2 Experiment 1 suggested that there were some modest effects of description on people’s judgments of euthanasia’s moral permissibility. However, there seemed to be remarkable consistency among judgments. For example, there were strong correlations between responses to the scaled items (rs > 0.83). Experiment 1 also suggested that voluntariness was an important factor in attitudes toward euthanasia

0.14 0.12

0.05

− 0.04

*p < 0.05 **p < 0.01

0.23

Area

− 0.24*

0.07

− 0.16

0.1

Politics 0.06

0.03

− 0.17

0

0.18

0.13

− 0.05

Age

0.08

0.12

− 0.15

Openness

0.03

0.03

0.09

− 0.14

− 0.22

− 0.21 − 0.37**

− 0.17

− 0.03

0.02

− 0.04

− 0.08

0.2

− 0.02

Aid non-voluntary

0.06

0.12

− 0.08

− 0.06

− 0.03

− 0.12 − 0.05

0.02 − 0.04

− 0.06

PAS voluntary

− 0.17

Gender

− 0.02

0.06

Emotional

Conscientiousness

Agreeableness

0.08

0.19

− 0.03

0.08

BNT

PAS non-voluntary

Euthanasia voluntary

Extraversion

Non-voluntary

Euthanasia

Table 9.2 Correlations for scenarios Experiment 1

− 0.13

− 0.21

− 0.04

0.15

0.12

0.13

0.2

− 0.12

− 0.26*

− 0.01

Aid voluntary

154 A. Feltz

0.12*

0.05

4. BNT

− 0.04

− 0.04

− 0.03

7. Conscientiousness

− 0.02

*p < .05, **p > .01

13. Area

− 0.04

0.02

− 0.02

− 0.3**

− 0.01

− 0.28**

11. Gender

12. Politics

0.06

0.01

− 0.03

10. Age

-0.05

− 0.02

− 0.33**

0.09

0.07

− 0.04

0.06

8. Emotional

9. Openness

− 0.01

− 0.07 − 0.04

− 0.05

− 0.08

− 0.05

0.06

5. Extraversion

0.08

1

3

6. Agreeableness

0.88*

0.86**

0.83**

2. PAS

1

2

3. Aid

1

1. Euthanasia

1

Table 9.3 Correlations for scales Experiment 1

− 0.05

− 0.09

− 0.09

0.03

0

0.09

0.04

0.01

0.21** 0.23**

0.04

0.08

0.07

1

5

− 0.04

0.01

− 0.04

− 0.03

1

4

− 0.02

− 0.05

0.08

0.19**

0.23**

0.43**

0.29**

1

6

0.01

0.1

0.05

0.2**

0.15**

0.43**

1

7

0.01

0.12*

− 0.22**

0.24**

0.18**

1

8

− 0.03

− 0.11*

0.04

− 0.03

1

9

0.05

0.1*

0.03

1

10

0.11*

− 0.03

1

11

0.1*

1

12

1

13

9 Everyday Attitudes About Euthanasia and the Slippery Slope Argument 155

156

A. Feltz

suggesting that attitudes toward euthanasia may form coherent clusters that center on the voluntariness of the treatment. But what could predict these attitudes across different descriptions? Experiment 2 was designed to help answer this question. In order to predict attitudes toward euthanasia, the Berlin Euthanasia Scale-6 (BE-6) was used. The BE-6 is a 6-item instrument that measures people’s general attitudes about the moral permissibility of euthanasia (Feltz and Cokely, submitted). Evidence from this scale suggests that people do not measurably distinguish active and passive euthanasia, but people do distinguish among three different kinds of voluntariness: Voluntary (i.e., the treatment is requested), non-voluntary (i.e., the person is unable to request treatment due to, for example, a coma), and involuntary (i.e., the person requests the treatment not be performed). The BE-6 uses two items to measure these three different types of euthanasia. In addition, the Moral Foundations Questionnaire (MFQ) was used (Graham et al. 2011). The MFQ measures five different foundations for people’s moral judgments: Harm/care, fairness/reciprocity, in-group/loyalty, authority/respect, and purity/sanctity. The elements of the MFQ have been argued to be major components in people’s general moral views. It was predicted that the BE-6 would be the major predictor of judgments about euthanasia even when considering other demographic variables and the MFQ.

9.4.1 Participants Four hundred and nine participants were recruited from Amazon’s Mechanical Turk. Twenty-two participants were excluded for not completing the survey. Two people were excluded for reporting an age less than 18. The mean age was 37.23, SD = 13.48, Range 18–74. Sixty-seven percent (N = 260) were female.

9.4.2 Materials Participants received the same three scales from Experiment 1, counterbalanced for order. Participants then randomly received only one of the six scenarios from Experiment 1. Next, participants completed the BE-6, MFQ, the TIPI, and the BNT. Finally, basic demographic information was gathered.

9.4.3 Results 9.4.3.1

Scenarios

Analyses proceeded in the same fashion as Experiment 1. Responses to the scenarios were analyzed first. Means and standard deviations are reported in Table 9.4. An

9 Everyday Attitudes About Euthanasia and the Slippery Slope Argument

157

Table 9.4 Means and standard deviations for scenarios in Experiment 2 Non-voluntary

Voluntary

Euthanasia

N = 65, M = 2.62, SD = 1.77

N = 73, M = 4.64, SD = 1.62

PAS

N = 47, M = 2.0, SD = 1.43

N = 66, M = 4.47, SD = 1.75

Aid in dying

N = 64, M = 2.64, SD = 1.79

N = 71, M = 4.42, SD = 1.71

ANOVA with the different scenarios as the independent variable and responses to the permissibility question as the dependent variable showed an overall difference between scenarios F (5, 380) = 29.85, p < 0.001, ηp 2 = 0.28. An ANOVA with voluntariness as the independent variable and responses to the permissibility question as the dependent variable revealed an overall difference between non-voluntary (M = 2.45, SD = 1.7) and voluntary (M = 4.51, SD = 1.69) conditions, F (1, 384) = 142.34, p < 0.001, ηp 2 = 0.27. Pairwise comparisons were next performed to determine differences in permissibility judgments as a function of voluntariness. ANOVAs revealed large differences for each description as a function of voluntariness: Euthanasia, F (1, 136) = 49.57, p < 0.001, ηp 2 = 0.27, physician assisted suicide F (1, 111) = 66.23, p < 0.001, ηp 2 = 0.37, aid in dying F (1, 133) = 35.01, p < 0.001 ηp 2 = 0.21. ANOVAs tested differences in judgments of permissibility as a function of the description. A statistically significant difference was found between non-voluntary euthanasia and non-voluntary physician assisted suicide F (1, 110) = 4.46, p = 0.04, ηp 2 = 0.04, and non-voluntary physician assisted suicide and non-voluntary aid in dying F (1, 109) = 4.71, p = 0.03, ηp 2 = 0.04, but not between nonvoluntary euthanasia and non-voluntary aid in dying (F < 1). Next, voluntary end of life decisions were analyzed using each description as the independent variable and response to the permissibility question as the dependent variable. There were no reliable differences between the different types of voluntary end of life decisions (Fs < 1).

9.4.3.2

Scales

The scales had excellent internal reliabilities and the mean responses were similar for each scale: Euthanasia Scale (M = 3.62, SD = 0.75, α = 0.92), physician assisted suicide scale (M = 3.65, SD = 0.79, α = 0.92), and aid in dying scale (M = 3.74, SD = 0.74, α = 0.92). A mixed-model ANOVA with responses to the three scales as within subjects factors and order of presentation as the between subjects factor revealed an overall significant difference F (2, 382) = 8.16, p = 0.001, ηp 2 = 0.04. However, this effect was qualified by a trend for an interaction of order F (2, 383) = 2.51, p = 0.08, ηp 2 = 0.01. To control for the order effect, first responses were analyzed. An ANOVA found an overall difference between the three scales, F (1, 383) = 4.59, p = 0.01, ηp 2 = 0.02. Pairwise comparisons found a small but reliable difference between euthanasia (M = 3.28, SD = 0.63) and physician assisted suicide

158

A. Feltz

(M = 3.46, SD = 0.63), F (1, 257) = 4.89, p = 0.03, ηp 2 = 0.02, and aid in dying (M = 3.5, SD = 0.58) F (1, 257) = 8.46, p = 0.004, ηp 2 = 0.03, but not between aid in dying and physician assisted suicide (F < 1).

9.4.3.3

Predicting Responses

A separate goal of Experiment 2 was to determine what predicted judgments about the 6 scenarios and 3 scales. Stepwise linear regressions were employed. Stepwise regressions proceed by finding the single best predictor. Then, the next step is to find the two best predictors. Then the next step is to find the 3 … n predictors until some pre-specified level of significant is not met by subsequent predictors. In these series of stepwise regressions, predictors that were significant at the p = 0.05 level were retained in the models. The predictor variables for all analyses involving responses to the 3 scales and 6 scenarios were the BE-6 (Voluntary M = 4.25, SD = 1.67, Non-voluntary M = 3.43, SD = 1.68, Involuntary M = 1.95, SD = 1.27), BNT, MFQ, political orientation, sex, age, and personality. See Table 9.5. for the stepwise regressions for the scenarios and Table 9.6. for the stepwise regressions for the scales. As predicted, the items for the BE-6 were the strongest predictors for the responses to both scenarios and scales. There were no other variables that consistently predicted in these models. Table 9.5 Stepwise regressions for scenarios in Experiment 2 Model Variable Euthanasia 1

PAS

Aid

BE voluntary

Adjusted df R2 0.41

F

1385 267.6

p

2 Rchange F change pF change

< 0.001 0.41

267.6

< 0.001

2

BE 0.42 non-voluntary

2384 141.33 < 0.001 0.01

9.29

0.002

3

Loyalty

3383

7.93

0.005

0.43

98.56 < 0.001 0.01

1

BE voluntary

0.48

1385 356.69 < 0.001 0.48

356.69 < 0.001

2

Gender

0.49

2384 184.52 < 0.001 0.01

6.9

0.009

3

BE 0.49 non-voluntary

3383 125.89 < 0.001 0.01

4.89

0.03

4.55

0.03

4

Fairness

0.5

4382

1

BE voluntary

0.45

1385 310.52 < 0.001 0.45

310.52 < 0.001

2

BE 0.46 non-voluntary

2384 167.82 < 0.001 0.02

14.35

< 0.001

3

Gender

3383 115.2

5.78

0.02

0.47

96.43 < 0.001 0.01

< 0.001 0.01

BE voluntary

BE voluntary

BE voluntary

BE non-voluntary

Emotional stability

Age

1

1

1

2

3

4

PAS voluntary

Aid non-voluntary

Aid voluntary

BE voluntary

BE involuntary

1

1

Gender

3

Euthanasia voluntary

BE non-voluntary

Emotional stability

1

2

Euthanasia

Non-voluntary

PAS non-voluntary

Variable

Model

Table 9.6 Stepwise regressions for scales in Experiment 2

0.58

0.56

0.54

0.46

0.14

0.8

0.24

0.65

0.2

0.16

0.07

Adjusted R2

466

367

268

169

162

164

146

171

161

162

163

df

25.26

30.73

41.83

61.08

11.59

260.39

15.7

131.99

6.37

7.06

6.05

F 0.02 0.001

0.002

< 0.001

< 0.001

< 0.001

< 0.001

0.001

< 0.001

< 0.001

< 0.001

P

0.03

0.03

0.08

0.47

0.16

0.8

0.25

0.65

0.05

0.1

0.09

2 Rchange

4.31

4.28

12.55

61.08

11.59

260.39

15.7

131.99

4.25

7.45

6.05

F change

0.04

0.04

0.001

< 0.001

0.001

< 0.001

< 0.001

< 0.001

0.04

0.01

0.02

pF change

9 Everyday Attitudes About Euthanasia and the Slippery Slope Argument 159

160

A. Feltz

9.5 Discussion Overall, the results from these two experiments suggested that there is some influence of descriptions on the judgments of permissibility for euthanasia. In Experiment 1, there was an overall modest effect of description in the scenarios for some voluntary, but not non-voluntary, end of life decisions. As predicted, physician assisted suicide was the least preferable option and aid in dying was the most preferable option. This modest effect was reproduced with the scales in Experiment 1. Again aid in dying was more permissible than physician assisted suicide. Experiment 2 found a modest effect for voluntary end of life decisions in the scenarios. Aid in dying was preferred to physician assisted suicide. For the scales, aid in dying was preferred to euthanasia. However, no measurable differences were found between aid in dying and physician assisted suicide. In sum, the different descriptions of euthanasia used in the current series of experiments had modest and intermittent effects on judgments of moral permissibility. These results suggest that if there is a stable, reliable effect of different descriptions of euthanasia, this effect is small. To illustrate, one of the largest differences occurred in Experiment 2 between scenarios involving non-voluntary physician assisted suicide and non-voluntary aid in dying. Differences in descriptions accounted for about 4% of the total variance in judgments (d ≈ 0.4). This is typically thought to be a small to medium effect size. To put the effect into perspective, only about 66% of responses to aid in dying were more favorable than the mean response to the physician assisted suicide scenario. In other words, 86% of the distributions of responses between the groups overlap. To illustrate in one final way, a person would have a 61% chance of being able to identify correctly which scenario the participant was responding to given the participant’s answer. Most of the other effects in the current series of studies were smaller. Hence, while the description of the end of life treatment may be a factor in some people’s judgments of the moral permissibility of euthanasia, these descriptions are not very important factors for most people. Even if the effect of description was modest, the current series of studies provide some relevant empirical evidence for some contemporary debates surrounding euthanasia. For one, they add to the evidence that some demographic variables are associated with some judgments about euthanasia. A number of demographic factors have been found to predict judgments about euthanasia including age (DeCesare 2000; Domino 2002; Ho and Penney 1992; Huber et al. 1992; Jorgenson and Neubecker 1981; Ostheimer 1980; Sawyer and Sobal 1987; Singh 1979), sex (Domino 2002; Emanuel 2002; Jorgenson and Neubecker 1981; Sawyer and Sobal 1987; Singh 1979), political orientation (Domino 2002; Emanuel 2002; Hains and Hulbert-Williams 2013; Jorgenson and Neubecker 1981; Sawyer and Sobal 1987; Singh 1979), and religious affiliation (Domino 2002; Emanuel 2002; Genuis et al. 1994; Hains and Hulbert-Williams 2013; Jorgenson and Neubecker 1981; Kemmelmeier et al. 1999; Meier et al. 1998; Ostheimer 1980; Singh 1979; Wasserman et al. 2005). However, these associations are not always consistently found and sometimes some of the associations go in the opposite direction (e.g.,

9 Everyday Attitudes About Euthanasia and the Slippery Slope Argument

161

Chong and Fok (2013) found a positive relation with age and acceptance of euthanasia contrary to other evidence suggesting a negative relation). Results from Experiment 1 supported some of these relations. While there were very few systematic associations with demographic variables for the six scenarios, there was a strong and consistent relation between political orientation and judgments concerning the scaled items. Judgments about the moral permissibility of euthanasia appear to be largely stable, yet varied. There were strong and persistent differences between voluntary and non-voluntary euthanasia independent of the descriptions of euthanasia (Ho 1998). Moreover, the BE-6 was the best predictor of the permissibility judgments for these end of life decisions (Feltz and Cokely, submitted). The BE-6 predicted better than any demographic variables and predicted better than domain general components of moral judgments measured by the MFQ. The lack of a substantive effect of descriptions along with the strong predictive ability of the BE-6 suggests that judgments about euthanasia are relatively stable and center on the voluntariness of the procedure—even if those descriptions refer to conceptually distinct types of euthanasia. In many instances, the BE scale was the only predictor of judgments about the moral permissibility of the end of life treatment. When there were multiple predictors, the BE-6 scale was the major predictor of those judgments accounting for the most variance. For example, the relation of BE-6’s voluntary subscale accounted for 41% of the overall variance in judgments about the permissibility of the voluntary euthanasia scenario in Experiment 2. Additional factors only accounted for about 1% of the variance. The influence of different descriptions of the end of life decision was small at around 3% of the total variance. These findings suggest that judgments about the permissibility of many end of life decisions form a coherent cluster that is relatively uninfluenced by framing. Judgments are much more influenced by the voluntariness of the procedure (about 12 times stronger). This coherent clustering could explain the relatively small influence of framing on permissibility judgments. As long as the procedure is voluntary (or non-voluntary or involuntary), it is relatively less important what kind of treatment it is. Not only do these data suggest that description is not likely to be an important factor in some people’s permissibility judgments, they also undercut some central premises in some arguments against the permissibility of voluntary active euthanasia. As mentioned, one prominent argument against allowing voluntary active euthanasia is the slippery slope argument. Take the logical slippery slope argument first. It appears that, on average, most people accept many forms of voluntary euthanasia. However, people are much less likely to judge as permissible non-voluntary (much less involuntary) types of euthanasia. This suggests that for many people, they have no problem making important moral and conceptual distinctions between different kinds of euthanasia. As such, it appears that they feel no conceptual or logical pressure to accept other kinds of euthanasia given that they accept voluntary euthanasia. Of course, it could be that people are inattentive or simply making a mistake. But to substantiate either of those two possibilities requires an argument that can account for the current body of evidence. The empirical version of the slippery slope argument could still be sound. Is it true that, as a matter of fact, if we allow voluntary active euthanasia, then that would

162

A. Feltz

cause a reduction in the judgments of impermissibility of less desirable forms of euthanasia (e.g., active involuntary euthanasia)? Some empirical data has already been used to help address the strength of the empirical slippery slope argument. Acceptance of active voluntary euthanasia does not necessarily lead to acceptance of any less desirable forms of euthanasia (primarily involuntary active euthanasia) (Lewis 2007). The data reported in this chapter support this claim. Overall, the mean responses to the BE-6 scale indicated that people agree that voluntary and nonvoluntary types of euthanasia are permissible whereas involuntary types of euthanasia are not. The correlation between Voluntary and Involuntary subscales of the BE-6 was very modest (r (386) = 0.15, p = 0.002, or about 2% of the total variance). However, the difference was very large (repeated measures ANOVA F (1, 385) = 542.92, p < 0.001, ηp 2 = 0.59, or about 35% of the total variance). As such, there is little reason to think that one who accepts voluntary active euthanasia will be led to accept involuntary active euthanasia. In fact, overall there are robust differences between Voluntary and Involuntary subscales. Hence, there is some good reason to think that there are already people who accept voluntary euthanasia but do not accept involuntary euthanasia, contrary to what would be predicted by the empirical slippery slope argument. In summary, different descriptions of euthanasia appear to have minimal impact on people’s judgments of the moral permissibility of euthanasia. Other factors like the voluntariness of the decision are more important. In addition, the BE-6 was the best predictor of a host of attitudes about end of life decisions. These data suggest that the BE-6 quickly assesses a cluster of related attitudes about euthanasia. This instrument thereby has the potential to be used to help ensure that patients’ autonomy is respected by following their treatment preferences. Rather than assessing any number of possible framings (e.g., aid in dying, euthanasia, physician assisted suicide), the BE-6 likely does a very good job of predicting these (and other) decisions about ending life.

Appendix Scenarios Instructions: The following scenario is meant to explore some of your feelings toward end of life decision making. It is not meant to test what you know. Euthanasia/Physician Assisted Suicide/Aid in Dying Not Voluntary Mr. Smith has a serious illness that is totally incurable given current knowledge. He is currently receiving the best possible treatment. He suffers atrociously and pain medication cannot relieve his suffering. He has never expressed a wish for (Euthanasia/Physician assisted suicide/Aid in dying). (Euthanasia/Physician assisted suicide/Aid in dying) is morally permissible in this case.

9 Everyday Attitudes About Euthanasia and the Slippery Slope Argument

163

Euthanasia/Physician Assisted Suicide/Aid in Dying Voluntary Mr. Smith has a serious illness that is totally incurable given current knowledge. He is currently receiving the best possible treatment. He suffers atrociously and pain medication cannot relieve his suffering. He has clearly and repeatedly requested (Euthanasia/Physician assisted suicide/Aid in dying). (Euthanasia/Physician assisted suicide/Aid in dying) is morally permissible in this case requested aid in dying. (Euthanasia/Physician assisted suicide/Aid in dying) is morally permissible in this case. Scales Instructions: The following scale is meant to explore some of your feelings toward end of life decision making. It is not meant to test what you know. (* indicates item to be reverse scored.) • (Euthanasia/Physician assisted suicide/Aid in dying) is acceptable if the person is old. • (Euthanasia/Physician assisted suicide/Aid in dying) should be accepted in today’s society. • There are never cases when (Euthanasia/Physician assisted suicide/Aid in dying) is appropriate.* • (Euthanasia/Physician assisted suicide/Aid in dying) is helpful at the right time and place (under the right circumstances). • (Euthanasia/Physician assisted suicide/Aid in dying) is a humane act. • (Euthanasia/Physician assisted suicide/Aid in dying) should be against the law.* • There are very few cases when (Euthanasia/Physician assisted suicide/Aid in dying) is acceptable.* • (Euthanasia/Physician assisted suicide/Aid in dying) should only be used when the person has a terminal illness. • (Euthanasia/Physician assisted suicide/Aid in dying) is acceptable in cases when all hope of recovery is gone. • (Euthanasia/Physician assisted suicide/Aid in dying) gives a person a chance to die with dignity. • (Euthanasia/Physician assisted suicide/Aid in dying) should be practiced only to eliminate physical pain and not emotional pain.

References Achille, Marie A., and James R.P. Ogloff. 1997. When is a request for assisted suicide legitimate? Factors influencing public attitudes toward euthanasia. Canadian Journal of Behavioral Science 29: 19–27. American Medical Association (AMA), Council on Ethical and Judicial Affairs. 1992. Decisions near the end of life. JAMA 267: 2229–2233.

164

A. Feltz

Barry, Vincent E. 2007. Philosophical thinking about death and dying. Belmont, CA: Thomson/Wadsworth. Battin, Margaret Pabst. 2005. Ending life: Ethics and the way we die. Oxford: Oxford University Press. Beauchamp, Tom L. 2006. The right to die as the triumph of autonomy. Journal of Medicine and Philosophy 3: 643–654. Brock, Dan W. 1992. Voluntary active euthanasia. Hastings Center Report 22: 10–22. Buhrmester, Michael, Tracy Kwang, and Samuel D. Gosling. 2011. Amazon’s Mechanical Turk: A new source of inexpensive, yet high-quality, data? Perspectives on Psychological Science 6: 3–5. Chong, Alice Ming Lin, and Shiu-Yeu Fok. 2013. Validation of the Chinese expanded euthanasia attitude scale. Death Studies 37: 89–98. Cokely, Edward T., Mirta Galesic, Eric Schulz, Saima Ghazal, and Rocio Garcia-Retamero. 2012. Measuring risk literacy: The Berlin Numeracy Test. Judgment and Decision Making 7: 25–47. DeCesare, Michael A. 2000. Public attitudes toward euthanasia and suicide for terminally ill persons: 1977 and 1996. Social Biology 47: 264–276. Domino, George. 2002. Community attitudes toward physician assisted suicide. Omega (Westport) 46: 199–214. Emanuel, Ezekiel J. 2002. Euthanasia and physician-assisted suicide: A review of the empirical data from the United States. Archives of Internal Medicine 162: 142–152. Feltz, Adam, and Edward T. Cokely. Submitted. The Berlin euthanasia scale. Frileux, S., C. Lelievre, M.T.M. Sastre, E. Mullet, and Paul C. Sorum. 2003. When is physician assisted suicide or euthanasia acceptable? Journal of Medical Ethics 29: 330–336. Gamliel, Eyal. 2013. To end life or not to prolong life: The effect of message framing on attitudes toward euthanasia. Journal of Health Psychology 18: 693–703. Genuis, Stephen J., Shelagh K. Genuis, and Wei-Ching Chang. 1994. Public attitudes toward the right-to-die. Canadian Medical Association Journal 150: 701–708. Gosling, Samuel D., Peter J. Rentfrow, and William B. Swann. 2003. A very brief measure of the big-five personality domains. Journal of Research in Personality 37: 504–528. Graham, Jesse, Brian A. Nosek, Jonathan Haidt, Ravi Iyer, Spassena Koleva, and Peter H. Ditto. 2011. Mapping the moral domain. Journal of Personality and Social Psychology 101: 366–385. Hains, Carrie A.M., and Nicholas J. Hulbert-Williams. 2013. Attitudes toward euthanasia and physician-assisted suicide: A study of the multivariate effects of healthcare training, patient characteristics, religion and locus of control. Journal of Medical Ethics 39: 713–716. Ho, Robert. 1998. Assessing attitudes toward euthanasia: An analysis of the subcategorical approach to right to die issues. Personality and Individual Differences 25: 719–734. Ho, Robert, and Ronald K. Penney. 1992. Euthanasia and abortion: Personality correlates for the decision to terminate life. Journal of Social Psychology 132: 77–86. Huber, Ruth, V.M. Cox, and W.B. Edelen. 1992. Right-to-die responses from a random sample of 200. The Hospice Journal 8: 1–19. Jackson, Emily, and John Keown. 2012. Debating euthanasia. Portland: Hart. Jorgenson, David E., and Ron C. Neubecker. 1981. Euthanasia—A national survey of attitudes toward voluntary termination of life. Omega-Journal of Death and Dying 11: 281–291. Kemmelmeier, Markus, Eugene Burnstein, and Kaiping Peng. 1999. Individualism and authoritarianism shape attitudes toward physician-assisted suicide. Journal of Applied Social Psychology 29: 2613–2631. Kroh, Martin. 2007. Measuring left-right political orientation: The choice of response format. Public Opinion Quarterly 71: 204–220. Levin, Irwin P., Sandra L. Schneider, and Gary J. Gaeth. 1998. All frames are not created equal: A typology and critical analysis of framing effects. Organizational Behavior and Human Decision Processes 76: 149–188. Lewis, Penney. 2007. The empirical slippery slope from voluntary to non-voluntary euthanasia. Journal of Law Medicine and Ethics 35: 197–210.

9 Everyday Attitudes About Euthanasia and the Slippery Slope Argument

165

MacDonald, William L. 1998. Situational factors and attitudes toward voluntary euthanasia. Social Science and Medicine 46: 73–81. McLachlan, Hugh V. 2010. Assisted suicide and the killing of people? Maybe. Physician-assisted suicide and the killing of patients? No: The rejection of Shaw’s new perspective on euthanasia. Journal of Medical Ethics 36: 306–309. Meier, Diane E., Carol-Ann Emmons, Sylvan Wallenstein, R. Timothy Quill, Sean Morrison, and Christine K. Cassel. 1998. A national survey of physician-assisted suicide and euthanasia in the United States. New England Journal of Medicine 338: 1193–1201. Ostheimer, John M. 1980. The polls: Changing attitudes toward euthanasia. Public Opinion Quarterly 44: 123–128. Paollacci, Gabriele, Jesse Chandler, and Panagiotis G. Ipeirotis. 2010. Running experiments using Amazon Mechanical Turk. Judgment and Decision Making 5: 411–419. Parkinson, Lynne, Katherine Rainbird, Ian Kerridge, Gregory Carter, John Cavenagh, John McPhee, and Peter Ravenscroft. 2005. Cancer patients’ attitudes toward euthanasia and physician-assisted suicide: The influence of question wording and patients’ own definitions on responses. Journal of Bioethical Inquiry 2: 82–89. Raz, Joseph. 2013. Death in our life. Journal of Applied Philosophy 30: 1–11. Rogers, James R. 1996. Assessing right to die attitudes: A conceptually guided measurement model. Journal of Social Issues 52: 63–84. Sawyer, Darwin, and Jeffery Sobal. 1987. Public attitudes toward suicide—Demographic and ideological correlates. Public Opinion Quarterly 51: 92–101. Singh, B. Krishna. 1979. Correlates of attitudes toward euthanasia. Social Biology 26: 247–254. Tversky, Amos, and Daniel Kahneman. 1981. The framing of decisions and the psychology of choice. Science 211: 453–458. Velleman, David J. 1992. Against the right to die. Journal of Medicine and Philosophy 17: 665–681. Verbakel, Wilko F.A.R., Johan P. Cuijpers, Daan Hoffmans, Michael Bieker, Ben J. Slotman, and Suresh Senan. 2009. Volumetric intensity-modulated arc therapy vs. conventional IMRT in head-and-neck cancer: A comparative planning and dosimetric study. International Journal of Radiation Oncology*Biology*Physics 74: 252–259. Wasserman, Jason, Jeffrey M. Clair, and Ferris J. Ritchey. 2005. A scale to assess attitudes toward euthanasia. Omega (Westport) 51: 229–237. Wolfe, Joanne, Diane L. Fairclough, Brian R. Clarridge, Elisabeth R. Daniels, and Ezekiel J. Emanuel. 1999. Stability of attitudes regarding physician-assisted suicide and euthanasia among oncology patients, physicians, and the general public. Journal of Clinical Oncology 17: 1274.

Adam Feltz serves as an associate professor of psychology at the University of Oklahoma and the Center for Applied Social Research. He specializes in theoretical and applied science for ethical and informed decision making.

Chapter 10

“You Got Me Into This …”: Procreative Responsibility and Its Implications for Suicide and Euthanasia Rivka Weinberg

Abstract This paper investigates connections between procreative ethics and the ethics of suicide and euthanasia. Regarding euthanasia/assisted suicide, we might think it too demanding to ask parents to help euthanize their terminally ill, suffering child, but had the parents not procreated, their child wouldn’t need euthanizing. If you need help killing yourself, shouldn’t your parents, who got you into life in the first place—without your consent—help you out of it? Yet knowing that your parents would help you kill yourself may increase your desire to die: a conundrum. Regarding suicide, the fact that we are forced into life should bolster the right to suicide, even for reasons that others might find wanting. The ways in which we are brought into life have moral implications for the ways in which we are entitled to get out of it. Keywords Terminal illness · Parental Obligation · Assisted suicide · Future person · Filial obligation

10.1 The Ethics of Starting and Ending Life: Are They Linked? Usually, when we think about the connections between the ethics of starting life and the ethics of stopping life, we are more focused on the morality of the end than we are on the morality of the beginning. We may be concerned about when we may permissibly end life, at its earliest and latest stages, i.e., the ethics of abortion and euthanasia. Questions regarding the permissibility of these two kinds of killing often center on issues related to personal identity and when people begin to have interests, particularly an interest in continuing to live. Regarding abortion and euthanasia, if we get clearer on how and when identity and interests form and disintegrate, we might be able to set parameters to personal identity and interests. And if we successfully set personal identity and interests parameters, we may think that we can permissibly kill R. Weinberg (B) Department of Philosophy, Scripps College, Claremont, CA, USA e-mail: [email protected] © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 M. Cholbi and J. Varelius (eds.), New Directions in the Ethics of Assisted Suicide and Euthanasia, The International Library of Bioethics 103, https://doi.org/10.1007/978-3-031-25315-7_10

167

168

R. Weinberg

outside those parameters (see McMahan 2003). But, setting our murderous impulses aside, what other connections may there be between the ethics of beginning and ending life? We might question whether these connections exist at all and, even if they do, whether they are of sufficient strength and interest to warrant our attention. We have reason to be skeptical because it seems clear that there is a difference between our evaluations regarding whether a life is worth starting and our evaluations regarding whether a life is worth continuing. For example, although we might regret the fact that we began to exist, that does not rationally compel us to kill ourselves (see Smilansky 2007; Benatar 2006) because we may value the continuation of something that we would not necessarily have chosen to begin. I would not have chosen to walk into that seedy bar but I do not want to break up the party so I go in with my friends. Once inside the seedy bar, I meet a really interesting person—a seedy person, admittedly, but one I find interesting anyway and I then choose to stay even after my group has left. I value the continuation of something even though I did not value starting it. If I had to do it over again, I would still prefer not to go to a seedy bar even though it worked out well for me this time. The difference between our ways of evaluating whether a life is worth starting and whether a life is worth continuing can seem to argue for a disconnect between the ethics of the beginning and the ethics of the end of life. However, if we analyze the reasons for distinguishing between what might make a life worth starting and what might make a life worth continuing, we will see that these reasons, while persuasive as far as they go, don’t provide us with reasons to sever the ethics of the beginning and end of life more generally. There are two main reasons for distinguishing between a life worth starting and a life worth continuing: · Risk: The nature of the risk we take by starting life and by continuing life is radically different. When procreating, we impose tremendous risk on the future person since we don’t know whether they will suffer a terrible birth defect or early life trauma. There is a lot we don’t know and can’t control about a future person’s birth circumstances, and birth circumstances can be very important to determining the course of a person’s life. Once a person is born, although life still poses great risks, some of those risks have passed. The person is alive, and we now know whether their birth circumstances have saddled them with terrible burdens or not. The risks of their being born with terrible problems may have been high but those risks may not have ripened into harms and, now, the risk of their continued life may be relatively low for burdens and high for benefits. It may have been a bad idea to create that person but it may still be a good idea for that person to continue living the life that no one should have started for them. The valuation is different. · Investment: Another major difference between the value of starting and continuing life is the investment people make in their lives. Before we exist, we have no interests, projects, or commitments and we therefore have no investment in the life we might lead should we start leading it. Once we start living, however, we usually take an interest in our own life; we invest in projects and pursuits and

10 “You Got Me Into This …”: Procreative Responsibility and Its …

169

we may want to see those through. We may have an interest in getting a return on our investments or simply in continuing to pursue them because we enjoy or value them. Personal relationships are another example of investments we make, once alive, that may provide us with reason to want to continue our lives. But these relationships don’t necessarily provide us with a reason to start our lives. Indeed, sometimes we don’t enjoy those relationship very much at all but we have committed to them and invested in them and that gives us reason to want to continue them, but it does not necessarily give us reason to have started them in the first place. We can accept all this and still wonder what implications our procreative ethics might have for euthanasia, assisted suicide, and suicide, assuming that these ways of ending life are at least sometimes permissible.1 That is what I would like to explore. I will begin with euthanasia, proceed onto assisted suicide, and end with some thoughts on suicide.

10.2 Procreative Responsibility and (Voluntary) Euthanasia In the Ibsen (1881, Act III) play, Ghosts, Oswald, a man suffering from inherited syphilis, begs his mother, Mrs. Alving, to euthanize him: Oswald: Well, now you have got to give me that helping hand, mother. Mrs. Alving (with a loud scream): I! Oswald: Who has a better right than you? Mrs. Alving: I! Your mother! Oswald: Just for that reason. Mrs. Alving: I, who gave you your life! Oswald: I never asked you for life. And what kind of a life was it that you gave me? I don’t want it! You shall take it back! Is Mrs. Alving obligated to take it back? Is she, as Oswald’s mother, particularly responsible to help him die? If someone is terminally ill, in unrelenting pain, and is begging to be euthanized—if someone deserves and wants to be euthanized, whatever you take that to entail—who ought to do the euthanizing? 1

I am not going to consider the arguments for and against suicide, assisted suicide, and euthanasia. Instead, I will explore the questions regarding the connections between procreative responsibility and suicide, assisted suicide, and euthanasia. For purposes of this discussion, I assume that these ways of ending life are sometimes justified and permissible. I am also not going to consider the possibility that children owe their parents help with ending life, out of gratitude toward their parents for having created or raised them (or for any other reason). In my view, the obligation children may have to their parents is more complex and controversial than the obligations parents have toward their children. I leave filial obligations for others to investigate.

170

R. Weinberg

A popular candidate for the job is The Doctor. This candidate is so popular that some people have all but renamed euthanasia “physician-assisted suicide.” But why doctors? Is it the angelic/scientific costumes—the white coats, the stethoscopes, the blood pressure cuffs, the note pads—that lend doctors the air of purity and gravitas that seem appropriate to the euthanizing task? That may explain why some think doctors should do our euthanizing but it does not justify it. A justification might be that doctors (perhaps especially palliative care or hospice doctors) have the knowledge and skills required to end life reliably and with minimal suffering. But that knowledge is not that hard to acquire, even for a layperson. Why should doctors shoulder the burden of killing? Killing people, even when done at their request and in order to avoid a more painful and prolonged death, is difficult and stressful. It’s not clear to me that doctors are obligated to undertake this burden. Remember executioners? We still have them, actually, though we don’t identify them the way we used to. We could hire the euthanizing job out. Surely, there’d be takers. But somehow this is off-putting. We don’t want eager volunteers killing our loved ones. We don’t want people who are not eager but are desperate for a job and therefore decide to become euthanizers either. So who should a person in need of euthanasia turn to for help? Why not their own parents?2 We can think of many good answers to that question. For starters, the job is likely to be more painful, stressful, awful, and difficult for the parents than for almost anyone else. We might think it nearly impossible for a parent to be able to kill their own child, even in order to relieve unremitting suffering and avoid a more painful and prolonged death for the child. That’s why god, clever dude that he is, tests Abraham’s loyalty by commanding him to kill his own son, Isaac. Killing your own child seems like the ultimate sacrifice, and one we should not ask of people. Besides, given the difficulty that euthanizing one’s own children would likely pose, parents are not the most reliable candidates for the job. But they might still be the ones most responsible to do it. Parents are more obligated, in most cases, to help their children than anyone else is, unless others have put that child in the position of needing the help or have made an explicit commitment to help. So if Harry throws Sally under the bus, he is more responsible to help her recover from her injuries (or to pay for her burial) than are Sally’s parents. But, if Sally is dying slowly and painfully from an incurable terminal illness and is in need of euthanasia, Sally’s parents seem more connected to and responsible for Sally’s predicament than anyone else. They are the ones who put Sally at risk of this outcome when they decided to toss their condoms out the window and have a child. It can seem particularly fitting, I suggest, to ask the people who put you into life to help you out of it. Just as drivers who hit pedestrians who are crossing the street in accordance with the traffic lights are responsible to alleviate harm or mitigate damages when the risk they pose to the pedestrian ripens into a harm, we may consider parents whose 2

Assuming, of course, that their parents are still alive. Given that terminal illnesses usually occur later in life, often this will not be the case. I am interested here in situations where it is the case that a parent of the person in need of euthanasia is still alive.

10 “You Got Me Into This …”: Procreative Responsibility and Its …

171

children are suffering from a painful and prolonged terminal illness responsible to alleviate the harm that results when the risk they pose to their child by creating her ripens into this sort of harm. Procreation imposes foreseeable risks on future people. Parents procreate knowing, or in the position to know, these risks, sometimes quite specifically (as is the case of many autosomal recessive diseases, for example, which run in carrier couples at a 25% risk to their children). But we don’t need specific knowledge of particular risk conditions in order to bear some responsibility for alleviating the resulting harm. Knowing, as we all do, that procreation imposes all of life’s risks on our children is enough to implicate parents in the harm their procreativity foreseeably imposes on their children. To clarify, the central source of the responsibility I am talking about is the responsibility we incur when we impose risks on others or expose others to risk of harm. We all impose risks on each other all the time—it’s the cost of doing business, of living life. Merely breathing near others exposes them to risk (of airborne contagions). As a society, we decide how to handle risk. In my view, this is contractual: we choose the rules that we think are fair to all to live by. Very roughly speaking, we generally: · Permit the risks that we deem worthwhile or necessary; · Set a standard of care that we require to be met in the imposition of those risks; · Deem those who fail to meet the standard to be negligent and liable for their negligence whether the risk ripens into a harm or not; · When the standard of care has been met but the risk ripens into a harm anyway, we usually require the risk imposer or exposer to mitigate damages and/or compensate the victim. To continue our driving example, we allow people to drive. We set a standard of care that includes being a certain age, being sober, passing a test about the rules of the road, etc. But, if the risk we pose to others by driving ripens into a harm anyway, e.g. we slide in the rain into a parked car, then we have to mitigate damages, compensate the victim, pay to fix the car. This risk model is what I have in mind when thinking about procreative responsibility for euthanasia. When people procreate, they expose their children to many life risks, including the risk of getting a painful and terminal illness. If the risk ripens into a harm, the parents may be obliged to mitigate/compensate (Might others be obligated as well, for whatever reason? Quite possibly. But that is not my concern here). I am concerned here about the implications of procreative responsibility and, by exposing children to the need for euthanasia, it is possible that parents incur the responsibility to euthanize their children who need it. What grounds the responsibility is not the fact that parents caused their child to be sick enough to need euthanasia and it is not the fact that parents caused unjustified harm to their child. Rather, it’s that creating persons exposes them to significant risks. One of those risks is the risk of a painful terminal illness. Even if one has not negligently procreated—you didn’t drive drunk—if the risk ripens into a harm, the imposer may be obligated to mitigate the risk or compensate the victim.3 3

David Boonin has suggested that this reasoning may apply to those who breed a dog and raise it as their pet.

172

R. Weinberg

I am not suggesting that parents are required to compensate their children for any and all life burdens. That is similar to a strict liability standard4 that we hold people to when they engage in very high risk behavior for no good reason. An example would be something like owning a pet lion. That’s an unnecessary pastime that puts your neighbors at high risk of injury. If injury does occur, you are responsible even if your pet lion was guarded by an electrical fence that it shockingly managed to breach. Having children, however, is something that people have a strong and legitimate interest in doing. Moreover, some harms that befall people are their own fault or someone other than their parents’ fault. These factors explain why procreativity is not an act held to a strict liability standard of care. So strict liability is not the reason why I think that parents may be obligated to euthanize their children, when euthanasia is called for. The reason is simply that if someone ought to help Sally by euthanizing her, her parents seem more connected to the risk that ripened into her need for euthanasia than anyone else (unless she is dying because she chose to take up smoking, for fun, in her thirties, say). And, for those who find talk of risk imposition speculative or beside the point, we can simply point to the fact that parents are generally more obligated to help their children than other people are. Just as we expect parents to help their children emotionally, physically, and financially, we may expect them to help their children euthanasia-lly, should the need arise. There are various views held about the source of parental obligations and how they are incurred (see Weinberg 2008). Whatever one’s view of the basis of procreative parental responsibility—be it risk imposition, gestation, causation, intent to raise, genetic relation—all theories of what parental responsibility includes, whatever their basis, seem to cover whatever significant needs children have that they are unable to provide for themselves and that parents are able to provide for them (at not undue cost). Thus, regardless of one’s view regarding how parental obligations are incurred, most people think that parental obligations usually include the obligation to care for and raise one’s children and to care for them, even once they are adults, in a variety of special ways (barring special circumstances or justified estrangement). If your child is in serious and legitimate need of something that they can’t provide for themselves and that you can provide for them, your special caring role as a parent, or your special obligations as a parent, or whatever view you take to be correct regarding parentalchild relations, will likely direct you to provide it. I am suggesting that euthanasia might be that serious and legitimate something that your child needs, that she cannot quite provide for herself, and that you can provide for her.

4

Shiffrin (1999) argues in favor of holding all parents to this sort of standard. I argue against Shiffrin’s view in Weinberg (2015).

10 “You Got Me Into This …”: Procreative Responsibility and Its …

173

10.3 Exceptions and Objections 10.3.1 Someone Else’s Fault As I noted earlier, an exception to holding parents particularly responsible to help euthanize their children, when euthanasia is appropriate, is the category of cases where the need for euthanasia is someone else’s fault (not the parents’). If Harry beat Sally within an inch of her life, causing her to need euthanasia to relieve her suffering so that she dies quickly and less painfully rather than slowly and more painfully, it is the person who beat Sally who is most obligated to help her. But, in that sort of case, Sally, her parents, and almost everyone else probably don’t want Harry anywhere near Sally and it would probably increase Sally’s suffering to have Harry involved in helping her die. Still, there may be other cases where the fact that Sally is in need of euthanasia is someone else’s non-criminal, non-sadistic fault and in those cases that person might be more obligated than Sally’s parents to help Sally die. Imagine that Harry introduced Sally to smoking when he was 22 and she was 19. She should have had the good sense to resist engaging in this sick-making act but Harry was really charming and charismatic. Sally succumbed. Now she’s dying, very slowly and agonizingly, of lung cancer. Between the two of them, Harry and Sally are more responsible for Sally’s dying than Sally’s parents are. Sally’s parents are off the risk imposing hook, though they are still, in some sense, responsible to help Sally for any reason and in any way—that’s what parents do and that’s what we think they ought to do (usually, anyway). If Sally’s dying is mostly her own fault, say due to her own risk taking, that too may distance her parents from Sally’s need for euthanasia that might result. If Sally has a sky diving or motorcycle accident that puts her in the position of needing euthanasia, we might think it is her own damn fault and that her parents have suffered enough as it is, by watching this whole process. We would not then find it fair to add to the parents’ suffering by requiring them to be the ones who euthanize their reckless daughter. Sally’s own risk taking distances her parents, in terms of the risk that ripened into a harm, from the harm that befalls her as a result of her self-imposed risk. The connection between Sally’s harm and her parents’ procreativity may be too weak at this point to obligate her parents to help euthanize her. But if Sally inherited her risk-taking proclivities from her parents, who are thrill-seekers themselves and know that thrill-seeking is a heritable trait (see Gower 2000; Friedman 2005), then maybe Sally’s parents are implicated in a close enough way as to reinstate the obligation. In any case, as I suggested earlier, even if we set the risk imposition claims aside, the obligation parents have to help their children, period, regardless of why help is needed, argues in favor of holding parents responsible to help their children with euthanasia, should that need arise. We might also wonder about the possibility of cases where people fight very hard to stay alive, and consciously choose to live.5 Years later, if they need euthanasia, 5

I owe this case to Justin Weinberg.

174

R. Weinberg

are they now solely responsible for risks posed by their own lives, including the risk of needing euthanasia? Are they “born again,” of their own choice, so to speak? I am not sure about this sort of case because it is hard to imagine a realistic instance that is not strongly influenced by the very strong biological survival instinct, which casts doubt on the possibility of true “born again of one’s own volition” type cases.

10.3.2 Too Demanding It may be too hard for parents to actively participate in the death of their own child. This may be the case (though maybe people should think about these potential obligations before they have children). On the other hand, when a person is suffering terribly and euthanasia would be the most merciful act to do for her, parents may want to do it, even though it is hard. Watching your child die a slower and more painful death might be even harder. Some of the most famous and influential lawsuits pressing for the right to die and demanding to be disconnected from life support were filed by parents on behalf of their children (as happened in the Karen Ann Quinlan case).6 Although parents, in those cases, were asking for hospital removal of life support, their lawsuits serve as reason to think that, if necessary, the parents would gladly have removed the life support themselves (though it does not give us reason to believe that the parents would prefer it was them rather than the hospital that euthanized their child). On the other hand, we also have cases where parents are suing hospitals to maintain life support even though the hospitals insist that the patient is brain dead and there is no life to continue to support (see Shoichet 2013). Even though these parents’ reactions to their child’s terminal illness seem to contradict each other, what they have in common is parents trying or thinking that they are trying to do what is best for their children. That’s what most parents try to do and they succeed often enough for me to conclude that, in many cases, when euthanasia is in their child’s best interests, parents will find that they are able to do the extremely difficult act of euthanizing their own child. As a parent, though, I find myself shuddering as I write this and wondering if there might not be something a little (or a lot) wrong, both morally and emotionally, with a parent who is up to the euthanizing task. You can kill your own kid? What’s wrong with you? The depth of feeling and attachment you should have for your own child, and the unique quality of parental love, should make it impossible for you to kill your child. On the other hand, if your kid needs killing and you don’t do it, what kind of mother (or father) are you? Who is your love and caring helping? Is this love for your child or your own selfish self-protection? Maybe both. With regard to euthanasia, parents may be in a dilemma: damned if they do and damned if they don’t.

6

In re Quinlan (1976).

10 “You Got Me Into This …”: Procreative Responsibility and Its …

175

Still more damned if they don’t, I think, given that euthanasia, if and when needed, is something one might do for the sake of another, despite the cost to oneself. That is consistent with the way that parents are often (though certainly not always) expected to treat their children.

10.4 Procreative Responsibility and Assisted Suicide The case for the parental obligation to help their child die is stronger if the child is not in need of euthanasia but, instead, requests assistance with their own suicide, as an adult, because they find life not worth living. The case against parental help with this sort of assisted suicide is stronger as well. It is a more vexing dilemma because the situation is more closely connected to parental decisions to procreate. Let me explain. Shiffrin (1999) famously argued that all procreativity is morally problematic because it violates children’s consent rights, since children are created without their consent even though it’s no harm or deprivation not to exist. I have argued elsewhere against Shiffrin’s view on the basis of children’s lack of consent and autonomy rights (Weinberg 2015). I won’t go through the entire argument here but the gist of it is that parents are entitled to procreate without their child’s consent because children, being not yet fully competent, do not have autonomy or consent rights. Just as parents are entitled to make many other decisions that affect their child without their child’s consent, parents can procreate without their child’s consent (so long as it is reasonable for the parents to think that being procreated will not be contrary to their child’s interests).7 But what if, despite the parents’ reasonable and justifiable procreativity, it turns out that the child, once grown, doesn’t enjoy life and doesn’t want to continue living it? Just as it may be reasonable and within one’s paternalistic authority to give your child violin lessons without asking them first (if they are very young, as they must be if they are to have any shot at getting good at it), it’s also reasonable for the child to stop playing the violin as she grows into an adult who doesn’t enjoy playing the violin. Not playing the violin is pretty easy to do on one’s own so no one needs to help their grown children abandon the violin in favor of pursuits that the grown children do enjoy. But if your children grow into adults who don’t enjoy living, it is not as easy for them to abandon life even if they really hate it. And since you got them into it, maybe you should help them out of it. Before the child becomes a grown up, paternalistic authority may be exercised to prevent her suicide—the child is not old enough to be sure that she really wants to end her life. She is not yet competent to make that sort of decision. But, if a child, once grown, realizes that not only does she no longer want to play the violin, because 7

I say “not contrary” to the child’s interests rather than “in the child’s interests” because I don’t think that anyone has an interest in being created. That’s why paternalism alone will not justify procreativity, in my view. My view is that parents procreate to further their own interests in becoming parents but they are permitted to do so despite not being able to obtain their child’s consent because the child does not have consent rights, among other reasons.

176

R. Weinberg

it is not worth the effort, she also no longer wants to live because it is not worth the effort, she may find walking away from the violin much easier to do on her own than walking away from life by killing herself. It can be hard to commit suicide even if one genuinely wants to die and wishes one were dead. We are biologically programmed for survival and that is an instinct that, like all strong instincts, is involuntary and hard to ignore even if, rationally, we want to ignore it. Moreover, suicide is a leap into the unknown, which is scary to contemplate and no less scary just because one may really not want to live anymore. At least life is the devil you know. Suicide also hurts those who love us, which is a very unfortunate side effect and one that many people who really really want to die are still loath to put their loved ones through. And a suicide attempt—that is, a failed suicide—can leave a person worse off than they were before. If you jump out of a fifth story window and live, the life you then live is likely to make you both more interested in dying and less able to kill yourself than you were before you tried to kill yourself by jumping out a fifth story window. As we see, there are serious obstacles to getting out of a life you don’t want and never agreed to lead. But with a little help from your friends, or, even better, your mama, these obstacles can be alleviated. It’s easier to do something scary with help and support from a friend or a family member. It can make the leap into the unknown feel less lonely and overwhelming. If your loved ones help you commit suicide, you might be justified in thinking that your suicide, while still difficult for them, will be less of a trauma than it would be if you slit your wrists alone at home in the tub, leaving them to find you only after they don’t hear from you for three weeks and get the police to open your door. And, finally, if you have help with your suicide, you are less likely to screw it up and therefore less likely to be worse off than you were before. All this adds up to a good case for wanting help with your suicide. It can make it a more bearable process. Who should help you? A natural answer is, “no one.” If you want to kill yourself just because you don’t like living, it is not so easy to see why anyone has to help you, given that it is not the kind of help that is given at little cost—well, not the kind of help given at little cost by anyone you might want help from, i.e., no Dr. Kevorkian or ax murderer types need apply. What you want is help from your loving and supportive family and friends (yes, even people who have loving and supportive family and friends may tire of living and wish to stop). If you are going to kill yourself regardless, and it would make your end so much easier for you if you could have some help with it, why shouldn’t your family, who are first in line to help you with all of your needs, help you with this one? If a person wants help killing themselves, shouldn’t their parents, who got them into life in the first place—and without their permission or consent—help them get out of life, if it turns out they don’t enjoy living? It might be too hard, even harder than the case of euthanasia (for the terminally ill). Whereas euthanasia presents parents with the tragic task of helping their child avoid a prolonged death, helping a child kill herself because she does not enjoy living is not only tragic but also possibly a huge and lethal mistake. Not everyone who wants to die at some point later regrets that they did not then die at that point. In fact, one of the few to survive jumping off the

10 “You Got Me Into This …”: Procreative Responsibility and Its …

177

Golden Gate Bridge described an experience common to attempted suicide. He said: “I still see my hands coming off the railing …. I instantly realized that everything in my life that I’d thought was unfixable was totally fixable—except for having just jumped.”8 And even if a suicidal person is correct in thinking that her death wish will not pass and that she genuinely will be better off dead, her loved ones might want to work to help her enjoy her life rather than help her end it. It’s hard to give up on your own child. I should hope! Isn’t believing in your child and always holding out hope for her one of a parent’s jobs? But, as a parent, you might still think that if your child is going to kill herself anyway, in order to exit a place you entered her into without her agreement, you should help make that exit easier. The problem is that knowing that your parents will help you kill yourself may, perversely, give you more reason to want to die: your own parents will help you kill yourself! You might as well die. It is reasonable to assume that, normally, if your parents love you as deeply and unconditionally as they are supposed to, they will not be able to participate in your suicide, even as a favor to you. It’s too much like giving up on you. This, I suggest, presents a dilemma for procreative ethics: As parents, you should help your child not want to die yet also help your child if it turns out that they would prefer to die, but the very knowledge that you would help your child die may contribute to your child’s desire to die. Enough dilemmas. Let’s get to the good news. The good news is that thinking about procreative responsibility will help justify your suicide, should you wish to kill yourself.

10.5 Procreative Responsibility and Suicide It is often thought that it is somehow morally or rationally remiss to kill yourself for no especially compelling reason. Over the course of human history, committing suicide has been considered, at one time or another, a sin, a crime, a sign of madness or some combination of all three. I am at a loss as to why suicide has gotten such a bad rap. If I am not enjoying a party, why am I an evil, criminal, maniac if I decide to leave? We may note that the fact that we are born without our consent may bolster our right to kill ourselves, even for reasons that others might deem insufficiently weighty. Simply being tired of living should suffice. Our parents may have been within their rights to create us if the odds of our enjoying life were good. But that doesn’t not mean that if we defy the odds and don’t enjoy life, that we need anything more than that as a reason for suicide. Just as we can stop playing the violin because we just don’t feel like it, we can kill ourselves because we just don’t feel like living. I concede that last bit of reasoning may have been too quick. Kantians may tell me to respect my rational agency and not use myself as a mere means to my own 8

Ken Baldwin, as quoted by Friend (2003).

178

R. Weinberg

happiness by killing myself. Consequentialists might point out that my suicide will decrease the happiness in the world because it will make people feel sad and guilty (though if I am miserable enough or make others miserable enough, I may be able to offset that decrease in happiness with the increase in happiness gained by ending my life). Aristotle might tell me that I am not being a good citizen and that I am certainly not faring well by contemplating suicide (though I may have already noticed this myself). All of these reasons against suicide may be good reasons though it is not clear that they are decisive, but it is not my aim here to show that. What I am suggesting here is that the nature of the reasons we may have to want to die may not need to be as strong as some may have thought in order to justify suicide. Life was thrust upon us and some of us might not appreciate it. We didn’t ask for it. We might not like it. In the words of Bernhard’s (2003) character in Amras, “Why do we still have to live?”9 As with euthanasia and assisted suicide, so too with suicide: the ways in which we are brought into life affect the ways in which we are entitled to get out of it. Acknowledgements My thanks to Cory Aragon, David Boonin, Michael Cholbi, Jukka Varelius, and the audience at the 2014 Rocky Mountain Ethics Congress for helpful comments on this paper.

References Ackerman, Terrence F., Glenn C. Graber, Charles H. Reynolds, and David C. Thomasma, eds. 1987. Clinical medical ethics: Exploration and assessment. Lanham: University Press of America. American Psychiatric Association. 2013. Diagnostic and statistical manual of mental disorders, 5th ed. Washington, DC: American Psychiatric Publishing. American Society for Bioethics and Humanities (ASBH). 2014. Code of ethics and professional responsibilities for healthcare ethics consultants. http://www.asbh.org/uploads/files/pubs/pdfs/ asbh_code_of_ethics.pdf. Accessed 24 July 2014. Anderson, Gary R., and Valerie A. Glesnes-Anderson, eds. 1987. Health care ethics: A guide for decision makers. Gaithersburg: Aspen Publications. Battin, Margaret Pabst. 1994. The least worst death. New York: Oxford University Press. Battin, Margaret Pabst, Rosamond Rhodes, and Anita Silvers, eds. 1998. Physician assisted suicide: Expanding the debate. New York: Routledge. Benatar, David. 2006. Better never to have been: The harm of coming into existence. Oxford: Oxford University Press. Bernhard, Thomas. 2003. Three novellas, trans. Peter Jansen and Kenneth J. Northcott. Chicago: University of Chicago Press. Boller, Francois, Marc Verny, Laurence Hugonot-Diener, and Judith Saxton. 2009. Clinical features and assessment of severe dementia: A review. European Journal of Neurology 9: 125–136. Bullock, Emma C. 2010. Informed consent as waiver: The doctrine rethought? Ethical Perspectives 17: 531–558. CBC News. 2014. Gillian Bennett, suffering with dementia, dies leaving right-to-die plea. CBC News, Aug 21. http://www.cbc.ca/m/touch/news/story/1.2742440. Accessed 30 Apr 2015. 9

The fact that we did not ask to live does not, by itself, tell us that we have no duties of any sort to continue to live. It’s possible that some duties are involuntary. My point here, however, is that the fact that we did not ask to live counts in favor of our being able to stop living, should we so desire.

10 “You Got Me Into This …”: Procreative Responsibility and Its …

179

Crisp, Johnny. 2013. Euthanasia bill runs out of time in Colombia’s senate. Colombia Reports. http:// colombiareports.co/colombia-euthanasia-bill-runs-out-of-time-in-senate/. Accessed 28 Apr 2015. Dees, Marianne, Myrra Vernooij-Dassen, Wim Dekkers, and Chris van Wee. 2010. Unbearable suffering of patients with a request for euthanasia or physician-assisted suicide: An integrative review. Psycho-Oncology 19: 339–352. D’Oronzio, Joseph C. 2002. The suicide note. Cambridge Quarterly of Healthcare Ethics 11: 422. Dworkin, Gerald, Raymond G. Frey, and Sissela Bok, eds. 1998. Euthanasia and physician assisted suicide: For and against. Cambridge: Cambridge University Press. Frankena, William K. 1973. Ethics, 2nd ed. New Jersey: Prentice-Hall. Friedman, Richard A. 2005. Thrills: What’s the lure of the edge? The answer is all in their heads. New York Times, 20 June 2005. Friend, Tad. 2003. Jumpers: The fatal grandeur of the Golden Gate Bridge. New Yorker, 13 Oct. Gill, Michael B. 2005. A moral defense of Oregon’s physician-assisted suicide law. Mortality 10: 53–67. Gower, Timothy. 2000. Are some guys just born reckless? Los Angeles Times, 10 Jan. Hendin, Herbert, and Kathleen Foley. 2004. The case against assisted suicide. Baltimore: The Johns Hopkins University Press. Hohfeld, Wesley N. 1923. Fundamental legal conceptions as applied in judicial reasoning. New Haven: Yale University Press. Huxtable, Richard, and Alex Mullock. Forthcoming. Voices of discontent? Conscience, compromise and assisted dying. Ibsen, Henrik. 1881. Ghosts. http://www.gutenberg.org/files/8121/8121-h/8121-h.htm. Accessed 29 Apr 2015. Legemaate, Johan, Marian Verkerk, Eric van Wijlick, and Alexander de Graeff. 2007. Palliative sedation in the Netherlands: Starting-points and contents of a national guideline. European Journal of Health Law 14: 61–73. Marcel, G.M., Olde Rikkert, Klodiana Daphne Tona, Lieneke Janssen, Alistair Burns, Anatonio Lobo, Philippe Robert, Norman Sartorius, Gabriela Stoppe, and Gunhild Waldemar. 2011. Validity, reliability, and feasibility of clinical staging scales in dementia: A systematic review. American Journal of Alzheimer’s Disease & Other Dementias 26: 357–365. McMahan, Jeff. 2003. The ethics of killing: Problems at the margins of life. Oxford: Oxford University Press. Quill, Timothy E., and Margaret Pabst Battin, eds. 2004. Physician-assisted dying: The case for palliative care and patient choice. Baltimore: The Johns Hopkins University Press. Regional Euthanasia Review Committees. 2010. Annual report 2009. http://www.rtde.eu/node/19. Accessed 29 Apr 2015. Regional Euthanasia Review Committees (EU). Annual reports. http://www.euthanasiecommissie. nl/archiefjaarverslagen.asp. Accessed 29 Apr 2015. Shiffrin, Seana V. 1999. Wrongful life, procreative responsibility, and the significance of harm. Legal Theory 5: 117–148. Shoichet, Catherine E. 2013. Jahi McMath’s mother: ‘I won’t stop fighting, for daughter on life support’. CNN, 23 Dec. Smilansky, Saul. 2007. Preferring not to have been born. In Ten moral paradoxes. Malden, MA: Blackwell. UK Police Firearms Officers Association (PFOA). n.d. Post-shooting procedures. http://www.pfoa. co.uk/211/post-shooting-procedures. Accessed 29 Apr 2015. Van Parijs, Philippe. 1995. Real freedom for all: What (if anything) can justify capitalism. Oxford: Clarendon Press. Veatch, Robert M. 1989. Clinical ethics, applied ethics, and theory. In Clinical ethics: Theory and practice, ed. Barry Hoffmaster, Benjamin Freedman, and Gwen Fraser, 7–25. Clifton, NJ: Humana Press.

180

R. Weinberg

Veteran Health Administration (VHA), National Center for Ethics in Health Care. Integrated ethics: Improving ethics quality in health care. http://www.ethics.va.gov/ECprimer.pdf. Accessed 24 July 2014. Weinberg, Rivka. 2008. The moral complexity of sperm donation. Bioethics 22: 166–178. Weinberg, Rivka. 2015. The risk of a lifetime: How, when, and why procreation may be permissible. New York: Oxford University Press.

Court Cases and Rulings Hoge Raad (1994) HR 21-06-1994, NJ 1994, 656, Chabot. In re Quinlan (70 N.J. 10, 355 A.2d 647 (NJ 1976)).

Laws and Legislation Oregon Death with Dignity Act. 127 Or Rev Stat §800–995 (1997). http://public.health.oregon. gov/ProviderPartnerResources/EvaluationResearch/DeathwithDignityAct/Documents/statute. pdf. Accessed 16 Dec 2013.

Rivka Weinberg is Professor of Philosophy at Scripps College, Claremont, CA. She is the author of The Risk of a Lifetime: How, When, and Why Procreation May Be Permissible, OUP 2015. She specializes in ethical and metaphysical issues regarding birth, death, and meaning, and is currently working on a book about death, time, and meaning.

Chapter 11

Due Care in the Context of Euthanasia Requests by Persons with Psychiatric Illness: Lessons from a Recent Criminal Trial in Belgium Marc De Hert, Sigrid Sterckx, and Kristof Van Assche Abstract Belgium is one of very few countries where euthanasia on the basis of psychiatric illness is legally possible. Three physicians involved in the euthanasia of a 38-year-old woman suffering from psychiatric illness recently faced a criminal trial for “murder by poisoning”, for allegedly having failed to comply with several requirements of the Belgian Euthanasia Law. Although none of the physicians were convicted, the case generated extensive debate, in the media and the general public as well as in the medical profession and among policy makers. In this chapter, we take this trial as the starting point for a critical analysis of the clinical-psychiatric, ethical, and legal issues involved in evaluating euthanasia requests in cases of complex psychiatric disorders. In these cases, the requirements of the Belgian Euthanasia Law could be improved by mandating the advice of two psychiatrists, who are experts in the treatment of the specific condition of the person requesting euthanasia. Simultaneously with the process of evaluating the euthanasia request, there should be a parallel treatment track in which all therapeutic and recovery-oriented options are explored. These two tracks should not be in the hands of the same physician.

M. De Hert (B) University Psychiatric Centre, Katholieke Universiteit Leuven, Kortenberg, Belgium e-mail: [email protected] Department of Neurosciences, Centre for Clinical Psychiatry, Katholieke Universiteit Leuven, Leuven, Belgium Antwerp Health Law and Ethics Chair, University of Antwerp, Antwerp, Belgium S. Sterckx Department of Philosophy and Moral Sciences, Bioethics Institute Ghent, Ghent University, Ghent, Belgium e-mail: [email protected] K. Van Assche Research Group Personal Rights and Property Rights, Faculty of Law, University of Antwerp, Antwerp, Belgium e-mail: [email protected] © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 M. Cholbi and J. Varelius (eds.), New Directions in the Ethics of Assisted Suicide and Euthanasia, The International Library of Bioethics 103, https://doi.org/10.1007/978-3-031-25315-7_11

181

182

M. De Hert et al.

Keywords Euthanasia · Belgium · Mental suffering · Psychiatric illness · Criminal law

11.1 General Introduction Active voluntary euthanasia and/or physician assisted suicide (PAS) are currently legally allowed in 10 countries and new legislation is pending or under consideration in various other jurisdictions (Mroz et al. 2021). In most jurisdictions that allow it, medical assistance in dying (MAID) is only possible for persons with a terminal illness (Calati et al. 2021). The number of countries where MAID is possible for nonterminally ill persons suffering from psychiatric illness is very limited. More specifically, for these persons both PAS and euthanasia are allowed in Luxembourg (since 2009; no reported cases until today) and the Netherlands (since 2002), euthanasia is allowed in Belgium (since 2002), and PAS in Switzerland (since 1942). A recent systematic review indicates that the numbers of patients requesting or receiving MAID based on a psychiatric disorder in countries where it is legal, are gradually increasing (Calati et al. 2021; see also De Hert 2017; De Hert et al. 2015; Dierickx et al. 2017). According to the Belgian Law on Euthanasia, the eligibility criteria to receive euthanasia on the basis of psychiatric illness are: (1) the request should be made by a legally competent adult patient; (2) the request needs to be voluntary, repeated, wellconsidered, and not the result of external pressure; (3) the patient has to be in a medical condition without prospect of improvement; (4) the patient has to experience constant and unbearable physical and/or psychological suffering that cannot be alleviated; and (5) the suffering should result from a serious and incurable disorder caused by illness or accident (Delbeke 2012; Raus et al. 2021). In the recent case of Mortier v. Belgium, the European Court of Human Rights held that the Belgian legal framework on euthanasia for psychological suffering caused by a psychiatric disorder complies with the rights protected under the European Convention on Human Rights (De Hert et al. 2022a).1 The physician who agrees to evaluate the euthanasia request (henceforth: the attending physician), performs a central role in the procedure. The attending physician needs to assess the (non-)fulfilment of the legal criteria in several conversations with the patient, spread over a reasonable period of time. This physician needs to inform the patient about the diagnosis, the possible remaining treatment options, and the possibility of palliative care. In addition, the attending physician needs to consult two independent physicians. The second physician (i.e., the first consulted physician) must be consulted to ascertain the serious and incurable nature of the disorder and the presence of constant and unbearable suffering that cannot be alleviated. The third physician (i.e., the second consulted physician), who needs to be an expert in the patient’s particular condition (a psychiatrist when the primary diagnosis is a psychiatric illness), must be consulted to ascertain the presence of constant and 1

Law of 28 May 2002 on Euthanasia. Belgian Official Gazette 22 June 2002, 2002/09590, 28515.

11 Due Care in the Context of Euthanasia Requests by Persons …

183

unbearable suffering that cannot be alleviated and the voluntary, well-considered, and repeated nature of the euthanasia request. If the attending physician is of the opinion that the euthanasia request meets the legal criteria, in case of non-terminal illness the euthanasia can only be performed after a one-month waiting period counted from the date the request was first made. If the euthanasia is indeed performed, the attending physician needs to inform the Federal Control and Evaluation Commission for Euthanasia (FCECE) within four working days. Since the Law on Euthanasia does not label the specific crime at issue if the legal criteria are not met, one has to resort to the general criminal law, meaning that any infraction of the requirements of the Law boils down to the crime “murder by means of poisoning” (Delbeke 2012; De Hert 2017). It should, however, be noted that the Belgian Constitutional Court has recently found the application of “murder by poisoning” to any violation of the Law to be disproportionate. As a result, the Belgian legislature will need to adopt a more diversified penalty regime, with lighter penalties for infringements upon procedural conditions that would be deemed to be less essential in guaranteeing the fulfilment of the eligibility criteria (De Hert et al. 2022b).2 MAID for psychiatric patients remains highly controversial and it was no surprise that a criminal trial of three Belgian physicians resulted in local and international media attention and debate in the general public as well as in the medical profession and among policy makers (Day 2018; De Hert 2021a; De Hert et al. 2023; Medscape 2019; The Guardian 2020).3 This trial was unique because, for the first time physicians were charged with “murder by means of poisoning” due to their involvement in euthanasia. The euthanasia of Tine Nys (TN) was performed on 27 April 2010, at age 38. Her original written euthanasia request was made on 22 December 2009. She herself had contacted the three physicians who would become involved in the euthanasia. The physician who performed the euthanasia was a general practitioner (GP), but not her regular GP. The consulted physicians were her regular GP and a psychiatrist. The case was decided before the Court of Assize, where jury members were confronted with the difficult task of judging the actions of physicians faced with the evaluation of a complex psychiatric disorder in the context of a euthanasia request. In the end, none of the physicians was convicted (De Hert 2021b). The attending physician, who performed the euthanasia, was acquitted because his guilt was found to be not proven beyond a reasonable doubt. The first consultant, TNs’ GP, was acquitted because he was considered not to have been fully aware that he was implicated in a euthanasia procedure and, therefore to have been misled by his colleagues. The second consultant, the psychiatrist, was acquitted because the jury found no elements indicating that she did not observe the requirements of the Euthanasia Law that were relevant to her role in the assessment of TN’s request. 2

Penal Code: Article 394 (murder); Article 375 (murder by poisoning); Article 66 (co-perpetrator). Consent of the victim is not a ground for justification under the Penal Code. 3 In the month before and after the trial 1215 news items regarding the trial were published in Belgian newspapers and online news platforms (De Hert et al. 2023).

184

M. De Hert et al.

However, the Court of Cassation found that the acquittal of the attending physician was insufficiently motivated by the jury.4 This physician is now being sued in a civil court case, which is currently on hold because the judge has referred preliminary legal questions to the Constitutional Court (De Hert 2021c). Although the physicians were acquitted by the Court of Assize, the trial has made clear that in this euthanasia case a number of legal requirements had not been met. Since the euthanasia of TN so powerfully illustrates many of the challenges in evaluating euthanasia requests in cases of complex psychiatric disorders, we take this euthanasia and the resulting criminal trial as the starting point for a critical analysis of the clinical-psychiatric, ethical, and legal issues involved. Our analysis is based on information presented during the criminal trial and the ensuing civil litigation (which were attended by the authors of this chapter) as well as published information regarding the euthanasia in question.

11.2 Legal Competence of TN Legal competence is one of the central elements of the debate about MAID in patients suffering from psychiatric disorders. With regard to both legal and medical matters, an adult person is considered to be competent until proof to the contrary is provided. Although a psychiatric disorder can impact on competence, there is no reason to presuppose that a person suffering from a psychiatric disorder lacks decisional capacity. The assessment of legal competence needs to be done on a caseby-case basis by a, preferably well-trained, medical professional (Appelbaum 2007; Berghmans et al. 2014; De Hert and Wampers 2018). It is widely agreed that the assessment of decisional capacity in psychiatric patients is a complex matter, and, as yet, there is no consensus on how this should be operationalised and what the minimal standards should be. Capacity is context-dependent and the complexity of the context/decision will determine which and how criteria should be used in the evaluation. This certainly applies to matters of life and death where the highest standard should be applied (den Hartogh 2016; Doernberg et al. 2016; VVP 2017). Notwithstanding this high standard, the possible impression that a person’s decision seems irrational or unwise in the eyes of the assessor or of other people concerned, should not be central to the assessment. Rather, a judgment on decisional capacity should be based on the quality of the process of decision-making and the values involved. In this regard, 4

Under Belgian law, the jury of the Court of Assize is required to provide a written justification of its decision. If a party disagrees with the decision, it can petition the Court of Cassation, which can quash the decision if it is based on a misinterpretation of the law or breaches an essential procedural requirement, such as, in this particular case, the obligation to sufficiently motivate the decision. Note, however, that an acquittal before the Court of Assize remains standing even if the Court of Cassation has quashed the acquittal. The case is then referred to a civil court, which on the basis of a new assessment of the facts may only decide on the civil liability of the acquitted person (i.e., whether compensations for damages should be awarded).

11 Due Care in the Context of Euthanasia Requests by Persons …

185

it should also be noted that (in)capacity is dynamic, task-specific, and not stable and unchangeable over time (Berghmans et al. 2014; De Hert and Wampers 2018; Ruissen et al. 2011; Thomas et al. 2021). At the criminal trial some issues were raised about the impact of TN’s substance use and depression. Although no formal assessments of capacity were carried out, different medical professionals (at least three GP’s, four psychiatrists, and three psychologists) had seen TN in the months prior to her death and none of them considered her to be lacking decisional capacity. Her death wish was documented in two formal written euthanasia requests (the last one on the day of the euthanasia). At the end of December 2009, she also wrote two legally binding advance directives (Cholbi 2015; Miller et al. 2019): one concerning the wish for euthanasia in case of irreversible coma and one abstaining from life-prolonging treatments and resuscitation.

11.3 Serious and Incurable Psychiatric Illness of TN, Without Prospect of Improvement 11.3.1 General Remarks Under the Belgian Euthanasia Law, a patient can only receive euthanasia on the basis of psychiatric illness if that illness is serious and incurable, and if, in addition, there is no prospect of improvement. Therefore, it had to be established that the psychiatric illness that TN was suffering from was serious. Considering that all serious psychiatric illnesses are incurable, it would in that case also need to be ascertained that a reasonable treatment perspective was not, or no longer available. In this regard, it is well documented that the long-term outcome and course of psychiatric disorders can be fluctuating and difficult to predict. Nevertheless, up to 30% of psychiatric patients over time do not or only partially respond to treatment under routine treatment conditions. For the majority of disorders current evidencebased treatment protocols and international guidelines exist that also include operationalised criteria to define treatment resistance. This assessment is complex and should be comprehensive and take into account aspects of the relevant disorder and its treatment options (i.e., effectiveness, availability, and accessibility), patient-related factors, clinicians- and teams-related factors, as well as environmental and societal influences (De Hert 2017; Levitt and Buchman 2020; van Veen et al. 2020).5

5

It is not without problems to evaluate the quality of assessment and diagnosis of complex psychiatric disorders and their potential treatment options in 2009–2010 (DSM-4). Thus, it is both a journey in time and an evaluation which is potentially tainted by current knowledge about the course and outcome of these disorders (DSM-5 as well as evolving treatment options over time).

186

M. De Hert et al.

Before a conclusion that there is no prospect of improvement can be reached, all reasonable evidence-based pharmacological, psychotherapeutic, and psychosocial treatments should be offered and tried. Reasonable in this context refers to acceptable odds of success, within a realistic time-frame, and an acceptable and/or tolerable risk of side-effects (NVvP 2009; VVP 2017; FMP NVvP 2018). In the euthanasia registration document that the attending physician (belatedly) sent to the FCECE, both Borderline Personality Disorder and Asperger Syndrome were listed as serious and incurable disorders for which no reasonable treatment options had been available for TN.

11.3.2 Borderline Personality Disorder (BPD) 11.3.2.1

Background

Personality disorders are a group of psychiatric disorders with enduring maladaptive patterns of behaviour, cognition, affect, impulse control, inner experience, and interpersonal functioning, exhibited across many contexts and deviating from those accepted by the individual’s culture. These patterns develop early in life (in that they can at least be traceable back to adolescence or early adulthood), are considered inflexible and pervasive, and are associated with significant distress and/or disability (Bateman et al. 2015; Bohus et al. 2021). Apart from the general criteria of a personality disorder, persons with BPD have a long-lasting pattern of instability in interpersonal relationships. They can experience rapid fluctuations in mood and can shift rapidly between love and hate. They have a tendency to perceive themselves and others as all good and all bad alternatingly, resulting in emotional instability, impulsivity, high levels of stress and anxiety, depression, feelings of abandonment, and isolation. Self-injury, suicidal tendencies, and suicidal behaviour/attempts are frequent (Grilo and Udo 2021; Mungo et al. 2020). In 2010 (when the psychiatrist involved in the assessment of TN’s euthanasia request referred her for psychodiagnostic testing), BPD was diagnosed on axis 2 (mainly the diagnosis of the different personality disorders) of the DSM-4 (APA 2000). Since the DSM-5 (APA 2013), axis 2 has been abandoned and personality disorders have moved to axis 1 (together with all other psychiatric disorders), but on the whole the diagnostic criteria have not changed (APA 2013). The diagnosis of personality disorder is made on the basis of the presence and course of specific symptoms, and psychodiagnostic testing can be useful in the differential diagnosis. The treatment of BPD remains challenging. BPD is still often considered as longlasting and not amenable to ‘curative’ treatment. However, the long-term outcome is more favourable than previously suspected (Bateman et al. 2015; Bohus et al. 2021). Short-term hospitalisations and pharmacological interventions can be needed

11 Due Care in the Context of Euthanasia Requests by Persons …

187

for crisis resolution. Pharmacotherapy can be useful to treat specific symptoms or comorbid conditions (e.g., impulsivity, anxiety, and depression) (Bateman et al. 2015; Gartlehner et al. 2021). Psychotherapy is the main treatment for BPD. Recent systematic reviews and meta-analyses confirm the beneficial effects of both comprehensive psychotherapies as well as non-comprehensive psychotherapeutic interventions for BPD core pathology and associated general psychopathology. Several psychotherapeutic approaches are more effective than ‘treatment as usual’ (mainly compared to individual support and crisis intervention). The most studied interventions for BPD, (e.g., dialectic behaviour therapy, schema-therapy, and mentalisation-based therapy) appear to be equally effective. The overall quality of trials is moderate to low and long-term outcome data remain sparse (Cristea et al. 2017; Jørgensen et al. 2021; Mungo et al. 2020; Storebø et al. 2020). Of particular interest in relation to the case of TN is that a number of these evidence-based treatments for BPD, as well as protocolised training in these approaches, were already available in ambulatory and day-hospital settings in 2009 in Belgium (NCCMH 2009; LSMRGGZ 2008).

11.3.2.2

Treatment of the BPD of TN

The treatment history of TN was long and difficult. She was hospitalised about ten times and had long stays in sheltered housing facilities in the community. Most of the admissions were short stays of crisis intervention for suicidal behaviour. The diagnosis BPD was established early and was confirmed at each hospital stay. She met the official criteria for functional disability in 1997, in part due to the physical consequences of a suicide attempt. Her last admission to hospital was at age 34, four years before the euthanasia. Considering the usual severity and turbulent course of BPD during the first 10 years after diagnosis, her global functional outcome in 2009 was not poor. For six years she was able to live independently or with a partner. She completed a graduate education and worked part-time until early 2010. Since her last admission to hospital, she had a regular follow-up with a psychologist in a community mental health centre. The frequency, intensity, and content of this follow-up were insufficiently clarified during the trial. At a minimum, it consisted of supportive counselling by a therapist not trained specifically to deal with persons with personality disorders. Face-to-face contacts lasted until the end of 2009. None of the physicians involved in the euthanasia of TN had contacted the psychologist prior to the euthanasia. The consulted psychiatrist was even convinced that the counselling had stopped in 2007. The consulted psychiatrist claimed that all reasonable treatment options for BPD had indeed been tried and had failed. The court-appointed psychiatry expert agreed that evidence-based treatments for BPD were tried by stating that: in hospital A, at age 19, TN had received a psycho-analytically oriented therapy; and in hospital B, at age 25, she had received behavioural therapy-based interventions (because the Head of the hospital was a known trainer in the latter approach). Although these

188

M. De Hert et al.

statements themselves are not incorrect, they do not support the claim that evidencebased treatment for BPD had been offered to or received by TN in the years before her death. Moreover, given the development of the state of the art as described earlier in this section, a psychiatrist evaluating a euthanasia request in 2010 could/should have been aware of the fact that different evidence-based approaches for BPD were indeed available, both ambulatory and in (semi)residential settings, in several locations in Belgium.6 The consulted psychiatrist should also have discussed and recommended these treatment alternatives to the patient (LSMRGGZ 2008; NCCMH 2009).

11.4 Asperger Syndrome (AS) 11.4.1 A Short History of an Elusive Syndrome A century ago, autistic traits and symptoms were part of the psychosis phenotype. The concept of autism as a separate neurodevelopmental disorder was first described by Leo Kanner (1943) and Hans Asperger (1944).7 Autistic Disorder, both in children and adults, was the prototypical pervasive developmental disorder according to DSM-4 (2000). Other developmental disorders were Asperger Syndrome (AS), Rett’s Disorder, Childhood Disintegrative Disorder, and Pervasive Developmental Disorder, Not Otherwise Specified. AS was considered to be the most important of these disorders (Lord et al. 2020; Rosen et al. 2021). The core features of AS are persistent difficulties in social interactions and communication, the presence of stereotypic behaviours and thinking, restricted interests, and resistance to change. Commonly, the diagnosis of AS is made during childhood, but a subsample of persons with AS, who are better functioning and of normal intelligence, are only diagnosed during adulthood (NICE 2011, 2012; Noens et al. 2018; NVvP 2013). A major shift in diagnosis occurred in DSM-5 (2013) with the introduction of Autism Spectrum Disorder (ASD) as a dimensional concept and continuum of traits and symptoms, with a specifier of severity (Lord et al. 2020, 2021; Rosen et al. 2021). The individual disorder AS was removed, in part due to the lack of diagnostic specificity. In the forthcoming version of the WHO International Classification of Diseases system (due to be published in 2022), AS will be merged into a new ASD diagnosis. Autism Spectrum Disorder is now considered as a complex and usually lifelong developmental disorder that can cause problems with thinking, 6

For the evidence-based psychotherapeutic interventions for BPD, handbooks and protocolised treatment manuals were available in 2010: dialectic behaviour therapy 1993, Dutch translations 1996 and 2002; schema therapy 1997 and 2003, Dutch translation 2005; mentalisation-based therapy 2004 and 2006, Dutch translation 2007. A comprehensive search on the effectiveness of these three approaches was performed by one of the present authors and is available upon request. 7 Recent historical research has discredited Hans Asperger because of his involvement with the Nazi-regime in Vienna (Czech 2018; Sheffer 2018).

11 Due Care in the Context of Euthanasia Requests by Persons …

189

feeling, and language, characterised by persistent difficulties with social communication and social interaction. The concept is much broader than that of pervasive developmental disorders in DSM-4 (cf. supra). A diagnosis of AS is made based on behaviour, clinical symptoms, developmental history, and specific psychodiagnostic assessment with both specific interviews and testing (both clinical and neuropsychological). Since the rate of psychiatric comorbidity is high in both children and adults with AS, a comprehensive diagnostic workout requires differential diagnostic testing for multiple disorders. The evidencebased assessment instruments of the current screening and diagnostic guidelines were available in Dutch in 2007 and 2008, and were not used in the assessment of TN. Because many persons with AS or ASD remain at risk of being undiagnosed, screening instruments have been developed, of which the Autism Spectrum Quotient (AQ) is the most widely studied self-reporting instrument. There is, however, limited evidence for the measurement properties of the screening and diagnostic tools used for AS in adults with a mean normal range of measured intelligence. The sensitivity of these tools is acceptable, but the specificity and negative predictive value are low (Baghdadli et al. 2017; Lord et al. 2020, 2021; Noens et al. 2018). In people with AS or ASD, comorbid symptoms of depression, suicidality, and suicide attempts are not infrequent. The problems for persons with AS to express emotions and thoughts make the diagnosis of suicidal ideation difficult and require adjustments to traditional assessments and psychotherapeutic interventions (Cassidy et al. 2018a, b). Recent studies confirm a high overlap between symptoms of BPD and ASD (Dudas et al. 2017; May et al. 2021). Given the developmental nature of the disorder, strict curative treatments are not (yet) available. Different psychosocial interventions have been developed for people with AS or ASD: ensuring access to care; coping with core features of ASD; focussing on life-skills; management of challenging behaviour; interventions for coexisting conditions; and family interventions (NICE 2011, 2012; NVvP 2013; Poon and Sidhu 2017). One of the effective psychological treatments for BPD is now being considered for suicidality in people with ASD (Huntjens et al. 2020).8

11.4.2 How Likely Is It that TN Had AS? During the evaluation of her euthanasia request, TN was referred for an additional psychodiagnostic evaluation. This resulted in a ‘new’ diagnosis of AS only weeks

8

Both in Belgium and The Netherlands mood and personality disorders are the most frequent diagnoses of persons with psychiatric illness requesting euthanasia and whose euthanasia is granted and performed. However, the percentage of people with autism spectrum disorders, in both the group of requesters and the group who received euthanasia, in Belgium is double the number reported for The Netherlands (Kammeraat and Kölling 2020; Kim et al. 2016; Nicolini et al. 2020; Tuffrey-Wijne et al. 2018; Waddell 2020).

190

M. De Hert et al.

before her death.9 The psychodiagnostic evaluation was performed only to confirm a hypothesis of the psychiatrist and was not designed to offer a differential diagnosis. TN obtained normal scores on five of the six tests applied (including normal scores on putative tests for autistic traits). Nevertheless, the conclusion of the two assessing psychologists was ASD (which was not an official diagnosis in 2010). Moreover, this conclusion was not the outcome of an analysis of the test results but a click-box score of DSM-4 criteria of AS, which was not substantiated by the test results. This seems to leave us with two hypotheses: (1) The AS diagnosis was accurate. How likely is this? In theory, DSM-4 allows for the co-morbidity of BPD and AS in the same person. An indication of another or a new diagnosis should have led to the exploration of new therapeutic options and ways to help the patient cope with this new situation. The problem was that no treatment whatsoever was tried for the condition, or for the potentially ensuing suffering. The psychiatrist claimed that an offer was made but refused by the patient because of long waiting lists and little prospect of improvement in a reasonable time-frame. The experts in the Assize trial initially found this lack of treatment offer, for the new diagnosis, highly problematic. However, after hearing the testimony of the psychotherapist from the community mental health centre (who had no specific expertise on either BPD or AS), surprisingly, they reconsidered their view, and concluded that it was sufficiently established that there were no therapeutic options. (2) The AS diagnosis was not accurate. TN possibly only had some symptoms or traits that are part of the ASD spectrum (the concept introduced by DSM5). Research has confirmed that there can be a substantial overlap between symptoms of BPD and ASD, but that the actual co-occurrence of these disorders is rare (cf. supra). Moreover, none of the numerous discharge notes had ever mentioned AS. Under this hypothesis, a patient who was already considering euthanasia, would have been given an additional (and irrelevant and incorrect) diagnosis of an incurable disorder.

9

During the trial as well as in the advices of the physicians enclosed with the registration instrument sent to the FCECE by the attending physician, there was a constant linguistic/semantic confusion between AS and ASD in the notes of the physicians. The report from the attending physician contained the oxymoronic term ‘pronounced’ AS (AS always being a disabling condition, but milder than autism). The impossibility of having a diagnosis of ASD in 2010 according to DSM-4 was initially acknowledged by the court experts during the criminal investigation but, remarkably, was disregarded by the experts in the Court of Assize. While in DSM-4 a comorbidity between BPD and AS was considered to be possible, according to DSM-5 a comorbidity between BPD and ASD is very unlikely.

11 Due Care in the Context of Euthanasia Requests by Persons …

191

11.5 Constant and Unbearable Suffering of TN, that Cannot Be Alleviated Before a euthanasia on the basis of a serious and incurable psychiatric illness, without any prospect of improvement, can be performed, the Belgian Euthanasia Law requires that this illness causes the patient to suffer constantly and unbearably and that this suffering cannot be alleviated. In this regard, it should be noted that a generally accepted definition and operationalisation of unbearable suffering is not available (Dees et al. 2010, 2011; Verhofstadt et al. 2019a). The personal experience of unbearable suffering is influenced by a complex interaction between illness-related factors, premorbid personality and coping styles, support from the environment, and other social and societal factors. Although it is up to the patient to judge her/his existential mental state, the circumstance that the suffering is both caused by the underlying serious and incurable disorder and that it cannot be alleviated, needs to be assessed by the physicians involved (Gopal 2015; Kissane 2012; Raus and Sterckx 2015; Strang et al. 2004; Varelius 2014). Although, at the time of the euthanasia requests and performance, TN may have been suffering unbearably as a result of at least one serious and incurable psychiatric illness, it is less clear whether there was no possibility of alleviation. As indicated above, evidence-based treatment options that might have offered improvement and, in this way, alleviation of her suffering, had not been tried after TN requested euthanasia. A major reason seems to have been that, according to some testimonies given at the trial, TN was at the time in a state of complete mental exhaustion following the breakup of a relationship mid-2009 which confronted her with prior problems in coping with life and a perceived impossibility to maintain relationships. Although no new treatments to improve her condition and alleviate her suffering were started, the consulted psychiatrist concluded that, because TN lacked resilience and the fact that she was diagnosed with AS at a very late stage, TN, as well as the psychiatrist herself, became convinced that the suffering could not be alleviated. This conclusion may not have been warranted since, despite major reluctance on the part of TN to explore remaining treatment options, the latter were available.

11.6 Role and Responsibilities of the Protagonists in the Euthanasia Procedure 11.6.1 The Attending Physician The attending physician was locally known as a physician willing to perform euthanasia and had partially followed an official training program on euthanasia and

192

M. De Hert et al.

other end-of-life decisions organised by a right-to-die society. During the assessment of the eligibility of her request, he saw TN three times and claimed to have had several telephone contacts with her. His medical file on TN was nearly non-existent, containing three short entries, and none of the required advices of the consultants (since he erroneously had also sent those to the FCECE). Notably, all the tasks that the Euthanasia Law assigns to the attending physician he ‘out-sourced’ to the consulted psychiatrist. At the trial, no conclusive evidence was provided that he informed the patient about her diagnosis and remaining treatment options. He also neglected to inform the FCECE within the legally specified timeframe of four working days (the files were only sent in months later). Moreover, when the file was finally transferred to the FCECE it contained several factual errors.

11.6.2 The First Consulted Physician The first consulted physician was TN’s GP. He had known her for many years and had been actively involved in the treatment of her psychiatric problems (by counselling and medication). This role legally excluded him from being a consultant in a euthanasia request from his patient.10 He testified on different occasions that he was unfamiliar with the Belgian Euthanasia Law. He had not been aware that a ‘letter’ he wrote on the day of the euthanasia was going to be used in the euthanasia approval procedure. He thought that euthanasia required an a priori approval of either the Belgian Medical Association or the FCECE. In his testimony (which he also repeated more recently under oath in the civil case) he maintained that he had been lured by the attending physician into writing this attestation and never agreed to participate in the evaluation of euthanasia.

11.6.3 The Second Consulted Physician (the Psychiatrist) The consulted psychiatrist is a nationally known ‘expert’ in euthanasia for persons with psychiatric disorders. She is the founder of an organisation to evaluate euthanasia

10

The consulted physicians should be independent from both the patient and the attending physician. “Independence” is not defined in the Euthanasia Law but is usually interpreted as implying that there should be no family tie or hierarchical relation between the attending and the consulted physician(s) and that the patient should not have a regular therapeutic relationship with the consulted physician. However, in its latest report, the FCECE stipulates that “independence” does not necessarily mean that the consulted physician should never have met the patient or should not know his or her medical history.

11 Due Care in the Context of Euthanasia Requests by Persons …

193

requests, mainly but not exclusively for people with a psychiatric illness, and she has published several books on the subject.11 She should have realised that the patient’s GP could not legally play a role in the evaluation of the request as a “consulted physician”. The consulted psychiatrist asked TN to see a second psychiatrist. The latter was not able to formulate an advice because he lacked specific knowledge on AS, could not evaluate whether all treatment options were exhausted, and also because TN had only been willing to see him once. The consulted psychiatrist has repeatedly stated that exploring euthanasia requests often opens new routes to treatment possibilities. Although this may well be true, being overly receptive to euthanasia requests may preclude the exploration of alternatives to euthanasia, which, given its irreversibility, is highly problematic from an ethical perspective. Specifically with regard to persons with personality disorders, various experts have argued and stressed the important role of transference and counter-transference (Calati et al. 2021; De Hert 2017; Groenewoud et al. 2004; Mehlum et al. 2020; Michaud et al. 2021; Nicolini et al. 2020). In the evaluation of AS, the consulted psychiatrist only requested a confirmative psychodiagnostic evaluation by a team which knew that a euthanasia request was pending. As explained earlier, the conclusion of the evaluation was not justified by the results of the tests done. The AS diagnosis was not even supported by the autistic traits screening instruments that were used. Although it is now generally acknowledged that self-reporting tests have limited value in adults of normal intelligence (see above), the psychiatrist according to her own statements continues to use these tests routinely in all patients attending her euthanasia evaluation organisation.

11

In a study the consulted psychiatrist has reported on 100 euthanasia requests by psychiatric patients (Thienpont et al. 2015). All patients were considered mentally competent and were suffering from longstanding and treatment-resistant disorders. In 48% of cases the request was granted and the euthanasia was performed on 35 patients. For the evaluation, on average four consultations were needed over a period of eight months. The large standard deviations from the means indicate that it often took more consultations but also that in a large number of cases only one or two consultations were caried out. The assessment of the unbearable, non-alleviable, suffering suggests a high predictive value (both positive and negative predictive value of 80). This is nevertheless sufficiently accurate in matters of life and death. Twenty percent of people for whom the request was approved were still alive two years later. Although all patients at baseline were considered treatmentresistant, the majority of survivors (57%) were able to cope with their illness with or without treatment. Based on published data from the FCECE an analysis estimates that this psychiatrist has been involved as a consultant in one third to half of all cases of euthanasia in Belgium for patients with a psychiatric disorder (Claes et al. 2015).

194

M. De Hert et al.

11.6.4 The Federal Control and Evaluation Commission for Euthanasia The head of the FCECE was a witness before the Court of Assize. He had been contacted for advice by the physicians on trial before the euthanasia was performed and, based on their information, he concluded that all due care criteria were met. It was only after the family had contacted him that it was discovered that the obligatory registration file had not been sent to the FCECE. When the file arrived, it was allegedly unanimously approved. The head of the FCECE was also contacted after the euthanasia for a mediation meeting between the family and the physicians on trial, which failed to bring reconciliation. In his testimony he explained the importance of the process of auto- and peer-control of the physicians in a euthanasia procedure (cf. supra). However, both these types of control were problematic in this case, and this was apparently not detected at the third stage of control (i.e., the a posteriori control) that the legislator has assigned to the FCECE (Raus et al. 2021).

11.7 Lessons to Be Learnt and Concluding Remarks Our analysis of the euthanasia of TN has highlighted several substantial problems in the evaluation and granting of the euthanasia request. The FCECE has repeatedly distinguished between three levels or kinds of control over the euthanasia procedure: (1) ‘auto-control’ (i.e., self-control by the individual physician who is considering whether or not to grant a euthanasia request and therefore needs to assess whether the legal criteria are fulfilled); (2) peer-control (i.e., the involvement of consulted physicians who control whether the other physician(s) involved observe the legal criteria); and (3) a posteriori review of the euthanasia by the FCECE. This three-steps control mechanism promoted by the Belgian Euthanasia Law and the FCECE appears to have failed at each level. Several of the requirements of the Belgian Euthanasia Law were not met in the euthanasia of TN. None of the three physicians involved met their obligations under that Law. A first and essential consideration is that physicians involved in the evaluation of euthanasia requests should know the Euthanasia Law and rigorously apply the legal requirements. The attending physician and the first consulted physician involved in the euthanasia of TN (who are both GP’s) have testified that they had limited knowledge of the Law. The Euthanasia Law requires that only a request from a legally competent person can be considered. Although a psychiatric condition can have a negative impact on competence, there is no reason to presuppose that a person suffering from a psychiatric disorder lacks decisional capacity. At the trial, ten medical professionals who had seen TN in the months before her euthanasia testified that she clearly had the required decisional capacity to make a euthanasia request. Whereas there were

11 Due Care in the Context of Euthanasia Requests by Persons …

195

no concerns regarding competence in this case, a cautious approach should be taken whenever doubts may arise. In this regard, it should be noted that the lack of a formal evaluation of capacity in MAID cases in psychiatric patients has been a recurrent finding in the literature (Kim et al. 2016; Nicolini et al. 2020; Thomas et al. 2021; Verhofstadt et al. 2020). Given the irreversibility of euthanasia, the observation that a formal assessment of capacity is as a rule not even considered when a person suffering from a serious psychiatric illness requests euthanasia, is highly problematic from both an ethical and a legal perspective. TN made repeated euthanasia requests. However, her alleged determination is not an unequivocal proof that her decision was well-considered, considering that during the evaluation process a ‘new’ diagnosis was given only weeks before the euthanasia was performed. There is no doubt that TN suffered from a complex, severe, and long-lasting psychiatric illness. Neither of the GP’s involved had the skills or the experience required to evaluate the diagnosis or possible treatment options. Both relied on the conclusions of a single psychiatrist, acting as the second consulted physician. There is also little doubt that TN’s suffering was severe and unbearable. Being confronted with a ‘new’ and ‘incurable’ disorder will surely not have helped to retain hope for any improvement. Regardless of whether this new diagnosis was accurate or not, the fact remains that appropriate and sufficient evidence-based treatments for her disorder(s) and the suffering related to the disorder(s) had neither been offered nor tried. The psychiatrist claimed that therapeutic options were offered, but were refused by the patient. As in most jurisdictions, a patient in Belgium can legally refuse treatment.12 However, the refusal of a potentially effective treatment (both for the disorder or the suffering induced or caused by the disorder) should not lead to a conclusion that the illness is without prospect of improvement, and that the suffering cannot be alleviated. Hence, a euthanasia request should not be granted merely on the basis of such refusal. A recent Delphi-consensus model has been established by Belgian and Dutch psychiatrists on the issues of essential and necessary diagnostic procedures and treatment criteria (van Veen et al. 2021). This model recommends that psychiatric diagnoses need to be established according to standard diagnostic criteria (e.g., DSM) and confirmed by two independent psychiatrists. Any clinical diagnosis should be accompanied by a narrative description, taking into account the contextual and systemic elements. In patients with long histories of psychiatric illness, additional diagnostic procedures should be limited to those that could lead to new treatment options. MAID should only be considered when all reasonable, guidelines-based, treatment options have been tried and have failed. This implies that suffering is not only time-, but also treatment-dependent and that it should persist over a long 12

Law of 22 August 2002 concerning the Rights of the Patient. Belgian Official Gazette 26 September 2002, 2002/22737, 43719.

196

M. De Hert et al.

period of time. In line with the bio-psycho-social model of psychiatric disorders, this entails both biological, psychological, and social treatments. Additionally, at least one recovery-oriented intervention should have been attempted. If needed, efforts should be made to improve the overall social situation of the patient. The experts, however, agreed that there should be a limit to the number of treatments a patient needs to undergo. Before concluding that psychiatric suffering cannot be alleviated, the following elements should be taken into account: the duration of the illness; the possibilities of responding to an intervention within a reasonable time-frame; the possible side-effects or negative consequences of an intervention; the experience and expertise of the involved physicians; and the process of joint decision-making. In recent years, different initiatives have been taken in Belgium to elaborate clinical guidelines for the evaluation of euthanasia requests in the context of psychiatric illness, some in response to the euthanasia of TN (Verhofstadt et al. 2019b; VVP 2017). Although most of these guidelines suggest additional legal safeguards, all agree that psychiatrists should adhere, not only to the requirements of the Euthanasia Law but also to the current recommendations of the Flemish and Dutch psychiatric associations (FMP NVvP 2018; VVP 2017). An important addition of the Flemish guideline is the recommendation of a two-track approach in the evaluation of a euthanasia request (VVP 2017). One track should comprehensively explore and evaluate the euthanasia request. Simultaneously, a separate treatment track should actively support the patient and explore all remaining therapeutic and/or recoverybased possibilities. This two-track approach is based on two ethical values: respect for the patient’s autonomy (taking the death-wish seriously) and the duty to protect life (exploring all meaningful life perspectives). The Flemish guideline also recommends that the evaluation of a euthanasia request in the context of complex psychiatric disorders should be performed by at least two independent psychiatrists who are experts in the specific condition of the person requesting euthanasia. Our analysis of the euthanasia of TN and the ensuing criminal trial fully corroborates and supports these recommendations and guidelines.

References Appelbaum, Paul S. 2007. Assessment of patients’ competence to consent to treatment. New England Journal of Medicine 357: 1834–1840. American Psychiatric Association. (APA). 2000. Diagnostic and statistical manual of mental disorders (4th ed., TR). Washington DC: American Psychiatric Publishing. American Psychiatric Association. (APA). 2013. Diagnostic and Statistical Manual of Mental Disorders (Fifth ed.). Arlington VA: American Psychiatric Publishing. Baghdadli, Amaria, Frédérick Russet, and Laurent Mottron. 2017. Measurement properties of screening and diagnostic tools for autism spectrum adults of mean normal intelligence: A systematic review. European Psychiatry 44: 104–124. Bateman, Anthony W., John Gunderson, and Roger Mulder. 2015. Treatment of personality disorder. The Lancet 385 (9969): 735–743. Berghmans, Ron, Donna Dickenson, and Ruud Ter Meulen. 2014. Mental capacity: In search of alternative perspectives. Health Care Analysis 12 (4): 253–263.

11 Due Care in the Context of Euthanasia Requests by Persons …

197

Bohus, Martin, Jutta Stoffers-Winterling, Carla Sharp, Annegret Krause-Utz, Christian Schmahl, and Klaus Lieb. 2021. Borderline personality disorder. The Lancet 398 (10310): 1528–1540. Calati, Raffaela, Emilie Olié, Débora Dassa, Carla Gramaglia, Sebastien Guillaume, Fabio Madeddu, and Phillipe Courtet. 2021. Euthanasia and assisted suicide in psychiatric patients: A systematic review of the literature. Journal of Psychiatry Research 135: 153–173. Cassidy, Sarah A., Louise Bradley, Erica Bowen, Sarah Wigham, and Jacqui Rodgers. 2018a. Measurement properties of tools used to assess suicidality in autistic and general population adults: A systematic review. Clinical Psychology Review 62: 56–70. Cassidy, Sarah, Louise Bradley, Rebecca Shaw, and Simon Baron-Cohen. 2018b. Risk markers for suicidality in autistic adults. Molecular Autism 9: 42. https://doi.org/10.1186/s13229-0180226-4. eCollection 2018. Cholbi, Michael. 2015. Medically enabled suicides. In New directions in the ethics of assisted suicide and euthanasia, ed. Michael Cholbi and Jukka Varelius, 169–184. Heidelberg, New York, London: Springer. Claes, Stephan, Ludi Van Bouwel, An Haekens, Marc Eneman, George Otte, Jan De Lepeleire, Marc Calmeyn, Ariane Bazan, and Willem Lemmens. 2015. Euthanasia for psychiatric patients: Ethical and legal concerns about the Belgian practice. BMJ Open. https://bmjopen.bmj.com/con tent/5/7/e007454.responses. Cristea, Ioana A., Claudio Gentili, Carmen D. Cotet, Daniela Palomba, Corrado Barbui, and Pim Cuijpers. 2017. Efficacy of psychotherapies for borderline personality disorder: A systematic review and meta-analysis. JAMA Psychiatry 74 (4): 319–328. Czech, Herwig. 2018. Hans Asperger, National Socialism, and “race hygiene” in Nazi-era Vienna. Molecular Autism 9: 29. Day, Michael. 2018. Three Belgian doctors are investigated over euthanasia of woman with Asperger’s. British Medical Journal 363: k516. De Hert, Marc. 2017. Euthanasie bij mensen met een psychiatrische stoornis, een brug te ver? Antwerpen: Universiteit Antwerpen. De Hert, Marc. 2021a. Renvoi d’une affaire d’euthanasie aux assises par la CMA. Tijdschrift voor Gezondheidsrecht/Revue de Droit de la Santé 2020/2021 (1): 320–326. De Hert, Marc. 2021b. Trois médecins euthanasieurs acquittés du chef d’assassinat par la Cour d’assises. Tijdschrift voor Gezondheidsrecht/Revue de Droit de la Santé 2020/2021 (1): 329–334. De Hert, Marc. 2021c. Le doute qui entoure la culpabilité d’un médecin ayant pratiqué une euthanasie et qui a donné lieu à son acquittement doit être motivé. Tijdschrift voor Gezondheidsrecht/Revue de Droit de la Santé 2020/2021 (1): 338–340. De Hert, Marc, and Martien Wampers. 2018. Bekwaam of onbekwaam? Dat is de (moeilijke) vraag Juridische aspecten in België. In Wilsbekwaamheid in de medische praktijk, ed. Irma Hein and Adger Hondius, 29–46. Utrecht: De Tijdstroom. De Hert, Marc, Liesbet Van Bos, Kim Sweers, Martien Wampers, Jan De Lepeleire, and Christophe U. Correll. 2015. Attitudes of psychiatric nurses about the request for euthanasia on the basis of unbearable mental suffering (UMS). PLoS One 10 (12): e0144749. https://doi.org/10.1371/jou rnal.pone.0144749. De Hert M, Thys E, Catthoor K, Van den Broeck K, Matthys F, Vansteenlandt K, Detraux. 2023. Media coverage of Belgium’s first criminal case concerning euthanasia for psychiatric patients: A content analysis of Flemish newspapers and magazines. Front Psychiatry. https://doi.org/10. 3389/fpsyt.2022.1050086. De Hert M, Loos S, Van Assche K. 2022a. Euthanasia of a person with a psychiatric disorder does not violate the European Convention on Human Rights (Mortier v. Belgium [no. 78017/17]). Eur Psychiatry 65(1):e80. https://doi.org/10.1192/j.eurpsy.2022.2342 De Hert M, Loos S, Van Assche K. 2022b. The Belgian euthanasia law under scrutiny of the highest courts. Lancet Reg Health Eur 24:100549. https://doi.org/10.1016/j.lanepe.2022.100549 Dees, Marianne K., Myrra J. Vernooij-Dassen, Wim J. Dekkers, and Chris van Weel. 2010. Unbearable suffering of patients with a request for euthanasia or physician-assisted suicide: An integrative review. Psychooncology 19 (4): 339–352.

198

M. De Hert et al.

Dees, Marianne K., Myrra J. Vernooij-Dassen, Wim J. Dekkers, Kris C. Vissers, and Chris van Weel. 2011. ‘Unbearable suffering’: A qualitative study on the perspectives of patients who request assistance in dying. Journal of Medical Ethics 37 (12): 727–734. Delbeke, Everlyn. 2012. Juridische aspecten van zorgverlening aan het levenseinde. Antwerpen: Intersentia. den Hartogh, Govert. 2016. Do we need a threshold conception of competence? Medicine Health Care and Philosophy 19: 79–81. Dierickx, Sigrid, Luc Deliens, Joachim Cohen, and Kenneth Chambaere. 2017. Euthanasia for people with psychiatric disorders or dementia in Belgium: Analysis of officially reported cases. BMC Psychiatry 17 (1): 203. https://doi.org/10.1186/s12888-017-1369-0. Doernberg, Samuel N., John R. Peteet, and Scot Y. Kim. 2016. Capacity evaluations of psychiatric patients requesting assisted death in the Netherlands. Psychosomatics 57 (6): 556–565. Dudas, Robert D., Chris Lovejoy, Sarah Cassidy, Carrie Allison, Paula Smith, and Simon BaronCohen. 2017. The overlap between autistic spectrum conditions and borderline personality disorder. PLoS One 12 (9): e0184447. https://doi.org/10.1371/journal.pone.0184447. eCollection 2017. Federatie Medisch Specialisten en Nederlandse vereniging voor Psychiatrie (FMP NVvP). 2018. Levensbeëindiging op verzoek bij patiënten met een psychische stoornis. In Startpagina—Levensbeëindiging op verzoek bij patiënten met een psychische stoornis. Richtlijn. Richtlijnendatabase. Gartlehner, Gerald, Karen Crotty, Sara Kennedy, Mark J. Edlund, Rania Ali, Mariam Siddiqui, Robyn Fortman, Roberta Wines, Emma Persad, and Meera Viswanathan. 2021. Pharmacological treatments for borderline personality disorder: A systematic review and meta-analysis. CNS Drugs 35: 1053–1067. Gopal, Abilash A. 2015. Physician-assisted suicide: Considering the evidence, existential distress, and an emerging role for psychiatry. Journal of the American Academy of Psychiatry and the Law 43 (2): 183–190. Grilo, Carlos M., and Tomoko Udo. 2021. Association of borderline personality disorder criteria with suicide attempts among US adults. JAMA Network Open 4 (5): e219389. Groenewoud, Johanna H., Agnes Van Der Heide, Alfons J. Tholen, Joost W. Schudel, Michiel W. Hengeveld, Bregje D. Onwuteaka-Philipsen, Paul J. Van Der Maas, and Gerrit Van Der Wal. 2004. Psychiatric consultation with regard to requests for euthanasia or physician-assisted suicide. General Hospital Psychiatry 26 (4): 323–330. Huntjens, Anne, Wies L.M.C. van den Bosch, Bram Sizoo, Ad Kerkhof, Marcus J.H. Huibers, and Mark van der Gaag. 2020. The effect of dialectical behaviour therapy in autism spectrum patients with suicidality and/or self-destructive behaviour (DIASS): Study protocol for a multicentre randomised controlled trial. BMC Psychiatry 20: 127. https://doi.org/10.1186/s12888-020-025 31-1. Jørgensen, Mie S., Ole J. Storebø, Jutta M. Stoffers-Winterling, Erlend Faltinsen, Adnan Todorovac, and Erik Simonsen. 2021. Psychological therapies for adolescents with borderline personality disorder (BPD) or BPD features—A systematic review of randomized clinical trials with metaanalysis and Trial Sequential Analysis. PLoS One 16 (1): e0245331. https://doi.org/10.1371/jou rnal.pone.0245331. Kammeraat, Monique, and Pieternel Kölling. 2020. Psychiatric patients at the Euthanasia Expertise Center. Online: Onderzoeksrapportage-Psychiatrische-Patiënten-ExpertisecentrumEuthanasie.pdf (expertisecentrumeuthanasie.nl). Kim, Scott Y., Raumond G. De Vries, and John R. Peteet. 2016. Euthanasia and assisted suicide of patients with psychiatric disorders in the Netherlands 2011 to 2014. JAMA Psychiatry 73 (4): 362–368. Kissane, David W. 2012. The relief of existential suffering. Archives of Internal Medicine 172: 1501–1505. Landelijke Stuurgroep Multidisciplinaire Richtlijnontwikkeling in de GGZ (LSMRGGZ). 2008. Multidisciplinaire Richtlijn Persoonlijkheidsstoornissen. Richtlijn voor de diagnostiek en behandeling van volwassen patiënten met een persoonlijkheidsstoornis. Utrecht: Trimbos-instituut.

11 Due Care in the Context of Euthanasia Requests by Persons …

199

Levitt, Sarah, and Daniel Z. Buchman. 2020. Applying futility in psychiatry: A concept whose time has come. Journal of Medical Ethics. https://doi.org/10.1136/medethics-2020-106654. Lord, Catherine, Traolach S. Brugha, Tony Charman, Jack Cusack, Guillaume Dumas, Thomas Frazier, Emily J.H. Jones, Rebecca M. Jones, Andrew Pickles, Matthew W. State, Julie Lounds Taylor, and Jeremy Veenstra-VanderWeele. 2020. Autism spectrum disorder. Nature Reviews Disease Primers 6 (1): 5. https://doi.org/10.1038/s41572-019-0138-4. Lord, Catherine, Tony Charman, Alexandra Havdahl, Paul Carbone, Evdokia Anagnostou, Brian Boyd, Themba Carr, Petrus J. de Vries, Cheryl Dissanayake, Gauri Divan, Christine M. Freitag, Marina M. Gotelli, Connie Kasari, Martin Knapp, Peter Mundy, Alex Plank, Lawrence Scahill, Chiara Servili, Paul Shattuck, Emily Simonoff, Alison Tepper Singer, Vicky Slonims, Paul P. Wang, Maria Celica Ysrraelit, Rachel Jellett, Andrew Pickles, James Cusack, Patricia Howlin, Peter Szatmari, Alison Holbrook, Christina Toolan, and James B. McCauley. 2021. The Lancet Commission on the future of care and clinical research in autism. The Lancet. https://doi.org/10. 1016/S0140-6736(21)01541-5. May, Tamara, Pamela D. Pilkington, Rita Younan, and Katrina Williams. 2021. Overlap of autism spectrum disorder and borderline personality disorder: A systematic review and meta-analysis. Autism Research 2021. https://doi.org/10.1002/aur.2619. Medscape. 2019. Criminal case spurs a rethink of euthanasia for mental illness. Medscape, 19 Feb. https://www.medscape.com/viewarticle/909233. Mehlum, Lars, Schmahl Christian, Berens Ann, Doering Stephan, Hutsebaut Joost, Kaera Andres, Kramer Ueli, Paul A. Moran, Babette Renneberg, Joaquim S. Ribaudi, Sebastian Simonsen, Michaela Swales, Svenja Taubner, and Ester di Giacomo. 2020. Euthanasia and assisted suicide in patients with personality disorders: A review of current practice and challenges. Borderline Personality Disorder and Emotion Dysregulation 7: 15. https://doi.org/10.1186/s40479-020-001 31-9. eCollection 2020. Michaud, Laurent, Kyle T. Greenway, Sylvie Corbeil, Céline Bourquin, and Stéphane RichardDevantoy. 2021. Countertransference towards suicidal patients: A systematic review. Current Psychology. https://doi.org/10.1007/s12144-021-01424-0. Miller, David G., Rebecca Dresser, and Scott Y. Kim. 2019. Advance euthanasia directives: A controversial case and its ethical implications. Journal of Medical Ethics 45 (2): 84–89. Mungo, Anaïs, Matthieu Hein, Phillipe Hubain, Gwendolé Loas G., and Phillipe Fontaine. 2020. Impulsivity and its therapeutic management in borderline personality disorder: A systematic review. Psychiatric Quarterly 91 (4): 1333–1362. Mroz, Sarah, Sigrid Dierickx, Joachim Deliens, and Kenneth Chambaere. 2021. Assisted dying around the world: A status quaestionis. Annals of Palliative Medicine 10 (3): 3540–3553. https:// doi.org/10.21037/apm-20-637. National Collaborating Centre for Mental Health (UK) (NCCMH). 2009. Borderline personality disorder: Treatment and management. Leicester: British Psychological Society. National Institute for Health and Clinical Excellence (NICE). 2011. Autism: Recognition, referral and diagnosis of children and young people on the autism spectrum. NICE clinical guideline. Retrieved from https://www.nice.org.uk/guidance/cg128. National Institute for Health and Clinical Excellence (NICE). 2012. Autism. Recognition, referral, diagnosis and management of adults on the autism spectrum. The British Psychological Society and The Royal College of Psychiatrists. Nederlandse Vereniging voor Psychiatrie (NVvP). 2009. Richtlijn verzoek om hulp bij zelfdoding door patiënten met een psychiatrische stoornis. Utrecht: De Tijdstroom. Nederlandse Vereniging voor Psychiatrie (NVvP), Nederlands Instituut van Psychologen (NIP). 2013. Multidisciplinaire richtlijn diagnostiek en behandeling van autismespectrumstoornissen bij volwassenen. Utrecht: De Tijdstroom. Nicolini, Marie E., John R. Peteet, Kevin D. Donovan, and Scott Y. Kim. 2020. Euthanasia and assisted suicide of persons with psychiatric disorders: The challenge of personality disorders. Psychological Medicine 50 (4): 575–582.

200

M. De Hert et al.

Noens, Ilse, Catharina Struyven, May Van Gool, and Peter Vermeulen. 2018. Classificerend Diagnostisch Protocol Autismespectrumstoornis bij volwassenen (Classification diagnostic protocol autism spectrum disorders in adults). Brussels: Kwaliteitscentrum voor Diagnostiek vzw in samenwerking met Vlaams Agentschap voor Personen met een Handicap. Poon, Kenneth K., and Dalvin J. Sidhu. 2017. Adults with autism spectrum disorders: A review of outcomes, social attainment, and interventions. Current Opinion in Psychiatry 30 (2): 82–83. Raus, Kasper, and Sigrid Sterckx. 2015. Euthanasia for mental suffering. In New directions in the ethics of assisted suicide and euthanasia, ed. Michael Cholbi and Jukka Varelius, 79–96. Heidelberg, New York, London: Springer. Raus, Kasper, Bert Vanderhaegen, and Sigrid Sterckx. 2021. Euthanasia in Belgium: Shortcomings of the law and its application and of the monitoring of practice. The Journal of Medicine and Philosophy 46 (1): 80–107. Rosen, Nicole E., Catherine Lord, and Fred R. Volkmar. 2021. The diagnosis of autism: From Kanner to DSM-III to DSM-5 and beyond. Journal of Autism and Developmental Disorders 51 (12): 4253–4270. Ruissen, Andrea, Gerben Meynen, and Guy A.M. Widdershoven. 2011. Perspectives on patient competence in psychiatry: Cognitive functions, emotions and values. Tijdschrift voor Psychiatrie 53 (7): 411–412. Sheffer, Edith. 2018. Asperger’s children: The origins of autism in Nazi Vienna. New York: W.W. Norton. Storebø, Ole J., Jutta M. Stoffers-Winterling, Brigit A. Völlm, Mickey T. Kongerslev, Jessica T. Mattivi, Mie S. Jørgensen, Erlend Faltinsen, Adnan Todorovac, Christian P. Sales, Henriette E. Callesen, Klaus Lieb, and Erik Simonsen. 2020. Psychological therapies for people with borderline personality disorder. Cochrane Database of Systematic Reviews 5 (5): CD012955. Strang, Peter, Susan Strang, Ragnar Hultborn, and Staffan Arnér. 2004. Existential pain—An entity, a provocation, or a challenge? Journal of Pain and Symptom Management 27 (3): 241–250. The Guardian. 2020. Doctors acquitted in Belgium’s first criminal euthanasia case. The Guardian, 31 Jan. https://www.theguardian.com/society/2020/jan/31/doctors-acquitted-belgiumfirst-criminal-euthanasia-case. Thienpont, Lieve, Monica Verhofstadt, Tony Van Loon, Wim Distelmans, Kurt Audenaert, and Peter P. De Deyn. 2015. Euthanasia requests, procedures and outcomes for 100 Belgian patients suffering from psychiatric disorders: A retrospective, descriptive study. BMJ Open 5 (7): e007454. https://doi.org/10.1136/bmjopen-2014-007454. Thomas, Elizabeth C., Shelly Ben-David, Emily Treichler, Stephanie Roth, Lisa B. Dixon, Mark Salzer, and Yaara Zisman-Ilani. 2021. A systematic review of shared decision–making interventions for service users with serious mental illnesses: State of the science and future directions. Psychiatric Services 72 (11): 1288–1300. Tuffrey-Wijne, Irene, Leopold Curfs, Ilora Finlay, and Sheila Hollins. 2018. Euthanasia and assisted suicide for people with an intellectual disability and/or autism spectrum disorder: An examination of nine relevant euthanasia cases in the Netherlands (2012–2016). BMC Medical Ethics 19 (1): 17. https://doi.org/10.1186/s12910-018-0257-6. van Veen, Sisco M.P., Andrea M. Ruissen, and Guy A.M. Widdershoven. 2020. Irremediable psychiatric suffering in the context of physician assisted death: A scoping review of arguments in the literature. Canadian Journal of Psychiatry 65 (9): 593–603. van Veen, Sico M.P., Natalie Evans, Andrea M. Ruissen, Joris Vandenberghe, Aartjan T.F. Beekman, and Guy A.M. Widdershoven. 2021. Irremediable psychiatric suffering in the context of physician assisted death: A Delphi-study. medRxiv. https://doi.org/10.1101/2021.07.19.21260430. Varelius, Jukka. 2014. On the relevance of an argument as regards the role of existential suffering in the end-of-life context. Journal of Medical Ethics 40: 114–116. Verhofstadt, Monica, Kenneth Chambaere, Roeslan Leontjevas, and Gjalt-Jorn Y. Peters. 2019a. Towards an assessment instrument for suffering in patients with psychiatric conditions: Assessing cognitive validity. British Journal of Psychiatry Open 5 (3): e35. https://doi.org/10.1192/bjo.201 9.25.

11 Due Care in the Context of Euthanasia Requests by Persons …

201

Verhofstadt, Monica, Kristof Van Assche, Sigrid Sterckx, Kurt Audenaert, and Kenneth Chambaere. 2019b. Psychiatric patients requesting euthanasia: Guidelines for sound clinical and ethical decision making. International Journal of Law and Psychiatry 64: 150–161. Verhofstadt, Monika, Kurt Audenaert, Kris Van den Broeck, Luc Deliens, Freddy Mortier, Koen Titeca, Koen Pardon, Dirk De Bacquer, and Kenneth Chambaere. 2020. The engagement of psychiatrists in the assessment of euthanasia requests from psychiatric patients in Belgium: A survey study. BMC Psychiatry 20 (1): 400. https://doi.org/10.1186/s12888-020-02792-w. Vlaamse Vereniging voor Psychiatrie (VVP). 2017. Hoe omgaan met een euthanasieverzoek in psychiatrie binnen het huidig wettelijk kader? Adviestekst van de Vlaamse Vereniging voor Psychiatrie (VVP) over te hanteren zorgvuldigheidsvereisten, 7–39. http://vvponline.be/uploads/ docs/bib/euthanasie_finaal_vvp_1_dec.pdf. Waddell, Michael M. 2020. Autism and assisted suicide. Journal of Disability and Religion 24 (1): 1–28.

Marc De Hert is a clinical psychiatrist and psychotherapist working at the University Psychiatric Centre KU Leuven, campus Kortenberg department of psychotic disorders, in Belgium. He is a full professor at the KU Leuven, department of neuroscience, research group clinical psychiatry. He studied medicine at the University of Antwerpen, where he also qualified as a psychiatrist. He had a formal training in different psychotherapeutic methods. He has a Ph.D. in biomedical sciences from the KU Leuven. He is currently a Ph.D. student in Law at Antwerp University, where he is affiliated with the Antwerp Health Law and Ethics Chair. Current research projects involve epidemiology, long-term outcome studies, somatic co-morbidities in people with severe mental illness and the evaluation of metabolic and other side effects of antipsychotic medication. Sigrid Sterckx is Full Professor of Ethics and Political and Social Philosophy at the Department of Philosophy and Moral Sciences of Ghent University. She is a founding member of the Bioethics Institute Ghent. She lectures courses in theoretical and applied ethics as well as social and political philosophy. Her current research projects focus on medical end-of-life practices; Big Data, AI and healthcare decision-making; second and third trimester abortions in Flanders (Belgium); and the use of patents as a tool to regulate the technology of genome editing. Sigrid has published widely on these issues, including the co-edited volumes Continuous Sedation at the End of Life: Ethical, Clinical and Legal Perspectives (Cambridge University Press, 2013) and Personalised Medicine, Individual Choice and the Common Good (Cambridge University Press, 2018). She is a member of the Belgian Royal Academy of Medicine. Kristof Van Assche is a Research Professor in Health Law at the University of Antwerp, Belgium, where he is also a member of the board of the Antwerp Health Law and Ethics Chair, which encourages independent research in the field of health law and ethics. His research focuses on rights and duties in the context of healthcare, and on the legal and ethical aspects of many important issues in biomedicine, including euthanasia, abortion, organ trafficking, genome editing, and assisted procreation. In addition, he is a Professor of Philosophy of Law and teaches European and International Health Law.

Chapter 12

Medical Futility and Physician Assisted Death Nancy S. Jecker

Abstract This chapter addresses the close association between withholding and withdrawing futile life-sustaining medical treatments and assisting patients with hastening ending their lives. Section 12.2 sets forth a definition of medical futility and places this concept in the broader context of bioethical principles of autonomy, beneficence, nonmaleficence and justice. Section 12.3 draws out futility’s ethical implications and considers the view that physicians are ethically permitted to refrain from medically futile treatments, should be encouraged to refrain, or have a duty to refrain. Section 12.4 examines ethical arguments that physicians also have a positive duty to assist patients with terminal conditions who are imminently dying to end their lives. It explores ethical arguments for assisted dying that appeal to the physicians’ significant responsibility and relationship with patients, the claimed moral equivalence of actions and omissions, and the medical goal of helping patients have a peaceful and dignified death. Keywords Assisted death · Medical futility · Dignity · Euthanasia · Physician-patient relationships

12.1 Introduction This chapter addresses assisted dying in the context of patients with terminal conditions who are imminently dying. It considers in particular, the setting of economically developed Western societies where a wide range of medical technologies are available to extend patients’ lives and asks, what special responsibilities do physicians have in such settings. Section 12.2 describes a concept of medical futility that I have defended (with co-authors Schneiderman and Jonsen) at length elsewhere. Section 12.3 considers the ethical responsibilities of physicians when an intervention qualifies as medically futile. Section 12.4 explores the ethics of physician assisted N. S. Jecker (B) Department of Bioethics and Humanities, University of Washington School of Medicine, Seattle, WA, USA e-mail: [email protected] © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 M. Cholbi and J. Varelius (eds.), New Directions in the Ethics of Assisted Suicide and Euthanasia, The International Library of Bioethics 103, https://doi.org/10.1007/978-3-031-25315-7_12

203

204

N. S. Jecker

dying for patients with terminal conditions who are imminently dying after futile interventions are withheld or withdrawn, after palliative and comfort care options have been exhausted, and when the patient prefers to hasten death. For the purposes of this chapter, I do not draw a bright line between self-administering a lethal dose of medication and having a physician administer this. Instead, my purpose is to make a more general case in support of physician assisted death that is relevant to both forms of assisted dying. The point is not to present a final argument, but a prima facie defense of the practice. I bracket the question of whether the arguments given extend beyond terminally ill and imminently dying patients and apply to other groups, such as individuals with chronic conditions, preclinical diagnoses, or older adults who are healthy, but consider their lives ‘complete.’

12.2 Deciding to Forego Futile Medical Interventions Decisions to forego medical treatment find ethical support in bioethical principles of autonomy, beneficence, nonmaleficence and justice (Beauchamp and Childress 2019). The principle of autonomy requires respecting a patient’s autonomous choices. Autonomous choices may be expressed in a variety of ways. First, at the time treatment is offered, a patient with decisional capacity may make an informed decision to refuse it. Second, at some prior time a now incapacitated patient may have made an informed decision that certain interventions should not be provided. Often, such a preference is expressed through a directive to physicians. Finally, a decision may be made by a surrogate who represents the prior preferences of a now incompetent patient to withhold or withdraw a particular intervention. A second and third ethical basis for foregoing medical treatment are the ethical requirements to promote the patient’s good (the principle of beneficence) and avoid harming the patient (the principle of nonmaleficence). There are at least two ways in which principles of beneficence and/or nonmaleficence may be invoked to justify foregoing treatment. First, a surrogate decision maker may decide to forego treatment because it is considered to be in the patient’s best interest. Second, benefitting the patient is at stake when treatment is foregone because it confers no significant benefit to the patent. A fourth basis for choosing to forego medical treatment appeals to a principle of justice. In health care, justice is typically understood in terms of the distribution of scarce resources under conditions where beneficial treatments are denied because the raw materials required to make them available are scarce or the dollars to pay for resources are limited. In both instances, there is not enough of a particular resource to provide it to everyone who stands to benefit, and the principle of justice requires that morally similar cases be treated similarly. The concept of futility is implicit within each of the four principles, but especially the principles of beneficence and nonmaleficence. To show this, it will be helpful to start with a general definition of futility and then draw out its implications for bioethics principles. The concept of futility comes from the Latin futilis. A futilis was a religious vessel that had a wide top and a narrow bottom. Its shape caused it

12 Medical Futility and Physician Assisted Death

205

to tip over easily when filled, rendering it useless for anything other than ceremonial purposes. The origin of the word reminds us of the symbolic power of rituals. In medicine, procedures that provide no significant benefit to patients are sometimes sought because of the symbolic importance of “doing everything possible,” which might be interpreted as an expression of caring for the patient (Schneiderman et al. 1994). Veatch and Spicer describe a patient who pleads, ‘Don’t let them give up on me,’ suggesting that irrespective of whether interventions serve any practical purpose in realizing the goals of treatment, they nonetheless carry meaning for patients, signifying that a patient is not being abandoned (Veatch and Spicer 1992). According to the Oxford English Dictionary, a futile action is “leaky, hence untrustworthy, vain, failing of the desired end through intrinsic defect,” which aptly conveys its objective uselessness, but not the purpose such an action might serve in religious, medical, or other contexts. In some instances, the positive symbolic meaning a futile intervention carries for a patient may overshadow the clinician’s judgment that it is futile; in such instances, patients may dig in their heels when medical futility is communicated and continue to equate the use of futile interventions with being cared for. In other instances, patient may question the physician’s judgment that an interventions offers no significant benefit. It is controversial whether futile interventions are justified in a narrow range of cases where they yield psychological benefits for dying patients. One way of framing this question is to liken futile interventions to placebos used in research settings, which may have an ameliorative effect on a symptom or disease but are ineffective for the condition being treated. Shapiro and Shapiro argue that for most of the history of medicine, the history of medical treatment was “essentially the history of the placebo effect,” because pre-scientific medicine had little else to offer (Shapiro and Shapiro 2000, 2). Lichtenberg et al. (2004) argue that placebo-type interventions continue to have a legitimate place in clinical practice in cases where no standard proven therapy exists and the placebo is reasonable expected to produce positive psychological benefits for the patient. Others maintain placebos can produce a “meaning response” in patients (Moerman and Jonas 2002). However, in many instances, using futile interventions only prolongs patient’s suffering and feeds false hope. Futile interventions often express disregard for professional standards, being seduced by technology, neglecting to focus on palliative care, being co-opted by a patient’s family, refusing to acknowledge medicine’s limits, or denying a patient’s impending death (Jecker 2013). Over time, futile intervention can create moral distress and burnout. There are other, better ways to show care and support for patients. In medicine, the concept of medical futility refers specifically to interventions that hold no reasonable prospect of helping the patient. Some use the term ‘futility’ more narrowly, to refer to an intervention that cannot bring about its physiological objective (Ardagh 2000). For example, the use of cardiopulmonary resuscitation (CPR) in the field hours after a patient stops breathing might be called, ‘physiologically futile’ in this sense. Part of what lies behind a physiologic definition of futility is the desire to practice medicine in a manner that is ‘value free.’ As White and Pope (2016, 72), argue,

206

N. S. Jecker

with physiological futility, the clinician does not make any evaluative assessment that the treatment’s effect is too unlikely, too small, or not worthwhile. There is no normative disagreement. Instead, clinicians can readily ascertain physiological futility based solely on their clinical knowledge. The basis for refusing treatment is a scientific and empirical one: the treatment simply will not work.

A problem with defining futility this narrowly, however, is that an intervention may produce effects on a patient’s body but still qualify as futile because it does not help the patient. Reasoning along these lines, Schneiderman et al. (1990) distinguish between effects and benefits, noting that during the early nineteenth century, many medications physicians used were effective—emetics could cause vomiting, purgatives could cause laxation, sudorific, sweating, and so on. In this sense, these interventions were not physiologically futile. Nonetheless, we regard them as medically futile because they did not benefit patients. Another way to say this would be that medicine’s goal is broadly speaking, to help the sick. Miller and Brody (2001) elucidate a plurality of more specific goals of medicine, including preventing disease and injury, relieving the pain and suffering caused by disease, caring for and curing those who are ill, preventing premature death, and helping a patient die peacefully. When medical interventions do not achieve these goals, which relate to helping patients and preventing disease and injury, they qualify as medically futile. The futility of an intervention can be evident in medicine in two distinct ways (Schneiderman et al. 1990). Quantitative futility refers to interventions where the likelihood of benefit falls well below a threshold considered minimal. The concept has ancient roots. The Hippocratic text, The Art, instructs physicians, “Whenever therefore a man suffers from an ill which is too strong for the means at the disposal of medicine, he surely must not even expect that it can be overcome by medicine” (Reiser et al. 1977, 6–7). The idea that when a patient is overmastered by their disease, medicine is powerless is expressed in modern parlance by the notion of quantitative futility. Quantitative futility indicates a situation where the probability that a medical intervention will achieve a desired outcome is extremely low. An example of quantitative futility might be offering cardiopulmonary resuscitation (CPR) to patients when the chances of survival after CPR are no better than 1 in 100. The quantitative dimension of futility draws on a standard of reasonableness that is the same as the standard used in the statistical evaluation of clinical trials (Schneiderman et al. 1996). In a clinical trial, an intervention is considered to have no significant benefit (and to be better than no intervention) when it produces its desired effect no more than 1 in 100 instances. The quantitative dimension of futility draws on a standard of reasonableness that is the same as the standard used in the statistical evaluation of clinical trials.1 Like quantitative futility, qualitative futility sets standards for medical interventions and appeal to the goal of helping the patient, rather than producing effects on

1

In the clinical trial, the question is not, does an intervention ever succeed, but rather, how often must a treatment fail before it is considered to provide no significant benefit. The answer is that it must succeed more than 1 in 100 times. See Schneiderman et al. (1996).

12 Medical Futility and Physician Assisted Death

207

a patient’s body. A qualitative conception of futility traces to ancient Greek philosophers, such as Plato, who deplored the kind of medicine that “pampers the disease” and praises the Asclepian physicians, who he (1981, 76–77) says, taught medicine for those who were suffering from a specific disease; he rid them of it…then ordered them to live as usual…For those, however, whose bodies were always in a state of inner sickness he did not attempt to prescribe a regimen…to make their life a prolonged misery…Medicine was not intended for them, and they should not be treated even if they were richer than Midas.

In the language of contemporary bioethics, an intervention is qualitatively futile whenever the quality of benefit associated with the intervention falls well below a threshold considered minimum. Whereas quantitative futility draws an analogy with the evaluation of statistical trials and appeals to the notion of reasonableness, qualitative futility draws on paradigm cases to illustrate. One paradigmatic example of qualitative futility is continuing to provide mechanical ventilation or nutrition and hydration for a patient in a permanent vegetative state. In this instance, a patient will never appreciate any benefit from the intervention. Another paradigm example of qualitative futility is an intervention associated with an overwhelming amount of suffering for a predictably brief period of time. A final paradigm is a patient who will never leave the intensive care unit and requires constant monitoring, ventilatory support, and intensive nursing care to the extent that they cannot participate in any other activities or goals. Various combinations of quantitatively and qualitatively futile interventions are possible. In some instances, an intervention is both quantitatively and qualitatively futile; for example, performing CPR in the field on a patient who is cold and blue. In other instances, an intervention achieves its physiologic goal, but is qualitatively futile. For example, administering artificial nutrition and hydration on a patient in a persistent vegetative state is a case of ‘hitting the target but missing the point.’ Quantitative and qualitative futility judgments reflect the values of beneficence and nonmaleficence (discussed in Sect. 12.3). When applied to a particular intervention for a particular patient, they should reflect a consensus of clinical judgment and conform to evidence-standards for best practice. They should not be based on individual standalone judgments rendered at the bedside. When substantial controversy exists about whether an intervention for a particular patient is futile, this should be transparent and the patient and family should be offered options, including being transferred to a different facility to receive interventions that a particular provider or institution is unwilling to offer or continue (Jecker 2015). Fair procedures can be particularly critical during such moments, because they serve to protect patients who are ill or injured and their families in situations where they have relatively less knowledge, power, and authority than the physician and healthcare institution. They also protect health professionals against the moral distress associated with providing a treatment that they object to on grounds of deeply held personal or professional beliefs (Wicclair 2016). Quantitatively and qualitatively futile interventions encompass a range of evaluations and interventions along a continuum of healthcare services. They include,

208

N. S. Jecker

for example, doing electroencephalography (EEG) for an uncomplicated headache, ordering a computer tomography (CT) or magnetic resonance imaging (MRI) scan for low-back pain in patients without any signs of a neurological problem, or placing a coronary-artery stent in patients with stable cardiac disease. Other common examples include the use of antibiotics for probable viral respiratory tract infections; and preoperative testing in low-risk patients, such as chest radiography (Hurley 2014). Nor is the use of futile interventions confined to high technology procedures in wealthy parts of the world. Instead, overuse is a problem around the world. According to Brownlee et al., “overuse,” defined as “the provision of medical services that are more likely to cause harm than good,” occurs in low- and middle-income countries, often in tandem with unmet healthcare needs (Brownlee et al. 2017). Brownlee cites accumulating evidence of widespread overuse of medical interventions in Australia, Spain, Israel, Brazil and Iran.

12.3 Ethical Implications of Medical Futility If a treatment is clearly futile in either the quantitative or qualitative sense, what should a responsible healthcare professional do? We focus here on the responsibilities of physicians because they are most typically in the role of issuing medical orders to initiate or remove treatments that might be deemed futile. In response to this question, there are at least three distinct positions one might hold (Schneiderman and Jecker 2011). First, it might be held that physicians are ethically allowed to withhold or withdraw a medically futile intervention, but not ethically required to do so. In other words physicians are ethically free either to provide or to refrain from providing futile treatments. This relatively weak ethical stance was apparently held by physicians in ancient Greece and Rome. Until the twelfth century, they did not belong to any professional body or practice in accordance with professional standards: “Bound by no duty to a licensing authority or professional organizations, the physician exercised his art at his own pleasure…completely free to treat or to refuse to treat” (Amundsen 1978, 24). Today, a difficulty this stance encounters is that medicine has become a profession and physicians are held to professional standards and expected to provide professionally competent care. This implies that certain interventions should and should not be used. A relatively stronger view holds that physicians should be encouraged to withhold or withdraw medically futile treatment. While physicians may present patient with options that are medically futile, they should recommend beneficial interventions and should be discouraged from offering futile ones. A still stronger stance holds that physicians should not use medically futile treatments and have a duty to withhold or withdraw them. According to the stronger view, offering or using interventions that do not benefit patients goes against widely recognized normative ends of the medical profession and should be avoided. This way of thinking presupposes professional

12 Medical Futility and Physician Assisted Death

209

standards that medical intervention must meet. Rather than empowering unilateral decision-making by physicians, this stance circumscribes the range of treatment options available to physicians to those that are professionally competent; it forbids physicians from offering or using interventions that fall outside this range. Since these are general stances, they admit of compassionate exceptions. For example, a duty to refrain from using futile interventions might take the form of a time delimited contract with a family aimed at achieving the goal of withdrawing a futile intervention after a series of steps that prepare the patient and/or family, reassess the patient, consult with outside experts, and so forth. Some clinicians may prefer not to use the term ‘futility’ explicitly in conversations with some patients or families, because doing so might be misunderstood or breed distrust. Other proposed language that conveys the futility of a particular intervention would be to say that the intervention in question is inappropriate, not medically indicated, nonbeneficial, or will only prolong the dying process. The account of quantitative and qualitative medical futility combined with the analysis of its ethical implications lends insight into the ethical principles discussed at the start of this section. First, respect for an autonomous person’s choices translates in the medical setting into a right of autonomous patients to choose from among a range of medically appropriate interventions, or to reject all interventions. Autonomy does not authorize ‘treatment on demand’, nor does it entitle patients to choose treatment options that are at odds with professional standards of competent medical care. Second, benefitting can be understood in the medical setting as requiring health professionals to act only in ways that help patients (or promote the health of society in the case of public health measures, like quarantine). It does not require them to use interventions the profession considers not to be of benefit for patients. Third, avoiding harming implies that physicians should forgo using interventions that are burdensome or impose harms on patients unless such burdens and harms can be justified by being offset by benefits. Finally, distributing benefits and burdens fairly among patients requires that health professionals allocate the resources they command to patients who stand to benefit in ways that help patients and avoid wasting resources by using them for patients who do not stand to benefit. Table 12.l summarizes the arguments of Sects. 12.2 and 12.3. Elsewhere, I discuss and reply to objections to these arguments at length (with co-authors, Schneiderman and Jonsen), and will not rehearse our arguments here (Jecker 2017; Schneiderman et al. 1996, 2017; Schneiderman and Jecker 2011).

12.4 Physician Assisted Dying In this section, the focus shifts to the aftermath of withholding and withdrawing medically futile treatment, focusing in particular, on life sustaining treatment (LST). It asks, if withdrawing or withholding medically futile LST is ethically permissible, does it follow that acting to end life is as well? If so, should a physician assist a patient who asks for help hastening their death? This concern is as old as medicine.

210

N. S. Jecker

Table 12.1 The 4 principles of bioethics and medical futility Principles

Definitions

Autonomy

Respect autonomous persons’ choices Autonomous patients choose from among a range of non-futile interventions, or reject all interventions

Implications

Beneficence

Benefit persons

Physicians act in ways that help patients, and have no duty to use interventions that do not help

Nonmaleficence Avoid harming persons

Physicians avoid harming patients, and do not use interventions that do not help patients because they carry risk and can harm patients

Justice

Physicians distribute scarce resources among patients who stand to benefit, which implies not wasting resources by offering them to people who do not stand to benefit

Distribute burdens and benefits fairly

The writings of Plutarch express a reportedly popular sentiment of the fifth century BCE, that “the best physician was the man who did not cause his patients to linger on but buried them quickly” (Amundsen 1978). Yet the reference here is foregoing treating patients “overmastered” by their disease and reportedly did not extend to hastening death. The Hippocratic Oath (1923) in fact explicitly forbids assisting dying: “I will do no harm or injustice…Neither will I administer a poison to anybody when asked to do so, nor will I suggest such a course.” For the ancients, the question of whether or not to hasten death may have seemed less urgent than it does today. While professional honesty demanded the physician to withdraw, the physician’s departure from a case was not tantamount to abandoning the patient; according to Gourevitch, “Far from feeling any liability for abandoning his patient, he would feel guilty if he undertook a cure he could not successfully carry out;” he explains that for the ancients, “the hope of a miracle never being extinguished, there is strictly speaking no question of abandonment or condemnation” (Gourevitch 1969, 503). From this standpoint, when the physician withdrew, control over death’s timing was left to the gods and fate; the physician’s work was done. Today the practice of medicine has of course changed dramatically. One way this is true is that in many economically developed countries, people no longer die at home. While the hospice movement has increased opportunities for dying in the location of one’s choosing, (Cross and Warraich 2019), the percent of people who die in hospitals remains high in many places (De Roo et al. 2014). In Europe, for example, the percentage of home deaths in 2014 ranged from 35.5 in Belgium to 50.6% in the Netherlands. When patients die in hospitals, managing the dying process presents a medical challenge. Today, there is widespread agreement among bioethicists that it is ethically permissible to cease LST when patients request it and a growing movement to discourage the use of medically futile interventions. While physicians and hospitals

12 Medical Futility and Physician Assisted Death

211

limit patients’ choices to a range of medically suitable options, defenders of physician assisted dying want to say that physician assisted death should be among the range of medically allowed options that patients may request after LST are withdrawn. The practice of physician assisted death may take the form of a physician prescribing a lethal dose of medication that the patient self-administers (sometimes referred to as ‘physician-assisted suicide’, although this terminology has fallen out of favor), or it might involve the physician administering a lethal dose of medication, usually for a patient who is unable to self-administer it (sometimes referred to as ‘active euthanasia’). As of January 2021, physician assisted dying of some form is legally allowed in six countries: Belgium, Canada, Luxembourg, the Netherlands, Spain, and Switzerland (Rada 2021), and in some U.S. jurisdictions (California, Colorado, District of Columbia, Hawaii, Maine, New Jersey, New Mexico, Oregon, Vermont, and Washington) (Oregon Health Authority 2021a, b). Regardless of the view one holds about it, there is agreement that physicians have a duty not to abandon patients and to aggressively pursue palliative and comfort care (Schneiderman et al. 1994). For this reason, providers should avoid saying, ‘there is nothing we can do for you,’ which may signify abandonment. They should also avoid referring to care as ‘futile,’ a situation as ‘futile,’ or a patient as ‘futile;’ instead, healthcare professionals should apply ‘futility’ to a particular intervention at a particular time for a particular patient. Ethical arguments in favor of physician assisted dying that link most directly with medical futility often appeal to the physician’s role prior to a request for assisted death. First, physicians who participated actively in prescribing treatments to extend a patient’s life bear some responsibility for bringing about the patient’s present predicament. Second, physicians who stand in a close patient–physician relationship may have a role-related responsibility to help. Third, if a physician is already omitting LST and if acting and omitting are morally equivalent, then actions to end life seem ethically defensible too. Finally, respect for patient dignity and the value of ensuring a dignified death as part of a dignified life lends support to physician assisted death. A counter argument to all of these claims is that patients requesting assisted death cannot be making a rational decision. They stand in need of help of a very different kind, for example, help treating clinical depression, or aid finding meaningful ways of living out their remaining life. Throughout the analysis I bracket these debates, assuming that the request for assisted dying can be rationally justified. This assumption is controversial. Mental health practitioners standardly hold that suicide is often the product of mental illness and contrary to the interests of the individual. However, it is increasingly recognized that the medical case of patients suffering from terminal illness who are imminently dying is ethically different and that for these cases, this standard view is out of place (Nelson and Ramirez 2017). Even if some terminally ill and imminently dying patients suffer from untreated clinical depression which determines or shapes their choice to die, it is not necessarily the case that a request to die is always arrived at in this way. Notwithstanding the significant concern to rule out and treat depression, this section assumes that for some patients, the request to die is informed and justifiable.

212

N. S. Jecker

12.4.1 The Role of Physicians In many respects, medicine has made it harder to die: when a person’s heart stops beating, they can be resuscitated, when someone’s lungs stop respiring, they can be ventilated, and when an individual’s kidneys cease functioning, they can be dialyzed. While these interventions can extend life in positive ways, they also can make a patient miserable. One reason for holding physicians responsible for assisted dying then is that in many instances, physicians bear significant responsibility for a patient’s predicament. For example, when oncologists extend the lives of cancer patients with chemotherapy, radiation, surgery, cardiopulmonary resuscitation, artificial ventilation, and other means, they contribute to that patient’s subsequent condition in the sense that absent these lifesaving treatments, the patient would probably have died sooner and been spared much misery. Although physicians do not cause a patient’s cancer, they repeatedly furnish medical treatments that extend the lives of cancer patients. It is in this respect that they help bring it about that a patient is in a situation where their quality of life has plummeted, and they endure prolonged suffering. In short, medicine can unwittingly make life worse for terminally ill patients. A patient with anal cancer describes the effects of medical treatment on the quality of her life this way: In order for people to maintain life in terminal illnesses, they very often have to be on wretched volumes of drugs that make them sick, that make them queasy, that make them extraordinarily sad, that sink them into a depression regardless. So then you’re on this whole superhighway of trying to balance the depressants, the drugs that they have to have to keep them alive is giving them, needs to be counteracted with drugs to try and lift their spirits like with [antidepressant]. (Karsoho et al. 2016, 192)

When medicine makes life worse for terminally ill patients in these ways, it is beholden upon physicians to ensure first, that futile interventions are not offered and are withdrawn if they are already underway; second, that palliative and comfort care are provided; third, in instances where pain and suffering cannot be adequately controlled, and more is needed to ensure a good death, physicians should offer other options, including hastening death for patients who request it. Even when pain is adequately controlled, hastening death might be preferrable for some patients. For example, Angell describes the situation of her husband, Arnold Relman, a physician and editor-in-chief of the New England Journal of Medicine, who suffered from end stage cancer and was concerned with having his mind and thinking affected by the drugs that would be needed to prevent his suffering; Angell argued that for her husband, a good death required other means, such as a lethal prescription of drugs to end his life (Angell 2014). For others, existential suffering, which includes feelings of hopelessness, futility, and meaninglessness; anxiety about death; and disruption of personal identity, detract from a good death. To address it, therapies that promote continued life as well as those that allow patients the ability to choose the timing and manner of their death, are needed (Gopal 2015).

12 Medical Futility and Physician Assisted Death

213

One way to formulate an argument on the basis of these reflections is to say that if assisted death is permissible, then the task of assisting rests with physicians because of their role in extending patients’ lives. The Significant Responsibility Argument 1. Physicians often play a significant role in extending the lives of patients to the point where patients’ quality-of-life plummets and they no longer desire to continue living. 2. For some patients, after life sustaining treatments are stopped, the available options (palliative and hospice care) are inadequate to ensure a good death. 3. In these cases, physicians bear responsibility to help patients who ask for assistance dying, because they acted in ways that helped bring about the patient’s present predicament. A variation of this argument points not to the causal contribution that physicians made to the patient’s condition, but to the significant relationship that physicians develop with patients during the course of care. The Significant Relationship Argument 1. Physicians are often by a patient’s side throughout the course of their illness, which, however long or short, often includes intense emotions, such as grief, fear, shock, pain, and suffering; hard choices about continuing or stopping treatment; and powerful memories of what was said and done. 2. This history creates a significant bond between doctors and patients, typically involving the patient knowing, trusting, relying on, and expecting their relationship with a particular physician (or with physicians as a group) to continue. 3. If it is ethically permissible to assist a patient’s death, physicians should be the ones to assist, given their history of being known, trusted, relied on, and expected to be there. 4. Therefore, the burden of proof lies with those who oppose physicians standing in this role. The Responsibility of Physicians argument and the Significant Relationship argument make several key assumptions which limit their scope of application. First, the arguments assume that patients have an ongoing illness, rather than an acute event or injury, leading to their current states and that a particular physician (or physicians) has been involved in their care. Second, the arguments assume that these relationships are positive, and the patient wants them to continue. Since these assumptions do not obtain in every case, the arguments do not apply to every case. However, they do cover a substantial subset of cases. For example, the arguments would apply to patients with chronic conditions such as cancer, chronic obstructive pulmonary disease, or cardiovascular disease who have positive relationship with the oncologists, pulmonologists, and cardiologists who care for them. The arguments would also apply to patients who suffer severe injuries, such as spinal cord injury, and have ongoing positive relationships with rehabilitation medicine physicians. Yet, even

214

N. S. Jecker

though the arguments are limited in scope, they represent one of several important paradigm situations. Against the Significant Responsibility and Significant Relationships arguments it could be said that ending life has not been part of the physician’s historic role. As noted already, it is explicitly excluded from the Hippocratic oath. According to the latest version of the American Medical Association’s code of medical ethics, helping a patient to die is “fundamentally incompatible with the physician’s role as healer” (American Medical Association 2001). In reply, as Miller and Brody note, the ends of medicine are not static and fixed, but evolving. Medicine’s ends are neither “Platonic forms, historically unchanging,” nor “invented more or less at whim;” instead, the goals of medicine “evolve along with human history and culture. At least some changes in the goals of medicine, to make the practice of medicine more congruent with contemporary social conditions, represent positive evolutionary changes” (Miller and Brody 2001, 585). The analysis given here makes evident that the responsibilities of contemporary physicians extend beyond the responsibilities that physicians have historically had. Prior to the nineteenth century, few means were available to extend life and the question of what physicians owed to patients they had repeatedly kept alive did not arise with any regularity. Moreover, while physicians from antiquity left their patients in the hands of deities, today’s providers lack these assurances. To take leave from a patient today is tantamount to abandoning them, which seems not only cruel, but unprofessional. Yet, another objection to the Significant Responsibility and Significant Relationship arguments holds that although physicians have a duty to help patients, they do not have a duty to help in a particular way. Therefore, the arguments establish only a prima facie duty to help while leaving open the possibility that a physician could help in another way, such as by means of terminal sedation. One way of responding to this objection is to say that there are many reasons why a patient (or their surrogate) may decide that a terminally sedated human life is not right for them and that the only ethically viable option is to hasten death. For example, for some patients, a terminally sedated life seems to lack dignity and to be worse than death. An objection to the Significant Responsibility argument is that the patient autonomously chose to undergo prior treatments, including LST; hence, the patient, not the physician, is responsible for their current predicament. The rejoinder to this objection is that medical decision making at the end of life is typically shared. The physician who stood in a fiduciary role with respect to the patient, directed the course of care, and prescribed a path. In this capacity, the physician is often a significant influencer and to the extent that they are, they cannot ethically extricate themselves when ‘the going gets rough.’ A final worry with the Significant Role and Significant Relationship arguments is that the process of dying is already overly medicalized and assigning physicians responsibility for patients’ deaths only exacerbates the problem. Rather than expanding medicine’s role, we ought instead to diminish it, to allow death to occur in a more ‘natural’ and less medicalized way. Just as the continued use of medically futile

12 Medical Futility and Physician Assisted Death

215

interventions extends medicine’s role beyond reasonable limits, physician assisted death gives too much power to physicians and medical institutions. According to Karsoho et al. (2016, 192), medicine is “increasingly colonizing ever more and more parts of the end of life, to the detriment of dying persons.” According to Illych (1995, 210), medicalization has unfortunately “brought the epoch of natural death to an end.” In reply to this last objection, it can be noted first, that the notion of ‘natural death’ has been overly romanticized; in contemporary society, technology is commonplace, and its use today seems ordinary and in many ways ‘natural’ (Seymour 1999). Second, even if some patients prefer a so-called ‘natural’ death, some do not. As noted, one reason for hastening death is that the alternative of dying ‘naturally’ may be more suffering, and this suffering may be the sort that cannot be alleviated by medications in a way that preserves dignity and maintains life’s meaning. Undeniably, there are losses that life’s ending and the dying process bring that palliative care is not designed to alleviate: loss of independence, loss of the ability to remember and think clearly, loss of the ability to communicate and affiliate in ways one did before. If these arguments are compelling, then physicians are ethically justified in offering assisted dying as a last resort option for terminally ill and imminently dying patients. Yet, a further question is what are the scope and limits of this duty in jurisdictions where physician assisted death is legally prohibited? In response, even if physicians are not required to violate the law to serve their patients’ interests, they should contribute to efforts to reform laws that prevent them from helping their patients die.

12.4.2 The No Difference View A second set of arguments that lend support to physician-assisted death holds that actions to end life are not significantly different from omitting sustaining life. Angell (2014) puts the point bluntly: “If we grant patients the choice to hasten their deaths by refusing dialysis, mechanical ventilation, antibiotics, or any other life-sustaining treatment… why deny them what is essentially the same choice” by denying them the option to hasten their death by obtaining a lethal dose of medication? The claim here is that even if there is a conceptual difference between omitting extending life and acting to end life, there is not a moral difference. One reason there may not be a moral difference is that the outcome of both is equivalent. No Difference in Consequences 1. Acting and omitting are morally equivalent when there is no significant difference in their consequences. 2. The consequences of omitting LST from a terminally ill and imminently dying patient are similar to the consequences of performing an action to bring about a patient’s death. In both instances, the consequence is the patient’s death.

216

N. S. Jecker

3. Therefore, if omitting LST from a terminally ill and imminently dying patient is morally permissible, them so is physician assisted dying. However, it might be thought that even if the outcome is the same, the reasons for action and specifically, the intention of the agent, matter. When a physician prescribes a lethal dose of medication, their intention is to deliberately hasten the patient’s death, but this differs from the intentions associated with withholding or withdrawing LST, which are, e.g., letting a terminal disease run its course or not using extraordinary means to continue a patient’s life. Although death is a foreseen and predictable outcome, it is not the physician’s aim. To address this objection we might modify the No Difference in Consequences argument to specify that moral equivalence applies only if the physicians’ intentions are similar in both the case of withholding or withdrawing LST and the case of assisted death. For example, it would pertain to cases where the intention of a physician who prescribes a lethal dose of medication is to end the patient’s suffering, respect the patient’s autonomy, or preserve the patient’s dignity. The argument can be recast as following. No Difference in Consequences and Intentions 1. Acting and omitting are morally equivalent when there is no significant difference in the consequences and intentions associated with them. 2. It follows that when the consequences and intentions of physicians omitting LST and physicians assisting with death are the same, they are morally equivalent. 3. Therefore, if omitting LST is ethically permissible and if moral equivalence applies, then physician assisted dying is morally permissible. Taken together, these two arguments establish that omitting futile life sustaining interventions is not morally different from assisted dying merely because the former is an omission, and the latter is an action. An objection to both arguments holds that even if it is not wrong to act to end a patient’s life, it undermines physicians’ personal or professional integrity. Integrity indicates “the condition of having no part or element taken away or wanting; undivided or unbroken state” and “the condition of not being marred or violated” (Oxford English Dictionary Online). To act with integrity indicates acing in ways that are consonant with one’s deeply held moral convictions. In the case of personal integrity, physicians violate their integrity if they assist with dying in opposition to their deeply held personal values, while in the case of professional integrity, violations occur when a person acting in the role of a physician acts contrary to the values and ends of medicine. Professional integrity in medicine “represents what it means normatively to be a physician; it encompasses the values, norms, and virtues that are distinctive and characteristic of physicians” (Miller and Brody 1995, 10). In reply to the integrity objection, it can be said first, that physicians who personally oppose assisted dying are free to abstain from providing it. Second, in response to the objection that assisted dying is contrary to the professional values inherent in medical practice, this view can be persuasively challenged. Amundsen (1978) notes that preserving biological life per se has never been part of the historical practice

12 Medical Futility and Physician Assisted Death

217

of medicine. Miller and Brody (1995, 12) argue that today, the goals of medicine include “healing, promoting health and helping patients achieve a peaceful and dignified death.” They go on to delineate the goal of helping patients achieve a peaceful and dignified death and hold that it lends support to assisted dying as a last resort medical therapy to achieve a peaceful and dignified death. To fully flesh out this argument requires unpacking the concept of “dignity” to which the third goal appeals. I have done this at length elsewhere (Jecker 2020), and provide a preview of this argument below. Another objection to the above arguments, which falls outside the scope of this chapter, is that although the moral equivalence thesis is widely accepted within Western-oriented bioethics, it is not universally accepted (Kadooka et al. 2014; Nakazawa et al. 2019). This may limit the force of the argument outside Western settings.

12.4.3 Dignified Lives Just as the continued use of futile medical interventions can result in a loss of patient dignity, the symptoms associated with end-stage terminal disease can bring about profound loss of dignity that persists after futile interventions are stopped. In fact, dignity is among the most frequently reported end-of-life concerns for patients with terminal illness who request assisted dying. In Oregon (USA), annual reports assessing the use of services permitted under the state’s Death with Dignity Act consistently find that patients rank loss of dignity among their chief concerns; in 2020, 72% of patients who requested a lethal dose of medication cited dignity as a central reason for their request (Oregon Health Authority 2021a, b). Dignity can be understood in terms of the central things human beings can do and be, such as the capacity to have an ongoing narrative or story of one’s life; be healthy and well nourished; have bodily integrity; affiliate with others; express a range of human emotions; exercise senses, imagination, and thought; deliberate about goals and plans for one’s life; play and recreate; and regulate one’s immediate environment. Serious illness can cause a person to lose these capabilities altogether or cause them to be markedly diminished. When central human capabilities fall below a threshold the patient considers minimal, they may express this by saying ‘I don’t have a life’, which may indicate that in some important respect the narrative or story of their life is over. When the symptoms associated with end stage disease have this outcome, patients perceive a loss of dignity. Evidence shows that concerns about diminished capabilities are central for dying patients. Chochinov et al. asked patients with terminal illness about their specific dignity concerns and found that they related precisely to a diminution of capabilities such as mental/thinking capacity; the ability to move independently from place to place and perform activities of daily living; loss of health and well-being due to physical pain and psychological anguish (Chochinov et al. 2002).

218

N. S. Jecker

To counter dignity threats, dignity conserving measures ought to include measures that help patients find a way of looking at their situation that can promote dignity and helping them engage in specific practices that compensate for capability shortfalls (Chochinov et al. 2002). For some patients, however, reasonable efforts to counter dignity threats are not enough. To avoid a loss of dignity, they request the ability to also have a measure of control over the manner and timing of their death. Patients may arrive at such a decision in a variety of ways. If they suffer a longstanding chronic progressive disease, the decision may be reached gradually over time, as their quality of living diminishes. If they experience a sudden devastating injury, the decision-making process will look different. Either way, patients in this situation have reached what we may call ‘a tipping point,’ which can be described as a point where there is significant diminishment of our ability to think coherently and feel effectually by physical maladies, by accompanying continuous suffering, and by the effects of medication intended to relieve our pain, erodes our personhood. It is rational to choose to die as oneself rather than survive as a lessened entity (Prado 2015, 442).

For some patients, a tipping point occurs when they lose significant capacities to affiliate, move independently from place to place, or have a measure of control over their immediate environment. While suffering might be part of the mix, it need not be. Likewise while losing the ability to think might be a central loss for some, for others what matters more is the ability to communicate and interact with their family. When serious illness leads to permanent capability loss in areas that matter most to a person, this is an assault on dignity that sometimes even our best efforts cannot forestall. Drawing on this analysis, we can summarize the Dignified Lives argument for assisted dying this way. The Dignified Lives Argument 1. Retaining dignity at the end of life requires retaining a threshold level of the central human capabilities that matter most to a patient. 2. Serious illness can interfere with these threshold human capabilities, and thereby undermine patient dignity. 3. Measures to counter illness-related loss of dignity may also prove inadequate. 4. When this is the case, protecting dignity for terminally ill and imminently dying patients may require, as a last resort, offering them the ability to control the timing and manner of their death. The Dignified Lives argument establishes that when patients lose threshold central human capacities that matter most to them, physician assisted dying should be considered as a treatment of last resort. The argument is notably distinct from arguments that appeal to autonomy to justify assisted dying. In contrast to autonomy-based arguments, dignity-driven arguments present a larger picture of what makes human life valuable. This picture assumes that there are a range of central capabilities that matter to us as human beings, which include not only the ability to think but also the ability to have an unfolding narrative or story of our lives, emote, affiliate, play, be

12 Medical Futility and Physician Assisted Death

219

healthy, move from place to place and relate to nature and other species. In short, dignity is a far broader concept than autonomy; while it includes the ability to plan our lives, it includes many other capabilities too. In this respect, the Dignified Lives argument is at odds with views that hold dignity is redundant with or reducible to autonomy (Macklin 2003). While it builds in the capacity for autonomy as one of the central things that people can do and be, it does not regard this as a standalone capability. Instead, the Dignified Lives argument better captures what is at stake, i.e., the loss of all the capacities that are central to human beings. Admittedly, those who argue for physician-assisted dying by appealing to respect for patient autonomy need not argue that autonomy is the only thing that makes human life valuable. Other goods, such as close relationships, bodily integrity, and health, might be highly valued too. However, what distinguishes the Dignified Lives defense from an autonomy approach is that it holds that a range of goods play a foundational role, whereas the autonomy account renders all goods other than autonomy secondary and conditional; they matter if and only if an autonomous person chooses them. The Dignified Lives argument is also distinct from meaning of life defenses of assisted dying which maintain that a healthy people should be allowed to die if their life has lost meaning. For example, members of the Dutch parliament backed a proposal to legalize assisted death for persons who determined that their lives were without purpose and were ‘completed’ (Government of the Netherlands 2016). This position relies on the individual’s subjective assessment of their life’s value to determine if life is ‘completed.’ The Dignified Lives argument, by contrast, assumes a partially objectivist conception, one that identifies dignity with possessing a list of threshold capabilities. The Dignified Lives argument is similar in key respects to rights-based defenses of assisted dying. For example, Battin (1994) calls assisted dying, “the least worst death.” She defends it by invoking a fundamental right to die, which she says is analogous to the right to life, liberty, freedom of speech and worship, education, political representation and the pursuit of happiness. Battin also hints at a deeper reason, rooted in dignity. She states that the right to die relies on the fact that we are entitled “to do certain things just because doing those things tends to be constitutive of human dignity” (Battin 1994, 280). In this way, dignity furnishes the philosophical underpinning for Battin’s rights claims, making her arguments well-matched with the Dignified Lives argument presented here. One concern a Dignified Lives argument raises is a slippery slope. A slippery slope objection might gain support by noting, as Gawande (2014, 245) does, that “assisted living is far harder than assisted death” and that ultimately, no matter how carefully crafted laws are to avoid potential abuse, the larger culture determines how the practice unfolds. Yet, in reply, it seems much easier to avoid a slippery slope by bringing assisted dying out in the open, setting standards that include checks and balances, and monitoring outcomes, than it is to offering assisted dying behind closed doors and without explicit oversight. Moreover, making available a right to die leaves individuals free to exercise or not exercise this right; no one has a duty to

220

N. S. Jecker

die (Jecker 2014). Likewise, no one has a moral duty to assist; a particular physician is morally free to refuse a request for physician assisted death if meeting it would violate their personal moral beliefs. Opponents of assisted dying also express concern about a related slippery slope, namely that racial and ethnic minorities might not be able to receive high quality medical care that would improve the quality of their life and they will instead be encouraged to use assisted dying instead (Cain and McCleskey 2019). However this argument relies on empirical predictions which have not been borne out in practice. For example, the state of Oregon has collected annual data about requests for assisted dying since its 1997 passage of the “Death With Dignity Act,” which allowed physicians to legally prescribe terminally ill and imminently dying patients a lethal dose of medication that they can choose to self-administer, provided certain conditions are satisfied. The state consistently reports that those requesting assisted death are white, educated, and affluent. In 2020, for example, 96% of patients were white; most (53%) were well educated, with at least a bachelor’s degree; and 75% were older (age 65 or over) (Oregon Health Authority 2020). These data suggest that the objection to assisted dying should be turned on its head. If the Dignified Lives argument is sound, then more should be done to ensure that all people can exercise a right to die, including non-white people, less educated people, and younger adults. Table 12.2 summarizes the arguments of Sect. 12.4. Table 12.2 Ethical bases for physician assisted dying Features

Values

Implications

Physician’s role

Recognizing the physician’s responsibility and the significance of the doctor–patient relationship

Physicians bear partial responsibility for the situation of some dying patients; they also stand in significant relationships with patients

Moral equivalence

Recognizing the moral equivalence of acting and omitting based on similar effects and intrinsic qualities

Omitting LST and assisting with dying are not morally different just because one is an omission while the other is an act

Loss of threshold central human capabilities

Patient dignity and rights

When patients lose threshold central human capacities that matter most for them, physician assisted dying should be available as a treatment of last resort

12 Medical Futility and Physician Assisted Death

221

12.5 Conclusion In conclusion, people who take on the role of doctoring have a duty to withhold and withdraw medically futile life sustaining treatment, and in some instances, they also have a duty to help patients die if they request help. These duties relate directly to the moral ends medicine serves.

References American Medical Association. 2001. AMA code of medical ethics, opinion 5.7. https://www.amaassn.org/delivering-care/ethics/physician-assisted-suicide. Amundsen, Darrel W. 1978. The physician’s obligation to prolong life: A medical duty without classical roots. Hastings Center Report 8 (4): 23–30. Angell, Marcia. 2014. The Brittany Maynard effect: How she is changing the debate on assisted dying. Washington Post, 31 Oct. Ardagh, Michael. 2000. Futility has no utility in resuscitation medicine. Journal of Medical Ethics 26: 396–399. Battin, Margaret P. 1994. The least worst death. New York: Oxford University Press. Beauchamp, Tom L., and James F. Childress. 2019. Principles of biomedical ethics, 8th ed. New York: Oxford University Press. Browlnee, Shannon, Kalipso Chalkidou, Jenny Doust, et al. 2017. Evidence for overuse of medical services around the world. Lancet 390 (10090): 156–168. https://doi.org/10.1016/S0140-673 6(16)32585-5. Cain, Cindy L., and Sara McCleskey. 2019. Expanded definitions of ‘good death’? Race, ethnicity, and medical aid in dying. Social Health and Illness 41 (6): 1175–1191. Chochinov, Harvey M., Thomas Hack, Susan McClement, et al. 2002. Dignity in the terminally ill: A developing empirical model. Social Science and Medicine 54: 433–443. Cross, Sarah H., and Haider J. Warraich. 2019. Changes in the place of death in the United States. New England Journal of Medicine 381: 24. De Roo, Maaike L., Guido Miccinesi, Bregje D. Onwuteaka-Phillipsen, et al. 2014. Actual and preferred place of death of home-dwelling patients from European countries. PLoS One 9 (4): e93762. https://doi.org/10.1371/journal.pone.0093762. Gawande, Atul. 2014. Being mortal: Medicine and what matters in the end. New York: Henry Holt and Co. Gopal, Abilash A. 2015. Physician-assisted suicide: Considering the evidence, existential distress, and an emerging role for psychiatry. Journal of the American Academy of Psychiatry and Law 43: 183–190. Gourevitch, Danielle. 1969. Suicide among the sick in classical antiquity. Bulletin of the History of Medicine 43 (6): 501–518. Government of the Netherlands. 2016. Government: Scope for assisted suicide for people who regard their life as completed, 10 Dec. https://www.government.nl/latest/news/2016/10/21/gov ernment-scope-for-assisted-suicide-for-people-who-regard-their-life-as-completed. Hippocratic Corpus, The Art. 1977 (400 BCE). In Ethics in medicine: Historical perspectives and contemporary concerns, ed. Stanley J. Reiser, Arthur Dyck, and William J. Curran, 6–7. Cambridge, MA: MIT Press.

222

N. S. Jecker

Hippocratic Oath. 1923 (5th century BCE). In Hippocrates of cos, the oath, trans. William H.S. Jones, 298–299. Loeb Classic Library, Harvard University Press. https://www.loebclassics.com/ view/hippocrates_cos-oath/1923/pb_LCL147.299.xml;jsessionid=25C42D950420953E4A01B 0B67AC69595. Hurley, Richard. 2014. Can doctors reduce harmful medical overuse worldwide? BMJ 349: g4289. https://doi.org/10.1136/bmj.g4289. Illych, Ivan. 1995. Limits to medicine, medical nemesis: The expropriation of health. London and New York: Marion Boyars. Jecker, Nancy S. 2013. The problem with rescue medicine. Journal of Medicine and Philosophy 38 (1): 64–81. Jecker, Nancy S. 2014. Against a duty to die. Virtual Mentor: American Medical Association Journal of Ethics 16 (5): 390–394. Jecker, Nancy S. 2015. Futility and fairness: A defense of the Texas advance directive law. American Journal of Bioethics 15 (8): 43–46. Jecker, Nancy S. 2017. Doing what we shouldn’t: Medical futility and moral distress. American Journal of Bioethics 17 (2): 41–43. Jecker, Nancy S. 2020. Ending midlife bias. New York: Oxford University Press. Kadooka, Yasuhiro, Atushi Asai, Miki Fukuyama, et al. 2014. A comparative survey on potentially futile treatments between Japanese nurses and laypeople. Nursing Ethics 21 (1): 64–75. Karsoho, Hadi, Jennifer R. Fishman, David K. Wright, and Mary E. Macdonald. 2016. Suffering and medicalization at the end of life: The case of physician-assisted dying. Social Science & Medicine 170: 188–196. Lichtenberg, Pesach, Uriel Heresco-Levy, and Uriel Nitzan. 2004. The ethics of the placebo in clinical practice. Journal of Medical Ethics 30: 551–554. Macklin, Ruth. 2003. Dignity is a useless concept. BMJ 327: 1419–1420. Miller, Franklin G., and Howard Brody. 1995. Professional integrity and physician-assisted death. Hastings Center Report 25 (3): 8–17. Miller, Franklin G., and Howard Brody. 2001. The internal morality of medicine. Journal of Medicine and Philosophy 26 (6): 581–599. Moerman, Daniel E., and Wayne B. Jonas. 2002. Deconstructing the placebo effect and finding the meaning response. Annals of Internal Medicine 136: 471–476. Nakazawa, Eisuke, Keiichiro Yamamoto, Reina Ozeki-Hayashi, and Akira Akabayashi. 2019. A global dialogue on withholding and withdrawal of medical care: An East Asian perspective. American Journal of Bioethics 19 (3): 50–52. Nelson, Lawrence J., and Erick Ramirez. 2017. Can suicide in the elderly be rational? In Rational suicide in the elderly, Kindle ed, ed. Robert E. McCue and Meera Balasubramaniam. Springer International. Oregon Health Authority. 2020. Oregon death with Dignity Act: 2019 data summary. https://www. oregon.gov/oha/PH/PROVIDERPARTNERRESOURCES/EVALUATIONRESEARCH/DEA THWITHDIGNITYACT/Documents/year22.pdf. Oregon Health Authority. 2021a. Death with Dignity Act. https://deathwithdignity.org/learn/deathwith-dignity-acts/. Oregon Health Authority. 2021b. Oregon death with Dignity Act: 2020 data summary. https:// www.oregon.gov/oha/PH/PROVIDERPARTNERRESOURCES/EVALUATIONRESEARCH/ DEATHWITHDIGNITYACT/Documents/year23.pdf. Plato. 1981 (375 BCE). In Republic, trans. G.M.A. Grube, 76–77. Indianapolis/Cambridge: Hackett Publishing. Prado, Carlos G. 2015. Ageism and elective death. Ethics, Medicine and Public Health 1: 442–449. Rada, Aser G. 2021. Spain will become the sixth country worldwide to allow euthanasia and assisted suicide. BMJ 372: 147. https://doi.org/10.1136/bmj.n147. Schneiderman, Lawrence J., and Nancy S. Jecker. 2011. Wrong medicine: Doctors, patients, and futile treatment, 2nd ed. Baltimore and London: Johns Hopkins University Press.

12 Medical Futility and Physician Assisted Death

223

Schneiderman, Lawrence J., Nancy S. Jecker, and Albert R. Jonsen. 1990. Medical futility: Its meaning and ethical implications. Annals of Internal Medicine 112: 949–954. Schneiderman, Lawrence J., Kathy Faber-Langendoen, and Nancy S. Jecker. 1994. Beyond futility to an ethic of care. American Journal of Medicine 96: 110–114. Schneiderman, Lawrence J., Nancy S. Jecker, and Albert R. Jonsen. 1996. Medical futility: Response to critiques. Annals of Internal Medicine 125 (8): 669–674. Schneiderman, Lawrence J., Nancy S. Jecker, and Albert R. Jonsen. 2017. The abuse of futility. Perspectives in Biology and Medicine 60 (3): 295–313. Seymour, Jane E. 1999. Revisiting medicalisation and ‘natural’ death. Social Science and Medicine 49: 691–704. Shapiro, Arthur K., and Elaine Shapiro. 2000. The placebo effect in medical history. In The powerful placebo: From ancient priest to modern physician, ed. Arthur K. Shapiro and Elaine Shapiro, 1–27. Baltimore and London: Johns Hopkins University Press. The Oregon death with Dignity Act, Oregon revised statutes, 127.800–127.995. https://www. oregon.gov/oha/PH/PROVIDERPARTNERRESOURCES/EVALUATIONRESEARCH/DEA THWITHDIGNITYACT/Documents/statute.pdf. Veatch, Robert M., and Cian M. Spicer. 1992. Medically futile care: The role of the physician in setting limits. American Journal of Law & Medicine 18: 15–36. White, Douglas, and Thaddeus Pope. 2016. Medical futility and potentially inappropriate treatment. In The Oxford handbook of ethics at the end of life, ed. Stuart J. Youngner and Robert M. Arnold, 65–86. New York: Oxford University Press. Wicclair, Mark R. 2016. Conscientious objection. In The Oxford handbook of ethics at the end of life, ed. Stuart J. Youngner and Robert M. Arnold, 87–108. New York: Oxford University Press.

Nancy S. Jecker is Professor of Bioethics and Humanities at the University of Washington School of Medicine. She holds Visiting Professorships at the University of Johannesburg and Chinese University of Hong Kong. Dr. Jecker’s research spans a range of topics, most notably justice and resource allocation, individual and societal aging, global perspectives in philosophy and bioethics, medically futility, and artificial intelligence and social robots. Dr. Jecker has published over 200 articles and 4 books, including a 2020 book, Ending Midlife Bias (Oxford University Press).

Chapter 13

Medical Aid in Dying: The Case of Disability Christopher A. Riddle

Abstract I argue that despite criticism from some disability rights organizations, aid in dying is morally permissible. First, I suggest that disability-related concerns can be classified as emerging from one of two kinds of harm: person affecting, and personhood affecting. Second, I examine whether person affecting harm has occurred within those jurisdictions that have legalized aid in dying. I conclude that despite suggestions to the contrary, there is no evidence to demonstrate that people with disabilities have been adversely impacted by legalized medical aid in dying. Third, I entertain these same arguments as a matter of principle and conclude the principled objections should not give us reason for concern. Fourth, I explore whether we have evidence to suggest the presence of increased personhood affecting harm where aid in dying is permissible. I suggest, through a focus on disability support services, that no evidence of social attitudes that diminish the moral worth of disabled people is present. Fifth, I examine whether we ought to have a principled concern with personhood affecting harm emerging from medical aid in dying. I conclude that to the contrary, to deny people with disabilities access to aid in dying constitutes a morally egregious form of personhood affecting harm. Keywords Assisted dying · Medical aid in dying · Disability · Autonomy · DAHE

13.1 Introduction Francis, a retired nurse, received a diagnosis of amyotrophic lateral sclerosis (ALS), or Lou Gehrig’s disease, after a two-year long diagnostic journey (Buchbinder 2021, 40–41). By the time she was interviewed about her experiences, she was communicating primarily through writing, as she had lost the ability to speak. She experienced difficulty eating and had limited mobility, with the exception of the use of three fingers (Buchbinder 2021, 41–42). While Francis began to think about aid in dying not long after receiving her diagnosis, her husband Tim, was initially reluctant to support her C. A. Riddle (B) Utica University, Utica, NY, USA e-mail: [email protected] © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 M. Cholbi and J. Varelius (eds.), New Directions in the Ethics of Assisted Suicide and Euthanasia, The International Library of Bioethics 103, https://doi.org/10.1007/978-3-031-25315-7_13

225

226

C. A. Riddle

wishes. After witnessing her physical decline, and inability to both manage pain, as well as retain autonomy, he came to accept her wishes (Buchbinder 2021, 42). The question of whether Francis and people like her should have access to aid in dying, is not a simple, straightforward, or uncontested one. I suggest mentally competent adults with a terminal diagnosis should be able to request, and receive access to, medical aid in dying. I do so, in spite of much criticism from within disability rights organizations and from disability rights advocates. Despite common conceptions however, there is not a homogenous perspective on whether aid in dying ought to be permissible within the disability and disability rights community. For example, within England, Wales, and Scotland, 44% of examined disability rights organizations have no mention of a position on assisted dying, 41% have no public position on aid in dying, 4% adopt a neutral position, 4% oppose it, and 9% are ambiguous (Box and Chambaere 2021, 2). This is of particular note, because given the historical significance of the United Kingdom for the disability rights movement, one might expect stronger opposition to legalized aid in dying.1 Also of note is that in at least in some regions of the United States where data was available, support for aid in dying appears to be approximately equal amongst disabled voters and non-disabled voters (Compassion and Choices 2014). Almost two-thirds of people, disabled or not, expressed support for aid in dying (Compassion and Choices 2014). That said, opposition to aid in dying from disability rights organizations takes on a variety of forms, but most notably focuses on a perceived lack of adequate safeguards and their potential for abuse, as well as the further devaluing of the lives of people with disabilities (Box and Chambaere 2021, 4). Conversely, those who support the practice tend to focus on support by their membership due to suitable safeguards, as well as the respect for autonomy (Box and Chambaere 2021, 5). I proceed by suggesting that despite criticism from some disability rights proponents and organizations, aid in dying is morally permissible. I do so in the following manner. First, in Sect. 13.2, I suggest that opposition to aid in dying can be rightly classified as being one of two kinds. I reintroduce a distinction I have made elsewhere between person affecting vulnerability and harm, and personhood affecting vulnerability and harm. Harm or vulnerability of the person affecting kind represents a threat to one’s physical well-being. On the other hand, harm or vulnerability of the personhood affecting kind damages the status of individuals as valuable members within a moral community. Second, in Sect. 13.2.1, I examine whether person affecting harm has occurred within those jurisdictions that have legalized aid in dying. I conclude that despite claims to the contrary, there is no evidence to suggest people with disabilities or other vulnerable groups have been adversely impacted by legalized medical aid in dying. Third, in Sect. 13.2.2, I entertain these same arguments as a matter of principle. While no person affecting harm has occurred against people with disabilities as a result of aid in dying, I ask whether the principled objections ought to 1

Here I am thinking of the significance of UPIAS, for example, in shifting public attitudes about disability away from a medical or individual pathological model of disability, to instead, a socially freeing and emancipatory, social model of disability (Union of the Physically Impaired Against Segretation 1976).

13 Medical Aid in Dying: The Case of Disability

227

give us reason for pause or concern. I suggest they do not. Fourth, in Sect. 13.2.3, I explore whether we have any evidence that suggests the presence of personhood affecting harm in those states where aid in dying is permissible. I suggest, through a focus on disability support services and expenditure, that no evidence of heightened or increased social attitudes that diminish the moral worth of people with disabilities is present. Fifth, in Sect. 13.2.4, I examine whether, despite evidence to the contrary, we ought to have a principled concern with personhood affecting harm emerging from medical aid in dying. I conclude that we have no such principled reason, and to the contrary, to deny people with disabilities (and others) access to aid in dying constitutes a morally egregious form of personhood affecting harm. Prior to doing so however, two points of clarification are in order. The first concerns the notion of ‘disability’ employed throughout the paper, while the second pertains to the practice of aid in dying. Throughout what follows, the discussion of disability is actually of a subset of the entire disabled population. While I have argued elsewhere how we ought to conceptualize disability (Riddle 2012, 2013), the discussion contained within this chapter excludes those people with disabilities who are unable to consent to care at the end of their life. Of course, people who fall within this category can have a wide range of impairments, but in order to avail oneself of aid in dying, one must be capable of consenting to the practice, and thus, we must exclude those unable to do so from our considerations. While others have suggested that those with impairments that might be thought to limit their capacity ought to be capable of seeking aid in dying (Schuklenk and van de Vathorst 2015), it is not my intention to defend this extension within this chapter. Relatedly, is clarification pertaining to medical aid in dying and its practice. Throughout this chapter, the practice of permitting a patient with a terminal diagnosis to hasten their own death is what is being defended: medical aid in dying. The administering of a life taking drug, or euthanasia, is irrelevant for the discussion that unfolds below, as is the hastening of the death experience for those without a terminal diagnosis.

13.2 Vulnerability and Harm That said, I suggest that a distinction I have previously made can be a useful heuristic to help understand the kinds of arguments that get used by disability rights organizations when launching opposition to medical aid in dying. A good deal of the opposition to medical aid in dying suggests that particularly vulnerable groups might be harmed by legalizing its practice. I have previously argued that when most think of vulnerability and harm to people with disabilities or other groups, they tend to conceptualize this harm in a person affecting manner. One is vulnerable in a person affecting way if there is an increased threat to their physical well-being (Riddle 2021b, 162). These forms of vulnerability are both inherent, insofar as they are intrinsic to the human condition, and situational, because some

228

C. A. Riddle

sources of vulnerability are context specific and caused or exacerbated by social, political, economic, or environmental factors (Rogers et al. 2012, 17; Mackenzie 2014, 38–39). However, often underemphasized is the kind of vulnerability and harm associated with one’s personhood. I suggest that “not all vulnerability emerges as a result of an increased threat to one’s person” (Riddle 2021b, 164). I suggest that social structures are such that people with disabilities are also vulnerable in a personhood affecting manner, or that they are more likely to experience stigmatization as a result of their impairment, to experience a lack of the social-basis of self-respect or dignity, or more generally, what I call the questioning of their personhood (Riddle 2021b, 164). Not all harm emerges as a result of threats to one’s person, but instead, prejudices or oppressive attitudes or actions can threaten to undermine the status of a disabled person as someone of moral worth on par with that of others (Mackenzie 2014, 41). In other words, personhood vulnerabilities or harms shift “attention to the manner in which social attitudes adversely impact social standing or membership to a community of valued individuals” (Riddle 2021b, 164). I suggest that individuals become personhood vulnerable, or suffer personhood affecting harm, when attitudes that perpetuate harmful stereotypes become prevalent enough to diminish the dignity and social basis of self-respect of disabled people (Riddle 2021b, 164). Opposition to aid in dying from disability rights organizations, as mentioned, often takes the form of a concern over a lack of adequate safeguards, or a concern about the devaluing of the lives of people with disabilities or other vulnerable groups. I suggest these concerns can be classified as person affecting harm, and personhood affecting harm, respectively. By dividing objections into kinds, and rebutting on the basis of the kind of argument given, it permits me the opportunity to handle multiple articulations of opposition (that reside within the same kind, nonetheless) with a single response. Take for example, the following multiple concerns that all reside within the kind of person affecting harm. Some are concerned that aid in dying might result in disabled people being “indirectly pressured into assisted dying” (Box and Chambaere 2021, 4). Others still, might express a concern about direct pressure to hasten the death experience, or the disabled receiving or feeling to have received, encouragement to give up (Box and Chambaere 2021, 4). Importantly, harm of this variety can impact both physical as well as mental well-being. Thus, person affecting harm need not impact someone solely in a physical manner, as one’s person could be harmed through a policy that adversely impacts one’s mental health, for example. Other disability rights organizations use stronger language and suggest that instead of pressure, the disabled might be otherwise coerced into seeking aid in dying (Box and Chambaere 2021, 4). An underlying thread amongst most of these objections concerns a lack of adequate safeguards being able to protect people with disabilities from harm (Stainton 2019, 4). All these kinds of arguments are of the person affecting harm variety. Similarly, multiple articulations of opposition might be said to be of the personhood affecting harm variety. Some disability rights organizations highlight how aid

13 Medical Aid in Dying: The Case of Disability

229

in dying might reflect or magnify prejudices against disabled people (Box and Chambaere 2021, 4). Aid in dying, it is suggested, demonstrates a negative valuation of disabled lives (Stainton 2019, 3). Others suggest a potential devaluing of people with disabilities and highlight the potentially detrimental way in which it might adversely impact social perceptions of the disabled (Box and Chambaere 2021, 4). Irrespective of the kind of harm disability rights organizations are concerned about, I suggest they ought not to be. In what follows, I address person affecting harm, and personhood affecting harm, in turn.

13.2.1 Evaluating Evidence: Person Vulnerability and Harm Concerns about person affecting harm permeate almost all disability rights organizations’ objections to aid in dying. The question at hand is the following: Is there any evidence in jurisdictions where aid in dying is legal that suggests harm befalls people with disabilities or other vulnerable populations to a greater extent than other states without legalized aid in dying? Prior to attempting to answer this question, an acknowledgement must be made. Our medical systems are failing not only, but principally, people with disabilities in wildly unacceptable and unjust ways. Healthcare disparities experienced by individuals with disabilities are abundant and appalling (Meade et al. 2015, 633). Sadly, despite the overwhelming data that suggest this fact, most assessments are also thought to underestimate or underreport the actual disparity present between disabled and non-disabled patients because of deficient understandings of the nuances of the disparities experienced by people with disabilities (Meade et al. 2015, 638). Moreover, within healthcare settings, people with disabilities often receive lesser care as a result of dominant discriminatory attitudes amongst health professionals (VanPuymbrouck et al. 2020, 102). A majority of healthcare providers were found to be ‘aversive ablests’, in so far as they demonstrated a preference for non-disabled patients (VanPuymbrouck et al. 2020, 101). In short, this acknowledgement is made to suggest that people with disabilities or disability rights organizations are not incorrect to be skeptical about medical facilities and healthcare more generally. Many people with disabilities are more familiar with healthcare settings than most, and are far more likely to have negative experiences than their non-disabled counterparts (Gill 2010, 35). In short, people with disabilities ought to be listened to. Not only are they experts about the care of their own bodies, but they have tremendously valuable knowledge about healthcare more generally, as a result of utilizing healthcare services more frequently. That said, even though the concern expressed by disabled people is of no doubt, a real concern, and a concern with a good cause, the question remains as to whether abuse of aid in dying has occurred and has resulted in harm to the disabled or other populations also commonly perceived to be person vulnerable. The short answer is that no harm appears to have befallen people with disabilities or others as a result of permitting aid in dying.

230

C. A. Riddle

Consider first, jurisdictions outside of America prior to moving to American ones. There exist more complexities with systems in most European jurisdictions than American ones, but nonetheless, these complexities do not give way to abuse. In the Netherlands, for example, there is no evidence that people with disabilities or other vulnerable groups are experiencing harm as a result of medical aid in dying. There is some well-documented concern over under-reporting within the Netherlands, but den Hartogh (2012) attributes this under-reporting to what he calls “a relic of prelegalization practice” (366), and not as a result of the legalizing of euthanasia. Opponents suggest that doctors and other medical professionals might be inclined to hasten the death experience against the wishes of a patient. Disability Rights Organizations suggest that this practice is more likely employed against people with disabilities who might be perceived to have a life not worth living. Disability Rights objectors cite concerns that the most likely manner in which lives might be ended without request would be for those outside of the terminal window to be killed (den Hartogh 2012, 365). The concern emerges from the claim that with normalizing the taking of lives, compassion might be thought to extend outside of the legal restrictions placed on aid in dying practices. Medical professionals might view suffering to be so bad that they hasten the death experience, even without death being imminent, as defined by the law. Perhaps obviously, given the ablest attitudes of many in society, disability rights organizations suggest this notion of suffering and a life not worth living might be more likely to be applied to people with disabilities. The concern here then, is that an under-reporting of the use of some drugs, such as morphine, might result in the use of it in large doses to kill those who have not expressed desire to die, and who have not navigated the legal process and the safeguards within. Opponents do not attribute malicious intent to medical professionals in all instances, but instead, can suggest such actions could emerge from a misbegotten effort to spare the disabled from lives perceived to be so dominated by suffering that they are not worth living. That said, there appears to be no evidence of this in the Netherlands that cannot be explained by a more general underreporting of morphine use (den Hartogh 2012, 366). In short, “no evidence for this causal nexus has ever been offered” (den Hartogh 2012, 365). In other words, while under-reporting of terminal sedative drugs exists, although degreasing in frequency (Onwuteaka-Philipsen 2012, 127), there is no evidence to suggest that this under-reporting constitutes an instance of abuse against disabled people or other vulnerable populations (Battin et al. 2007, 597). Similar conclusions can be drawn in Belgium. Indeed, little or no opposition exists to Belgian laws from Belgian disability rights organizations or people with disabilities. Concern about abuse is not present in Belgium and there appears to have been little or no opposition to the legalizing of euthanasia from disabled people (Fitzpatrick and Jones 2017, 147). There is no evidence to suggest abuse of any kind, and indeed, there remains to be no organized disability-rights-based opposition to even a mere hypothetical risk of harm, let alone any attempt to suggest actual harm exists (Fitzpatrick and Jones 2017, 149). In Canada, where aid in dying has been legal since 2016, and where 2% of all accounted deaths were attributed to the practice in 2019, there is no evidence of person affecting harm emerging from either abuse of the system, or the system itself

13 Medical Aid in Dying: The Case of Disability

231

(Martin 2021, 137). Indeed, recent data from Canada, the Netherlands, and Belgium are consistent with the claim that there is “no indication that individuals who may be vulnerable to undue influence are accessing assistance in dying” (Martin 2021, 142). Within the United States, and Oregon in particular, the jurisdiction with the oldest assisted dying laws in America, there is no evidence of vulnerable populations of any kind experiencing person affecting harm. No Oregonians with disabilities have, since 1997, experienced person affecting harm from aid in dying. No people without a terminal diagnosis confirmed by two physicians have died in Oregon (Battin et al. 2007, 594). More pointedly, “no one received such assistance for disability alone” (Battin et al. 2007, 594). Robert Lindsay has concluded that a “decade after implementation of the ODWDA [Oregon Death with Dignity Act], the weight of evidence suggests that these predictions of dire consequences were incorrect” (Lindsay 2009, 19). Indeed, a good deal of data points to rejecting many people who requested aid in dying, who were not deemed capable of consenting to such action. Almost 20% of requests for aid in dying came from patients deemed to be experiencing depression, and exactly none of them progressed to medical aid in dying (Battin et al. 2007, 596). More generally, no people availing themselves of aid in dying were concluded to have a mental illness influencing their decision (Battin et al. 2007, 596).2 Indeed, not only has no disparate impact on those perceived to be vulnerable been detected, but there has been no slippery slope, and there has been, more generally, the effective prevention of abuse (Lindsay 2009, 22–23). Indeed, some strong opponents to medical aid in dying have publicly expressed that the concerns they previously stated have not materialized (Coombs Lee 2014, 97–98). In short, there is no evidence of abuse or coercion, and there is no evidence to suggest the misuse of the carefully crafted policies supporting aid in dying (Coombs Lee 2014, 99). To support this point further, consider that 87.8% of individuals availing themselves of medical aid in dying were in a hospice setting (Al Rabadi et al. 2019, 5). If patients were typically placed in a hospice care setting prior to initiating medical aid in dying requests, there is an additional layer of protection to confirm terminal diagnosis, and to thus, avoid or mitigate the potential for the sort of abuse opponents suggest is present.3 That said, data “supports the overall safety and reliability of the lethal medications used in MAID [medical aid in dying]” (Al Rabadi et al. 2019, 5). But, in the absence of evidence of person affecting harm, is there still cause to be concerned about this possibility as a matter of principle? I suggest there is not.

2

It is relevant to note that this study has received critical appraisal (Finlay and George 2011). That said, the critical remarks focused on the study’s perceived failure to identify all possible forms of vulnerability, and thus, do not undermine the claims pertaining to any potential person affecting harm to people with disabilities. 3 Indeed, palliative care, has appeared to have improved in jurisdictions permitting aid in dying (Lindsay 2009, 19).

232

C. A. Riddle

13.2.2 Arguments: Person Vulnerability and Harm I suggest that arguments concerned about person affecting harm ought not to be regarded as justifying a prohibition on aid in dying for at least the following two reasons. First, disability rights organizations that suggest person affecting harm constitutes a sufficient threat to prohibit medical aid in dying are guilty of moral inconsistency. Second, these arguments also fail because of moral disproportionality. First, let us examine the form that these person affecting arguments take. Some forms of the argument suggest that there is an illusion of free choice when seeking medical aid in dying, and thus, people with disabilities will be harmed because they will be forced or coerced to avail themselves of it. The claim is that for some vulnerable populations, it is not a free choice, but instead, a forced one (Scoccia 2010, 481). It is suggested that when choices are made in the context of pervasive inequality, or under a structure of oppression, free choices cannot exist (Scoccia 2010, 481). Indeed, actions taken by people with disabilities to seek aid in dying might be thought to be suitably likened to those taken by others when under duress (Feinberg 1989, 98–219). Additionally, not only is the concern that people with disabilities might avail themselves of aid in dying due to social pressures, but that they themselves might be viewed as preferential subjects or objects of euthanasia and be killed against their will (Somerville 2001, 263). In short people with disabilities might not only be pressured to invoke the legal process for aid in dying, but they might be killed, against their will, in spite of safeguards or laws designed to protect them. Some suggest the only method to ensure the prevention of person affecting harm as a result of aid in dying is its prohibition: ‘[S]afeguards cannot be established to prevent abuses resulting in the wrongful death of death of numerous disabled persons, old and young.’ Indeed, the only true safeguards against abuse ‘is that assisted suicide remain illegal and socially condemned for all citizens equally’. (Bickenbach 1998, 125)

Sumner (2018, 105) calls arguments of this variety, ‘arguments from abuse’, and suggests the common thread is a concern over safeguards being inadequately established, or monitored and enforced. Similarly, but more generally, I have previously called these arguments, ‘avoidance of harm’ arguments (Riddle 2019, 188–190). I believe arguments of this kind suffer from a moral inconsistency that renders them ineffective. First, consider how many people die as a result of aid in dying. This numbers differs significantly depending upon jurisdiction, but ranges from 0.05% of deaths, to as high as 1.7% of total deaths (Emanuel et al. 2016, 85). By any measure, this number represents a very low percentage of total deaths. In other words, not many people are dying from aid in dying. In contrast, consider those who have opted to refuse or remove life sustaining treatment. Approximately 85% of critical care physician respondents acknowledged that they had withdrawn or withheld life support in the preceding year (Way et al. 2002, 1342). An American study indicated that between 1992 and 1993, over 90% of deaths in intensive care units resulted from a decision to withdraw or withhold life

13 Medical Aid in Dying: The Case of Disability

233

support (Way et al. 2002, 1342). This is true in most countries, where most deaths in intensive care units occur as a result of a decision to stop or refuse life sustaining treatment (Way et al. 2002, 1342). Importantly, all the reasons that can be invoked to support the refusal or removal of life sustaining treatment, can be applied, with equal force, to medical aid in dying. Because we value patient autonomy, and relief of suffering, we permit patients to make choices about the kind of care they receive, or do not receive, at the end of their life. These reasons are so powerful, that we permit people the autonomy to make them even when it will surely result in their death. Why then, are disability rights organizations not in favor of denying patients the right to remove or refuse treatment? I suggest that for their argument against aid in dying to be morally consistent, they must. Surely the potential for abuse that can emerge with aid in dying is also present in current practices. People with disabilities or other vulnerable groups, such as the elderly, or even those living in poverty, might be thought to be pressured into hastening their death experience. They could just as likely be subjected to an unjust death from a medical professional exercising a wrongful notion of compassion to rid them of a life perceived to be so dominated by suffering or misery, that it is not worth living. In short, our current practices that permit people to make choices about care at the end of their life are not subject to the same scrutiny that medical aid in dying is, and many more people are forced to make choices pertaining to the refusal or removal of care, than those who will be eligible, or who will seek, aid in dying. If disability rights organizations were genuinely concerned about abuse of healthcare systems and person affecting harm against people with disabilities, they should be equally as concerned about granting any autonomous decision-making ability at the end of life, due to both its equal potential for abuse, and its more frequent use. They are silent on this matter however. To fail to apply their moral logic in this case constitutes a moral inconsistency that is both unjustifiable as a matter of principle, as well as inexplicable. But aside from the arguments concerned with person affecting harm being morally inconsistent, I argue they are also morally disproportionate. By morally disproportionate, I mean to suggest something like the following: as a result of a moral wrong or harm, actions to be taken must be proportionate to that harm, and similar to analogous cases where moral wrong or harm has been done. In other words, despite the fact that no demonstrable person affecting harm emerges from permitting aid in dying, if it could, it would be insufficient to point to an instance or instances or harm, and suggest that on the basis of that harm, a proportionate response is its prohibition. Instead, one must demonstrate, again, counterfactually, that not only will harm emerge, but that it is of a sufficient quality and quantity that it justifies an outright refusal to permit the action leading to that harm. To be clear, such an argument has not been made with reference to aid in dying, nor can it be. To demonstrate this, consider other actions that have risks. All medicine carries risk. For example, it is thought to be the case that as high as 10% of patients admitted into a hospital setting will suffer an adverse reaction, or acquire a new ailment, often as a result of medical error, by virtue of being in the hospital (Riddle

234

C. A. Riddle

2019, 190). This number is startling, and demonstrates the risk that we endure to receive medical treatment. No one suggests we ought to prohibit hospital visits as a result of medical error and the person affecting harm that emerges as a result of it. The reason this is not suggested is because it is not morally proportionate to do so. It is neither proportionate to risk aversion strategies employed in morally similar situations, nor would it be proportionate to the actual quality and quantity of harm or risk, more generally. We can now circle back to a discussion of the refusal or removal of life sustaining care. I argue that this represents a much greater threat to people with disabilities and other vulnerable populations than aid in dying. That said, no one has argued for a denial of autonomy at the end of a patients’ life in this regard. As a result of moral proportionality, disability rights organizations cannot argue for a moral prohibition on aid in dying. At most, opponents to aid in dying can argue for safeguards to be enacted, as we do with reference to hospital visits more generally, and to end of life decisions to refuse or remove life sustaining care. To suggest a prohibition on aid in dying is justifiable, is to make a morally disproportionate argument. In other words, opponents to aid in dying overstate the implication of their argument, and suggest a prohibition on the practice, when at best, their principled case can justify the enacting of safeguards, which have already been established, and proven to be reliable.

13.2.3 Evaluating Evidence: Personhood Vulnerability and Harm That said, not all harm that can emerge from legalized aid in dying is person affecting harm. Personhood affecting harm is also a real concern of disability rights organizations. The prejudices that exist against people with disabilities are in fact, harmful and abundant (Morin et al. 2013). Disability rights organizations suggest that by legalizing aid in dying, people with disabilities will be further devalued and harmful stereotypes will be ignited, rather than extinguished. If people with disabilities are at present, devalued, which we have sound reason to believe is true, the concern is that legalizing medical aid in dying would be even more “detrimental to the way that [the disabled] are viewed by society as a whole” (Box and Chambaere 2021, 4). Measuring social attitudes is difficult, especially when subjects are asked about attitudes or dispositions that they know they ought not to have, or that are not socially favorable, such as discriminatory or ablest ones (LaPiere 1934, 230). Indeed, it is often thought that actions are more representative of attitudes or dispositions (LaPiere 1934, 237). The adage, ‘actions speak louder than words’ is perhaps helpful to bear in mind here. If what we aim to discover is if people with disabilities are devalued to a greater extent in states that have legally permissible aid in dying, than examining how those states treat people with disabilities, and not just reported attitudes, is perhaps a good starting point. In other words, if disability rights organizations suggest that

13 Medical Aid in Dying: The Case of Disability

235

disabled people are devalued by legalized aid in dying, it should be the case that support services and spending on people with disabilities is less in states with medical aid in dying than in those without. In fact, the opposite appears to be true.4 Just as with person affecting harm, let us start by examining non-American jurisdictions first. Public expenditure on disability (PED) is a measure commonly used in Europe to assess welfare programs for people with disabilities across different, and seemingly incommensurable, social welfare models. European models of disability welfare are startlingly different in their approaches and application, but have a common solidarity and commitment to both social justice more generally, as well as the provision of resources to mitigate and eliminate social exclusion, more specifically (Boeri et al. 2001; Hemerijck 2002). People with disabilities are thought to represent approximately 17% of the population of Europe for people between the ages of 16 and 64 (Navarro et al. 2021, 1481). Given the significance of this number, PED is an especially important measure. The typical manner in which PED is assessed is as a percentage of total social expenditures. Thus, the higher the percentage of total social expenditure absorbed by PED, the more resources allocated to people with disabilities, and in my estimation, the greater the social value placed upon disabled people. After all, if people with disabilities were devalued, presumably the policies within those States would reflect those values, and public expenditure would at least trend in a direction that reflected those social values. Conversely, if disabled people were thought to deserve provisions necessary through the law, public expenditure would also reflect this positive disposition (or at least not a negative one) towards the disabled. The European Union (EU) average is 7.38% of total social expenditure on PED. Countries that are thought to correspond to the Nordic typology, perhaps unsurprisingly, do remarkably well in this regard. Denmark, Sweden, Holland, and Finland, are all significantly higher than the other EU countries (Navarro et al. 2021, 1481). That said, Belgium and the Netherlands have a PED as a total percentage of social expenditure much higher than average. Belgium’s PED as a percentage of total social expenditure is just below 9%, and the Netherlands is just above 9%—significantly above the European Union average (Navarro et al. 2021, 1481). In contrast, countries such as France, Ireland, Greece, and the United Kingdom are all below the European Union average. In short, if it were true that legalized aid in dying causes the further devaluing of disabled people, and if it were also true that this devaluing would manifest itself in harmful social policy and less expenditure on support for people with disabilities, then those countries that permit aid in dying should be spending less on the disabled. This is not the case. Let us shift our focus now to American jurisdictions. In the United States, the most common measure utilized for our present purposes is disability-associated health 4

It is important to note that I am not suggesting a causal effect between aid and dying and support for people with disabilities. I am not suggesting that legalizing aid in dying increases support services for disabled people. That said, it is at least possible that in light of a concern over abuse of aid in dying, states increase support services for people with disabilities. This has not been established however.

236

C. A. Riddle

expenditures (DAHE). In 2015, for example, DAHE were $868 billion nationally (Khavjou et al. 2021, 441). This number accounted for 36% of total health care expenditure nationally, and it ranged from 29 to 41% across states (Khavjou et al. 2021, 441). Oregon spent 40% of total health expenditures on DAHE (Khavjou et al. 2021, 444). This number is bested only by two other states. Washington, which has the second oldest aid in dying laws in the country, also has a DAHE that is above the national average (Khavjou et al. 2021, 444). Vermont, the next state to legalize medical aid in dying, is on par with the national average (Khavjou et al. 2021, 444). California, although aid in dying was legalized the same year as the data was gathered, has a DAHE two points above the national average. Indeed, no state that had legalized aid in dying had a DAHE as a percentage of total health expenditures less than the national average at the time the data was collected. More pointedly, the data in the United States suggests that states with legalized aid in dying have a DAHE as a percentage of their total expenditure of health services, that is equal, or greater than, the national average. States that permit aid in dying are not devaluing people with disabilities or under-funding support services to any greater extent than states that do not permit medical aid in dying. Just as in Belgium and the Netherlands, there appears to be a correlation between those states with aid in dying, and a higher DAHE as a percentage of total health expenditure. Indeed, between 2003 and 2015, DAHE per capita spending increased well above the national average in Oregon. While nationally, the increase represented a 28% change, in Oregon it was 64% (Khavjou et al. 2021, 448). In other words, Oregon appears to be increasing its DAHE as a percentage of total health expenditure at a rate much fast than other states. Presumably, if aid in dying caused the devaluing of disabled people, this would result in DAHE per capita spending decreasing, or at least increasing slower than national trends, and not much faster. Again, every state with legalized aid in dying at the time these data were collected is above the national average with respect to increased spending on DAHE between 2003 and 2015 (Khavjou et al. 2021, 448). In short, there appears to be no factual evidence to support the claim that legalizing aid in dying causes the greater devaluing of people with disabilities. More pointedly, personhood affecting harm does not emerge as a result of aid in dying. It is simply inaccurate to suggest it does.

13.2.4 Arguments: Personhood Vulnerability and Harm However, do arguments pertaining to personhood affecting harm have any principled merit? I argue that they do not. I argue that respect for people with disabilities and their autonomy demands access to aid in dying, and not its denial or prevention. Arguments of the personhood affecting kind share a common sentiment: permitting aid in dying devalues the lives of people with disabilities. A primary concern is that harmful stereotypes become further ingrained into society when we suggest

13 Medical Aid in Dying: The Case of Disability

237

some lives are not worth living (Gill 2010, 35). More strongly, opponents can suggest that even if medical aid in dying provided a benefit to everyone, including people with disabilities, and did not introduce person affecting harm into their lives, that it would still be impermissible because the very practice harms people with disabilities as a group, or class (Scoccia 2010, 480). An analogy can be drawn between arguments of this sort, and arguments against something like sex work, for example, that suggest even if it were not harmful to sex workers themselves, the very practice harms women, more generally (Scoccia 2010, 480). Arguments of the personhood affecting harm kind are such that even without harm actually befalling people with disabilities, there is a greater social harm being done in the perpetuating of harmful attitudes or demeaning stereotypes against the disabled. These arguments suggest that an already marginalized or oppressed group is only bound to have those harmful attitudes magnified if aid in dying is encouraged or allowed. Given that many of us tend to think we have even stronger obligations to avoid further harming already disadvantaged populations, it only stands to reason, they might suggest, that the argumentative force behind a denial of access to aid in dying is even stronger when couched as being a matter of importance for disabled people. People with disabilities are devalued, disrespected, and denied access to many things they are owed as a matter of justice.5 To suggest otherwise would be an error. People with disabilities are more likely to live in poverty—to be prone to economic deprivation (Palmer 2011). Disabled people are also far more likely to experience difficulty receiving education and educational resources (Riddle 2021a, 91). That said, simply because people with disabilities are denied many opportunities to flourish as a result of socially dominant harmful attitudes, does not imply that aid in dying perpetuates these attitudes, or further ingrains these injustices. Opponents to aid in dying suggest that it results in an affront to the dignity of disabled people that manifests itself through social policy and laws. As a result of this personhood vulnerability, disability rights organizations argue that aid in dying ought not to be permissible. To the contrary, I suggest that respect for people with disabilities demands the guaranteeing of autonomy and the recognition that people with disabilities, like people without disabilities, are best suited to make decisions about their own life and their own medical care. To suggest that people with disabilities ought to be denied the ability to control what happens with their own bodies at the end of their lives is an overly paternalistic attitude that cannot be justified, and that itself, does harm. Indeed, respect for patient autonomy and compassion for patient suffering are claimed to provide powerful pro tanto reasons for permitting aid in dying (Sumner 2018, 103). The very manner in which these arguments are constructed demonstrates a lack of respect for the autonomy of people with disabilities. People with disabilities are individuals, and not an amorphous group of insignificant parts. Instead, rightly regarded, disabled people are capable of making important decisions on their own (Nelson 2003, 3). 5

I have argued this more thoroughly in Riddle (2014, 2016, 2017b, 2020).

238

C. A. Riddle

I have previously argued that denying “people with disabilities the right to exercise autonomy over their own life and death says powerfully damaging things about the disabled, their abilities, and their need to be protected” (Riddle 2017a, 487). The late Silvers (1998) has forcefully stated that “characterizing people with disabilities as incompetent, easily coerced, and inclined to end their lives places them in the roles to which they have been confined by disability discrimination” (133). The attitude that people with disabilities need protecting from themselves is in itself, demeaning and patronizing. Thus, if disability rights organizations want to promote the dignity and rights of people with disabilities, denying medical aid in dying is not the proper means of doing so. The patronizing and paternalistic attitudes displayed by opponents to aid in dying cause personhood affecting harm, rather than prevent it. By acknowledging that people with disabilities do not need protecting from themselves and that they are capable of making choices about their own care, even if pressured from ablest social attitudes, we can begin to undo the negative stereotypes that have followed disabled people even after the enacting of human rights provisions such as the Americans with Disabilities Act, or the UN Convention on the Rights of Persons with Disabilities. If true regard or care is to be given to endorsing actions that promote accurate, positive dispositions towards people with disabilities, disabled people need to stop being painted as helpless, pitiable individuals, requiring the care and protection of others. Such a disposition plays into ablest preconceptions of disability and further entrenches attitudes of disability as a state of suboptimal or inferior functioning. Importantly, the general practice of aid in dying, or the particular laws surrounding its implementation, make no judgments about what kinds of life are worth living. The only inherent values in the practice of medical aid in dying are ones concerning compassion for suffering, and perhaps most importantly, respect for autonomy. Neither the practice nor the laws force anyone to seek aid in dying and to suggest that people with disabilities are especially vulnerable to social nudging is to perpetuate the myth that people with disabilities cannot make decisions of their own and need to be protected from themselves. Similarly, these laws offer no guidance as to who ought to consider such a practice, aside from those with terminal conditions. Negative valuations about people with disabilities are not perpetuated or brought to the forefront through legalized medical aid in dying. Denying its practice as a result of the perceived personhood vulnerability of disabled people, however, does.

13.3 Concluding Remarks I have argued that despite criticism from some disability rights organizations, aid in dying is morally permissible. First, I suggested that disability-related concerns could be classified as emerging from one of two kinds of harm: person affecting, and personhood affecting. Second, I examined whether person affecting harm has occurred within those jurisdictions that have legalized aid in dying. I concluded that despite objections to the contrary, there is no evidence to suggest people with

13 Medical Aid in Dying: The Case of Disability

239

disabilities have been adversely impacted by legalized medical aid in dying. Third, I entertained these same arguments as a matter of principle and concluded the principled objections should not give us reason for concern. Fourth, I explored whether we have evidence to suggest the presence of increased personhood affecting harm where aid in dying is permissible. I suggested, through a focus on disability support services, that no evidence of social attitudes that diminish the moral worth of disabled people was present. Fifth, I examined whether we ought to have a principled concern with personhood affecting harm emerging from medical aid in dying. I concluded that to the contrary, to deny people with disabilities access to aid in dying constitutes a morally egregious form of personhood affecting harm. In short, everyone should have the autonomy to make choices pertaining to their own medical care at the end of life. More pointedly, aid in dying should be permissible to all those desiring it, and capable of consenting to it. Opposition from disability rights organizations and advocates do not constitute sufficiently robust arguments to justify an outright prohibition on medical aid dying.

References Al Rabadi, Luai, Michael Leblanc, Taylor Bucy, Lee M. Ellis, Dawn L. Hershman, Frank L. Meyskens, Lynne Taylor, and Charles D. Blanke. 2019. Trends in medical aid in dying in Oregon and Washington. JAMA Network Open 2 (8): e198648. Battin, Margaret P., Agnes van der Heide, Linda Ganzini, Gerrit van der Wal, and Bregje D. Onwuteaka-Philipsen. 2007. Legal physician-assisted dying in Oregon and the Netherlands: Evidence concerning the impact on patients in ‘vulnerable’ groups. Journal of Medical Ethics 33 (10): 59–597. Bickenbach, Jerome E. 1998. Disability and life-ending decisions. In Physician assisted suicide: Expanding the debate, ed. Margaret P. Battin, Rosamond Rhodes, and Anita Silvers, 123–132. New York: Routledge. Boeri, T., A. Borsch-Supan, and G. Tabellini. 2001. Would you like to shrink the welfare state? A survey of European citizens. Economic Policy 32: 9–50. Box, Graham, and Kenneth Chambaere. 2021. Views of disability rights organizations on assisted dying legislation in England, Wales and Scotland: An analysis of position statements. Journal of Medical Ethics 47 (12): e64. Buchbinder, Mara. 2021. Scripting death: Stories of assisted dying in America. Oakland: University of California Press. Compassion and Choices. 2014. People living with disabilities support the option of medical aid in dying. https://www.compassionandchoices.org/take-action/community-engagement/disabi lity-community/people-in-support. Coombs Lee, Barbara. 2014. Oregon’s experience with aid in dying: Findings from the death with dignity laboratory. Annals of the New York Academy of Sciences 1330 (1): 94–100. den Hartogh, Govert. 2012. The regulation of euthanasia: How successful is the Dutch system. In Physician-assisted death in perspective: Assessing the Dutch experience, ed. Stuart J. Youngner and Gerrit K. Kimsma, 351–391. New York: Cambridge University Press. Emanuel, Ezekiel J., Bregje D. Onwuteaka-Philipsen, John W. Urwin, and Joachim Cohen. 2016. Attitudes and practices of euthanasia and physician-assisted suicide in the United States, Canada, and Europe. Journal of the American Medical Association 316 (1): 79–90. Feinberg, Joel. 1989. Harm to self . Oxford: Oxford University Press.

240

C. A. Riddle

Finlay, Ilora, and Rob George. 2011. Legal physician-assisted suicide in Oregon and The Netherlands: Evidence concerning the impact on patients in vulnerable groups—Another perspective on Oregon’s data. Journal of Medical Ethics 37 (3): 171–174. Fitzpatrick, Kevin, and David Albert Jones. 2017. A life worth living? Disabled people and euthanasia in Belgium. In Euthanasia an assisted suicide: Lessons from Belgium, ed. David Albert Jones, Chris Gastmans, and Calum MacKellar, 133–149. New York: Cambridge University Press. Gill, Carol J. 2010. No, we don’t think our doctors are out to get us: Responding to the straw man distortions of disability rights arguments against assisted suicide. Disability and Health Journal 3 (1): 31–38. Hemerijck, A. 2002. The self-transformation of the European social model. Internationale Politik Und Gesellschaff 4: 39–66. Khavjou, Olga A., Wayne L. Anderson, Amanda A. Honeycutt, Lurel G. Bates, NaTash D. Hollis, Scott D. Grosse, and Hilda Razzaghi. 2021. State-level health care expenditures associated with disability. Public Health Reports 136 (4): 441–450. LaPiere, Richard T. 1934. Attitudes vs. actions. Social Forces 13 (2): 230–237. Lindsay, Ronald A. 2009. Oregon’s experience: Evaluating the record. The American Journal of Bioethics 9 (3): 19–27. Mackenzie, Catriona. 2014. The importance of relational autonomy and capabilities for an ethics of vulnerability. In Vulnerability: New essays in ethics and feminist philosophy, ed. Catriona Mackenzie, Wendy Rogers, and Susan Dodds, 33–59. New York: Oxford University Press. Martin, Stevie. 2021. Assisted suicide and the European convention on human rights. London: Routledge. Meade, Michelle A., Elham Mahmoudi, and Shoou-Yih Lee. 2015. The intersection of disability and healthcare disparities: A conceptual framework. Disability & Rehabilitation: An International, Multidisciplinary Journal 37 (7): 632–641. Morin, D., M. Rivard, A.G. Crocker, C.P. Boursier, and J. Caron. 2013. Public attitudes towards intellectual disability: A multidimensional perspective. Journal of Intellectual Disability Research 57 (3): 279–292. Navarro, Maria José Portillo, Gabriela Lagos Rodríguez, and Maria Leticia Meseguer Santamaría. 2021. Public expenditure on disability (PED) in Europe: An efficiency analysis. Regional Science Policy & Practice 13 (4): 1479–1495. Nelson, Lawrence J. 2003. Respect for the developmentally disabled and forgoing life-sustaining treatment. Mental Retardation and Developmental Disabilities Research Reviews 9 (1): 3–9. Onwuteaka-Philipsen, Bregje D. 2012. The unreported cases. In Physician-assisted death in perspective: Assessing the Dutch experience, ed. Stuart J. Youngner and Gerrit K. Kimsma, 123–136. New York: Cambridge University Press. Palmer, Michael. 2011. Disability and poverty: A conceptual review. Journal of Disability Policy Studies 21: 210–218. Riddle, Christopher A. 2012. Defining disability: Metaphysical not political. Medicine, Health Care and Philosophy 16 (3): 377–384. Riddle, Christopher A. 2013. The ontology of impairment: Rethinking how we define disability. In Emerging perspectives on disability studies, ed. Matthew Wappett and Katrina Arndt, 23–40. New York: Palgrave Macmillan. Riddle, Christopher A. 2014. Disability and justice: The capabilities approach in practice. Lanham: Lexington Books/Rowman & Littlefield. Riddle, Christopher A. 2016. Human rights, disability, and capabilities. New York: Palgrave Macmillan. Riddle, Christopher A. 2017a. Assisted dying & disability. Bioethics 31 (6): 484–489. Riddle, Christopher A. 2017b. Disability & doing justice. In The ethics of ability and enhancement, ed. Jessica Flanigan and Terry Price, 41–55. New York: Palgrave Macmillan. Riddle, Christopher A. 2019. Assisted dying, disability rights, and medical error. International Journal of Applied Philosophy 32 (2): 187–196.

13 Medical Aid in Dying: The Case of Disability

241

Riddle, Christopher A. 2020. Disability and disadvantage in the capabilities approach. In The Oxford handbook of philosophy and disability, ed. Adam Cureton and David Wasserman, 229–324. New York: Oxford University Press. Riddle, Christopher A. 2021a. Applying the capabilities approach to disability and education. Philosophical Inquiry in Education 28 (2): 83–94. Riddle, Christopher A. 2021b. Vulnerability, disability, and public health crises. Public Health Ethics 14 (2): 161–167. Rogers, Wendy, Catriona Mackenzie, and Susan Dodds. 2012. Why bioethics needs a concept of vulnerability. The International Journal of Feminist Approaches to Bioethics 5 (2): 11–38. Schuklenk, Udo, and Suzanne van de Vathorst. 2015. Treatment-resistant major depressive disorder and assisted dying. Journal of Medical Ethics 41 (8): 577–583. Scoccia, Danny. 2010. Physician-assisted suicide, disability, and paternalism. Social Theory and Practice 36 (3): 479–498. Silvers, Anita. 1998. Protecting the innocent from physician-assisted suicide. In Physician assisted suicide: Expanding the debate, ed. Margaret P. Battin, Rosamond Rhodes, and Anita Silvers, 133–148. New York: Routledge. Somerville, Margaret. 2001. Death talk: The case against euthanasia and physician-assisted suicide. Montreal: McGill-Queen’s University Press. Stainton, Tim. 2019. Disability, vulnerability and assisted death: Commentary on Tuffrey-Wijne, Curfs, Finlay and Hollins. BMC Medical Ethics 20 (1): 1–6. Sumner, L. W. 2018. Disability, death, and self-determination. In From disability theory to practice: Essays in honor of Jerome E. Bickenbach, ed. Christopher A. Riddle, 101–120. New York: Lexington Books/Rowman & Littlefield. Union of the Physically Impaired Against Segregation. 1976. Fundamental principles of disability. London: UPIAS. VanPuymbrouck, Laura, Carli Friedman, and Heather Feldner. 2020. Explicit and implicit disability attitudes of healthcare providers. Rehabilitation Psychology 65 (2): 101–112. Way, Jenny, Anthony L. Back, and J. Randall Curtis. 2002. Withdrawing life support and resolution of conflict with families. British Medical Journal 325 (7376): 1342–1345.

Christopher A. Riddle Ph.D., is Professor and Chair of Philosophy at Utica University, NY. In 2016, he was a visiting researcher at the Foundation Brocher in Hermance, Switzerland and in 2020, he was a visiting scholar at the Hastings Center in New York. He is the author of Disability and Justice (2014), Human Rights, Disability, and Capabilities (2016), The Ethics of Assisted Dying (2023), The Ethics of Disability (2024), and the editor of From Disability Theory to Practice (2018), and Disability and Death.

Chapter 14

Feminist Approaches to Medical Aid in Dying: Identifying a Path Forward Jennifer A. Parks

Abstract This essay addresses feminist approaches to medical aid in dying (MAID), considering whether it is a practice that should be supported for women and other marginalized groups. Some feminists have raised rights and justice-based arguments in support of MAID; others have taken a care-based approach to suggest that the practice violates relationships of care and only worsens distrust between marginalized groups and the medical establishment. I argue that we need to adopt both justice and care approaches to develop a robust feminist account of MAID. I defend a woman’s right to choose death in cases of terminal illness or suffering but argue that relational concerns of care, trust, and trustworthiness must be addressed at the same time to ensure that MAID is justly practiced and accessible to all those who want it. Keywords Medical assistance in dying · Feminism · Disability · End of life · Right to die

J. A. Parks (B) Department of Philosophy, Loyola University Chicago, Chicago, IL, USA e-mail: [email protected] © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 M. Cholbi and J. Varelius (eds.), New Directions in the Ethics of Assisted Suicide and Euthanasia, The International Library of Bioethics 103, https://doi.org/10.1007/978-3-031-25315-7_14

243

244

J. A. Parks

14.1 Introduction In 2000 I published an article in the Hastings Center Report (Parks 2000) that addressed gender and end of life choices, considering the gender implications of medical aid in dying (MAID) policies.1 In it, I questioned Susan Wolf’s argument that women’s death requests in cases of terminal illness are more likely to be fulfilled than those made by men (Wolf 1996). Wolf reasoned that gender role stereotypes and expectations inform physicians’ thinking in connection to women’s requests for medical assistance in death. She asserted that physicians are more likely to heed women’s requests to die based on a shared cultural assumption that women should be caregivers—not care recipients—and that women’s altruistic concern with not burdening their loved ones is a reasonable ground for supporting their requests for MAID. Responding to Wolf’s concerns that women are more likely to be heeded in their death requests, I argued that there are reasons to believe women would be taken less seriously. Given the studies available at that time, there was reason to think that physicians were not listening to women’s pain reports and that their self-regarding choices were not being respected. Given these gendered credibility concerns, I posited that physicians would be less likely to adhere to women’s death requests, not more likely to do so. As a result, I argued that to undercut the sexist tendency to dismiss women’s pain reports in relation to their death requests, we need to support a woman’s request to die. In the 20-plus years since this article was published the field of bioethics has continued to evolve and change. While in 2000 there was much less research done on pain bias and medical responses to women’s pain reports, there is now a more robust body of literature on these concerns. And while two decades ago bioethical concerns with race and health disparities were starting to be raised, we now see a more concerted effort to address them, especially following the Covid-19 pandemic. 1

I will adopt the term medical aid in dying (MAID) as used by the group Compassion and Choices to refer to the practice in which terminally ill patients voluntarily take life-ending medications, thus accessing the means to end their own lives. By using this terminology, Compassion and Choices distinguishes medical aid in dying from euthanasia. On their website they note that “While both practices are designed to bring about a peaceful death, the distinction between the two comes down to who administers the means to that peaceful death. Euthanasia is an intentional act by which another person (not the dying person) administers the medication. By contrast, medical aid in dying requires the patient to be able to take the medication themselves and therefore always remain in control. Euthanasia is illegal throughout the United States” (https://compassionandchoices.org/res ource/assisted-suicide/). This terminology is like Canada’s, which refers to medical assistance in dying (MAiD). Yet in Canada, MAiD includes the act by a physician or nurse practitioner of directly administering a substance that causes death, such as an injection of a drug. For the purposes of my argument, the person who applies the means of death is not a central concern, since my focus is on the grounds for supporting medical aid in dying for women, and not who fulfills it. In this paper I will adopt the acronym MAID to refer generally to the act of aiding or assisting terminally ill or suffering patients in achieving their death. My paper will limit itself to the current laws in the U.S. and Canada. However, my comments concerning MAID address general feminist concerns relating to the practice and its impact on women globally.

14 Feminist Approaches to Medical Aid in Dying: Identifying a Path Forward

245

In what follows, I will consider how these developments might impact my argument in support of adhering to women’s death requests. Feminist bioethicists that support MAID have had to reckon with charges that their accounts often reflect the concerns of white, middle-class women and that feminist theory has failed Black and other minority women and persons with disabilities.2 For example, by advancing arguments in favor of MAID white feminists might be emphasizing the concerns of the privileged while overlooking the harms that such laws pose to those who are less well-endowed. And as Pullman (2020) has recently noted, feminist responses to MAID also tend to reflect a preoccupation with individual autonomy and rights. He claims that feminists who used to emphasize relationships of care and a “different voice” (a la Carol Gilligan) have adopted “the familiar voice of personal autonomy, individual rights and justice” (12). I take these concerns seriously and pursue their implications in the rest of this essay. I acknowledge that feminist support for MAID must take account of objections that come from marginalized bioethical perspectives. However, I argue that this doesn’t lead to the conclusion that feminists should not support this practice for women, or that it is dangerous for women to pursue it. On the contrary, as I will discuss, the studies done on pain bias and discriminatory attitudes toward women in health care over the past couple of decades provide further support for the concern I raised in 2000 about the rejection of women’s death requests. Paternalism is still alive and well in medicine, and it continues to affect the care and treatment of female patients. This article will consider a way forward as feminists reckon with two seemingly opposing viewpoints: the view that women’s wishes about MAID should be heeded and followed, and the view that feminists should reject such requests out of concern for the implications for marginalized patients.

14.2 What Is a Feminist to Do? Feminists take these contextual questions seriously regarding women and other marginalized groups in health care. As Susan Wolf notes in her 1996 article, feminists “who feel in control of their lives, [and] are less subject to subordination by age or race or wealth,” may pursue MAID as “yet another option to add to their many [options]” (Wolf 1996, 308). Laws echoes Wolf’s concern by claiming that “At a time when the U.S population is aging and browning, and the healthcare system is in a heightened state of flux, it is relevant to ask: Who benefits from the political pursuit

2

The article in this collection by Christopher Riddle specifically addresses concerns about disability in medical aid in dying. But as feminists like Kimberle Crenshaw have long observed, individuals have complex and multiple identity points such that they may experience discrimination based on more than just one identity (Crenshaw 1989). My feminist analysis in this paper will address concerns with race, disability, and other marginalized identities because they impact women’s use of (and perspective on) MAID in important ways.

246

J. A. Parks

of physician-assisted suicide legalization?” (Laws 2019).3 These charges of unequal harms and benefits based on a woman’s wealth, race, or age status are difficult to dismiss: feminist bioethicists have been concerned about the impact that these various identity points have upon a woman’s care and treatment. But at the same time, feminists must consider the implications of refusing MAID to women who are unduly experiencing pain and suffering in the final months of terminal illness or who are not terminally ill but face of lifetime of unremitting pain caused by untreatable chronic illness. It seems that if we take seriously the concerns raised by Wolf and Laws, we risk ignoring the legitimate appeals to MAID that some female patients are making, and we also risk reverting back to the paternalistic treatment of female patients. Rather than seeing feminists who support MAID and those who reject it as diametrically opposed, we can see them as being motivated by the same concerns regarding the historical discrimination that has influenced the paternalistic treatment of women, people with disabilities, and minorities. Feminists who support MAID for rights and justice-based reasons must at the same time recognize the health disparities and other care-related issues that still plague women and other marginalized patients. Addressing these concerns would help to ensure that requests for MAID from patients in marginalized situations are not being influenced by lack of decent medical and palliative care. If it is true, as some feminist, race, and disability theorists have asserted, that requests to die may come from a place of loneliness, isolation, depression, financial want, power inequality, lack of access to medical care, and gender role socialization (to name a few),4 then feminist bioethicists must recommit to the goals of women’s gender equality and the reduction of racialized health disparities and disability discrimination that plague the U.S. and other countries.5 Conversely, those who reject MAID should consider the implications of denying women access to it now because they believe the current unjust features of societies may lead women to seek MAID for reasons they presumably would not have if those societies were more just. It doesn’t seem to follow, though, that women should be barred from legal access to MAID in present day circumstances (and/or that women must await more just circumstances in order to have legal access). The feminist opposition to MAID supposes that what we should be permitted to do in non-ideal circumstances 3

Note that Laws’s terminology is inaccurate in the U.S. context, since as the Compassion and Choices website notes, “Medical aid-in-dying laws on the books in California, Colorado, the District of Columbia, Hawai’i, Maine, Oregon, Vermont and Washington expressly state: ‘Actions taken in accordance with [the Act] shall not, for any purpose, constitute suicide, assisted suicide, mercy killing or homicide.’ And in Montana, where assisted suicide is specifically illegal, the Montana Supreme Court ruled in Baxter v. Montana that medical aid in dying provided to terminally ill, mentally competent adult patients in no way violates established state law [including Montana’s assisted suicide statute] or the principles of public policy.” (https://compassionandchoices.org/res ource/assisted-suicide/). 4 Note that even if these factors are in play to some degree when a patient invokes MAID, I do not think that should automatically eliminate MAID as an option if the patient is seriously ill and experiencing unremitting suffering. This will be addressed in more detail later in this paper. 5 Indeed, feminist bioethicists are concerned about these issues and much feminist literature addresses issues of access to health care for marginalized patients. See Sherwin (1992), Wendell (1996), and Rogers (2006) as examples of this literature.

14 Feminist Approaches to Medical Aid in Dying: Identifying a Path Forward

247

should be guided by the reasons we would have in ideal circumstances—but this is unconvincing given that no social practices or policies are implemented under ideal circumstances. The prospect of allowing women (and others) to suffer unduly by refusing MAID on paternalistic grounds hardly seems like an appropriate response while we are working toward that ideal future.

14.3 Women and MAID: What Is at Stake? Feminist critics of MAID have shared several concerns that lead to questions regarding the advisability of laws and policies that support it. These include the fact that women’s greater longevity means they are more likely to develop diseases and disabling conditions that could motivate the desire for assisted death; and that, given this greater longevity, women are more likely to suffer the death of a partner or spouse, resulting in them living in isolation and thus experiencing loneliness.6 Furthermore, as critics argue, women tend to have fewer economic resources when they are older, and economic disadvantage limits women’s ability to access care or pursue care options, making it more likely that they will request death as an “out.” Besides this, as Wolf has argued, women may be more self-sacrificial than men: cultural expectations that they be caregivers, not care recipients, may make them more likely to request death if they believe they will be a burden on their families (1996). Finally, another important objection is the concerns raised around the language of autonomy and choice that surrounds MAID, which tends to reflect the priorities and concerns of those who value—and enjoy—control over their lives (Beaudry 2020; Wolf 1996; Pullman 2020). Recently there has also been more discussion of the impact that epistemic injustice (testimonial and hermeneutical injustice), implicit bias, racism, ageism, ableism, pain bias and other wrongs can have on the clinical care of queer, female, disabled, and minority patients. These concerns are related to the objections raised against MAID, so I consider them briefly below.

6

This issue for elderly women of surviving their partners and living in isolation and loneliness was especially acute during the Covid-19 pandemic. The need to shut down nursing homes in order to contain the spread of the virus had seriously damaging physical, emotional, and psychological effects for the nursing home resident population. See Parks and Howard (2021) for more on this.

248

J. A. Parks

14.4 Women and Pain Bias It is fair to say that women as a group experience the effects of pain bias, a phenomenon that has been widely studied over the past couple of decades. Recent studies demonstrate that women’s pain claims are not taken as seriously, and that women’s pain is not treated to the same extent as men’s (Northwell Health 2021; Zhang et al. 2021; Samulowitz et al. 2018; Check 2010). But, even worse, when factors beyond gender are at work, such as being a racial minority, impoverished, or suffering from the cumulative effects of substandard health care and lack of insurance, then the bias only increases. A 2019 Washington Post article raises the question “Is bias keeping female, minority patients from getting proper care for their pain?” to which the answer seems to be yes (Consumer Reports 2019). When Black women’s pain claims are encountered, health care providers can be even more dismissive of them. Recent narratives by Black women have highlighted their experiences; one example is Tressie McMillan Cottom’s account of her acute pain during labor and delivery. In response to her report that she was experiencing excruciating pain the attending staff attributed it to her being obese, constipated, and incompetent in her reporting. Her care was significantly delayed and, as a result, her baby died. As McMillan Cottom writes, “When the medical profession systematically denies the existence of black women’s pain, underdiagnoses our pain, refuses to alleviate or treat our pain, healthcare marks us as incompetent bureaucratic subjects. Then it serves us accordingly” (2019, 86). Commentators have noted that even the most famous and wealthy Black women’s self-reports of pain and/or medical urgency are ignored, as demonstrated by tennis star Serena Williams’s experiences during her labor and delivery. When Williams reported her symptoms and her history of pulmonary embolism, her self-report was all but ignored until a CT scan she demanded was finally provided and they detected several small blood clots that had settled in her lungs. Even though she had requested a heparin drip much earlier, it was only instituted after the delayed CT scan (Lockhart 2018).7 The delayed medical care and medical errors that result from bias against women’s pain reports place women at an obvious unfair disadvantage when accessing health care. When considering how this relates to women’s reports of pain in terminal illness, 7

Studies have determined that in the clinical setting Black patients experience biases against their pain claims to a much higher degree than white patients. See Hoffman et al. (2016), Meghani et al. (2012), Mende-Siedlecki et al. (2019). Since Black women’s experiences intersect race and gender, commentators raise concerns about the way they may encounter unique forms of pain bias based on the combination of their race and gender. See Bassett (2021), Harrigan-Farrelly (2022), Omeish and Kiernan (2020). More studies are needed to address the intersections of race and gender where pain bias is concerned. Research has established that pain bias affects the care of women and Black patients, but little has been done to address how the intersecting identities of being Black and female may worsen the biases they experience.

14 Feminist Approaches to Medical Aid in Dying: Identifying a Path Forward

249

it is not difficult to see how sexism might intrude upon the medical establishment’s uptake of a woman’s death request if her claims to pain are simply not believed.8 These recent pain studies uphold my argument in my 2000 article that women are not more likely to be recommended (or to receive) MAID services than men.9

14.5 Epistemic Injustice One of the factors attributable to the pain bias against women relates to epistemic injustice: the way that women’s knowledge claims in healthcare (and beyond) go unheeded. Miranda Fricker’s work on epistemic injustice has resulted in a burgeoning body of literature on the way that it shapes the view and treatment of marginalized persons. As Fricker notes, one way in which epistemic injustice manifests itself is in terms of testimonial injustice, where a speaker’s word is not given any credence, or their testimony is simply not believed, based on a credibility deficit10 (Fricker 2007, 25). Prejudice and gender role stereotypes surrounding women’s ability to know—and to accurately articulate their knowledge—has impacted the way the medical establishment responds to their claims. Testimonial injustice is evident in the concerns with pain bias noted above, since women experience a credibility deficit in relation to their reports of pain. Relatedly, women may experience another manifestation of epistemic injustice: hermeneutical injustice, where they may lack the language to describe what they are experiencing, and the concepts to make it intelligible (to themselves and to others) because we still do not understand women’s bodies well or how they may differently manifest symptoms. To take just one medical example of hermeneutical injustice, consider patients’ experiences with chronic fatigue syndrome: for a long time before this condition was given a name, 8

In researching this paper, I could find no studies that address the impact that pain bias (or other factors, such as loneliness or greater longevity) might have on women’s access to or requests for MAID. The data on women and pain bias alone does not support the claim that women will be more likely to be assisted in dying. Since women experience a credibility deficit where pain is concerned, there is no reason to think their reports are more believable—and more likely to be given uptake— than men’s. One might still think that women’s un-treated or under-treated pain might drive them to request MAID. But “Multiple studies have shown that most patients who seek MAID do so not because of unrelieved symptoms, such as chronic pain, but because of more existential suffering, such as loss of autonomy and an inability to enjoy life” (Weiss 2018, 249). 9 Indeed, 2021 data reported from the Oregon Death With Dignity Act indicates that women are not over-represented in the number of patients who have died from ingesting a lethal dose of medication. On the contrary, in 2021 out of 238 cases of MAID, 133 (55.9%) were men and 105 (44.1%) were women. And in 2020, out of 259 cases 131 (50.6%) were men and 128 (49.4%) were women. In total, since the Death With Dignity Act began collecting data, 53% of deaths have been men and 47% have been women (Oregon Health Authority Public Health Division 2021). 10 Note that “testimony” in this case does not refer to an individual giving evidence in court; rather, it refers to the general practice of speaking, reporting and offering observations. In the case of women, I am arguing that testimonial injustice is committed because they are denied believability based partly on gender norms that depict women as flighty, unreliable, hypersensitive and demanding.

250

J. A. Parks

sufferers who reported their symptoms were referred for psychiatric evaluation, diagnosed as depressed, or told it was “only in their heads.”11 As Drew Foster notes, “For a long time, it was pretty standard in the medical community to repudiate and dismiss chronic fatigue syndrome. [As a Stanford Medicine brief notes] ‘it’s not uncommon for CFS patients to face several mischaracterizations of their condition, or even suspicions of hypochondria, before receiving a diagnosis of CFS’” (Foster 2014). Such responses stem from, as Fricker puts it, “the injustice of having some significant area of one’s social experience obscured from collective understanding owing to a structural identity prejudice in the collective hermeneutical discourse” (Fricker 2007, 155). Only after chronic fatigue syndrome was named and assigned a set of symptoms were sufferer’s self-reports given any credence. Indeed, some patients report relief, not sadness, at receiving their diagnoses of this condition, since the diagnosis allows them to give a name to their health issue and requires doctors to take their self-reports seriously (Fennell et al. 2021). Women’s lack of credibility within the medical realm feeds into the way that their pain reports are assessed—which then leads to the experience of pain bias that I have presented above. An injustice is committed against women when their pain reports are undermined, ignored, or denied and they must suffer in silence—whether in the context of giving birth, experiencing the symptoms of chronic fatigue syndrome, or in dying. For my purposes in this paper, I simply want to note these gender-related concerns that may feed into the feminist debate surrounding medical aid in dying. Women should not be recommended or denied MAID based on a factor like pain bias; and this bias against women must be addressed to ensure they receive proper and robust clinical care.

14.6 Feminist Social Justice Concerns Some recent articles on MAID present concerns with the hyper-emphasis on rights and individual autonomous choice that tends to support arguments in favor of it. According to these critics (Buchbinder 2018; Sikka 2021), we ought to shift the grounds for the debate about medical aid in dying from a focus on formal rights and the negative liberty to choose without government interference to a concern with social justice. For example, Buchbinder addresses the literature on the “right to die,” claiming that A central premise of these movements is that patients should have the right to seek AID [aidin-dying] to control the time and circumstances of their death when death is imminent because of terminal illness…they have been persuasive because they mirror dominant bioethical perspectives that privilege patient autonomy above other competing ethical principles. (755)

11

While some men suffer from chronic fatigue syndrome, the large proportion of CFS patients are women. Studies indicate that women are 3–4 times more likely than men to have the condition (US Institute of Medicine 2015; Faro et al. 2016) and that “patients do not know how to describe their condition to themselves or others” (Fennell et al. 2021).

14 Feminist Approaches to Medical Aid in Dying: Identifying a Path Forward

251

Buchbinder argues for a social justice approach that stems from the reproductive justice movement. The concern is to look beyond “a right to choose” to consider the social factors and health inequalities that impact an individual’s choice.12 Similarly, Tina Sikka rejects autonomy arguments concerning MAID to instead focus on materialist feminist theory and critical feminist approaches. In so doing, Sikka addresses the structural impediments, implicit biases, and overlapping marginalizations that inform the dying experience for patients experiencing intersecting or interlocking forms of identity-based oppressions (5). Both authors are concerned with the way that MAID policies fail to ensure that minorities and other marginalized populations have access to assistance in dying. For example, as Buchbinder notes, “terminally ill people can face a range of barriers to accessing AID [assistance-in-dying] in permissive jurisdictions” (754); and Sikka claims that “there are a considerable number of barriers marginalized groups face with respect to accessing medically assisted dying” (21). Within the U.S. context, the best statistics available regarding the use of MAID come from Oregon, where the Death With Dignity (DWD) Act has been in place since 1997. If one wants to investigate the degree to which minorities utilize it, this data is the most indicative, since the state of Oregon has been collecting it for decades. The most recent 2021 data indicates that of the 238 users of the act, 226 (95%) were white, 0 were Black, American Indian, or Pacific Islander, and 6 (2.5%) were Hispanic. 133 were male and 105 were female; and 71.9% of users were well educated, with a bachelor’s degree at minimum (Oregon Health Authority Public Health Division 2021, 10). This data raises a host of important questions, including whether these disparities show that Blacks and other minorities lack access to MAID, or whether they simply reject it.13 As noted above, Buchbinder and Sikka read these statistics as suggesting that minorities face barriers to accessing MAID, raising social justice questions about equal access to medical care. Buchbinder claims that the safeguards put in place in Vermont to ensure the voluntary participation of both patient and provider results in creating undue burdens to accessing MAID for minority and economically disadvantaged groups in that state. Sikka similarly notes that “regulatory and legal barriers can also pose difficulties for groups who may not possess the cultural capital, communication skills, or legal knowledge to perform their illness in ways that are acceptable within the context of the program…and launch an appeal if refused” (13). 12

Of course, addressing justice in health care requires ensuring that everyone has a substantive right to certain goods—like decent health care, access to palliative and hospice care, access to medical aid in dying, etc. Buchbinder’s account recognizes the connection between justice and rights, but she adopts an approach that rejects a “thin” formal approach to rights. Thus, the title of her article (“Access to Aid-in-Dying in the United States: Shifting the Debate from Rights to Justice”) is somewhat misleading because she is concerned with rights, and she doesn’t separate rights from justice; she just demands a more substantive view of rights in the context of MAID. 13 The problem is that there are a variety of reasons that could explain why few minorities are taking advantage of MAID (where legal) in the U.S. Minority groups might reject MAID for religious or cultural reasons, for reasons of distrust, or because they are not properly informed of the option. More research is required to determine which of these factors is impacting the use of MAID by minority groups to determine if there are equal access issues.

252

J. A. Parks

Sikka and Buchbinder may be right in their assessment that the very low numbers of minorities accessing MAID could indicate that marginalized patient populations lack fair access to it. The statistics from Oregon would at least suggest that this might be the case. And given that these citizens are supposed to have equal access to MAID services in the states in which it is legal, the lack of use by minorities might indicate a problem. Yet critics who reject this interpretation note that minorities and people with disabilities distrust the medical establishment and that they may not practice MAID because (in the case of African Americans and LatinX patients) it may run against their religious or cultural beliefs.14 Furthermore, some critics worry that a focus on medical aid in dying detracts from the need to address access to health care that would help to undercut the health disparities experienced by marginalized populations. As such, I worry that focusing at this time on MAID access for minority, disabled, and other marginalized populations might not be taken by these communities as a social justice initiative, but rather as evidence that our medical institutions are attempting to eliminate them. Rather than upholding justice, emphasizing access to MAID for minority and marginalized populations might be viewed as violating respect and equal protection for populations that view themselves as particularly susceptible to elimination. As evidence for this concern, consider the disability rights activist group Not Dead Yet, whose platform is focused on preventing the passage of medical aid in dying laws that are being proposed in various parts of the United States. As stated on their website: In judging that an assisted suicide request is rational, essentially, doctors are concluding that a person’s physical disabilities and dependence on others for everyday needs are sufficient grounds to treat them completely differently than they would treat a physically able-bodied suicidal person. There’s an established body of research demonstrating that physicians underrate the quality of life of people with disabilities compared with our own assessments. Nevertheless, the physician’s ability to render these judgments accurately remains unquestioned. Steps that could address the person’s concerns, such as home care services to relieve feelings of burdening family, are not explored. In this flawed world view, suicide prevention is irrelevant. (https://notdeadyet.org/disability-rights-toolkit-for-advocacy-against-legalizat ion-of-assisted-suicide)

The way that Buchbinder and Sikka characterize the problem with MAID may not resonate with the concerns of a group like Not Dead Yet, which is not bemoaning their lack of access to medical-aid-in-dying, but rather the very existence (and spread) of the practice. The problem Not Dead Yet identifies is with the ableist attitudes, beliefs, and values that they see supporting the practice of MAID. Their goal is not to ensure that people with disabilities are equally able to access it, but rather to fight the practice entirely in order to protect themselves against the right to die.15 14

Indeed, Buchbinder acknowledges this concern and notes that “Some racial and ethnic minority groups, particularly African Americans, harbor deep mistrust of hospice and palliative care and prefer aggressive end-of-life treatment. Among patients holding such views, access to AID will not be desirable, and different patterns of access will not raise justice concerns” (2000, 757). 15 Not Dead Yet (NDY) represents a disability rights perspective on MAID, but not the perspective on this issue. Disability theorists who reject the claims of NDY claim that denying people with

14 Feminist Approaches to Medical Aid in Dying: Identifying a Path Forward

253

Some Black bioethicists similarly address the lack of value placed on Black lives in medicine, raising concerns about how it may filter into discussions of Black patients’ “right to die.” As Laws claims, There is danger in continued distrust of the medical establishment, a distrust which will not improve when the same hospital-based physicians who are responsible for giving medication to help patients die–as a treatment service–are also responsible for informing other patients that there is nothing more that can be done for them. (2019)

Arguing for better access to MAID in the U.S. for marginalized and minority groups will, I argue, not likely be hailed as a social justice initiative by them: rather, it will likely be viewed suspiciously, as an attempt to encourage those with “costly” and/or “unwanted” lives to end them rather than investing in the means to accommodate their illness or their dying process. This fear has been realized by many in minority and disability communities during the Covid pandemic, where reports indicate that Black patients were becoming ill and dying at much higher rates than white patients. As the Kaiser Family Foundation (2021) notes, “Age-standardized data show that Hispanic, Black, and AIAN [American Indian and Alaska Native] people are at least twice as likely to die from COVID-19 as their White counterparts and that Hispanic and AIAN people are at nearly two times greater risk of COVID-19 infection than White people.”16 In addition, triage protocols that were put in place were seen by some minorities and disability activists as threats given that some protocols resulted in refusal of access to ventilators and ICU beds for those who were the worst off: primarily, minorities and people with disabilities. Some critics thus see the push for MAID as one piece of a bigger problem in which the medical establishment denies care to patients from these communities yet advocates for the advancement of laws that support their medically assisted death.

14.7 Feminist Disagreements Regarding MAID: The Canadian Context To provide context for this tension surrounding feminist responses to MAID, consider Canada’s controversy surrounding its recent extension of MAID laws. In June 2016 Canada’s Parliament passed federal legislation that allows eligible Canadian adults to request medical assistance in dying. This law was implemented after many years disabilities the right to make end of life decisions is unjustified paternalism that treats them in exactly the infantilizing ways to which NDY objects. See, for example, Silvers and Francis (2016) and Riddle (2017). Downie and Schuklenk (2021) also question whether vocal groups like NDY can speak for all persons with disabilities, given that two of the cases that prompted the revised Canadian law were patients with disabilities who were petitioning for MAID access using equal protection concerns. 16 See Kaiser Family Foundation report (2021): https://www.kff.org/racial-equity-and-health-pol icy/issue-brief/covid-19-cases-and-deaths-by-race-ethnicity-current-data-and-changes-over-time/ [Accessed December 3, 2021].

254

J. A. Parks

of resistance to the practice. On October 5, 2020, the Minister of Justice and Attorney General of Canada introduced Bill C-7 in Parliament: an Act which proposed changes to Canada’s 2016 law on medical assistance in dying. The Canadian Parliament passed this revised legislation on March 17, 2021, making important changes to who may be eligible to obtain medical assistance in dying and to the process of assessment, with changes taking effect immediately (Bill C-7 2021). The extension of the law has been met with vigorous support, disagreement and debate, including widely differing views expressed by feminist and other critical bioethicists. Bill C-7 removes the stipulation that Canadians must have a reasonably foreseeable death in order to be eligible for medical assistance in dying. In so doing, patients who are intolerably suffering, but whose death is not imminent, have been extended the right to seek MAID. Additionally, those patients with a reasonably foreseeable death who have been assessed and approved for MAID, but who may lose their capacity to consent prior to enacting it, have the option of a waiver of final consent. This extension of the 2016 law has thus been controversial in that it allows patients suffering from progressive dementia to utilize it. There is currently a two-year mental illness exclusion such that patients who are solely suffering from severe mental illness cannot access MAID; after this exclusionary period, individuals in this circumstance will be extended access to it. During this period an expert panel has been convened to advise on the safeguards and protocols that should be put in place to protect people with mental illnesses from exploitation or harm. According to Downie (2020), examining Canada’s MAID law through the lens of feminist bioethics indicates the degree to which “MAID in Canada is infused with feminist philosophy” (25). Downie highlights past calls by feminists “to have input into the formation of MAID policies and precedents to ensure the exploration of issues around MAID from an explicitly feminist perspective” (22). She notes the degree to which feminists had a place at the public policy table in the wave of the most recent law reform, naming the key academic, legal, and political figures that struck down Canada’s prohibition on MAID. In addition to approving the participation of feminists in constructing current MAID policies, Downie praises the way that reform discussions attended to multiple forms of oppression. As she observes, …when the legislation was being debated in Ottawa, three [female] senators…spoke out passionately and powerfully about the potential implications for poor immigrant women, the elderly, and persons with disabilities – all speaking against: a) barriers to access that would particularly harm marginalized groups (e.g. the insurmountable costs if one had to go to court to gain judicial preauthorization for MAID); and b) stigmatization and discrimination implicit in some proposed restrictions on access (e.g. assumptions that persons with disability don’t have decision-making capacity and need to be protected from themselves). (2020, 23)

Downie also emphasizes the focus on voluntariness that has been part of the public policy debates, noting feminist concerns about the “particular vulnerability women face with respect to voluntariness” that was evident in the debates about the federal MAID legislation. In these discussions concerns were raised both against MAID— addressing pressures for disabled individuals to access MAID (given concerns that they will internalize messages to not be a burden or messages that their lives are not worth living)—and in support of MAID, with arguments that it “reduces suffering;

14 Feminist Approaches to Medical Aid in Dying: Identifying a Path Forward

255

it respects relational autonomy; it sends positive messages about the capacity of women, elderly and persons with disabilities …” (23–24). Downie thus addresses the ways in which feminists have informed and directed the current MAID law in Canada, allowing for the inclusion of an important alternate approach. In his article “In a Familiar Voice: The Dominant Role of women in Shaping Canadian Policy on Medical Assistance in Dying,” Pullman agrees with Downie that feminists have “played a central and even dominant role…in advancing this cause” (2020, 12). Yet Pullman criticizes feminists for failing to adopt the “different voice” that Gilligan identifies in her groundbreaking book on an ethic of care. As he notes, Gilligan’s In a Different Voice identifies a female-oriented moral approach and attitude that begins with an interdependent, connected self who places relationships and responsibilities of care at the forefront.17 And while he acknowledges the important role that Downie has played in Canada’s acceptance of medical assistance in dying, he depicts her as failing to uphold a feminist care ethics approach: Legal scholar Jocelyn Downie has been amongst the most vocal and prolific advocates [of medical assistance in dying]. Again, however, she champions the familiar voice of autonomy rights and fundamental justice in forwarding her cause. “I believe the case for a permissive regime with respect to voluntary assisted death hangs largely on the principle of respect for autonomy,” she writes in Dying Justice, her 2004 book setting forth a case for decriminalizing euthanasia and assisted suicide in Canada. Although Downie has written elsewhere in favour of a broader notion of ‘relational autonomy’ in interpreting and applying health law, and even cites Gilligan in support of this perspective, that broader notion and the voice with which it is ostensibly spoken is seldom heard in her substantive discussions of MAID. (2020, 15)

Pullman’s critique extends to other feminist judges, lawyers, doctors, and scholars who have weighed in on MAID in Canada. He points out the degree to which concerns for the principle of autonomy in medicine (and its related concern with individual rights and justice-based reasoning) is prevalent in the concerns that they voice. And, as he further notes, it is the “powerful and articulate” women in Canada who have silenced that “different voice” (18). Pullman harkens back to Gilligan’s care voice in advancing his concerns with the extension of MAID in Canada. While he acknowledges that feminists like Downie have rejected the liberal individualistic notion of the self in their writings, he observes that these views have not made their way into their public policy and legal discussions:

17

Pullman’s account of care ethics draws on Gilligan’s empirical claims about women’s different moral sensibilities to argue that pro-MAID feminists are failing to speak in that “different” voice. Yet Gilligan’s empirical claims regarding women’s moral orientation toward care and the care perspective have been roundly criticized by some scholars who see it as essentializing and reductive of women’s perspectives. See Pollit (1992) and Senchuk (1990) as examples of such criticisms. By contrast, when I appeal to care ethics in this paper I will be referring to it as a feminist stance that challenges rights-based approaches to MAID, one that emphasizes the relational nature of autonomy and the social and political value of care and caregiving relationships.

256

J. A. Parks

….it is one thing to recognize a relational conception of the self: it is quite another to operationalize it in the context of a legal and social discourse that is continually mediated through the Charter of [Individual] Rights and Freedoms. As the foregoing discussion has amply illustrated, the liberal individualistic self and the familiar voice with which it speaks continually overwhelms the different voice of care, even for some feminists who ostensibly recognize and value that latter perspective. (2020, 18)

Pullman emphasizes a care perspective according to which we recognize and honor the care needs of our ill, dying, vulnerable citizens. He notes that defenders of MAID in Canada ignore evidence that many Canadians lack access to adequate palliation; instead, they emphasize that it is abandoning patients to not fulfill a patient’s request for MAID. But, as Pullman notes, “Advocates for greater access to palliative care might take the opposite view, arguing that offering MAID as opposed to adequate palliation, especially when resources for palliation are often woefully inadequate, is the ultimate act of abandonment” (2020, 16–17). He ends his inquiry from a care ethics perspective as follows: While we do not want the interests of the state routinely to supersede those of individual citizens, we nevertheless need to make room in our social and political discourse for that different voice of care that recognizes our interdependence and values our mutual responsibilities to one another. That voice of care insists there is no shame in either asking for or receiving care. Failing to harken to that voice may leave Canadian bioethicists asking the same question Daniel Callahan asked of American bioethics some 25 years ago, namely ‘can the moral commons survive autonomy?’ (2020, 19)

Pullman’s objections are themselves problematic. When dealing with important legal and public policy issues like those raised by MAID, one must use the language of rights since laws are constructed around individual rights. His critique of Downie as being a feminist “sellout” because of her emphasis on justice and rights is thus misdirected, since there may be no way around this bias when one engages public and legal policy debates. However, as I will note in the next section, Pullman’s point about care ethics is worth pursuing given that people cannot access their rights under conditions of distrust. This relational ethic addresses the problem of distrust—a relational problem if there ever was one—which can lead minorities, women, and people with disabilities to perceive rights not as guaranteed goods but as things that can be bad for them.

14.8 Feminist Social Justice Versus Care Ethics: Irreconcilable Differences? Is there an irresolvable tension between the feminist social justice accounts of medical aid in dying that are offered by scholars like Sikka, Buchbinder, and Downie and the feminist care ethic account presented by Pullman? At first glance it might seem so, given Pullman’s indictment of the way in which concerns about justice and rights

14 Feminist Approaches to Medical Aid in Dying: Identifying a Path Forward

257

represent the “familiar voice” of autonomy in bioethics. If the issue of MAID is limited to this individualist autonomy-based approach, then the sorts of relational, care-based concerns that arise from care ethics might just disappear. Yet an unqualified appeal to care ethics can result in the importation of very gendered notions of care and caregiving that may not serve feminists well. As a number of feminist critics of care ethics have observed, this moral framework can result in appeals to traditional gender role stereotypes of women as “sharer and carer,” and as “guardian of all the small rituals that knit together a family and community” (Pollit 1992), placing responsibility on women for addressing caregiving needs in society. While Pullman’s concerns are not without merit, he appeals to care ethics as though it unproblematically reflects an approach that feminists ought to endorse. Yet Gilligan herself addressed the need to adopt both a justice perspective and a care perspective so that they are equally valued and balanced (Gilligan 1982). Overemphasizing a care perspective might also mean that feminists lose sight of the important justice-based concerns that Downie and others raise.18 I suggest that rights and justice-based concerns are not incompatible with care-based concerns because the two address very different aspects of the debate surrounding MAID. Feminists need a rights and justice-based approach if we are to address equal access to health care, a patient’s right to proper palliation, or any other inequality experienced by women, minorities, and people with disabilities. Indeed, I cannot imagine addressing policies like MAID without a rights-based approach, since individual rights provide the rationale for the laws and policies that support the practice. By contrast, we need care ethics to address the relational problems that plague health care and create distrust around MAID. As noted previously, the sense of distrust that has been fostered by a medical history of discrimination and abuse must be addressed, and that is the purview of care ethics. Such a feminist moral framework is necessary for those who experience this lack of trust if we are to address their belief that MAID may be used to deny them care and end their lives. This care ethics framework also places appropriate responsibility on the medical establishment to correct for the past treatment that has led to these legitimate concerns. It is a call to the medical community to be sensitive, to address and take responsibility for implicit biases and to do what it takes to build trust and to be a community that is viewed as being trustworthy. Thus, Pullman’s critique from a care ethics approach is legitimate, but it is formulated wrongly: he cannot fault policymakers for formulating MAID in the language of autonomy and rights, since that is what laws are based on; but he can apply his care ethical approach to address the significant relational concerns that must simultaneously be addressed. The rights and care approaches can also be seen as compatible with each other in that the former focuses more on the current world in which the aims of the care approach have not (yet) been achieved while the latter emphasizes measures to earn and build trust from marginalized patient populations. On this compatibilist view, we should work to adequately meet patients’ needs for care and to address the relational problems plaguing health care, but not deny access to MAID while that work 18

For excellent accounts of care in the social/political context, see Kittay (1999) and Tronto (2013).

258

J. A. Parks

is ongoing. For, given that the relational problems and lack of adequate care are not swiftly remediable, denying access to MAID could make things even worse for patients who are already faring badly. This concern has been addressed in the Canadian context, where courts have encountered the argument that no citizen should have access to MAID until palliative care services are available to all as needed in a timely fashion. In response to this claim, a judge responded: the argument that legalization should not be contemplated until palliative care is fully supported rests…on a form of hostage-taking. In other words, this argument suggests, the suffering of grievously-ill individuals who wish to die will serve as leverage for improving the provision of adequate palliative care. (Downie and Schuklenk 2021, 667)

MAID services should not be entirely denied to patients out of concern for the social justice concerns cited by critics: we must avoid paternalistically taking hostage those patients who are dying and suffering. As societies, we can address these social justice and health inequalities concerns while allowing MAID to take place within the limits of the law.

14.9 Returning to Women’s Requests for MAID Notwithstanding the concerns raised against MAID policies noted in this essay, I argue that we should heed those women who claim to be experiencing pain and suffering at the end of life. Given the factor of pain bias as previously addressed, and the general tendency to question women’s rational capacity, we do them no favors in rejecting their MAID requests out of a concern for protecting their interests. Such paternalistic interventions are not justified when they may only make women (and other marginalized patients) worse off. As noted earlier in this paper, concerns with epistemic injustices that might undercut the knowledge claims of marginalized patients regarding their own pain and suffering must also be addressed by these justice and rights-based concerns. Appropriate clinical care for sick and dying women should include everything from acute medical interventions to appropriate palliation and admission into hospice care; but sometimes these measures will simply be insufficient. For those women who persist in the request to die, it is irresponsible to not take them seriously and to not extend to them the choice to end their own lives. Thus, for justice and rightsbased reasons (as noted by Downie and others) I support the Canadian government’s decision to legalize MAID in Canada. Yet the pandemic experience of Covid has laid bare the impact that a lifetime of substandard or nonexistent health care has upon a patient’s ability to survive. The health inequalities experienced by minority communities leading into the pandemic—high rates of comorbidities such as diabetes, heart disease, high blood pressure, and other chronic untreated conditions—demonstrate the need to address health care for minorities, people with disabilities, and the poor, if we are ever to

14 Feminist Approaches to Medical Aid in Dying: Identifying a Path Forward

259

achieve a semblance of health equity. Projects and programs to address these inequalities need to be prioritized, and feminist bioethicists should take the lead in advocating for changes so that the health inequalities that lead to unnecessary illness and death can be addressed.19 Doing this would also begin to address the distrust that minority patients are reported to express with regard to their healthcare teams and would avoid the perceived hyper-emphasis on MAID that commentators like Laws have criticized.

14.10 Conclusion And so, I end where I began, by returning to the article I wrote in 2000 that defended respect for women’s death requests. I reassert my concern as noted in that article, that “contingencies such as poverty, poor education, age, and race are important to a feminist account” of MAID and that “a person’s access…should not be affected by conditions over which she has no control: it is alarming to think that in a sexist society, only some members’ claim to pain and psychic suffering will be taken seriously” (Parks 2000, 36). As a feminist I am called to respect women’s death requests out of concern for their rational capacity and their reports of pain and suffering, while at the same time addressing the trust and care concerns of those who criticize MAID. These commitments are not incompatible, and both must be part of any robust feminist account of MAID.

References Bassett, Abigail. 2021. Why isn’t women’s pain taken seriously? https://www.shondaland.com/live/ body/a37397056/why-isnt-women-pain-taken-seriously/. Accessed 12 Apr 2022. Beaudry, Jonas-Sebastien. 2020. Bill C-7, assisted dying and “lives not worth living”. Policy Options. https://policyoptions.irpp.org/magazines/december-2020/bill-c-7-assisted-dying-andlives-not-worth-living/. Accessed 10 Jan 2022. Bill C-7. 2021. An act to amend the criminal code (medical assistance in dying), 2nd Session, 43rd Parliament. Ottawa. https://parl.ca/DocumentViewer/en/43-2/bill/C-7/royal-assent. Accessed 4 Dec 2021. Buchbinder, Mara. 2018. Access to aid-in-dying in the United States: Shifting the debate from rights to justice. American Journal of Public Health 108 (6): 754–759. Check, Hayden E. 2010. Sex bias blights drug studies. Nature 464: 332–333.

19

This is a feminist issue because, within each marginalized group, women tend to represent the worst-off members of those groups. For example, Black and Hispanic women and women with disabilities tend to fare worse than men from those same groups when various factors are assessed, such as income, education, and other factors. According to the U.S. Bureau of Labor Statistics (2021), Black women’s average weekly salary in 2020 was $764 compared to Black men’s average of $830. And the Department of Labor Office of Disability Employment Policy (2021) reports that for 2021, 24.7% of disabled women between the ages of 25–54 lived below the federal poverty level versus 20.9% of disabled men.

260

J. A. Parks

Consumer Reports. 2019. Is bias keeping female, minority patients from getting proper care for their pain? Washington Post, July 29. Crenshaw, Kimberle. 1989. Demarginalizing the intersection of race and sex: A black feminist critique of antidiscrimination doctrine, feminist theory and antiracist politics. University of Chicago Legal Forum 1 (8): 139–167. Department of Labor Office of Disability Employment Policy. 2021. Spotlight on women with disabilities. https://www.dol.gov/sites/dolgov/files/ODEP/pdf/Spotlight-on-Women-with-Disabi lities-March-2021.pdf. Accessed 12 Apr 2022. Downie, Jocelyn. 2020. Why feminist philosophy (especially Sue Sherwin’s) matters: Reflections through the lens of medical assistance in dying. IJFAB: International Journal of Feminist Approaches to Bioethics 13 (2): 21–27. Downie, Jocelyn, and Udo Schuklenk. 2021. Social determinants of health and slippery slopes in assisted dying debates: Lessons from Canada. Journal of Medical Ethics 47: 662–669. Faro, Monica, Naia Saez-Francas, Jesus Castro-Marrero, et al. 2016. Gender differences in chronic fatigue syndrome. Reumatología Clínica 12 (2): 72–77. Fennell, Patricia, Nancy Dorr, and Shane S. George. 2021. Elements of suffering in myalgic encephalomyelitis/chronic fatigue syndrome: The experience of loss, grief, stigma, and trauma in the severely and very severely affected. Healthcare (Basel). https://www.ncbi.nlm.nih.gov/pmc/ articles/PMC8150911/. Accessed 5 Jan 2022. Foster, Drew. 2014. A study gives mysterious chronic-fatigue syndrome some legitimacy. The Atlantic. https://www.theatlantic.com/health/archive/2014/11/study-gives-mysterious-chr onic-fatigue-syndrome-some-legitimacy/382617/. Accessed 3 Jan 2022. Fricker, Miranda. 2007. Epistemic injustice: Power & the ethics of knowing. Oxford: Oxford University Press. Gilligan, Carol. 1982. In a different voice: Psychological theory and women’s development. Cambridge: Harvard University Press. Harrigan-Farrelly, Joan. 2022. For black women, implicit racial bias in medicine may have far-reaching effects. US Department of Labor Blog. https://blog.dol.gov/2022/02/07/for-blackwomen-implicit-racial-bias-in-medicine-may-have-far-reaching-effects. Accessed 29 Mar 2022. Hoffman, Kelly M., Sophie Trawalter, Jordan R. Axt, and M. Norman Oliver. 2016. Racial bias in pain assessment and treatment recommendations, and false beliefs about biological differences between blacks and whites. Proceedings of the National Academy of Sciences of the United States of America 113 (16): 4296–4430. Kaiser Family Foundation. 2021. COVID-19 cases and deaths by race/ethnicity: Current data and changes over time. Available at https://www.kff.org/racial-equity-and-health-policy/issue-brief/ covid-19-cases-and-deaths-by-race-ethnicity-current-data-and-changes-over-time/. Accessed 3 Dec 2021. Kittay, Eva F. 1999. Love’s labor: Essays on women, equality, and dependency. New York: Routledge. Laws, Terri. 2019. Race, religion, and physician assisted suicide. Religion & Politics. Washington University in St. Louis. https://religionandpolitics.org/2019/09/03/how-race-matters-in-the-phy sician-assisted-suicide-debate/. Accessed 14 Oct 2021. Lockhart, P.R. 2018. What Serena Williams’s scary childbirth story says about medical treatment of black women. Vox. https://www.vox.com/identities/2018/1/11/16879984/serena-williams-chi ldbirth-scare-black-women. Accessed 8 Nov 2021. McMillan Cottom, Tressie. 2019. Thick and other essays. New York: The New Press. Meghani, Salimah H., Eeeseung Byun, and Rollin M. Gallagher. 2012. Time to take stock: A metaanalysis and systematic review of analgesic treatment disparities for pain in the United States. Pain Medicine 13 (2): 150–174. Mende-Siedlecki, Peter, Jennie Qu-Lee, Robert Backer, and Jay J. Van Bavel. 2019. Perceptual contributions to racial bias in pain recognition. Journal of Experimental Psychology: General 148 (5): 863–889.

14 Feminist Approaches to Medical Aid in Dying: Identifying a Path Forward

261

Northwell Health Gaslighting in women’s health: No, it’s not just in your head. https://www.nor thwell.edu/katz-institute-for-womens-health/articles/gaslighting-in-womens-health. Accessed 8 Nov 2021. Not dead yet disability rights toolkit for advocacy against legalization of assisted suicide. https://notdeadyet.org/disability-rights-toolkit-for-advocacy-against-legalization-of-ass isted-suicide. Accessed 24 Sept 2021. Omeish, Yousof, and Samantha Kiernan. 2020. Targeting bias to improve maternal care and outcomes for Black women in the USA. eClinicalMedicine 27: 1–2. Oregon Health Authority Public Health Division. 2021. Oregon death with dignity act: 2021 data summary. https://www.oregon.gov/oha/PH/PROVIDERPARTNERRESOURCES/EVALUA TIONRESEARCH/DEATHWITHDIGNITYACT/Documents/year24.pdf. Accessed 25 Mar 2022. Parks, Jennifer A. 2000. Why gender matters to the euthanasia debate: On decisional capacity and the rejection of women’s death requests. Hastings Center Report 30 (1): 30–36. Parks, Jennifer A., and Maria Howard. 2021. Dying well in nursing homes during COVID-19 and beyond: The need for a relational and familial ethic. Bioethics 35 (6): 589–595. Pollit, Katha. 1992. Marooned on Gilligan’s Island: Are women morally superior to men? The Nation, 28 Dec. Pullman, Daryl. 2020. In a familiar voice: The dominant role of women in shaping Canadian policy on medical assistance in dying. Canadian Journal of Bioethics 3 (3): 11–20. Riddle, Christopher A. 2017. Assisted dying & disability. Bioethics 31 (6): 484–489. Rogers, Wendy A. 2006. Feminism and public health ethics. Journal of Medical Ethics 32 (6): 351–354. Samulowitz, Anke, Ida Gremyr, Erik Eriksson, et al. 2018. “Brave men” and “emotional women”: A theory-guided literature review on gender bias in health care and gendered norms towards patients with chronic pain. Pain Research and Management. https://doi.org/10.1155/2018/635 8624. Accessed 27 Oct 2021. Senchuk, Dennis M. 1990. Listening to a different voice: A feminist critique of Gilligan. Studies in Philosophy and Education 10: 233–249. Sherwin, Susan. 1992. No longer patient: Feminist ethics & health care. Philadelphia: Temple University Press. Sikka, Tina. 2021. Barriers to access: A feminist analysis of medically assisted dying and the experience of marginalized groups. Omega (Westport) 84 (1): 4–26. Silvers, Anita, and Francis L. 2016. Death, dying, and the disabled. In The Oxford handbook of ethics at the end of life, eds. Stuart Younger and Robert M. Arnold. https://doi.org/10.1093/oxf ordhb/9780199974412.001.0001. Accessed 23 Apr 2022. Tronto, Joan. 2013. Caring democracy: Markets, equality, and justice. New York: NYU Press. US Bureau of Labor Statistics. 2021. Report 1094 highlights of women’s earnings in 2020. https:// www.bls.gov/opub/reports/womens-earnings/2020/home.htm. Accessed 12 Apr 2022. US Institute of Medicine. 2015. Beyond myalgic encephalomyelitis/chronic fatigue syndrome: Redefining an illness. Washington, DC: The National Academies Press. Weiss, Edward S. 2018. Taking unnecessary aim at MAID. Canadian Family Physician 64 (4): 249. Wendell, Susan. 1996. The rejected body: Feminist philosophical reflections on disability. New York: Routledge. Wolf, Susan. 1996. Gender, feminism, and death: Physician-assisted suicide and euthanasia. Feminism & bioethics: Beyond reproduction. New York: Oxford University Press. Zhang, Lanlan, Elizabeth A. Reynolds Losin, Yoni K. Ashar, Leonie Koban, and Tor D. Wager. 2021. Gender biases in estimation of others’ pain. The Journal of Pain 22 (9): 1048–1059.

262

J. A. Parks

Jennifer A. Parks Ph.D., is Professor of Philosophy and Director of the Undergraduate Bioethics Minor Program at Loyola University Chicago. Her research focuses on health care ethics with special interests in reproduction and reproductive technologies, end of life, and aging and longterm care. She is currently working on ethical issues relating to elder care and technology, and global issues in reproductive technologies.

Chapter 15

Envisioning Markets in Assisted Dying Michael Cholbi

Abstract Ethical debates about assisted dying typically assume that only medical professionals should be able to provide patients with assisted dying. This assumption partially rests on the unstated principle that assisted dying providers may not be motivated by pecuniary considerations. Here I outline and defend a mixed provider model of assisted dying provision that contests this principle. Under this model, medically competent non-physician professionals could receive fees for providing assisted dying under the same terms and conditions as physicians can in those jurisdictions where medically assisted dying is lawful. The mixed provider model blunts objections to assisted dying rooted in supposed clashes with medical values. In addition to generating a market likely to expand access to assisted dying, the mixed provider model would not create markets that are unjust because they are “noxious” in Satz’ sense or because they raise “semiotic” concerns about the value of human life. Keywords Assisted dying · Medical values · Justice in markets · Debra Satz · Jason Brennan Ethical debates about assisted dying have long focused on medically assisted dying, and all the more on assisted dying enabled or undertaken by physicians (Ost 2010: 503ff). In large measure, this focus reflects the fact that in most societies, physicians have a monopoly on legal access to the technologies that those who desire assisted dying wish to take advantage of. Only physicians have the right to prescribe or administer the lethal medications that enable the safe, fast, and relatively pain free deaths sought by those who desire assisted dying (Cholbi 2015). One result of this focus on physicians aiding patients to die is that much of the ethical debate about assisted dying has addressed whether it is compatible with physicians’ professional roles or with the values thought to undergird the medical profession (Callahan 1992; M. Cholbi (B) School of Philosophy, Psychology and Language Sciences, University of Edinburgh, Edinburgh, UK e-mail: [email protected] © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 M. Cholbi and J. Varelius (eds.), New Directions in the Ethics of Assisted Suicide and Euthanasia, The International Library of Bioethics 103, https://doi.org/10.1007/978-3-031-25315-7_15

263

264

M. Cholbi

Baumrin 1998). Skeptics about assisted dying, for example, have often argued that the practice violates the principle that physicians may not kill (or knowingly contribute to killing) their patients; that it would undermine the trust needed for a healthy physician–patient relationship; that it would render physicians complicit in injustices toward vulnerable groups; etc. Many defenses of assisted dying address these objections about physicians’ professional roles or the medical profession’s values head on, aiming to show that assisted dying is a threat to neither (Miller and Brody 1995; Seay 2005; Cholbi 2011: 145–148). But an alternative dialectical strategy is to concede the justifiability of assisted dying in principle but to deny that physicians must have a monopoly over its provision. If individuals sometimes have a right to others’ assistance in hastening their deaths, it does not follow that doctors alone may provide that assistance. And if there were good reason for that assistance not to be the exclusive province of doctors, the aforementioned objections to assisted dying would be muted. My purpose here is to defend just this possibility, namely, a regulatory regime in which physicians are one legal option for those seeking assisted dying. I will call this regulatory regime a mixed provider model for the provision of assisted dying. Again, this regime would enable those ethical objections to assisted dying relating to physicians’ roles and medical values to be sidestepped. However, my case for the mixed provider model does not rest on this consideration alone. My case also questions a principle that has gone largely unstated in debates about assisted dying, namely, that providers of assisted dying may not be motivated by pecuniary considerations. Being human, physicians are of course economic actors, with the same material interests as anyone else. But the thought that a physician might have an economic motivation for providing assisted dying offends against the image of physicians as disinterested professionals entrusted with their patients’ health. Indeed, medicine has long had an ambivalent relationship to the commercial sphere. Some ancient Mediterranean physicians took no fees for their services, and Hippocrates advised that physicians should forego payment from patients in dire financial straits (DeCou 2019). English practitioners in the medieval period were legally barred from billing patients for fees; payments from patients were instead conceptualized as voluntary “honoraria” for services rendered (Hall and Schneider 2008). No doubt such reservations have diminished over time. (In many parts of the world, physicians loudly advertise their fees, particularly for non-essential or cosmetic procedures). But the aversion to physicians’ being financially motivated has persisted in connection with assisted dying, I would suggest. The Swiss organization Dignitas, for example, has provided what it calls “accompanied suicide” to over 3000 individuals. It operates as a non-profit organization in which individuals first pay a membership. Dignitas will sometimes waive the fees for assisting in death in cases of financial hardship. The group’s practices are clearly meant to conform with Swiss law, which permits assistance in dying so long as the assistance is not selfishly motivated. However, Dignitas’ model is also likely to echo public distaste for the notion that anyone—but especially a medical professional—should profit from another’s death. My mixed provider model rejects the notion that the provision of assisted dying must rest on such pure or morally impeccable motives. The model instead envisions

15 Envisioning Markets in Assisted Dying

265

a market in assisted dying, wherein those jurisdictions in which assisted dying is lawful allow its provision by physicians but also by clinically qualified non-physician providers who may charge fees for their services. (Hereafter, I will call these ‘clinical providers’ for ease of exposition).1 These clinical providers can be analogized to other specialized medical service providers whose practice is far more restricted than that of physicians’: dialysis clinics, medical imaging centers, testing labs, and the like, each of which has a fairly limited relationship with their patients, and correspondingly, a more limited set of moral responsibilities than physicians have to their patients. The provision of assisted dying on an explicitly for profit basis is compatible with these responsibilities, I shall argue. Moreover, a market in assisted dying would redress certain difficulties regarding access to assisted dying that have emerged in those jurisdictions where it is the sole prerogative of physicians. Finally, many philosophers believe that markets have moral limits, i.e., that not everything should be for sale. As I see it, markets in assisted dying need not be “noxious” nor give rise to what have been called “semiotic” concerns. Markets in assisted dying thus fall within the parameters of morally defensible markets. If I am correct, then those otherwise sympathetic to the legal permissibility of assisted dying should therefore welcome the prospect of a market for assisted dying in which clinical providers ‘compete’ with one another and with physicians. The mixed provider model answers classic objections to assisted dying emanating from the role played by physicians in the process, would expand access to those eligible for assisted dying, and is not unjust in itself.

15.1 The Mixed Provider Model My mixed provider model introduces market-oriented actors into the provision of assisted dying. Crucially however, the market I envision is not an unregulated one. I will first outline the model I have in mind before proceeding (in Sects. 15.2, 15.3 and 15.4) to defend it. Jurisdictions in which assisted dying is legal vary in the exact roles that different individuals may take vis-à-vis the individual who is helped to die. US jurisdictions tend to follow the template set by Oregon’s Death with Dignity Act. In Oregon, physicians alone can evaluate an individual’s eligibility for assisted dying (determining if they are terminal and are competent to decide this matter for themselves) and prescribe the lethal agent. With respect to the patient’s death, physicians may witness the death but may not administer, or assist in the administration of, the lethal agent. This contrasts with, for instance, practices in the Netherlands, where physicians evaluate individuals’ eligibility for assisted dying, prescribe lethal agents for patients who seek it, and may administer the lethal agent themselves. As in Oregon, Dutch physicians may be witnesses to patient self-administration. Both 1 Crisp (1987) proposes to call these professionals ‘telostricians’, corresponding to the Greek terms for ‘end’ and ‘artisan’.

266

M. Cholbi

systems also allow patient’s families or volunteers to play a part as witnesses either to self-administration or physician administration of the lethal agent or as assistants to that administration. A chief difference between systems like Oregon’s and systems like the Netherlands is that the former mandate self-administration, whereas the latter allows for selfadministration. It is not among my purposes here to assess the merits of assisted dying in which death is merely ‘aided’ by means of prescribing a lethal medication versus assisted dying in which active euthanasia is available. I merely wish to highlight that where assisted dying is legal, the law ascribes different roles to different individuals within the process of assisted dying. My mixed provider model would allow clinical providers to play whatever role a given jurisdiction presently assigns to physicians. The central difference, though, is that assisted dying and the medical needs associated with it delimit the bounds of the professional relationship between clinical providers and those seeking assisted dying. Clinical providers would not be authorised to treat whatever underlying medical conditions might be responsible for individuals wanting assistance in dying, for example. They would not have the authority to treat cancer, recommend pain management regimens to patients, or provide counselling to patients with depression. Akin to allied health professionals, clinical providers would be medical specialists whose relationships with patients would begin (and end) with patients’ soliciting their assistance in dying. Of course, it is very likely that for most patients, physicians will have already played a large role in their medical histories. After all, many will have terminal or chronic illness for which physicians have supervised their treatment. Some measure of cooperation between physicians and clinical providers would be necessary under the mixed provider model. For example, clinical providers would be entitled to access the medical records of those who seek their services. But the collaboration between physicians and clinical providers need not extend any further. Clinical providers would therefore take fees from their clients seeking assisted dying and be responsible for whatever aspects of the process of assisted dying physicians may also oversee in that jurisdiction. This could include determining patients’ eligibility for assisted dying within the relevant jurisdiction, ascertaining the patients’ competence if doubts arise due to psychological impairment, prescribing lethal agents, administering lethal agents, and witnessing their administration. As the assisted dying market evolves, it is likely that clinical providers would augment this ‘basic package’ of assisted dying services with other options of patients’ choosing. Clinical providers could provide individuals with specific settings in which their deaths could occur (their homes, but also forests, meadows, beaches, etc.), as well as other accoutrements meant to foster a ‘good death’ for the patient. Clinical providers could also play a logistical role, making travel arrangements for patients or their loved ones. In this respect, clinical providers may come to resemble pre-mortem versions of funeral homes, offering ‘euthanatic’ experiences to those eligible for assisted dying. And as with funerals, we might expect clinical providers to ‘brand’ themselves in order to attract clientele (a ‘green’ clinical provider, a feminist clinical provider, providers with specific language or cultural competencies, etc.).

15 Envisioning Markets in Assisted Dying

267

Clinical providers are not physicians, but nor would they be medical amateurs. Like other non-physician medical practices, the clinical provision of assisted dying would require licensure, and in particular, the licensure of the providers themselves. With regard to knowledge of physiology, the requisite training would include understanding of the physiological nature of the dying process, the dosages of lethal agents needed to induce death, signs of bodily distress, and the jurisdictional criteria for declaring death. Part of clinical providers’ training would ethical or legal. They would need to know what the relevant jurisdictional standards are for assisted dying (whether it is restricted to the terminally ill or also extends to those with conditions involving incurable and unremitting suffering, for instance), the relevant jurisdictional standards for determining psychological competency and how to assess competency, and laws regarding the handling and disposition of corpses. Clinical providers would also benefit from training designed to establish a ‘bedside manner’ appropriate to the needs of those who seek assisted dying. Here their training might involve communication and listening skills, the ability to detect anxiety and other psychological stressors, and awareness of the psychology of illness, dying, and grief. Particularly important in this regard would be training to recognize ambivalene and rescinded consent to assisted dying. Finally, some measure of specialization within this specialization is possible: Some within a given clinical provider setting might be responsible for assessing patients’ eligibility for assisted dying, others for discussing the circumstances under which the patient desires their death to occur, still others for the actual supervision of the patient’s death, and so on. A final aspect of a regulated market in assisted dying is the protections afforded to patients and providers. Clinical providers would presumably be subject to all the same reporting requirements as physician assistors, as well as being subject to criminal or civil sanctions for malpractice, the administration of assisted dying in the absence of valid and informed patient consent, etc. A further protection for patients might be to require that all deaths be witnessed by an individual unaffiliated with the assisted dying provider. A regulated market would also likely need financial protections for both patients and providers. Clinical providers could charge fees for the various preliminary steps in the process of assisted dying in order to recoup their costs. Likewise, a patient’s fee could be held in trust until their deaths, and providers would be entitled only to partial payment if the patient rescinds their consent. Such rules would reduce provider incentives to ensure patients’ deaths while also compensating them for the services they provide even when the patient does not end up dying thanks to their assistance.

15.2 Market Demand and Barriers to Accessing Assisted Dying As domain-specific medical professionals, clinical providers would have a distinct and carefully circumscribed part to play in the assisted dying process. Their presence

268

M. Cholbi

would likely increase the ‘supply’ of assisted dying options. Is there reason to think that there is demand for assisted dying that would be responsive to this supply? This section offers evidence for an affirmative answer to that question. For one, jurisdictions in which assisted dying is lawful—and keep in mind that these are almost entirely jurisdictions in which physicians alone can provide assistance—have witnessed steady increases in the prevalence of assisted dying over time. One UK publication has reported in 2019 that assisted dying rates are “soaring” worldwide (Davis 2019). The number of euthanasia cases in the Netherlands, for example, has risen from about 2% of all deaths in 2002 to approximately 4% of all deaths in 2019 (Groenwoud et al. 2021). Similar statistical findings emerge from studies in Belgium (Raus et al 2021). The number of prescriptions written and deaths registered under Oregon’s Death with Dignity Act has increased two- to three-fold since 2015 (Oregon Health Authority 2021). The number of participants under Washington’s assisted dying law has doubled since 2014 (Washington State Department of Health 2021). Reported deaths with medical assistance in Canada increased 700% since 2015 (Health Canada 2020). These findings are not atypical. Virtually whenever assisted dying has been legalised, deaths authorised by such laws remain small in number. Yet the numbers of individuals who take advantage of such laws often exceed what even supporters of legalization predict and their popularity grows over time. Opponents of assisted dying sometimes point to such statistics as providing reasons to second guess or revisit its legalization. Certainly increases in the prevalence of assisted dying should invite scrutiny. Some have proposed that in Belgium, the increases reflect physicians not honouring legal safeguards (Raus et al. 2021). Such possibilities should be acknowledged and investigated. Nevertheless, the overall pattern of data across time and across multiple jurisdictions suggests that the better part of the explanation for these increases is straightforward: Many people prefer assisted dying to other pathways toward death. The growing popularity of assisted dying should in fact be viewed as a point in favor of its legalization. In addition, we ought not assume that all those who desire assisted dying have ready access to it. Critics of legalized assisted dying often worry that it will be used in ways that fail to respect the rights of minorities and socially marginal groups. If anything however, we might have the contrary worry—not that minorities and socially marginal groups are too often compelled toward assisted dying but that they do not have sufficient access to it where the practice is lawful. Statistics from Oregon paint a picture wherein assisted dying is almost exclusively the province of the least disadvantaged social groups: whites, males, well-educated, and affluent (Oregon Health Authority 2021: 8). Marginalized groups in the United States, for example, may lack the relationships with health care providers necessary to facilitate assisted dying. Indeed, they may face structural barriers and implicit bais such that their low level of participation in assisted dying conceals an underlying interest in it (Sikka 2019). It is therefore likely that there is sufficient patient demand for assisted dying that at least some clinical providers would be able to function successfully under the mixed market model I propose. Still, one might wonder whether there are specific

15 Envisioning Markets in Assisted Dying

269

reasons to expect that clinical providers would attract a significant clientele. Why might someone prefer to use a clinical provider rather than a physician for assistance in dying? As we observed earlier, it seems likely that commercial competition among clinical providers will, over time, make it possible for patients to have highly tailored assisted dying experiences. But an additional set of reasons concerns a patient’s relationship with their physician. On the one hand, some patients may not have a relationship with a physician within which they are comfortable discussing or pursuing assisted dying, and they may furthermore not wish to seek out a new physician solely for the purpose of discussing or pursuing assisted dying. Clinical providers would presumably be visible and identifiable to such patients. Conversely, some patients may have strong relationships with physicians that they might not wish to cloud or complicate by discussing or pursuing assisted dying with them. They might not want a physician who has competently and compassionately provided them life-extending care to switch roles so as to facilitate their deaths. Patients may also value their physicians retaining their disposition to save life (Crisp 1987). Assisted dying may thus complicate a patient’s care relationship with their physician and hence provide them with incentives to seek assisted dying from a clinical provider. A patient’s relationship with a clinical provider of assisted dying would not be uncaring. But at the same time, it would be a single purpose transactional relationship of relatively short duration, and therein lies some of the appeal of seeking assisted dying from a clinical provider. In relationships with physicians, assisted dying must necessarily be a chapter within the context of a larger professional relationship— typically a final chapter—a relationship that has the extension of life as one of its recognized goals. In contrast, a patient’s relationship with a clinical provider rests on a narrower basis: that the patient seeks (or is at least seriously contemplating) assisted dying provides the foundation for the relationship and demarcates its professional bounds. Unlike their physicians in many cases, the patient can be confident that the clinical provider is not ambivalent about the prospect of assisting them to die. Hence, the very brevity and lack of ambiguity in patients’ relationship with clinical providers is likely to appeal to some patients.

15.3 The Moral Limits of Markets, Part I: Satz on ‘Noxious’ Markets Even if I am correct that the mixed provider model answers to legitimate patient interests that are not currently addressed where physicians have a monopoly on assisted dying, some philosophers may argue that a market in assisted dying is unjust in itself. Some things, according to these philosophers, should not be for sale, and perhaps assisting in the deaths of others is among these. In this section and the next, I consider two frameworks for the morality of markets and show that a mixed provider market in assisted dying does not transgress the limits of morally permissible markets.

270

M. Cholbi

In her Why Some Things Should Not Be for Sale: The Moral Limits of Markets, Satz (2010) proposes four parameters for what she calls “noxious” markets. A market for a good is noxious to the extent that it a. results in harmful outcomes for market transactors or identifiable 3rd parties, b. results in harmful outcomes to society at large, c. involves transactors with “weak agency” (due, for instance, to lacking information relevant to transactions involving the good), or d. involves transactors in highly vulnerable positions in relation to the good in question. Satz’s account of noxious markets has a broadly republican pedigree. For as she sees it, markets characterized by these four features undermine or prevent individuals from relating to one another as moral equals by contributing to relations of inferiority or domination among them. Noxious markets generate problems “relating to the standing of parties before, during and after the process of exchange” (Satz 2010: 93). Echoing Adam Smith, Satz argues that markets should instead contribute to creating “a society of horizontal relationships based on free interaction, equality and reciprocal self-interest” (2010: 42). What does Satz’s account of noxious markets imply about the justifiability of a market in assisted dying based on mixed provider model? Let us address each of the four features of noxious markets in turn. a. Harms to Transactors or to 3rd Parties We have reason to expect that transactions within a regulated market in assisted dying would be Pareto optimal. Clinical providers of assisted dying profit from its provision, for one. A mixed market would likely expand access to assisted dying overall, enabling patients to avoid harms they might otherwise suffer, as well as expanding the variety of circumstances, settings, etc., in which assisted dying can occur, allowing patients to enjoy deaths that better answer to their preferences or values. There is also no obvious class of 3rd parties who would be harmed (or harmed unjustifiability) by the mixed market. Granted, physician providers of assisted dying could be said to be harmed inasmuch as some individuals who might have sought their assistance would turn to clinical providers instead. But it is not clear that physicians are entitled to avoid such ‘competitive’ harms. Markets will of course have winners and losers, yet the losers in market competition (especially those who would otherwise have a monopoly on the provision of the relevant good) do not have the right to veto a market because that market does not accord them competitive advantage. b. Harms to Society at Large To assert that a regulated mixed market in assisted dying would never harm anyone would be foolish. Yet it is difficult to discern any large scale societal harms that would arise from the introduction of such a market. A society would have to bear the bureaucratic costs of regulating such a market. The mixed market in assisted dying would not cause pollution or other transactional externalities,

15 Envisioning Markets in Assisted Dying

271

undermine anyone’s basic rights, consign anyone to a lower status, or amplify the power of some citizens or groups at the expense of others. c. Weak Agency Worries about individuals’ ability to rationally opt to hasten death have long been central in debates about the ethics of assisted dying. Should we worry that a mixed market in assisted dying would exhibit what Satz calls weak agency, i.e., will transactors lack the knowledge or understanding needed to properly consent to such transactions? No doubt the decision to hasten one’s own death is fraught and complex. It will rarely be crystal clear (for example) whether a person with advanced stage cancer is better off dying sooner rather than later. My mixed market model does not pretend to simplify these questions, but nor does it complicate these questions beyond the level already present in end of life decision making. A patient who struggles to determine whether she is better off dying sooner rather than later is not likely to struggle more if opting for the former means she would later end up paying a clinical provider to help her die. Recall also that my mixed provider model does not offer one set of patient protections regarding competency or agency for patients who seek assisted dying from physicians and another (weaker) set of protections for those who seek assisted dying from clinical providers. The same protections apply regardless. Most jurisdictions have in place various safeguards meant to ensure that patients who opt for assisted dying do so carefully, under full information, with adequate forethought and free for undue pressures. For instance, patients are typically required to have multiple medical certifications that they meet the operative medical standards for eligibility for assisted dying (are terminally ill, etc.); to make multiple requests for assisted dying separated by an established time interval; to have their requests witnessed by others; etc. These same safeguards would apply to assisted dying via clinical providers. In addition, under the regulatory transparency I propose, clinical providers would also have strong motivations to ensure that their clients opt for assisted dying under conditions of full information and robust agency. Like physicians, they could be subject to criminal or civil liability in cases of malpractice or malfeasance. But clinical providers would have the further incentive that the success of their commercial enterprise may well depend on their clients exhibiting robust agency in the course of deliberating about whether to procure assisted dying from them. d. Transactor Vulnerability A market in assisted dying might seem to invite exploitation of those interested in assistance in dying. Many of those seeking assisted dying might be in desperate straits, willing to pay someone to end a life whose quality is in rapid decline. Does the seemingly vulnerable position of such patients speak against my mixed market model? My mixed provider model is of course predicated on clinical providers acting from financial motives. Its critics might then reason that clinical providers, in their eagerness to turn a profit, will take advantage of desperate clients. Some of the measures to counteract this were mentioned in Sect. 15.1: allowing fees to be

272

M. Cholbi

collected for patient consultations, partial payment for patients who rescind their consent, liability to criminal or civil action in cases of malpractice or malfeasance, etc. And again, we should keep in mind that my model establishes a right to purchase assisted dying services from clinical providers under the same medical circumstances as a jurisdiction permits individuals to seek assisted dying from physicians. Jurisdictions would require that clinical providers follow the same protocols as physicians in determining patients’ eligibility and willingness to receive assistance in dying (including medical examinations, waiting periods, etc.) Clinical providers and physicians who offer assistance in dying would thus operate on the same competitive turf. No doubt those who seek assisted dying are vulnerable in various ways, but a regulated mixed market is unlikely to exploit that vulnerability in unjust ways. On balance then, the case for a mixed market in assisted dying being noxious is unconvincing. No doubt there would be some failures in this market. But absent evidence to the contrary, we cannot assume that such failures would be more common or more egregious than they are when physicians monopolize access to assisted dying. Considered holistically and over the long run though, my mixed market model is likely to provide many of the benefits markets can confer while minimizing (albeit not eliminating) harms or abuses. In Satz’s terms, a regulated market in assisted suicide would neither presuppose conditions of unequal status nor contribute to conditions of equal status. It would not exploit conditions in which individuals can ‘push around’ each other (or foster such ‘pushing around’).

15.4 The Moral Limits of Markets Part II: Brennnan and Jaworski on ‘Semiotic’ Limits to Markets In their Markets Without Limits: Moral Virtues and Commercial Interests, Brennan and Jaworski (2016) defend a more permissive view of markets than Satz’. In their pithy slogan, “if you may do it for free, you may do it for money” (2016: 10). Supposing that this slogan is correct, its implications for my mixed provider model for assisted dying are not straightforward. For one, rightly or wrongly, in many parts of the world, a person may not ‘do it for free’ when it comes to assisted dying; doing so would run afoul of laws against assisting in suicide. Moreover, where medically assisted dying is lawful, physicians do not provide their services ‘for free’. They are often compensated indirectly, via patients’ insurers or their own salaries. Nevertheless, the presumption is that even though physicians financially benefit from providing their services (assisted dying services included), they are not motivated by financial benefit in the sense that their reasons for providing particular services are financial. A physician ‘profits’ from the overall course of care provided to patients, not from their services viewed as distinct economic transactions. This ‘transactional’ picture is of course where my mixed provider model parts ways with convention: Clinical providers could be motivated by the financial gains associated with each assisted dying ‘transaction.’

15 Envisioning Markets in Assisted Dying

273

Hence, Brennan and Jaworski’s slogan does not lend unequivocal support to a market in assisted dying. However, in defending their position, they address an objection to markets that is particularly salient in connection with the prospect of a market in assisted dying. What they call the ‘semiotic’ objection to a market rests on the observation that market activity seems to symbolically communicate something about how market transactors value goods. The objection maintains that markets can disrupt or alter ethically valuable meanings we attach to those certain goods. Markets thus come to “express or communicate certain negative attitudes” or prove “incompatible with holding certain positive attitudes” toward a class of goods (2016: 21). The commodification of some goods can ascribe a meaning to certain goods that “is essentially disrespectful or degrading” (2016: 83) to these goods or to what is valuable about them. For instance, some argue that paid gestational surrogacy signals that children are a commodity and so undermines the value of the intimacy characteristic of a pregnant woman’s relationship to her fetus (Anderson 1990). Those skeptical of my mixed provider model are likely to be attracted to such semiotic worries about markets in assisted dying. They may view paying a clinical provider, a professional with whom a patient has no wider professional relationship, as cheapening the process of assisted dying, as failing to reflect the value of human life, or as at odds with the solemnity of dying. Paying someone explicitly to kill oneself seems to communicate that a person’s life is a kind of commodity. Ultimately, Brennan and Jaworski reject the semiotic objection altogether. Not only are the meanings we impute to various goods contingent and subject to revision over time, they argue, the consequences of assigning ‘semiotic’ significance to different goods vary, and it is the consequences of assigning such significance that should govern whether a good is made available for market exchange (2016: 83). In virtually every case, the adverse consequences of restricting market access to a good are greater than whatever benefit results from treating goods as having a significance or value that is disrespected or degraded by market exchange, according to Brennan and Jaworski. In the spirit of dialectical charity, I will here assume the contrary—that semiotic objections can have force against the existence of markets in various goods—and proceed to consider whether such objections are convincing when directed against my own mixed provider market in assisted dying. Evaluating semiotic objections to particular markets is a complex matter. It requires identifying (i) what ‘pre-market’ value the good in question has, (ii) what a market in such a good would express, and (iii) how this expression would be incompatible with the pre-market value. In the case of assisted dying, it seems likely that opponents of markets would hold that such a market expresses repugnant views about the value of human life—that a market in assisted dying in effect puts a price on human life by attaching a monetary value to life’s cessation. (Kass 2002: 234ff.) A central challenge to this reasoning is isolating the precise role that markets in assisted dying as such might play in undermining or expressing disrespect for the value of human life. No doubt many opponents of my mixed provider model oppose physician-assisted dying (and indeed, all forms of assisted dying) on semiotic grounds—that it fails to accord with the sanctity of human life. But then what additional objectionable statement about the value of human life would be made

274

M. Cholbi

if assisted dying were made available via market transactions, as I have proposed? Doing X may express a disrespectful stance regarding some valuable good, but it is not obvious that doing X from at least partially monetary motives expresses further disrespect. Unpaid gestational surrogacy may express a “negative attitude” toward pregnancy and motherhood. Does paying the surrogate add to the negativity thus expressed? Murder expresses disrespect toward its victims. Do paid assassins express greater disrespect than do other murderers? My defense of the mixed provider model is not meant to defend assisted dying as such. It aims only to consider whether, if assisted dying is in principle defensible, the provision of assisted dying in a regulated market is also defensible. My opponents may well believe that assisted dying as such raises semiotic concerns about the value of human life. But it difficult to see that market-provided assisted dying raises additional semiotic concerns. In addition, as applied to markets in assisted dying, the semiotic objection misidentifies the good in question. Tempting as it is to describe market-provided assisted dying as others (or ‘the market’) putting a price on someone’s human life, a more accurate description is that individuals would be putting a price on, i.e., making a cost judgment regarding, what they are willing to pay for their own good death. A free and voluntary market transaction occurs at a price reflecting both the purchaser’s willingness to buy and the seller’s willingness to sell. In this regard, markets in assisted dying would be to some extent dependent on what those besides the patient care about. A patient considering the services of a clinical provider could not decide unilaterally how much to monetarily value the death she might hope to attain. Yet securing such services will have to reflect the value that, ethically speaking, should prevail in such choices: the value that the patient assigns to her dying by her own lights, a value measured by the monetary resources she is willing to forego to acquire such death on the terms she seeks. That we have a right to expend our own resources to provide ourselves with what we take to be a good death is indisputable. A person with the means to (for example) die in hospice has the presumptive right to do so, and more generally, patients have the right to make their own determinations about what should be exchanged in order for them to acquire the death they prefer. A patient has the right to forego pain medications in order to be more lucid in the days and hours prior to her death; the right to forego additional treatments for underlying illnesses in order to minimise time in hospitals prior to death; etc. In these cases, it is implausible to describe this as the patient (or anyone) expressing any negative attitude at odds with whatever value human life has. Admittedly, when the patient’s judgments have implications for how long she is likely to live (whether more or less than she otherwise would), she will perform a cost–benefit analysis one ingredient of which is the value of continued life. But this need not reflect any judgment about the value of her life, nor a fortiori, any societal or putatively objectively judgment about her life’s value (Kamm 2020). In sum then, semiotic objections to market-provided assisted dying do not succeed. It is not clear that market provision of assisted dying expresses more objectionable messages regarding human life (or death) than does assisted suicide offered outside

15 Envisioning Markets in Assisted Dying

275

the context of market exchange, and it is equally unclear that it expresses an objectionable stance regarding the value of a person’s life (or the significance of their death).

15.5 Conclusion As I see it, a mixed market in assisted dying, where clinical providers co-exist with physicians who provide assisted dying, answers to legitimate interests individuals have in connection with how they die, is likely to make assisted dying available to a wider population of those eligible for it, and would not devolve into an unjust market. No doubt certain opponents of assisted dying will not be moved by such arguments. Those who believe that actively hastening one’s death is immoral or who maintain that we do not have a right to shape the circumstances of our deaths will not find my appeal to the merits of a market in assisted dying compelling. But I hope to have shown such opponents that a market in assisted dying does not introduce new ethical concerns not already present in physician-assisted dying. Such opponents are likely to reject my proposal for a mixed market in assisted dying simply because it makes possible assisted dying—not because it does so through market mechanisms. But I hope to have persuaded those largely sympathetic to the cause of assisted dying that a mixed market represents a possible improvement over the status quo. Moreover, perhaps some opponents of assisted dying will be converted to support its legalisation under precisely the market-based provision I have outlined. The opponents I have in mind are those who assert that assisted dying may be something which patients may rightfully claim but may not rightfully claim from their physicians because it is incompatible with physicians’ professional duties or with the values that undergird medicine. These opponents might welcome the introduction of clinical providers as a way to ‘siphon off’ patients who desire assistance in dying to professionals whose responsibilities are explicitly crafted to allow for the provision of assisted dying. Let me conclude by answering two final objections. The first objection is this: Health care is a right, and if assisted dying is a form of health care, then individuals should be entitled to it without regard to their ability to pay. If the clinical provision of assisted dying were not part of the standard bundle of health care to which all patients are entitled, then it would not be equally open to all, since its availability to patients would depend on their ability to pay. Some individuals may not be able to afford the services that clinical assisted dying providers offer. Hence, clinical provision of assisted dying offends against the claim that health care is a right. That health care is a right is a claim I accept. Yet it is implausible that individuals have an unlimited right to health care or a right to whatever health care they wish to receive. Regardless of exactly how health care is financed or delivered within different systems, each system must establish limits concerning what health care individuals may receive. These limits will depend on the benefits of various treatments, the cost of their provision, etc. It may well be that clinical provision of assisted dying

276

M. Cholbi

would turn out not to fall within those limits. In other words, it may turn out to be a health care luxury rather than a health care need. Yet this fact does not speak against markets in which clinical providers offer assisted dying services. A society that meets its members’ health care needs should not necessarily object to some of its members exercising their ability to pursue health care luxuries. A society that meets those needs does not have prima facie reason to object to individuals using their private resources to receive non-essential care or to undergo cosmetic procedures, for instance. As I argued in Sect. 15.3, a mixed market is unlikely to harm to third parties. Nor will it undermine access to other forms of health care over which physicians retain a monopoly. Here is the second objection: Some may worry that clinical providers will have a greater incentive to ‘bend the rules’ regarding eligibility for assisted dying. Not content to be able to provide assisted dying services to whatever class of individuals who have legal access to assisted dying in the relevant jurisdiction, clinical providers will agree to provide assisted dying to those who are not legally eligible: where available only to the terminally ill, clinical providers will be tempted to provide assisted dying to those with chronic and unbearable conditions that are nevertheless not terminal; where available only to patients deemed psychologically competent, clinical providers will be tempted to provide assisted dying to the psychologically incompetent; etc. Clinical providers may thus function as a shadowy back door for individuals to receive assisted dying illegally. My mixed market model, critics may allege, thus increases the likelihood of ‘slippery slopes’ wherein assisted dying is offered to those who, legally or morally, ought not have access to it. In response, candor demands the acknowledgement that it is impossible for each and every such abuse to be prevented. But the possibility of abuse is not sufficient to ground a compelling objection to my mixed provider model. For one, there is no obvious reason to expect that such abuse will be more likely with clinical providers than with physicians—and some reason to expect it could less likely in the former case. Physicians are occasionally charged with criminal misconduct in connection with assisted dying, but clinical providers would be subject to additional financial incentives not to bend the rules regarding patient eligibility. Secondly, we should not assume that there is no ‘black market’ in assisted dying provision as is. There appears to be significant illicit trade in the sedative pentobarbital, nicknamed “death in a bottle,” by individuals wishing to end their lives (Costa et al. 2017). Some of these individuals may satisfy the criteria for lawful assisted dying where they live, and given that their evident willingness to pay to access lethal agents, they are likely to be among those who would seek out the service of clinical assisted dying providers. But over the longer term, this might well have a salutary effect on the overall market for assisted dying: A market in which some can purchase assisted dying could exert economic pressure against unlawful provision of assisted dying, and as a result, make it more difficult for those ineligible to access it through market means while expanding access to those legally eligible. A regulated market, with requirements related to provider training, safety, etc., is also likely to reduce risks to those seeking assisted dying. On balance then, we may be optimistic that my mixed provider market in assisted dying will have distinct advantages over a status quo in

15 Envisioning Markets in Assisted Dying

277

which individuals’ only options are either physician assisted dying or assisted dying with the help of black market providers with dubious scruples. Acknowledgements I gratefully acknowledge the useful feedback I received on an earlier version of this chapter from Christopher Cowley, Peter Schaber, Tatjana von Solodkoff, Isra Black, Naomi Richards, and other participants at a workshop on Ethical Issues of Non-physician Assisted Suicide held at University College, Dublin. The Society for Applied Philosophy generously sponsored the workshop.

References Anderson, Elizabeth. 1990. Is women’s labor a commodity? Philosophy and Public Affairs 19: 71–92. Baumrin, Bernard. 1998. Physician, stay thy hand! In Physician Assisted Suicide: Expanding the Debate, eds. Margaret P. Battin, Rosamond Rhodes, and Anita Silvers, 177–181. New York: Routledge. Brennan, Jason, and Peter M. Jaworski. 2016. Markets Without Limits: Moral Virtues and Commercial Interests. New York: Routledge. Callahan, Daniel. 1992. When self-determination runs amok. The Hastings Center Report 22: 52–55. https://doi.org/10.2307/3562566. Cholbi, Michael. 2011. Suicide: The Philosophical Dimensions. Peterborough, Ontario: Broadview. Cholbi, Michael. (2015) The right to die and the medical cartel. Ethics, Medicine, and Public Health 1: 486–493. Crisp, Roger. 1987. A good death: Who best to bring it? Bioethics 1: 74–79. Davis, Nicola. 2019. Euthanasia and assisted dying rates are soaring. But where are they legal? The Guardian. Available at: https://www.theguardian.com/news/2019/jul/15/euthanasia-and-ass isted-dying-rates-are-soaring-but-where-are-they-legal. Accessed April 4, April. DeCou, Christopher. 2019. The ancient doctors who refused payment. Wellcome Collection. Available at: https://wellcomecollection.org/articles/Xc6-2REAACgABY_t. Accessed April 9, 2022. Groenewoud, A Stef, Femke Atsma, Mina Arvin, Gert P. Westert, and Theo A. Boer. 2021. Euthanasia in the Netherlands: A claims data cross-section study of geographical variation. BMJ Supportive and Pallative Care. https://doi.org/10.1136/bmjspcare-2020-002573. Hall, Mark A., and Carl E. Schneider. 2008. Learning from the legal history of billing for medical fees. Journal of General Internal Medicine 23: 1257–1260. https://doi.org/10.1007/s11606-0080605-1. Health Canada. 2020. Second Annual Report on Medical Assistance in Dying in Canada 2020. Available at: https://www.canada.ca/en/health-canada/services/medical-assistance-dying/annual-rep ort-2020.html. Accessed March 26, 2022. Kamm, Frances M. 2020. Five easy arguments for assisted suicide and the objections of Velleman and Gorsuch. In Almost Over: Aging, Dying, Death, 208–244. New York: Oxford University Press. Kass, Leon R. 2002. Life, Liberty, and the Defense of Dignity: The Challenge for Bioethics. San Francisco: Encounter Books. Melo, Paula, Costa Pedro, Maria Jose Quintas, et al. 2017. Pentobarbital in the context of possible suicides: Analysis of a case. Forensic Science International 274: 109–112. https://doi.org/10. 1016/j.forsciint.2016.11.042. Miller, Franklin G., and Howard Brody. 1995. Professional integrity and physician-assisted death. Hastings Center Report 25: 8–17.

278

M. Cholbi

Oregon Health Authority. 2021. Oregon Death with Dignity Act: 2021 Data Summary. Available at: https://www.oregon.gov/oha/PH/PROVIDERPARTNERRESOURCES/EVALUA TIONRESEARCH/DEATHWITHDIGNITYACT/Documents/year24.pdf. Accessed March 22, 2022. Ost, Suzanne. 2010. The de-medicalisation of assisted dying: Is a less medicalised model the way forward? Medical Law Review 18: 497–540. https://doi.org/10.1093/medlaw/fwq025. Raus, Kasper, Bert Vanderhaegen, and Sigrid Sterckx. 2021. Euthanasia in Belgium: Shortcomings of the law and its application and of the monitoring of practice. Journal of Medicine and Philosophy 46: 80–107. https://doi.org/10.1093/jmp/jhaa031. Satz, Debra. 2010. Why Some Things Should Not Be for Sale: The Moral Limits of Markets. Oxford: Oxford University Press. Seay, Gary. 2005. Euthanasia and physicians’ moral duties. Journal of Medicine and Philosophy 30: 517–533. https://doi.org/10.1080/03605310500253071. Sikka, Tina. 2019. Barriers to access: A feminist analysis of medically assisted dying and the experience of marginalized groups. Omega (Westport) 84: 4–26. Washington State Department of Health. 2021. 2020 Death with Dignity Act Report. Available at: https://doh.wa.gov/sites/default/files/2022-02/422-109-DeathWithDignityAct2020pdf? uid=6264c8c52f4ba. Accessed April 18, 2022.

Michael Cholbi is Chair in Philosophy at the University of Edinburgh. He has published widely in ethical theory, practical ethics, and the philosophy of death and dying. His books include Suicide: The Philosophical Dimensions (Broadview, 2011) and Grief: A Philosophical Guide (Princeton University Press, 2022). He is the editor of several scholarly collections and co-editor of the textbook Exploring the Philosophy of Death and Dying: Classic and Contemporary Perspectives (Routledge, 2020).

Chapter 16

Robots, AI, and Assisted Dying: Ethical and Philosophical Considerations Ryan Tonkens

Abstract The focus of this chapter is on some of the ethical and philosophical issues at the intersection of robotics and artificial intelligence (AI) applications in the health care sector and medical assistance in dying (e.g. physician-assisted suicide and euthanasia), including: (1) Is there a role for robotic systems/AI to play in the orchestration or delivery of assisted dying?; (2) Can the use of robotic systems/AI make the orchestration of assisted dying more ethical?; and (3) What insights can be generated in the ethical debate on physician assisted suicide and euthanasia from considering the prospect of robotic systems/AI assisting with the provision of or providing assistance in dying? The prospect of including robotic systems/AI in the context of assisted dying provides opportunity to revisit longstanding philosophical and ethical issues under new light. Indeed, reflecting on these questions may invigorate debate, for example in reconsidering the de-medicalization of assisted dying, reconsidering whether assisted dying is within the proper scope of medicine, and reconsidering which normative approach to the ethics of assisted dying is the most appropriate. Keywords Artificial Intelligence (AI) · Deep learning systems · Euthanasia · Medical assistance in dying · Robotics

16.1 Introduction Not much attention has been given to possible applications of research in robotics or artificial intelligence (AI) in the context of end of life care, including assistance in dying (e.g. voluntary euthanasia and physician-assisted suicide). Yet, given the enthusiasm in which medical applications of robotic systems/AI are being pursued R. Tonkens (B) Centre for Health Care Ethics, Lakehead University, Thunder Bay, ON, Canada e-mail: [email protected] NOSM University, Thunder Bay, Canada Department of Philosophy, Lakehead University, Thunder Bay, Canada © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 M. Cholbi and J. Varelius (eds.), New Directions in the Ethics of Assisted Suicide and Euthanasia, The International Library of Bioethics 103, https://doi.org/10.1007/978-3-031-25315-7_16

279

280

R. Tonkens

in other areas, and given that versions of assistance in dying are being legalized in a growing number of countries, are there morally compelling reasons (not) to allow robotic systems/AI applications in the context of assisted dying? The prospect of including robotic systems/AI in the context of assisted dying has the potential to expand the ethical and philosophical issues that require attention (e.g. who should be held responsible if a robot accidentally or on purpose ‘euthanizes’ a patient who does not wish to die?). It also provides opportunity to revisit longstanding philosophical and ethical issues under new light. Reflecting on these questions may invigorate debate, for example in reconsidering the de-medicalization of assisted dying, reconsidering whether assisted dying is within the proper scope of medicine, and reconsidering which normative approach to the ethics of assisted dying is the most appropriate, etc. With this in mind, the central focus of this chapter is on the following questions: Is there a role for robotic systems/AI to play in the orchestration or delivery of assisted dying? Can the use of robotic systems/AI make the orchestration of assisted dying more ethical? What insights can be generated in the ethical debate on physician assisted suicide and euthanasia from considering the prospect of robotic systems/AI assisting with the provision of, or providing, assistance in dying? Two broad categories for the inclusion of robotic systems/AI in the context of assisted dying will be considered, namely (1) in a supportive role, i.e. in assisting with assisted dying, and (2) in a more direct role, i.e. in providing assistance in dying to eligible people. Some examples of indirect applications include assessing a patient’s eligibility for receiving assistance in dying, and providing information about medical assistance in dying to patients who request it. Some examples of direct roles include prescribing a lethal dose of sodium pentobarbital, dispensing of such or similar, and euthanizing a person at their sound, legal request (e.g. through administering a lethal dose of nitrogen gas). In the final Sect. (16.5), I consider how certain applications of robotics systems/AI may help advance debate in the ethics of assisted dying without the machines themselves providing or assisting in the provision of assisted dying. The central conclusions drawn below include that the use of robotic systems/AI could be beneficial in terms of verifying eligibility for assistance in dying and patient consent, has the potential to make the provision of assisted dying more ethical (e.g. by augmenting protection for vulnerable people), and has the ethical advantage that it removes medical personnel from the picture without compromising patient care. While more work is needed in this area, it seems that the main ethical issues about the prospect of robotic assisted dying are not attached to the introduction of robotic systems/AI in this context, but rather are about the ethics of assisted dying itself, or the development and use of (semi-)autonomous robotic systems more generally.

16 Robots, AI, and Assisted Dying: Ethical and Philosophical Considerations

281

16.2 Recent Technological Developments and the Prospect of Robotic Assisted Dying There is some risk of the imagination running wild about what form such machines could take.1 A good gauge is the advanced applications of robotic systems/AI in the health care, elder-care and military sectors. These three areas of recent development motivate the ethical and philosophical questions discussed in this chapter, and highlight the importance and timeliness of such considerations.

16.2.1 Robotics and AI in Health Care The first recent development is the rapidly expanding scope of applications of robotic systems/AI in the health care sector, at least in many advanced health care systems in developed countries, where the requisite resources are available. Some of the main areas of application so far have included rehabilitation, diagnostics, and robotic surgery (e.g. Bartneck et al. 2021, 73). Recently, Stevie the robot has been used in the Intensive Care Unit (ICU) at the University of Pretoria hospital, to conduct rounds during the covid-19 pandemic Mnguni (2021).2 Elsewhere, AI systems have been developed for the purpose of helping to prevent the unnecessary escalation of care (Li et al. 2022). Another such application, which will be discussed later in this chapter, is the use of deep learning technology and predictive modelling to assist medical professionals in predicting mortality and illness progression, such that patients who would benefit from earlier discussions about options for palliative care and advanced care planning can be identified and approached accordingly. Also, research and development in the area of robotic surgery has been extensive, although not without attracting ethical debate. For example, some researchers are calling for human surgeons to stay “in-the-loop” in well-defined ways, “thereby safeguarding patients undergoing operations that are supported by surgical machines with autonomous capabilities” (O’Sullivan et al. 2018). Some others, however, are discussing the prospect of autonomous surgical robots for certain applications (e.g. Mughal 2022). Similarly, developments in the area of telemedicine abound. There could be ways in which robotic systems/AI could be used to assist trained medical professionals in delivering end of life care/assistance in dying remotely. One example of many is iRobot’s “Remote Presence Virtual + Independent Telemedicine Assistant”, or RPVITA. This robot has the capacity for navigation of the standard clinical setting (via mapping and obstacle avoidance technology), and can be connected with diagnostic 1

I encourage the reader to think more Bicentennial Man, rather than The Terminator. “Stevie has a privacy handset which is a live phone to aid confidential communication during ward rounds; a stethoscope port where it can remotely relay information while a patient is being examined, and it enables visualisation of detail for close-up diagnosis and patient care oversight with high-definition pan-tilt-zoom cameras”.

2

282

R. Tonkens

devices such as ultrasound via Bluetooth, and “can be integrated with live patient information from electronic health records” (Ackerman et al. 2010). With such technological advancement and widespread application to the health care setting, there is abundant opportunity for robotic systems/AI to connect with the context of assisted dying as well.

16.2.2 Robot Carers There has also been a lot of recent attention given to the use of robotic systems in the context of care for people who are elderly, in both long term care facilities and at the home of the person. In places like Canada and Japan, the elderly population is growing, and there are longstanding concerns about not having the requisite human and other resources to be able to provide adequate care for those people. One example is the use of robotic systems such as Pepper the robot. For example, Dr. Arshia Khan’s lab at the University of Minnesota has been researching applications for Pepper to be able to assist people living with dementia, and are equipped with the ability to detect signs of the early onset of dementia (e.g. through ability to identify slight modification in gait). Consider also the fictional example presented in the film Robot and Frank (2012). (The robot portrayed in the film is inspired by the Honda Asimo robot). Robot does daily chores for Frank, who is experiencing symptoms of dementia. The robot serves as a support for Frank, including providing reminders for activities and to take their pills, and also functions as a companion. Imagine that Robot could also provide information to Frank about medical assistance in dying (if Frank lives somewhere where this is legal, and if Frank requested such information), via connection to the internet and a tablet display monitor. Given advancement in sophistication in the future, perhaps Robot could be equipped with face recognition software and other similar detection capacities, in order to be able to conduct a capacity to consent assessment on Frank, as part of the screening process for formal requests for assisted dying. If it was ethically and legally appropriate, perhaps Robot could also assist more directly in the provision of assisted dying at Frank’s voluntary and informed request (and assuming Frank met all eligibility criteria for assistance in dying within their jurisdiction). There are certainly many ethical issues raised by the prospect of including robotic systems in elder care, and it is beyond the scope of this chapter to discuss them directly. I concur with van Wynsberghe (e.g. 2013, 2021) in advocating for the responsible development and use of such care robots, and think that their “carecentered value sensitive design” approach is promising for making sure that this happens. Elsewhere (Tonkens 2015) I have argued that the case for including robotic systems/AI in the context of assisted dying is stronger if such care robots are already being used, because part of the ethical and socio-political task is to make a compelling case for the research and development of the technology in the first place. If care robots are already being developed and used, an extension of their application to now

16 Robots, AI, and Assisted Dying: Ethical and Philosophical Considerations

283

include assistance in dying (or assistance with assisted dying), as part of a holistic approach to elder care/care at the end of life, is not as resource-intensive, compared to independently developing robots for the express purpose of assistance in dying.

16.2.3 Robots with Lethal Capabilities The third recent development is that robotic systems/AI are already being used in other contexts where the death of humans is a direct consequence. Indeed, in some cases the very purpose of the machine is to kill humans. Modern day media headlines tell often of the presence of semi-autonomous robotic systems being used in the context of armed conflict across the globe, some of which are designed with lethal capabilities. It may be responded that just because robotic systems are being used for the purpose of killing humans in warfare it does not mean that this is a proper use for technology, or that similar robots ought to be used in the context of assisted dying. I agree with this reply. However, this development does attest to technological feasibility, and encourages important ethical and philosophical debate in both the area of military robotics and the area of medical robotics. One of the main drivers behind the expanding automation of military operations is cost effectiveness and efficiency. The same sort of reasoning is present in the health care setting as well. In the context of modern health care delivery (in developed countries with robust health care systems), one of the driving forces behind introducing robotic systems/AI is that doing so promises to be more efficient, for example by robotic systems outperforming humans at particular tasks, or by opening up more human health care workers for different tasks by having robots conduct the more mundane ones. Estimates have been made to the effect that one (human) nurse monitoring blood pressure remotely via telemedicine technology can handle thousands of patients at once, rather than a couple dozen or less without such technological assistance (see e.g. Omboni and Ferrari 2015). While there are predictions being made to the effect that ‘machine learning will create $150 billion annually in health care saving in the United States by 2026’ (Masters 2020), there have also been calls for caution, highlighting issues about fairness in the distribution of the automation of health care (e.g. Henschke 2015). An early proponent of the use of semi-autonomous military robotic systems with lethal capabilities is Arkin (e.g. 2009), who argues that the use of such systems could elevate the humaneness of warfare. Ironically, replacing or supplementing human combatants with lethal robotic systems has the potential to make warfare and behaviour on the battlefield more humane (less inhumane), compared to warfare where such systems are not involved. It is interesting to consider whether this would be possible in the context of robotic assisted dying. On one hand, some may think that the inclusion of robotic systems/AI in the context of assisted dying could decrease the warmth and the potential for meaningful connection at the end of life, thereby establishing an isolating, potentially less dignified setting in which people would

284

R. Tonkens

experience the end of their lives (see e.g. discussion in Sharkey and Sharkey 2012). The degree to which the use of robotic systems in elder care could impact human dignity will of course depend on our conception of “dignity” (see e.g. Zardiashvili and Fosch-Villaronga 2020), and the specific uses of the technology and the degree to which its involvement is direct or indirect (see e.g. Laitinen et al. 2019). On the other hand, depending on the degree of sophistication of the machine, perhaps genuine relationships could be formed, as in the fictional story presented in Robot & Frank (see e.g. Weber-Guskar 2021). A recent survey of the use of social assistance robots during the covid-19 pandemic concludes that such technology holds “great promise in helping to reduce loneliness and improve psychological well-being, while also helping to minimize the burden on healthcare staff and caregivers” (Getson and Nejat 2021). Some of Arkin’s reasons for advocating the development of lethal robotic systems for military use could transfer to the context of assisted dying, whereas others do not transfer. For example, for Arkin, one example of behaviour on the battlefield which contributes to the inhumaneness of warfare is that sometimes innocent civilians and other non-combatants are either directly targeted or killed as collateral damage. According to Arkin, robotic systems of a certain sophistication could be better at discriminating between legitimate combatants and non-combatants amidst the fog of war. Part of the reason for this is their ability to process information extremely quickly, and also because they do not necessarily need to rush in making such decisions in order to ‘save their own life’. Generally speaking, human physicians have not reported difficulty in discerning which patients making a request for assistance in dying actually meet the eligibility criteria and which ones do not, and do not need to make such determinations quickly, and so this reason in favour of military robotics does not work in favour of robotic assisted dying. An example of a reason for using robotic systems in order to make warfare less inhumane that does transfer to the context of assisted dying is keeping certain human fallibilities and limitations out of the situation, such as fatigue and undesirable emotions (e.g. greed, revenge, etc.). Importantly, one reason why the use of robotic systems/AI in the context of assisted dying may be less ethically controversial than the use of lethal robotic systems in war is because providing a good death to those who have decided to die (voluntary euthanasia) is morally better than killing humans against their will (murder on the battlefield), all else being equal. Because of the three technological developments discussed above, the prospect of the use of robotic systems/AI in the context of assisted dying is a legitimate one, and important ethical and philosophical questions emerge and ought to be addressed. In what follows, I first consider the prospect of indirect (supportive) applications of robotic systems/AI in the context of assistance in dying (Sect. 16.3), before considering the more radical prospect of ‘killer robots’ in the context of assisted dying (Sect. 16.4).

16 Robots, AI, and Assisted Dying: Ethical and Philosophical Considerations

285

16.3 Indirect Applications of Robotic Systems in Assisted Dying There are several ways in which robotic systems/AI may be incorporated into the context of end of life care and assistance in dying that do not require the machine to do any killing. Elsewhere, I have suggested that applications for robotic systems/AI in the context of assisted dying could include the following: having the capacity “to (a) inform patients about their diagnosis, prognosis, and treatment options (something Watson is currently being programmed to do), (b) assess the patient’s degree of rational consent to the assisted dying […] and assess the level of understanding of the patient with regards to the relevant information—something busy human physicians may not always have the time to do satisfactorily; (c) ensure that the patient can self-administer the drugs; (d) notify the patient’s next of kin; and (e) distribute or prescribe the means and instructions necessary for aiding the patient in reaching their termination effectively, and in a way that helps maintain and promote the patient’s dignity” (Tonkens 2015, 211, emphasis in the original). Building on this earlier work, in this section I describe three specific ways in which robotic systems/AI may have an indirect role in the context of assisted dying. I have chosen to focus on these three possible applications because of their potential to augment the ethical delivery of end of life care in some way, for example by aligning practice more closely with current law on medical assistance in dying (e.g. in Canada), promoting the protection of the most vulnerable people, and augmenting precision in determining whether someone meets the eligibility criteria for receiving assistance in dying.

16.3.1 Reducing Vulnerability In Canada and other countries that have legalized versions of medical assistance in dying, there is a strong emphasis on doing what is necessary to ensure (as much as possible) that requests for assisted dying are free from external pressure, i.e. that the request is a voluntary one. One way in which robotic systems like Watson,3 Stevie, Pepper and Frank’s Robot could promote ethical delivery of health care in the context of assisted dying is to protect patients from being unduly pressured or coerced to request, or to follow through with a request, to receive assistance in dying. Imagine Robot in a surveillance role for determining whether there is any suspicious behaviour exhibited by other people in Frank’s life, who might be a catalyst for unduly pressuring or influencing Frank’s decision to request assistance in dying. Of course, there are important ethical issues that would accompany this prospect, 3

Since Watson’s achievements on the trivia game show Jeopardy, its roles have expanded significantly. At the time of writing, the applications listed on the Health section of the IBM Watson AI website include patient engagement, diagnostics, and clinical decision support.

286

R. Tonkens

including to do with the privacy of the people who are being monitored by the robotic system. Yet, this could be a straightforward way in which people like Frank could be protected from undue influence, protection built into the very robotic system that is already in use by Frank for other purposes to do with their ongoing care. Depending on the particular situation, some care robots may come to be very close companions to their users, and be in a privileged position to provide intricate details of the circumstances under which their user has come to the decision to make a formal request for assistance in dying. One issue worth flagging here is about trust. For example, many are concerned about whether the decisions made by robotic systems/AI are trustworthy (see e.g. Hatherley 2020), and are not instead based on inadequacies or mistakes in the machine’s programming (see e.g. Ratti and Graves 2022). While there is ongoing debate about the ‘black box of machine learning’ and lack of transparency in the decision making activity of AI, some arguments are appearing to suggest that such concerns are beyond the standards set for traditional medical technologies and, more importantly, that “ability to explain how results are produced can be less important than the ability to produce such results and empirically verify their accuracy” (London 2019). These issues are important, and cannot be settled here. Worth noting is that these issues are about the development and use of robotics/AI in general, rather than anything specific about the context of end of life care or assisted dying. Another way in which robotic systems like Watson and Pepper could promote ethical delivery of health care in the context of assisted dying is to help reduce susceptibility to vulnerability in this context. In anticipation of the decriminalization of medical assistance in dying (voluntary euthanasia and physician-assisted suicide) in Canada in 2016, a group of scholars and other experts developed the “Vulnerable Persons Standard” (VPS), which is a set of “safeguards to help to ensure that Canadians requesting assistance from physicians to end their life can do so without jeopardizing the lives of vulnerable persons who may be subject to coercion and abuse” (Vulnerable Persons Standard 2017). The law on medical assistance in dying in Canada was then measured against these safeguards, and recommendations were made by this group of experts about the extent to which the law met the standards, and where changes were needed or recommended. The use of robotic systems/AI could help with the delivery of assistance in dying to be more in line with the VPS, and because of this their use could help to promote the ethical delivery of health care and the protection of vulnerable people at the end of life. For example, the authors argue that the (2016) MAiD law is not compliant with the VPS because patient requests do “not require a prior review by an independent body” to determine whether the patient actually meets the eligibility requirements. Rather, the law only “requires two medical/nurse practitioners to provide written opinions confirming that the patient meets all criteria” which, according to VPS, is insufficient for appropriately reducing patient vulnerability. I contend that a sophisticated robotic system could serve as such an independent body. Outfitted with robust surveillance technology, a complete understanding of the law, and trained with case-based deep machine learning, etc., such a system could assess the competency of the person making the request, and confirm whether or

16 Robots, AI, and Assisted Dying: Ethical and Philosophical Considerations

287

not the person meets the eligibility criteria including diagnosis and prognosis, etc., such that it could act as a qualified independent body reviewing the opinions of the two medical professionals. This is not to propose the removal of the requirement of two human medical practitioners to also assess the patient’s eligibility, but rather to suggest that this could be an option for additional oversight without requiring additional human resources, and also meeting the VPS call for independent oversight in this area.

16.3.2 Assessment and Monitoring As mentioned above, robotic systems could help with the screening process of people who are requesting assistance in dying. This is especially useful in cases where more than one practitioner needs to make an independent assessment, and such practitioners are busy, and perhaps there are few or none of them available locally, for example in rural and remote communities. Such assessments could even be orchestrated remotely, and thus could be coordinated across distances and with fewer human resources required. Some communities could benefit from this as an option, in case they do not have easy access to medical professionals, or if those practitioners hold a conscientious objection to assisted dying. Insofar as the inclusion of robotic systems/AI made access easier or more likely for some groups of people, then this could increase the ethical delivery of health care, and boost equity and fairness. This boost in access to ‘qualified medical personnel’ could be supportive, such as the provision of information or assistance in contacting non-local medical professionals who do not hold a conscientious objection to assisted dying, or also more direct roles as well.

16.3.3 Predicting End of Life A foundational tenet of the law on medical assistance in dying in Canada is that the person requesting assistance in dying must have consented to it only after receiving all of the information necessary to make an informed and voluntary decision, including information about their diagnosis, treatment options available, and “available options to relieve suffering, including palliative care” (Government of Canada 2021). Importantly, the patient must be aware of options for pain relief and palliative care, aside from pursuing a hastened death (including terminal palliative sedation). One promising line of robotics/AI research that is especially relevant to this aspect of the Canadian law is current work being done at Stanford University on the use of AI and deep learning for predicting probability of mortality during in-hospital stay. In a 2020 article in the Journal of the American Medical Association, researchers reported that ‘a machine learning algorithm was integrated into the electronic health record to generate accurate predictions in real time of short-term (3–12 months) mortality for

288

R. Tonkens

patients with cancer, and outperformed routinely used prognostic indices’ (Manz et al. 2020). According to these authors, “this algorithm may be used to inform behavioral interventions and prompt earlier conversations about goals of care and end-of-life preferences among patients with cancer” (Manz et al. 2020). According to Courtright et al. (2019): “Targeting hospital-based palliative care using an [Electronic Health Record] mortality prediction model is a clinically promising approach to improve the quality of care among seriously ill medical patients”. While more research is needed, there is promise of potential practical use of such machine learning in the context of palliative care and advanced care planning. To this extent, the use of robotic systems/AI could help medical practitioners more robustly conform to the stipulations of the law on medical assistance in dying by offering more timely and precise information about the probable timing of the patient’s death, and thereby promoting patient autonomy. Not only can such technology more precisely identify those whose death is and is not “reasonably foreseeable”, but they would also be in a better position to discuss options sooner with those in need, including options for palliative care. In this section, we have seen several potential ways that robotic systems/AI could play a role in assisting with the provision of assistance in dying, focusing on specific ways that could augment the ethical delivery of health care in this context. It is fair to speculate that there are more possibilities beyond these. However, when so many people continue to go without adequate access to even basic health care, including consistent access to qualified medical professionals, might there be reasons grounded in fairness, equity and the just allocation of medical resources which speak against spending the money and time necessary for developing such robotic systems/AI, especially for use in assistance in dying (in developed health care systems in more affluent countries)? These resources can certainly be used elsewhere, including in ways that are meant to be to the benefit of the least well-off, including those with least access to the social determinants of health. At least, proponents of developing the technology for robotic systems/AI applications in the context of assisted dying would need to show that the overall benefits are compelling, compared to spending those resources to help combat other important issues in health care. Because of this, I continue to hold that the case in favour of such applications of robotic systems/AI is augmented to the extent that those systems are already being used in the relevant contexts (e.g. elder care, ICU settings, telemedicine, etc.). To this I would add that there is a moral imperative that the development and use of such technology be done so responsibly, and that an important component of this is that, ideally, the motivations for developing the technology and applications of it be thoroughly inclusive, and are not restricted to the problems of affluent and privileged people.

16 Robots, AI, and Assisted Dying: Ethical and Philosophical Considerations

289

16.4 Robotic Assisted Dying? Machines have been used in the context of health care delivery at the end of life for a very long time. For example, the mechanical ventilator, going at least as far back as the “iron lung tank respirator” being used in the early 1900s. Modern day mechanical ventilators control a patient’s body in such a way so as to make breathing possible for that patient, and in that sense assist the patient in remaining alive (e.g. Schneiderman and Spragg 1988). A (human) medical professional connects the patient to the ventilator, and the ventilator will continue to maintain the patient’s breathing artificially, unless it is disconnected or turned off (or if the mechanical ventilator malfunctions in some way). Inclusion of mechanical ventilation systems in the context of patient care has largely been welcomed. There is of course a significant gap between the human use of machines such as mechanical ventilation and the use of robotic systems/AI in some way to help facilitate assistance in dying. It may be surprising to hear that, while the mechanical ventilator is being widely used to help promote life and the restoration of health, other technologies are being developed to help provide, or assist in the provision of, assistance in dying. An early iteration of such machines is Kevorkian’s Thanatron and, later on, the Mercitron. These are mechanical contraptions that are controlled by a human user; the human user (e.g. the patient) seeking to die presses a button or releases a latch and a substance or gas is emitted in a manner that is lethal to that patient. In line with Kevorkian’s machines but with a higher degree of automation (and extravagance in design!), Philip Nitschke has been working to develop an “assisted suicide pod” called “Sarco”. For decades, Nitschke has been working as an advocate for patient self-determination at the end of life, and for allowing people to orchestrate the timing and manner of their death. One element of Nitschke’s advocacy is a book called “The Peaceful Pill Handbook” (originally published in 2006, co-authored with Fiona Stewart), which provides information to people considering hastening their death, including information about potential methods for doing so effectively. In 2019, what had previously seemed to be nothing more than a controversial art instillation being displayed in a couple of museums in Europe, received international media attention as a prototype device for facilitating suicide. The user enters the pod and rests horizontally (as if lying on a sofa), then “will be asked a number of questions and when they have answered, they may press the button inside the capsule activating the mechanism in their own time” (O’Dea 2021). When the button is pressed, this activates a release of a lethal dose of nitrogen gas. At the time of writing, the Sarco device is still in the prototype stage, but, according to Nitschke, its use is in line with law about assistance in dying in Switzerland. The Sarco suicide pod is 3-D printable, which means that anyone with access to a 3-D printer and the necessary materials could print their own device.4 In the remainder of this section, I focus on two potential benefits of including robotic systems/AI directly in the context of assistance in dying. First, developing 4

It is not clear where the user will acquire the nitrogen.

290

R. Tonkens

advanced robotic systems/AI for providing assistance in dying could allow opportunity for those who do not otherwise qualify for access to legal provision of assistance in dying to have access. Secondly, the inclusion of robotic systems/AI in the provision of assistance in dying adds distance between the medical profession, whose traditional ethos is thought by many to preclude the intentional killing of patients, and the provision of assistance in dying, since robots, rather than human medical professionals, would be the ones assisting people to die.

16.4.1 Widening the Scope of Eligibility for Assisted Dying Physician assisted suicide and euthanasia are illegal in most countries in the world. Even in countries where these practices are legally available (e.g. Australia, Canada, Columbia), access to assisted dying is typically restricted to people who have been diagnosed to be near the end of their life (e.g. patients who are suffering from a terminal condition, or whose death is “reasonably foreseeable”). Exit International (Nitschke’s non-profit organization) is founded on the belief that ‘all adults of sound mind have a fundamental human right to implement plans for the end of their life so that their death is reliable, peaceful and at a time of their choosing’ (Exit International, n.d.). The view at the foundation of this claim, of course, continues to be one of the most contentious aspects of ongoing debate in the ethics of physician assisted suicide and euthanasia. In places like Canada, while medical assistance in dying is now legal, current law does not represent the view that all adults have a fundamental human right to access assistance in dying. One of the central catalysts for the perceived need for Kevorkian’s Thanatron and Nitschke’s Sarco is that some people desire to hasten their death, but pursuing this option through the mainstream medical system is unavailable to them. This could be for one of two reasons: Either the person does not meet the eligibility conditions enforced in their country (and cannot travel elsewhere), or in case assisted dying in any form is not legally available where they reside (and they cannot travel elsewhere). One potential role for (appropriately sophisticated) robotic systems/AI could be to provide assistance in dying to people who fall outside of the scope of the provision within the dominant medico-legal paradigm, and yet who have decided that death would be a good thing for them. Perhaps they are suffering immensely but their death is not “reasonably foreseeable”. Perhaps they have a debilitating disease and would like the peace of mind that they have the option of painless and effective method of hastening their death at a time of their choosing, without unduly implicating their family or friends.5 5

There are huge philosophical questions here about how to decide whether death is a good for someone. Answering this question is not requisite in order to determine the ethics of robotic assisted dying, and attaches to the legal provision of assistance in dying more generally. I assume that there is nothing about someone dying by robotic assisted dying, rather than standard medical assistance in dying, that would make it less likely that their death can be a good for them.

16 Robots, AI, and Assisted Dying: Ethical and Philosophical Considerations

291

Of course, there are alternatives to the inclusion of robotic systems/AI that could also allow for the expansion of the kinds of people who have access—it is not suggested here that robotic systems are the only way to give access to those who do not otherwise qualify. In Switzerland, for example, it is volunteers outside of the medical establishment who provide assistance in dying, e.g. through organizations such as Dignitas, Exit Deutsche Schweiz, or Ex International. These volunteers are “clergy persons, social workers, and nurses assisting in the final act of suicide, rather than doctors” (Ost 2010, p. 498). Philosophically, the prospect of including robotic systems in the context of the provision of assistance in dying re-introduces longstanding and important questions about the scope of human rights in the context of assistance in dying (i.e. should all adults have access to assistance in dying?), the justification for restricting access to people whose death is not reasonably foreseeable, and the justification for (not) restricting the legally sanctioned provision of assistance in dying to hospitals and health care practitioners—especially in the face of historical and contemporary hesitation by some medical professionals to include “killing” patients within the proper scope of those professions. If robotic systems/AI can (help to) perform the necessary actions and provide adequate protection and oversight, are there morally compelling reasons to (not) allow this practice, as a possible alternative stream for safe and effective assistance in dying to those who do not qualify for access to the services available within the dominant medico-legal paradigm regulating medical assistance in dying? Importantly, given the (future) prospect of this alternative stream, the status quo ought not be assumed out of hand, and thus an answer to the above questions would be needed.

16.4.2 The De-medicalization of Assisted Dying This introduces a second potential reason in favour of including robotic systems/AI as the providers of assistance in dying to eligible patients: Doing so would thicken the wedge between satisfying the preferences of people who want to die, and the potential implications doing so has with respect to tarnishing the reputation and traditional ethos of the medical profession. In the inaugural issue of the journal Bioethics in 1987, Roger Crisp wrote a short article entitled “A Good Death: Who Best to Bring it?” (Crisp 1987), in which they asked whether physicians ought to be the ones to provide euthanasia (‘the good death’), or else if there are other parties who are more suitable for this role. Crisp considered several candidates to bring a good death to eligible patients, including the family or friends of the patient, and what Crisp calls “ordinary medical practitioners”, by which they mean doctors and nurses. On Crisp’s view, family or friends do not typically have the requisite knowledge and qualifications, and there is the threat of experiencing retrospective doubt or anxiety, and possible trauma resulting from being the cause of the death of a loved-one.

292

R. Tonkens

For Crisp, since euthanasia is understood to be a medical procedure, perhaps it is best that qualified medical professionals be the ones to provide it. However, there are potential problems with having medical professionals be the ones to provide assistance in dying. According to Crisp, being trained in the medical profession and being less acquainted with the patient (compared to the patient’s family) might make the act of killing the patient even more traumatic for the ordinary medical practitioner, compared to the family. Moreover, Crisp (1987) argues that “it is in the interests of other patients, among whom we might find ourselves, that the dispositions to save lives of medical practitioners not be weakened” (77). In addition to these concerns noted by Crisp, there are of course additional concerns with restricting the provision of assisted dying to within the medical establishment. For example, there is typically a ‘wiling physician clause’ in laws that permit assistance in dying (see e.g. Dworkin et al. 1997), such that the putative right to determine the timing and manner of one’s death depends crucially on the willingness and availability of qualified medical professionals. In places like Canada, the legal provision of medical assistance in dying must involve at least two qualified medical professionals. Many medical professionals and medical institutions do not support the current decriminalization of medical assistance in dying, and current law explicitly includes a stipulation that no one can be forced to provide it—medical professionals and institutions are free to not accommodate requests for assisted dying for reasons of conscience.6 Because of the concerns with family/friends and with ordinary medical professions being the providers of euthanasia, Crisp suggests that a new sub-specialization within medicine would be more suitable for providing euthanasia, although recognizes that these specialists may come to be viewed by the public as “executioners”. Crisp calls members of this proposed new sub-specialty in medicine “telostricians”. Of course, Crisp imagined human telostricians. The discussion above motivates a reconsideration of the extent to which assisted suicide and euthanasia ought to be “medicalized”, or else if there is good reason to shift to a model of de-medicalized assistance in dying. De-medicalization of assisted dying means its orchestration would be the responsibility of some entity other than the state medical system and its institutions and licensed professionals. De-medicalization of death could mean, for example, dying at home rather than in a hospital. (Care robots could be a useful tool here.) According to Ost (2010, 498–99): “legal, ethical, and social discourses surrounding assisted dying and laws that have permitted assisted dying have tended to focus on the assistance of doctors, the provision of medicine to cause death, and medical grounds for requesting death, that is, pain and suffering derived from medical conditions. As such, medicine has provided the main frame of reference, a vital component of the phenomenon of assisted death” (503).

6

“The federal legislation does not force any person to provide or help to provide medical assistance in dying”. However, in provinces such as Ontario, conscientiously objecting medical professionals are required to make a swift referral on to a qualified medical professional who will not object.

16 Robots, AI, and Assisted Dying: Ethical and Philosophical Considerations

293

A profound argument in favour of the decriminalization of assisted dying, and a main argument in the ethical debate on this topic, is that determining the timing and manner of one’s death is thought to be a right (protected liberty) that people have. The medical institution has tended to want to keep distance between its work and the deliberate killing of people, in part because it is believed that such practices are not in line with the proper purpose of the medical profession. The tension that emerges is one based on deeply held and competing values, which it has been challenging to harmonize, perhaps even impossibly so (see e.g. Cholbi and Valerius 2015). Moreover, in places where voluntary euthanasia and/or physician assisted suicide are legally permitted, such laws are often accompanied with restrictions, attempting to balance these competing values without letting either side become too pronounced (lest the other side object). Allowing for medical professionals to conscientiously object to the provision of medical assistance in dying is a good example of this attempt at balancing patient rights and also respecting the personal and professional values of physicians. Another example is limiting eligibility to people who are already dying, for the work of the medical professions is more futile in those cases. It seems that deviating from this “compromise view” would require at least some concession from one or even both sides. However, this is only the case if the legal provision of assistance in dying is restricted to its provision by medical professionals. Robotic systems/AI need not be medical professionals, per se, even though it may be possible for them to be sufficiently qualified to provide or assist with the provision of assistance in dying. And, robotic systems/AI do not have—or need to be programmed to have—personal or professional “conscience”. As long as there is no logical inconsistency, it is straightforward to program a robotic system to demonstrate ‘compassion’ and care towards human patients in their care and to perform services that assist that patient in dying. In the context of robotic assisted dying, conscientious objection does not emerge. One upshot of this is that the use of robotic systems/AI in the provision of assistance in dying need not compromise either patient autonomy or the fundamental values of the medical profession. (It seems to me that this is so even more than would be the introduction of a new sub-specialty within medicine, like Crisp’s telostricians, and more so than volunteers who are current or former practicing health care professionals, as in Switzerland). This is because robotic assisted dying could be open to all competent adults, and the robot has no allegiance to the profession of medicine and no values or interests of its own that conflict with providing safe and effective assistance in dying to eligible patients, at their voluntary and informed request. This would be the reason, or one of the reasons, for the existence of such robotic systems/AI in the first place, and ingrained at the very basis of their programming. Therefore, in places that decide to shift to a de-medicalized model of assisted dying, including places that decide to include an alternate stream for legal assisted dying outside of the dominant medico-legal paradigm and its institutions, robotic systems/AI could have an important role to play.

294

R. Tonkens

16.5 Ethics, Machine Learning, and Euthanasia Consideration of the inclusion of robotic systems/AI in the context of assisted dying holds potential for revitalizing certain aspects of longstanding debate on the ethics of assisted dying. For example, there is a lot riding on which normative framework is appealed to when examining the ethics of assisted suicide and voluntary euthanasia, but also in how the determination of ethical acceptability/impermissibility is generated (e.g. by human philosophers and bioethicists), and how concepts are defined and values weighed, etc. A classic example of a moral framework that could be implemented into the design of medical robotic systems/AI is Asimov’s Laws of Robotics: Law 1: A robot may not injure a human being or, through inaction, allow a human being to come to harm; Law 2: A robot must obey the orders given it by human beings except where such orders would conflict with the First Law; Law 3: A robot must protect its own existence as long as such protection does not conflict with the First or Second Laws. Consistency would require that the normative framework implemented into the design of robotic system itself allow for the type of action performed by that actor (Tonkens 2009). At first glance, Asimov’s Laws of Robotics would not allow for the use of robotic systems/AI for the purpose of voluntary euthanasia or assisted suicide, insofar as these activities entail injuring or harming a human being, and thus would constitute violations of the First Law of Robotics. This introduces a longstanding philosophical issue in the context of assisted dying, namely what counts as a good death: Does voluntary medical assistance in dying entail injuring or harming a human being? Of course, if one normative framework does not allow for the creation or use of robotic systems for the purpose of assisted dying, then we could look for a different framework to implement into such robots to guide their behaviour and learning, one that does allow for this. One concern with this is that, once the machine is running its own learning and considering the moral prescriptions that are generated by its own analysis of new cases, the robot may come to disagree. Indeed, it may come to very different conclusions about the ethics of assisted dying, including the values that ought to be considered when making such determinations and how they ought to be ranked. This could be a good thing. If the Watson robot or Deep Blue were trained to analyze the ethics of assisted dying and voluntary euthanasia, what would they decide, after (say) millions or billions of case scenarios are applied to a specific normative framework? We already see in these systems the amazing capacity to analyze complex cases and concepts, and this could be extended to include moral principles and values. It might be a very fruitful line of research to run machine learning and analysis on the longstanding ethics questions connected to voluntary euthanasia and physician assisted dying, to see what results the machine generates.

16 Robots, AI, and Assisted Dying: Ethical and Philosophical Considerations

295

One hypothesis is that machines may be better equipped to discover answers to the trickiest questions in the grey areas of ethics, compared to humans. One example here is the attempt at balancing or harmonizing the principles of respect for autonomy and beneficence (let alone to also balance these with other principles like justice, and perhaps with considerations of moral character, etc.). One important critique of the principlism of Beauchamps and Childress is that it does not offer any (clear, uncontentious) guidance in the grey cases where conflict emerges between respect for autonomy and beneficence, for example when there are very high, significant autonomy interests at stake and very high, significant potential benefit of intervening to overrule patient autonomy. The work of Anderson and Anderson (2007, 2015) is a good illustration of this potential use of AI, although admittedly it is still quite early in its development. These researchers are optimistic that applications of machine learning techniques can discover “philosophically novel and interesting principles in ethics because the learning system is general enough that it can be used to learn relationships between any set of prima facie duties where there is a consensus among ethicists as to the correct answer in particular cases” (2007). Anderson and Anderson have developed two machine learning systems that are worth noting. The first is called MedEthEx, which is a medical ethics advisor system, able to discover and articulate new ethical principles (such as “A health care worker should challenge a patient’s decision if it isn’t fully autonomous and there’s either any violation of nonmaleficence or a serious violation of beneficence”), as well as apply that principle to novel cases and generate recommendations for action. The second is called EthEl, which “is a system in the domain of elder care that determines when a patient should be reminded to take medication and when a refusal to do so is serious enough to contact an overseer”. Humans are certainly capable of making decisions and offering moral justification for those decisions. However, even those most familiar with the concepts, values, principles, and details of the cases in the context of the ethics of assisted dying, and the best trained to critically engage with the philosophical and ethical issues that emerge in this area, may pale in comparison to the abilities of a supercomputer with capacity for deep learning and data analysis. Either the robotic system/AI would come to confirm what we humans have determined about the ethics of assisted dying, or else it would generate a different view. No matter what, the findings would be momentous.

16.6 Conclusion The prospect of robotic assisted dying is forged out of many fires, some of which are new, and others which have been burning for a long time. Recent developments in robotics/AI research in other sectors (e.g. surgery, military, etc.) mean that the technology is becoming sophisticated enough such that robotic systems/AI for the purpose of providing, or assisting in the provision of, assisted dying is a realistic possibility in the future. Seemingly never-ending debate on old disagreements about

296

R. Tonkens

(e.g.) whether euthanasia and assisted suicide are morally acceptable practices, and whether the intentional killing of patients can have a legitimate place within the mainstream medical profession, render alluring the prospect of new ways to attempt to promote people having control of the timing and manner of their death and respecting the personal and professional values of medical professionals. Once we have turned to consider this new possibility, some important and unexpected insights emerge, three of which have been central to the analysis here. First, there are several concrete ways—and many more not considered here, or not yet conceived of—for robotic systems/AI to contribute to the orchestration of end of life care that could increase the ethical delivery of health care, including the protection of vulnerable people, and which do not directly involve the robotic system doing any killing. Secondly, direct inclusion of sufficiently sophisticated robotic systems/AI in the provision of assisted dying means that certain pronounced aspects of the ethics of assisted dying on a medicalized model—and the ethical debate based on that model—do not come up at all, including strict restrictions on the scope of people eligible for receiving assisted dying in order to protect the values and tradition of the medical profession and its constituents, and practical and ethical issues to do with the ‘willing physician clause’ and allowing for conscientious objection to the provision of assisted dying. Lastly, with respect to the theoretical aspects of debate on the ethics of voluntary euthanasia and physician assisted suicide, designing robotic systems/AI with the purpose of examining the ethics of assisted suicide is an interesting line of research at the intersection of applied robotics/machine systems learning and philosophy/bioethics, and it is exciting to think of the possibility that the outputs generated could add important nuance, deepened understanding and perhaps even answers to debate on these important and complex topics.

References Ackerman, M.J., R. Filart, L.P. Burgess, I. Lee, and R.K. Poropatich. 2010. Developing next-generation telehealth tools and technologies: Patients, systems, and data perspectives. Telemedicine and e-Health 16 (1): 93–95. Anderson, S.L., and M. Anderson. 2015. Towards a principle-based healthcare agent. In Machine Medical Ethics, 67–77. Cham: Springer Anderson, M., and S.L. Anderson. 2007. Machine ethics: Creating an ethical intelligent agent. AI Magazine 28 (4): 15–15. Arkin, R. 2009. Governing Lethal Behavior in Autonomous Robots. Chapman and Hall/CRC. Bartneck, C., C. Lütge, A. Wagner, and S. Welsh. 2021. An Introduction to Ethics in Robotics and AI. Springer Nature. Cholbi, M., and J. Varelius. 2015. New Directions in the Ethics of Assisted Suicide and Euthanasia, vol. 64. Springer. Courtright, K.R., C. Chivers, M. Becker, S.H. Regli, L.C. Pepper, M.E. Draugelis, and N.R. O’Connor (eds.). 2019. Electronic health record mortality prediction model for targeted palliative care among hospitalized medical patients: A pilot quasi-experimental study. Journal of General Internal Medicine 34 (9): 1841–1847. https://doi.org/10.1007/s11606-019-05169-2. Epub 2019 Jul 16. PMID: 31313110; PMCID: PMC6712114. Crisp, R. 1987. A good death: Who best to bring it? Bioethics 1 (1): 74–79.

16 Robots, AI, and Assisted Dying: Ethical and Philosophical Considerations

297

Dworkin, R., T. Nagel, R. Nozick, J. Rawls, T. Scanlon, and J.J. Thomson. 1997. Assisted suicide: The philosophers’ brief. The New York Review of Books 41–47. Exit International (n.d.). https://www.exitinternational.net/about-exit/our-philosophy/. Accessed January 31, 2023. Getson, C., and G. Nejat. 2021. Socially assistive robots helping older adults through the pandemic and life after COVID-19. Robotics 10 (3): 106. Government of Canada. 2021. Medical Assistance in Dying: https://www.canada.ca/en/health-can ada/services/medical-assistance-dying.html#consent1 Hatherley, J.J. 2020. Limits of trust in medical AI. Journal of Medical Ethics 46 (7): 478–481. Henschke, A. 2015. Opportunity costs: Scarcity and complex medical machines. In Machine Medical Ethics. Intelligent Systems, Control and Automation: Science and Engineering, vol. 74, eds. S. van Rysewyk, M. Pontier. Cham: Springer. https://doi.org/10.1007/978-3-319-081 08-3_9 Laitinen, A., M. Niemelä, and J. Pirhonen, J. 2019. Demands of dignity in robotic care: Recognizing vulnerability, agency, and subjectivity in robot-based, robot-assisted, and teleoperated elderly care. Techné: Research in Philosophy and Technology 23 (3): 366–401. Li, R.C., M. Smith, J. Lu, A. Avati, S. Wang, W.G. Teuteberg, K. Shum, G. Hong, B. Seevaratnam, J. Westphal, M. Dougherty, and N.H. Shah. 2022. Using AI to empower collaborative team workflows: Two implementations for advance care planning and care escalation. NEJM Catalyst Innovations in Care Delivery 3 (4): CAT-21. London, A.J. 2019. Artificial intelligence and black-box medical decisions: Accuracy versus explainability. Hastings Center Report 49 (1): 15–21. Manz, C.R., J. Chen, M. Liu, et al. 2020. Validation of a machine learning algorithm to predict 180-day mortality for outpatients with cancer. JAMA Oncology 6 (11): 1723–1730. https://doi. org/10.1001/jamaoncol.2020.4331. Master, B. 2020. The AI doctor will see you now. Financial Times. Available at: https://www.ft. com/content/d0aeeec8-5703-11ea-abe5-8e03987b7b20 Mnguni, H. 2021. University of Pretoria News. Available at: Stevie the robot joins intensive-care team at UP Faculty of Health Sciences and Steve Biko Academic Hospital. University of Pretoria. Accessed May 2, 2022. Mughal, H. 2022. 310 ethical challenges facing autonomous surgical robots. British Journal of Surgery 109 (Supplement 1): znac039–205. O’Dea, C. 2021. Sarco suicide capsule hopes to enter Switzerland. Swiss Info. Available at: https:// www.swissinfo.ch/eng/sarco-suicide-capsule--passes-legal-review--in-switzerland/46966510 Omboni, S., and R. Ferrari. 2015. The role of telemedicine in hypertension management: Focus on blood pressure telemonitoring. Current Hypertension Reports 17 (4): 1–13. Ost, S. 2010. The de-medicalisation of assisted dying: Is a less medicalised model the way forward? Medical Law Review 18 (4): 497–540. O’Sullivan, S., N. Nevejans, C. Allen, A. Blyth, S. Leonard, U. Pagallo, K. Holzinger, A. Holzinger, M.I. Sajid, and H. Ashrafian. 2018. Legal, regulatory, and ethical frameworks for development of standards in artificial intelligence (AI) and autonomous robotic surgery. The International Journal of Medical Robotics and Computer Assisted Surgery 15 (1): e1968. Ratti, E., and M. Graves. 2022. Explainable machine learning practices: opening another black box for reliable medical AI. AI and Ethics 1–14. Schneiderman, L.J., and R.G. Spragg. 1988. Ethical decisions in discontinuing mechanical ventilation. New England Journal of Medicine 318 (15): 984–988. Sharkey, A., and N. Sharkey. 2012. Granny and the robots: Ethical issues in robot care for the elderly. Ethics and Information Technology 14 (1): 27–40. Tonkens, R. 2009. A challenge for machine ethics. Minds and Machines 19 (3): 421–438. Tonkens, R. 2015. Ethics of robotic assisted dying. In Machine Medical Ethics, 207–221. Cham: Springer. van Wynsberghe, A. 2013. Designing robots for care: Care centered value-sensitive design. Science and Engineering Ethics 19 (2): 407–433.

298

R. Tonkens

van Wynsberghe, A. 2021. Social robots and the risks to reciprocity. AI & SOCIETY 1–7. Vulnerable Persons Standard. 2017. Available at: http://www.vps-npv.ca/read-the-standard-20 Weber-Guskar, E. 2021. How to feel about emotionalized artificial intelligence? When robot pets, holograms, and chatbots become affective partners. Ethics and Information Technology 23 (4): 601–610. Zardiashvili, L., and E. Fosch-Villaronga. 2020. “Oh, dignity too?” said the robot: Human dignity as the basis for the governance of robotics. Minds and Machines 30 (1): 121–143.

Ryan Tonkens earned a Ph.D. in Philosophy from York University (Toronto) in 2012. Dr. Tonkens is a bioethicist and researcher at the Lakehead University Centre for Health Care Ethics, Associate Professor in Bioethics and Philosophy at the Northern Ontario School of Medicine, and Adjunct Professor in the Department of Philosophy at Lakehead University. The scope of Dr. Tonkens’ research in theoretical ethics, applied ethics and bioethics is wide, including topics in reproductive ethics, ethics of AI and robotics, ethics of human gene editing (e.g. “Vulnerable groups and the hollow promise of benefit from human gene editing”, Bioethics 2021), and ethical issues in the context of organ recruitment and transplantation (e.g. “The most good you can do with your kidneys: Effective altruism and the organ-shortage problem”, Journal of Medicine & Philosophy 2021). Dr. Tonkens is an award-winning teacher.

Chapter 17

Dying to Live: Transhumanism, Cryonics, and Euthanasia Adam Buben

Abstract It might seem counterintuitive to think transhumanists, who are typically characterized by extreme techno-optimism and hope for radical life-extension, would be interested in assisted dying. Because the technological enhancements they long for will probably not be available during their natural lifetimes, many transhumanists at least entertain the idea of having themselves cryonically preserved to buy some additional time for real-world technology to catch up to their dreams. However, since an ideal preservation would take place before serious cellular deterioration sets in, controlling the time and manner of death would be very attractive for transhumanist cryonicists. This chapter identifies some circumstances under which they might be justified in seeking assistance in dying (even though they would describe it as something more akin to hibernating). After providing a fundamental overview of both transhumanism and cryonics, and then making my case for what some scholars have called “cryothanasia” (i.e. assisted dying for the purpose of improved preservation quality), I proceed to consider some scenarios meant to test the limits of this justification. While things are admittedly murkier when considering the possibility of cryothanasia outside the context of imminent medical collapse, I can see no insurmountable criticisms of cryothanasia as a last resort in cases of (currently) terminal and irreparable neurodegeneration or other forms of progressive brain damage. Keywords Alcor · Assisted dying · Cryonic wager · Reanimation · Suicide

17.1 Introduction At first glance, it might seem that transhumanists would have little interest in receiving assistance in ending their lives. Such assistance is typically regarded as a last resort for relieving otherwise inescapable and intolerable suffering, so why would people characterized by tremendous optimism about what near-future technology might make possible—which includes hope for radical life-extension—focus their attention A. Buben (B) Leiden University, Leiden, The Netherlands e-mail: [email protected] © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 M. Cholbi and J. Varelius (eds.), New Directions in the Ethics of Assisted Suicide and Euthanasia, The International Library of Bioethics 103, https://doi.org/10.1007/978-3-031-25315-7_17

299

300

A. Buben

on ways to end it all? This is not to suggest that transhumanists are unconcerned about suffering and all the ways it shows up in the world; in fact, concern about the misery ordinary humans have experienced throughout history is among the primary motivators for much of the transhumanist agenda. What sets them apart, however, is that most transhumanists think just-over-the-horizon technological developments will put a stop to much, if not all, of our misery without needing to put us out of our misery. Of course, problems arise when that horizon turns out to be a bit farther away than they had initially hoped and predicted, and it is in response to these problems that a connection between transhumanism and various forms of assistance in dying comes into view. For those of us unlikely to survive long enough to see the promised technological wonders, transhumanists think there might be a way to buy some extra time: cryonic preservation. As in the case of the procurement of functional organs for transplant, however, the preservation process works best if it is begun before serious deterioration of the tissue takes place, and this fact makes controlling the time and manner of death very attractive for transhumanists interested in cryonics. Building on a fairly small body of relevant existing literature, this chapter argues that—at least under certain circumstances—these transhumanists might be justified in seeking assistance in dying (and perhaps even dying via the cryonics procedures themselves) in order to increase the quality of preservation. But before getting into this argument, it will be necessary to provide a fundamental overview of transhumanism, cryonics, and the relationship between them.

17.2 What Is Transhumanism? Transhumanism can be difficult to characterize given the wide range of interests and individuals associated with it. As the term itself suggests, most of the scientists, philosophers, futurists, and science fiction enthusiasts who consider themselves transhumanists think humans can (and should be trying to) transcend their ordinary limitations through technological enhancements.1 In the words of one commentator: “While definitions and strains of transhumanism vary, at the most basic level, transhumanism is an intellectual and cultural movement premised upon the idea that human beings can use science and technology to significantly enhance their capabilities and thereby overcome many of the limitations of human biology” (Huberman 2018: 50). While this basic description might be sufficient for many self-proclaimed transhumanists, plenty of others would add “…with the ultimate goal of transitioning into some kind of posthuman being.” This posthuman being might very well constitute a 1

Although transhumanists encourage technological enhancement, nearly all of them also advocate for individual freedom and autonomy when it comes to biomedical choices and enhancements, which makes it highly unlikely they would oppose assisted dying (see e.g. Manzocco 2019: 34–5; Sandberg 2013; Swan 2019: 726–7). On their general openness to choosing suicide, see, among many others, Jenny Huberman (2018: 57) and Tena Thau (2020: 641).

17 Dying to Live: Transhumanism, Cryonics, and Euthanasia

301

new, likely inorganic, species with abilities far beyond those we biological humans currently possess. As Roberto Manzocco (2019: 33) explains, A possible post-human being should…possess an intelligence superior to that of any human genius past or present, as well as being far more resistant than us to diseases and aging….a post-human being should also have direct control over their own desires and moods; the ability to avoid tiredness, boredom, unpleasant emotions and sensations; to adjust their sexual inclinations to their liking; to accentuate their hedonistic and aesthetic experiences; to experience brand-new states of consciousness inaccessible to…Homo sapiens.

Some transhumanists even go so far as to posit, and long for, the eventual immersion of all posthuman minds into a universal artificial consciousness that will expand into the cosmos and reorganize it into a more perfect form (cf. Geraci 2008: 149–52; Graham 2002: 72–3). For the purposes of this discussion, we can probably focus on the more down-toearth desires and ideas associated with transhumanism, setting aside the somewhat extreme (and less commonly held) notions about the emergence of posthuman species and “Apocalyptic AI.” On this more relatable level, what is it that transhumanists hope to achieve, and what technologies would be necessary to make such achievements possible? When they speak of “the radical enhancement of human being,” what they seek is “the enhancement of all its psycho-physical capacities and functions in the way that specifically presupposes the application of non-traditional means, those of biomedicine (neuroscience, genetics, pharmacology) and those of technology (molecular nanotechnology, informational technology, artificial intelligence, robotics)” (Agatonovi´c 2018: 430). More specifically, some common transhumanist goals include increased intelligence, physical strength and endurance, and longevity. In the latter case, the hope is that developments in cellular medicine, cyborgization, and mind-mapping/-uploading will lead to much longer lives and eventually to overcoming necessary mortality.2 More than a few transhumanists think it will not be long before medical interventions can “cure aging” and eradicate various deadly diseases that now seem almost inevitable—e.g. the numerous forms of cancer and neurodegeneration.3 Geraci (2011: 143) states that “purely biological solutions…include neuro-pharmacology, to…enhance mental abilities, stem cell research, to regenerate limbs and organs, and genetic engineering, for therapeutic and enhancement purposes.” There are quite a few serious problems to solve along each of these lines of research, but the biggest hurdle to clear when it comes to slowing, stopping, or reversing aging is basic cell 2

The most optimistic speculations (see e.g. Kurzweil 2005: 358, 486) even suggest advancing to the point of being able to “hop” to other universes in order to avoid the inevitable destruction of this one. Reaffirming my commitment to down-to-earth ideas, I will not pursue this notion any further here. 3 In the meantime, most transhumanists happily recommend following the latest research on the benefits of exercise and a healthy diet, while some make more controversial lifestyle choices, such as adopting a regimen of “caloric restriction.” This practice finds limited support in laboratory experiments showing that the lives of mice can be radically extended (roughly doubled) by drastically cutting caloric intake (by about 50%) without sacrificing certain “fundamental nutrients”; similar effects have not yet been demonstrated in humans (Manzocco 2019: 92–3).

302

A. Buben

degeneration. The hopeful call attention to existing species—e.g. Turritopsis dohrnii, a.k.a. “the immortal jellyfish”—that are better at maintaining cellular robustness (and might provide clues to unlocking our own potential), to experiments that have succeeded in rejuvenating laboratory worms and mice (suggesting that analogous human therapies might be right around the corner), and to speculation about sophisticated nanotechnology that could repair cell damage on a molecular level (see e.g. Dvorsky 2013; Jabr 2021; Manzocco 2019: ch. 5; Piore 2021; Rich 2012).4 However, even if these biomedical dreams come true, and death is no longer necessary, humans will still be susceptible to accidents so long as they depend on these fragile organic bodies. This problem could be partially addressed by cyborgization (i.e. the gradual replacing of biological organs and tissues with mechanical components) or by transplanting brains into more durable robotic bodies (see e.g. Hughes 2004; Kurzweil 2005: 309–10; Manzocco 2019: 28–9). Unfortunately, these strategies would only spare us from relatively small-scale dangers; we would still be at risk from larger calamities and cataclysms. Enter “mind-uploading,” which would end human reliance on bodily form through the detailed mapping/recording of neural membranes, synapses, and their functions, and the transferring of this information onto a computer processing platform (see e.g. Goldstein 2012; Huberman 2018; Koene 2013; Manzocco 2019: ch. 7.3). One of the purported benefits of purely digital existence is almost total invulnerability (especially if one is constantly “backed up” and saved in various locations around the world/galaxy/universe), but the suggestion that the subjective experience of selfhood could survive the uploading process after being reduced to mere patterns of information raises some difficult questions about the nature of personal identity (cf. Harle 2002; Hughes 2013; Zimmerman 2013: 100–1).5 Setting aside these questions for now, uploading is one more speculative longevity-enhancing technology eagerly awaited by many transhumanists. But what about those on “short clocks” due to age or illness who cannot wait for medicine, robotics, and neuroscience to advance at their current pace? It is cryonics companies that give these people the purported chance to buy some time while all that troublesome R&D plays out.

4

Some of the standard objections to transhumanist enthusiasm for radical longevity have to do with overpopulation, strain on limited resources, and unequal access to new technologies (see e.g. Glannon 2002: 347; McNamee and Edwards 2006: 514–5; Noonan 2016: 40–1). The standard response to these concerns usually falls along the lines of: “a civilization with technology advanced enough to exacerbate these problems by extending lives will probably be able to solve or mitigate them (e.g. by colonizing uninhabited planets).” A more thoughtful response (see Cutas 2008; cf. Moen 2015: 680) is to argue that the living do not have some moral duty to die so that others (especially others who do not yet exist) can be accommodated. I might not be justified in doing absolutely anything to survive, but others’ rights to reproduction and at least some resources might just be eclipsed by my right to avoid death, when possible, even if it is not possible for everyone. 5 Of course, for those interested in the emergence of posthumans and a universal artificial consciousness, the potential loss of personal identity might not be much of a problem (and personal identity is a fraught philosophical concept anyway, as James Hughes and others (e.g. Swan 2019: 710–3) are careful to point out), but this is a somewhat different line of inquiry from the one currently under consideration.

17 Dying to Live: Transhumanism, Cryonics, and Euthanasia

303

17.3 A Miracle on Ice? Transhumanist goals and strategies certainly receive their share of criticism, even mockery, and this is especially true when it comes to cryonics6 —the practice of preserving human (and sometimes other animal) bodies or body parts at extremely low temperatures in the hope of having them reanimated or resurrected at some point in the future. In order to understand which critiques (or jokes) are well-placed and fair, it will be helpful to clarify what exactly cryonic preservation entails. Fortunately, there is widespread agreement on the details, and the handful of (mostly US-based) cryonics companies seem to provide roughly the same services (although some of them outsource certain elements of the process—e.g. preparation and transportation of the body—to other companies). To begin with, the prospective patient (or designated surrogate) must decide whether to pursue full-body preservation or instead opt for neuropreservation (i.e. preservation of only the head/brain); the latter costs less than half as much but creates obvious additional complications for reanimation (cf. Manzocco 2019: 114). Without one’s original body, one would have to hope that future technology will sort out issues such as head/brain transplants in addition to reanimation and other more straightforward biomedical interventions aimed at preventing/reversing degeneration (at least in heads/brains). Of course, the development of uploading technology would render a lot of these other advances unnecessary, but this possibility comes with its own problems, as mentioned above, and it seems to be a less desirable option for many people hoping to come out on the other side of cryonic preservation someday (see Swan 2019: 708). So, let us suppose I am interested in full-body preservation; how would I pay for it? A common misconception is that cryonics is only for the rich. It is certainly true that preservation is not cheap, but that does not mean it is inaccessible to all but the fabulously wealthy. Full-body preservation at Alcor (in Scottsdale, Arizona), the most prominent cryonics company, costs $200,000 with everything included (but there is a much cheaper option in Russia, and prices in between these extremes elsewhere in the US). That is not an insignificant amount of money, but there are some mitigating factors to consider. For example, money otherwise spent on cremation or burial could instead go toward the costs of this alternative approach to handling one’s remains. More significantly, cryonic preservation can be paid for through a special life-insurance policy (with one’s chosen cryonics provider as the beneficiary) that might only cost a few dollars per day, depending on one’s age and health history. 6

Not every cryonicist is a transhumanist, or vice versa, but there is a lot of overlap in that Venn diagram. In one notable example, Max More, one of the most influential transhumanists, spent a decade as president and CEO of Alcor, the largest cryonics company. In the small non-overlapping parts of the diagram, one might find people who choose cryonic preservation simply hoping to reach an ordinary human lifespan prevented by the early onset of an extraordinary (and currently incurable) illness; such people would not necessarily be seeking any sort of transhumanist enhancement. See, for instance, the 2019 documentary, Hope Frozen, which details one family’s decision to have their three-year-old daughter’s body preserved after she succumbed to brain cancer.

304

A. Buben

This kind of financial planning might not be available to absolutely everyone, but it certainly makes cryonics a viable option for most members of the middle-class (cf. Manzocco 2019: 119; Moen 2015: 678–80; Shaw 2009: 516). Once the necessary arrangements have been made, the next step—dying—is easy enough in the abstract, but the specific circumstances surrounding it are massively important and get right to the core issue of this chapter. Ideally, a preservation team is waiting nearby at the moment death is declared (it also helps if you can die in a hospital near your chosen cryonics facility). Their job is to begin cooling the body immediately while maintaining respiration and blood flow artificially to slow the onset of cell degeneration. They also inject anti-coagulants to prevent blood-clots and replace much of the body’s fluids with a kind of antifreeze solution. The goal is to limit the formation of destructive ice crystals as they gradually lower the body’s temperature even further. Instead of freezing, the body vitrifies, reaching a glass-like state in which there is little molecular activity and cell deterioration is brought to a virtual standstill. The final step in the process involves transferring the body to the chosen cryonics facility and submerging it upside down in regularly replenished liquid nitrogen at −196 °C. There is no electricity required, but if the liquid nitrogen refills are neglected, the container housing the body will begin to thaw starting at the top, which is why the all-important head/brain is at the bottom (cf. Manzocco 2019: 118; Minerva and Sandberg 2017: 527; Moen 2015: 677; Shaw 2009: 515; Swan 2019: 701–2). Provided the cryonics company stays in business and keeps the liquid nitrogen flowing, the preserved have all the time in the world to wait for technology to advance far enough to be able to fix what was ailing them prior to preservation and make reanimation possible. But how likely is it that this will ever happen? Honestly, not very, and this extreme unlikelihood is probably the main source of the criticism and mockery cryonicists face. In response, they offer what David Shaw (2009: 520–1) calls the “Cryonic Wager,” which is similar in structure to Pascal’s more famous version. Among the available postmortem options—various forms of burial and cremation vs. cryonic preservation—the latter seems to provide the best chance of living again (and possibly a lot longer if some of transhumanism’s other technological dreams are also realized). In fact, assuming there will be no divine intervention, dissolution into one’s constituent elements provides no chance whatsoever of returning to life in the world (in the relevant sense), so even the very limited possibility of cryonic preservation paying off triumphs in comparison. Of course, it does come at a premium, which has to be weighed against one’s other financial responsibilities, but people make questionable purchases that are not “potentially life-saving” all the time and we generally think they are free to spend (at least some of) their money as they wish (Moen 2015: 680). But is it really accurate to say that being resurrected, young and healthy, from a vitrified state is any less of a fantasy than the sort of resurrection Pascal believed in? The reason the Cryonic Wager has some rational pull is that no one can definitively rule out the possibility of advancing science and technology eventually being able to

17 Dying to Live: Transhumanism, Cryonics, and Euthanasia

305

solve all the pertinent problems (cf. Minerva and Sandberg 2017: 531–2).7 If truth be told, many of the required technologies already exist in some rudimentary form, which suggests to some that successful reanimation might not be such a far-fetched idea. I have already mentioned biomedical research in genetic engineering, the therapeutic use of stem cells, etc., and there are several examples of proto-cyborgization to be found in the medical devices we currently implant in human bodies. When it comes to cryonics specifically, “mainstream cryobiology” has “routinely cryopreserved, thawed and recovered,” the “sperm, eggs and tissues” of humans (Moen 2015: 678). There have also been experiments that have succeeded in reanimating nearly frozen rats and pigs whose brain activity had come to a halt, and there have even been examples of this kind of thing in cases where humans drowned in very cold water. More recently, hospitals have begun using some of the same cooling techniques employed in cryonic preservation to buy extra time for trauma surgeons to repair serious damage in patients (cf. Moen 2015: 677; Swan 2019: 702). However promising the different lines of research might be, there are obviously still many battles to be fought on each front. For example, in the case of cryonics, even an ideal preservation will result in some micro-fracturing of the tissues and organs that will need to be repaired before a successful reanimation can take place.8 Once again, sophisticated nanotechnology seems to be the “magic bullet” that most transhumanists predict will solve some of these very difficult lingering problems (cf. Manzocco 2019: 118, 122, 127; Moen 2015: 678). Whether this technology is ever developed remains to be seen, but the most important takeaway from this overview of cryonics, for my purposes, is that successful repair, rejuvenation, and reanimation is made easier (and thus, more feasible) when the preserved body is as undamaged as possible. This is the observation that finally brings the conversation around to transhumanist interest in assisted dying.9

7

However, even if the technology becomes available, some have pointed out that there might still be social or political obstacles to the reanimation of preserved bodies. For example, future societies might not see any good reason to devote resources to reanimating and caring for the people of a bygone era that could be better spent on the people of their own (cf. Moen 2015: 678; Shaw 2009: 517). On the other hand, Thau (2020: 643–4) considers the disturbing possibility that future societies would be motivated to reanimate the preserved people of the past, but only to use or abuse them. 8 Since this damage can occur in the brain, it is also a problem for the proponents of future uploading technology. It is hard enough to map accurately and understand one’s neural structures without having to contend with structural damage as well. 9 Just to be clear, I do not mean to discount the various social, political, or economic practicalities I have mentioned along the way that would surely have some bearing on the likelihood of successful repair, rejuvenation, and reanimation; all of these issues are taken into consideration when coming to the conclusion that cryonic preservation is extremely unlikely to work out. (I have said less about aesthetic and psychological uncertainties related to how a reanimated individual might adapt to the future world, but I doubt these kinds of worries will essentially alter the Cryonic Wager math, especially if suicide remains a possibility under unsatisfactory conditions (cf. Thau 2020: 641, 643).) However, my focus is on the medical, physiological, and technological practicalities because those have the most direct relevance to the issue of assisted dying in the cryonics context.

306

A. Buben

17.4 Cryothanasia What I suggested above is that the longer a dead body sits before the preservation process is begun, the worse the damage will be at the cellular level. The worse the damage is, the more sophisticated the repair techniques will need to be. This point is of particular concern when it comes to brain damage, given the singular importance of the brain in most accounts of personal continuity. If the damage is severe enough, we might reasonably say the person is gone, and no repair techniques will be sufficient to bring the person back into being.10 In an extreme, but illustrative, example, it seems accurate to claim that the brain of a cremated person is damaged beyond repair (again, barring divine intervention, or that of some AI with knowledge and capabilities well beyond human comprehension). But we do not need to reach this catastrophic level of damage to come to similarly pessimistic conclusions about the chances of successful restoration. In fact, it is not even necessary for a brain to be dead before we start worrying about the onset of irreparable damage. Consider the situation of otherwise healthy people with degenerative brain conditions. For someone struggling with progressive dementia due to Alzheimer’s disease, the more time passes, the less there is of this person to preserve or restore. In the case of (mathematician and noted cryonics advocate) Thomas K. Donaldson, the problem was an expanding brain tumor, rather than Alzheimer’s, but he was thinking about the same “clock is ticking” problem when he asked the California courts (in the early 1990s) to allow him access to pre-mortem cryonic preservation. Donaldson’s doctors argued that if he were to wait “until his natural death to be suspended, future reanimation will be futile because the tumor will have destroyed his brain” (Pommer 1993: 590). The lawsuit was ultimately unsuccessful because the courts in question “found that recognizing a right to pre-mortem cryonic suspension was tantamount to sanctioning assisted suicide,” which is “a legislative matter rather than a judicial one” (Pommer 1993: 600, 603).11 Whether or not the legal reasoning that resulted in this decision was sound, it is imperative to acknowledge that cryonicists would not necessarily describe the situation in these terms. Donaldson himself states, “I would not see myself as committing suicide, nor would the people who froze me see me as committing suicide. They’d see me as going through a draconian treatment that was my only chance to stay alive, however slim that chance might be” 10

Many transhumanists/cryonicists speak of “information-theoretic death,” in which the personality-generating memories and other information encoded in one’s neural structures are irreparably lost due to damage/deterioration so severe that no technology would be able to recover it (cf. Manzocco 2019: 116–7; Minerva and Sandberg 2017: 527; Moen 2015: 677–8; Shaw 2009: 518–9). According to this understanding of death, a cryonics patient with a well-preserved brain need not be considered entirely dead, even though most people (relying on our current legal and medical definitions of death) would not consider such a person alive either. To borrow a quote from The Princess Bride’s Miracle Max, the cryonically preserved might turn out to be “only mostly dead.” For further discussion of this strange situation, see Hershenov (2003: 91–4). 11 Despite this legal setback, things apparently worked out ok for Donaldson in the end (see Minerva and Sandberg 2017: 526), but his good fortune does little to diminish the force of the hypothetical situation his case represents.

17 Dying to Live: Transhumanism, Cryonics, and Euthanasia

307

(Pommer 1993: 601n98). This different way of describing things suggests that, even where the language of assisted dying remains useful, requesting it for the purposes of better cryonic preservation could, and maybe should, be distinguished from ordinary requests aimed at preventing otherwise inescapable and intolerable suffering. At the risk of momentarily sounding a bit theological or soteriological, dying to (hopefully) live again seems importantly different from dying just to avoid the pain of being alive any longer. Ole Martin Moen (2015: 680) and Shaw (2009: 519) both offer brief arguments supporting this distinction, but it is the more recent work of Minerva and Sandberg (2017: 529–30) that really digs into the ethical issues surrounding these two motivations for assisted dying. They coin the term “cryothanasia” to refer to the kind of pre-mortem cryonic preservation Donaldson sought, and they claim that the standard arguments against euthanasia (broadly construed here, it seems, to include various sorts of assisted dying), which might not be all that compelling to begin with, simply do not apply to cryothanasia. The strongest of these arguments have to do with ethical prohibitions (both within society in general and within the medical profession specifically) on the taking of human lives, regardless of the reasons for wanting to take them.12 However, since it remains to be seen if cryonic preservation will pay off, it is unclear if cryothanasia qualifies as taking a life or merely pausing a life. Perhaps one could make a case that cryonics is so unlikely to work out that, for all practical purposes, cryothanasia amounts to taking a life. Setting aside the conjecture required to make such a case, it would still be necessary to explain how cryothanasia differs from other potentially deadly but widely accepted medical choices—for example, undergoing experimental treatments that many jurisdictions already allow as last resorts. As Minerva and Sandberg (2017: 531–2) see things, what is relevant in the context of people affected by terminal diseases, and for whom no other treatment is available, is that the probability is still greater than 0. From a moral perspective, the use of experimental treatment for some patients is indeed justified on the basis that it is the only alternative to their dying within a short time, so it is in their best interest to use a potentially unsafe treatment in order to have a chance to survive.

12

On the weaker end of the spectrum are arguments based on religious belief, or on arbitrary and contingent views about what is “natural” or “dignified.” Minerva and Sandberg do an admirable job of briefly responding to religious arguments on their own terms and showing that support for cryothanasia need not conflict with certain, mostly Judeo-Christian, religious beliefs. While I am unsure that dreams of life-extension via cryonic preservation would be as compatible with other religious traditions, especially certain versions of Buddhism (see Buben 2019), the main problem with religious arguments is that they often depend on additional, highly dubious, metaphysical and epistemological claims about the existence of divinities and knowledge of their will. Something similar could be said about the questionable assumptions made by those who attempt to define what is natural or dignified for human beings. This is a tried, if not true, strategy employed by the conservative-minded in response to new technological developments throughout history. In most cases, their concerns are gradually ignored as the new developments are normalized and deemed compatible with the ever-shifting socio-cultural understanding of what is appropriate for humans.

308

A. Buben

If a probability greater than zero is all it takes to justify other last resorts, then why could the same not be said of cryothanasia? The point of this mostly rhetorical question can be made even stronger through further consideration of the Cryonic Wager. As previously explained, the only thing at stake in post-mortem cryonic preservation is money—a not insignificant, but also not impossibly large, amount of money. Things are a little more complicated when it comes to cryothanasia because, in addition to the costs of preservation, it is also necessary to put some actual life on the line. Granted, the quality of the remaining days of the terminal cancer or Alzheimer’s patient might not be so great, but it is possible, and perhaps likely, that someone opting for cryothanasia would end up sacrificing a little of the good stuff as well—e.g. extra time with loved ones, or one more scoop of ice cream (cf. Minerva and Sandberg 2017: 531–2; Shaw 2009: 519). Now, Pascal’s Wager suggests that where the potential benefits of making a finite sacrifice are infinite—“an infinite life of infinite happiness,” in Pascal’s (2005: 213) words—it really does not matter how likely getting those benefits is; so long as there is even a remote possibility, making the sacrifice is sensible. Given that the potential (though unlikely to materialize) benefits of cryonic preservation for the transhumanist include indefinite life-extension, one might argue that the merely finite sacrifice/wager of time and money involved in cryothanasia is similarly sensible (cf. Minerva and Sandberg 2017: 533n29). Although the analogy with Pascal’s Wager is imperfect, the potential benefit of indefinite life-extension means that cryothanasia would compare quite favorably, even if there is a substantial divergence in probability of success, to other experimental last resorts that might only buy someone a few more years. And there is still more to be said in defense of cryothanasia. On the topic of money, I can think of another mitigating factor to take into account. Both Minerva and Sandberg (2017: 532) and Moen (2015: 680) mention that cryothanasia would come with certain economic benefits, but what they do not get into explicitly is how these benefits impact the Cryonic Wager. The end of life is typically quite expensive, given the costs of medicine, ambulance rides, hospice care, stays in intensive care units, etc. In the Cryonic Wager, as originally formulated, the costs of post-mortem cryonic preservation would be incurred in addition to standard end of life expenses. In the case of cryothanasia, however, these costs would simply replace the expense of weeks or months of hospice care or ICU stays. Combined with the savings that come from avoiding burial or cremation, cryothanasia might easily pay for itself in many instances. One could certainly imagine insurance providers wanting to give cryothanasia some serious consideration once these economic issues are factored into the equation. Support from insurance companies might not be all that reassuring to most people, but the economic pros of cryothanasia do seem to mitigate (even if only slightly) the cons associated with sacrificing a few (mostly and/or increasingly miserable) weeks or months in the hope of gaining indefinite life-extension. Since economic concerns were an important element of the original Cryonic Wager, it seems significant that these concerns can be more or less neutralized in cases of cryothanasia.

17 Dying to Live: Transhumanism, Cryonics, and Euthanasia

309

17.5 Going Too Far? These supportive comments about how cryothanasia fares in connection with the Cryonic Wager are not the end of the conversation, however, as they leave room for other worries about the circumstances under which cryothanasia is compelling. Even if it is justifiable to sacrifice some finite amount of time in pursuit of indefinite life-extension, we might still wonder how much time it is justifiable to sacrifice. For example, would it be justifiable for a healthy 43-year-old to sacrifice a likely 40 more years of life (given current demographic information about life-expectancy)? After all, though we tend not to put it in the same category as cancer, aging is currently a terminal condition and 40 years is not so different from four months in comparison with the life that awaits if transhumanists get what they want out of cryonics (cf. Minerva and Sandberg 2017: 532). The obvious reply here, especially considering the probability of success based on where technology currently stands, is that cryothanasia makes the most sense as a last resort, when waiting any longer would cause serious irreparable damage. As long as one is healthy, Minerva and Sandberg (2017: 533n29) explain, “staying alive has a high likelihood of being a safer way of getting to the future than undergoing cryonics.” If, on the other hand, one’s health is failing and technology has not yet advanced to the point of being able to push the end of life off further into the future without the use of cryonics, then I can see little reason why cryothanasia should not be a live option (so to speak) in cases where natural death would come too late for a good preservation. Although I accept that health and probability of successful preservation and reanimation are key factors in determining when cryothanasia makes the most sense, I would like to consider a couple of scenarios that complicate matters a little. The first scenario gets back to the reservation I mentioned earlier about neuropreservation. Having only a head/brain preserved means that additional and/or more sophisticated technologies will be required to restore an individual to something like the kind of life enjoyed prior to preservation. Alternatively, if a preserved individual already has a body, that is one thing no one needs to worry about before thawing this individual out. This shorter to-do list makes it easier to bring a person back, which in turn makes it more likely that it will happen sooner. Could something similar be said about aging and rejuvenation technologies? If it turns out that merely arresting the aging process will be much easier in the future than rejuvenating someone who has already reached an advanced age, then not having reached an advanced age prior to cryonic preservation will come with certain benefits that could lead to an earlier and more desirable reanimation. Just like the extra technologies associated with regrowing bodies or transplanting brains in the neuropreservation scenario, rejuvenation technologies might constitute an additional hurdle to clear before an aged person could be fruitfully reanimated.13 So, thinking about the healthy 43-year-old, it might actually make some sense to consider cryothanasia before the aging-related damage can get any worse. This scenario is not among the 13

For a transhumanist preserved at age 90, the goal is not to be reanimated only to go on living someday in the body of an ordinary 90-year-old.

310

A. Buben

more frequent worries of transhumanists because it is commonly assumed that any technology (by which they primarily mean molecular nanotechnology) sophisticated enough to reanimate a preserved body will also be able to reverse the effects of aging, but this is not necessarily true. If these technologies come apart, then youth might also be a key factor, alongside health and probability of success, in determining when cryothanasia makes the most sense. Another troubling scenario arises in connection with basic human fragility. Most of our activities involve taking physical risks that, if things go poorly, could render cryonic preservation and resurrection difficult or even impossible. This seems like an uncontroversial claim given the variety of destructive accidents that might befall us from moment to moment. Knowing how susceptible we are to accidents, even when we are not engaging in particularly risky behavior, I wonder if there is a case to be made for having ourselves preserved before we become victims and lose our chance to have our lives prolonged by future technology. It might turn out that this technology will render our activities less risky than they are at present, and so it might be smart to put at least some of these activities off until it does. What I am suggesting here is a more thoughtful and risk-averse YOLO (“You Only Live Once”) mentality than what is usually intended by this expression in contemporary popular culture.14 In fact, if we take certain precautions with cryonic preservation in mind, perhaps we will, in some sense, end up living more than once. Consider the following example: at the height of the Islamic State turmoil, I went on a road-trip through rural Turkey with my cousin, let us call him “Phil Kierney,” an Alcor bracelet-wearing cryonicist.15 Now, I do not mean to exaggerate our adventurousness (or our stupidity), but along the way we did wander alone through a complex set of narrow caverns, face the business end of a farmer’s rifle during some unintentional trespassing (for which we apologized profusely), and manage to get our rental car stuck in the mud at night in the middle of nowhere.16 We also discussed what would happen to his plans for cryonic preservation if he were to fall victim to our questionable decision-making during the trip. Scottsdale is a long way from Turkey, and my options for lowering his body temperature while maintaining circulation would have been fairly limited and hard to manage in such a remote setting. Suffice it to say, Phil’s body would have been in pretty bad shape by the time it was delivered to the professionals (if it was delivered at all). For the sake 14

I highly recommend The Lonely Island’s song “YOLO” for a humorous rebuttal of the standard “don’t sweat the future; live in the present” interpretation of the expression. 15 As an Alcor member, one wears a “stainless steel bracelet…which reports the phone numbers to call in case of an emergency and the protocols to follow by whoever finds the body of this person” (Manzocco 2019: 117–8). 16 Although we were far from any particularly dangerous areas near the Syrian border (we were trying to find the ancient ruins of Oenoanda, which housed some extensive Epicurean inscriptions, when we got lost and then stuck), Phil was convinced that IS insurgents were lurking out there in the darkness ready to take us hostage. Fortunately, a lovely local family in a small car stumbled upon us first and helped pull us out of the muck.

17 Dying to Live: Transhumanism, Cryonics, and Euthanasia

311

of a quality preservation and his hopes for future reanimation, would he not have been better off staying close to home (which just happens to be a few miles away from Alcor) and out of harm’s way? This is not to suggest that Phil should never experience the joys of road-tripping through the Turkish countryside, but maybe experiencing these joys could wait until, for example, developments in cyborgization provide him with a more durable body or easily replaceable component parts. And if such developments are more than a few decades away, then perhaps he would be better off having himself cryonically preserved before pursuing certain travel opportunities. Furthermore, because analogous claims can be made about virtually all human activities—from riding a bike on a busy street to teaching in person at a university in a Republican-leaning state during a pandemic to sleeping at home in a poorly maintained Florida condominium complex—preventative cryothanasia before some catastrophic accident or another can get us might have a certain macabre appeal. Given the countless ways one could suffer irreparable damage before the preservation process can be initiated, there are at least some reasons to want to initiate it while one is still relatively young and healthy. As I said above about youth and health, durability might be another key factor in determining when cryothanasia makes the most sense. To be clear, I do not actually think it would be a very good idea for a relatively young and healthy person to give up decades of life now in the hope of preserving youth or avoiding danger later. I am simply not confident enough in advancing technology to make that wager at this point. When aging, disease, or injury have drastically reduced both the quantity and the quality of my life, such that I no longer need to gamble very much, then my level of confidence in a successful outcome of cryonic preservation will not be as important, and I will be more open to the potential benefits of cryothanasia. However, remarkable leaps forward in technological capability and scientific understanding might change my tune a bit. I could imagine seriously considering cryothanasia in the kinds of scenarios I just discussed if it suddenly looked a lot more likely (even if not yet entirely certain) that the technology of transhumanist dreams will become reality. So, my lingering resistance to these scenarios appears to be a merely contingent matter.

17.6 Conclusion Transhumanists interested in cryonics have some compelling reasons to seek out assistance in dying, and they also have good reasons to describe the situation using other terms. “Cryothanasia” is a more appropriate way to describe the rather odd situation of choosing an earlier death, or what they hope is just a kind of pre-mortem hibernation, in order to improve their chances of living again, and possibly a lot longer, one day. Whether or not things ultimately work out for them, I have not encountered any insurmountable objections to cryothanasia as a last resort in cases of (currently) terminal and irreparable neurodegeneration or other forms of progressive brain damage. Things are admittedly murkier when considering the possibility

312

A. Buben

of cryothanasia outside the context of imminent medical breakdown. One of the most crucial issues in these more complicated cases is the likelihood of realizing the technologies necessary for successful repair, rejuvenation, and reanimation. Although an argument can be made for cryothanasia at nearly any point in life, given the potentially indefinite benefits of even a vanishingly small chance of a positive outcome, I would not recommend wagering the “bird in the hand” until its song is less sweet. What is “in the bush” may well be quite enticing, but you could end up with both prizes (i.e. a full natural lifespan and, should sophisticated enough technology ever be developed, an indefinitely extended lifespan after reanimation from a cryonically preserved state) if you can hold out a while longer before placing your bet.17 Leaning entirely into this combination of metaphors, for now I would advise paying more attention to the hedges than the bushes.18

References Agatonovi´c, Miloš. 2018. The case of transhumanism: The possibility of application of Nietzsche’s ethics and critique of morality today. Philosophy and Society 29 (3): 429–439. Buben, Adam. 2019. Personal immortality in transhumanism and ancient Indian philosophy. Philosophy East and West 69 (1): 71–85. Cutas, D.E. 2008. Life extension, overpopulation and the right to life: Against lethal ethics. Journal of Medical Ethics 34. https://doi.org/10.1136/jme.2007.023622 Dvorsky, George. 2013. Do these startling longevity studies mean your lifespan could double? io9. Accessed July 5, 2021. http://io9.gizmodo.com/do-these-startling-animal-studies-mean-your-lif espan-co-486041314 Geraci, Robert M. 2008. Apocalyptic AI: Religion and the promise of artificial intelligence. Journal of the American Academy of Religion 76 (1): 138–166. Geraci, Robert M. 2011. There and back again: Transhumanist evangelism in science fiction and popular science. Implicit Religion 14 (2): 141–172. Glannon, Walter. 2002. Extending the human life span. Journal of Medicine and Philosophy 27 (3): 339–354. Goldstein, Evan. 2012. The strange neuroscience of immortality. The Chronicle of Higher Education. Accessed July 5, 2021. http://chronicle.com/article/The-Strange-Neuroscience-of/132819/ Graham, Elaine. 2002. ‘Nietzsche Gets a Modem’: Transhumanism and the technological sublime. Literature and Theology 16 (1): 65–80. Harle, Robert F. 2002. Cyborgs, uploading and immortality—Some serious concerns. Sophia 41 (2): 73–85. Hershenov, David. 2003. The problematic role of ‘irreversibility’ in the definition of death. Bioethics 17 (1): 89–100. Huberman, Jenny. 2018. Immortality transformed: Mind cloning, transhumanism and the quest for digital immortality. Mortality 23 (1): 50–64.

17

This idea is a bit like trying to cheat Pascal’s Wager by concluding the life of an unbeliever with a deathbed conversion, albeit without having to worry about facing divine wrath for attempting to pull a fast one on the Almighty. 18 I am grateful to Bill Tierney for informative conversations about his own experiences in the cryonics world, and to Jukka and Michael for their very helpful comments on an earlier draft of this chapter.

17 Dying to Live: Transhumanism, Cryonics, and Euthanasia

313

Hughes, James J. 2004. Citizen Cyborg: Why Democratic Societies must Respond to the Redesigned Human of the Future. Cambridge, MA: Basic Books. Hughes, James. 2013. Transhumanism and personal identity. In The Transhumanist Reader: Classical and Contemporary Essays on the Science, Technology, and Philosophy of the Human Future, eds. Max More and Natasha Vita-More. Chichester: Wiley-Blackwell. Jabr, Ferris. 2021. How long can we live? The New York Times Magazine. Accessed May 3, 2021. https://www.nytimes.com/2021/04/28/magazine/human-lifespan.html Koene, Randal A. 2013. Uploading to substrate-independent minds. In The Transhumanist Reader: Classical and Contemporary Essays on the Science, Technology, and Philosophy of the Human Future, eds. Max More and Natasha Vita-More. Chichester: Wiley-Blackwell. Kurzweil, Ray. 2005. The Singularity is Near: When Humans Transcend Biology. New York: Viking. Manzocco, Roberto. 2019. Transhumanism—Engineering the Human Condition. Cham: Springer. McNamee, M.J., and S.D. Edwards. 2006. Transhumanism, medical technology and slippery slopes. Journal of Medical Ethics 32: 513–518. Minerva, Francesca, and Anders Sandberg. 2017. Euthanasia and cryothanasia. Bioethics 31: 526– 533. Moen, Ole Martin. 2015. The case for cryonics. Journal of Medical Ethics 41: 677–681. Noonan, Jeff. 2016. The debate on immortality: Posthumanist science vs. critical philosophy. The European Legacy 21 (1): 38–51. Pascal, Blaise. 2005. Pensées. Translated by Roger Ariew. Indianapolis, IN: Hackett. Piore, Adam. 2021. Can blood from young people slow aging? Silicon Valley has bet billions it will. Newsweek. Accessed April 14, 2021. https://www.newsweek.com/2021/04/16/can-bloodyoung-people-slow-aging-silicon-valley-has-bet-billions-it-will-1581447.html Pommer II, Robert W. 1993. Donaldson v. Van de Kamp: Cryonics, assisted suicide, and the challenges of medical science. Journal of Contemporary Health Law and Policy 9: 589–604. Rich, Nathaniel. 2012. Can a jellyfish unlock the secret of immortality? The New York Times Magazine. Accessed July 4, 2021. http://www.nytimes.com/2012/12/02/magazine/can-a-jellyf ish-unlock-the-secret-of-immortality.html?hp&gwh&pagewanted=all&_r=0 Sandberg, Anders. 2013. Morphological freedom—Why we not just want it, but need it. In The Transhumanist Reader: Classical and Contemporary Essays on the Science, Technology, and Philosophy of the Human Future, eds. Max More and Natasha Vita-More. Chichester: WileyBlackwell. Shaw, David. 2009. Cryoethics: Seeking life after death. Bioethics 23 (9): 515–521. Swan, Melanie. 2019. Worldwide cryonics attitudes about the body, cryopreservation, and revival: Personal identity malleability and a theory of cryonic life extension. Sophia 58: 699–735. Thau, Tena. 2020. Cryonics for all? Bioethics 34: 638–644. Zimmerman, Dean. 2013. Personal identity and the survival of death. In The Oxford Handbook of Philosophy of Death, eds. Ben Bradley, Fred Feldman, and Jens Johansson. Oxford: Oxford University Press.

Adam Buben is a University Lecturer (Universitair Docent 1) in philosophy at Leiden University in the Netherlands. His primary areas of research are nineteenth and twentieth century continental philosophy, and the philosophy of death across various traditions. He is the author of two books: Existentialism and the Desirability of Immortality (Routledge, 2022) and Meaning and Mortality in Kierkegaard and Heidegger (Northwestern University Press, 2016). His work has appeared in journals such as European Journal of Philosophy, International Journal for Philosophy of Religion, and Journal of the American Philosophical Association. He is currently the series co-editor for Lexington Books’ New Kierkegaard Research.

Chapter 18

When Is Self-perceived Burden an Acceptable Reason to Hasten Death? Michael B. Gill

Abstract Many terminally ill patients perceive themselves to be a burden to loved ones who care for them. The self-perception of being a burden can play a significant role in terminal patients’ decisions to take courses of action, such as ceasing lifesustaining treatment or requesting physician-assisted suicide, that hasten death. I will use the term ‘burden-based decision’ as a shorthand for cases in which a terminal patient’s perception that she is a burden to her loved ones influences her decision to hasten death. When should we view a terminal patient’s inclination to make a burden-based decision to be an ethical problem or a failure of treatment? And when should we view it to be an acceptable response rather than a problem or failure? I argue here that such decisions are acceptable more often than many who write on this topic imply. Keywords Self-perceived burden · Terminal illness · Hastening death

18.1 Introduction Many terminally ill patients perceive themselves to be a burden to loved ones who care for them. The self-perception of being a burden can play a significant role in terminal patients’ decisions to take courses of action, such as ceasing life-sustaining treatment or requesting physician-assisted suicide, that hasten death. I will use the term ‘burden-based decision’ as a shorthand for cases in which a terminal patient’s perception that she is a burden to her loved ones influences her decision to hasten death. When should we view a terminal patient’s inclination to make a burden-based decision to be an ethical problem or a failure of treatment? And when should we view it to be an acceptable response rather than a problem or failure? I argue here that such decisions are acceptable more often than many who write on this topic imply. The self-perception of being a burden to loved ones is one of the most prevalent sources of distress of the terminally ill. One study found that over three-fourths M. B. Gill (B) Department of Philosophy, University of Edinburgh, Edinburgh, UK e-mail: [email protected] © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 M. Cholbi and J. Varelius (eds.), New Directions in the Ethics of Assisted Suicide and Euthanasia, The International Library of Bioethics 103, https://doi.org/10.1007/978-3-031-25315-7_18

315

316

M. B. Gill

of patients with advanced cancer had the sense of self-perceived burden, with about a third experiencing moderate to extreme distress as a result (Wilson et al. 2005). Similar findings have been reported for numerous other late-stage diseases (Bausewein et al. 2013; Cahill et al. 2009; Chio et al. 2005; Chochinov et al. 2005; Chochinov et al. 2002; Chochinov et al. 2007; Cohen and Leis 2002; Cousineau et al. 2003; de Faye et al. 2006; Dyeson 2000; Eitel et al. 1995; Foley et al. 2016; Ganzini et al. 2002; Kowal et al. 2012; McPherson et al. 2007a; McPherson et al. 2007b; McPherson et al. 2010; Morita et al. 2000; Morita et al. 2004; Pearlman et al. 1993; Singer et al. 1999; Vig and Pearlman 2004). Not being a burden is commonly cited as one of the most important elements of a good death (Chio et al. 2005; Chochinov et al. 2002; Chochinov et al. 2007; Ganzini et al. 2002; Geng et al. 2017; McPherson et al. 2007b; McPherson et al. 2010; Steinhauser et al. 2000; Wilson et al. 2005). Concern about being a burden is a significant factor in many patients’ decisions to hasten death. Numerous studies have found this concern to be one of the top reasons for refusing or discontinuing life-extending treatment (Ashby et al. 2005; Chochinov et al. 2005; Cohen-Mansfied et al. 1992; Gudat et al. 2019; Lee et al. 2015, Lulé et al. 2014; Malpas et al. 2012; McPherson et al. 2007b; Murray et al. 2003; Patrick et al. 1997; Wilson et al 2005; Zweibel and Cassel 1989). It is also among the most common considerations associated with requests for physician-assisted suicide (Akechi et al. 2004; Back et al. 1996; Berkman et al. 1999; Filiberti et al. 2001; Ganzini et al. 1998; Ganzini et al. 2002; Ganzini et al. 2006; McGlade et al. 2000; Morita et al. 2004; Street and Kissane 2000; Sullivan et al. 2001; Wilson et al. 2005; Wilson et al. 2013; Virik and Glare 2002). Some who write on this topic suggest that burden-based decisions are virtually always regrettable. Their idea is not just that it’s sad that a patient has made that choice. It’s that burden-based decisions are inseparable from a moral problem or a failure to treat patients as they ought to be treated. Rodriguez-Prat et al. maintain that “the law, clinical policy and education in virtues and values should all lay the foundation so that it is impossible to ever feel like a burden to others” (419). Metselaar and Widdershoven write that “many professionals doubt whether alleviating the perception that one is a burden constitutes an acceptable reason for requesting the withdrawal or withholding of treatment or even physician-assisted suicide” (432). In Oregon, physicians were less likely to grant a request for assisted suicide if they deemed it to be based on self-perceived burden (Ganzini et al. 2000). Undoubtedly some burden-based decisions are regrettable results of moral problems or treatment failures. But I think we should resist the notion that burden-based decisions are regrettable by default, that they are virtually always indicative of problems or failure. The alternative view is well-expressed by Gudat et al.: “Ethically it is not plausible that WTDs [wishes to die] in the context of SPB [self-perceived burden] and advanced disease are always bad. They can be a way to accommodate to an inevitable ending of life… It would not meet the needs of the patients and their families to generalize a view that WTDs, if they are founded on SPB, can (or even must) be resolved by care and support. Sometimes, the patient has important personal reasons for rejecting care… Patients and families should be reassured that having a WTD due to feeling one is a burden is nothing abnormal and, above all, it is not

18 When Is Self-perceived Burden an Acceptable Reason to Hasten Death?

317

morally wrong” (428–9). I think Gudat et al. are right that some burden-based decisions can be reasonable responses to the inevitable rather than regrettable results of problems or failure. Even when care is effective in improving a patient’s condition, it may still be burdensome. The appropriate response to some burden-based decisions may not be intervention or recrimination, but acceptance. Here is one reason this point is important to make. Many patients with terminal illness feel they are a burden to their loved ones. Many of them consider hastening death as a result. Many loved ones are aware that a terminal patient is considering hastening death because she feels she is a burden. And in many cases everyone feels terrible about all of it. Loved ones feel bad because they believe that if the patient is considering such a thing, it must be because they have done something wrong. Aware of this, the patient feels bad about imposing on her loved ones not only physical, financial, and social burdens but also the meta-burden of realizing that she feels herself to be a burden. Not wanting to impose that meta-burden, the patient may feel pressure to dissemble, or to choose life-extending treatment even though the first-order burdens are real and pressing and matter tremendously to everyone concerned. Then too, because there is a taboo against endorsing burdenbased decisions, loved ones feel guilty about even considering that what the patient is considering is acceptable, let alone engaging in frank and open discussion about it (see Gudat et al. 2019; Rehmann-Sutter 2015). All of which adds complex interpersonal and intrapersonal anguish to a situation that is the last one that needs any more anguish. If we can see clearly how a burden-based decision can be acceptable, it may diminish the taboo, and may make one of the hardest things any of us will ever have to face a little bit easier. In Sect. 18.2, I make an initial or prima facie case for the acceptability of burdenbased decisions. In Sect. 18.3, I address the objection that burden-based decisions are vitiated by mistaken beliefs. In Sect. 18.4, I address the objection that burden-based decisions are vitiated by mistaken value-judgments.

18.2 Initial Case In this section I sketch a prima face case for the acceptability of some burden-based decisions. It’s only an initial sketch. Much of the substance of the case will be made in two subsequent sections, where I’ll address reasons for thinking burden-based decisions to be indications of moral problems or treatment failures. Let’s start with two examples of decisions not to prolong life. Example 1. A person in her mid-70 s is writing out an advance directive. She has had her fair share of caring for sick and dying loved ones. She is healthy, but she knows that cannot last forever. She considers what to say in the advance directive about situations in which she is irreversibly unconscious or otherwise significantly cognitively compromised. Should physicians resuscitate her if she goes into cardiac arrest when she is in one of those states? Should she be intubated

318

M. B. Gill

or undergo a gastrostomy if it is necessary to keep her alive? Should she be given life-extending treatment? Or should physicians forgo attempts to prolong her life? The person writes in her advance directive that in such situations she should receive only palliative care. And one of her most important reasons is that she does not want to be a burden to her loved ones. Example 2. A person in mid-70 s is reaching the end of a long battle with illness. Her physician has told her she has six months to a year to live. The person’s condition is debilitating and painful, but she has no significant cognitive impairment. Her family are profoundly affected by her illness, but she does not think they are unduly burdened. Now her physician tells her that because of a new development, it appears that she may die sooner than previously expected, in a month or two, unless she undergoes intensive invasive treatments that can keep her alive for six to eight months. The person decides against treatment. She judges that her pain is too great, her quality of life too poor, to make them worthwhile. She has battled illness long enough. These two examples are not unusual. Such decisions are typical among people in those situations. I venture most of us will allow that these decisions can be reasonable and acceptable. They do not invariably indicate a moral problem or treatment failure; they are not regrettable by default. There is not necessarily something to try to reverse in these cases. The appropriate response can be acceptance rather than intervention or recrimination. What do the examples show? Example 2 shows it is not necessarily a problem or failure when a mentally competent person decides to hasten death, that a person can have reasons for such a decision that we deem acceptable. Example 1 shows that it is not necessarily a problem or failure when a person puts in her advance directive instructions to hasten death because she desires not to be a burden to her loved ones. The combination of these two examples gives us an initial reason for the acceptability of burden-based decisions. If we think it is acceptable that a person decides to hasten death when she is competent; and if we think it is acceptable that a person writes into an advance directive instructions not to prolong life because she does not want to be a burden to her loved ones when she is incompetent; then it seems at least initially plausible that it can be acceptable that a person hasten death because she does not want to be a burden to her loved ones when she is competent. The combination of the two examples suggests that underlying such a decision can be considerations that are not indicative of a problem or a failure. Underlying those decisions can be reasons often deemed acceptable in end-of-life situations. Here is an additional consideration. Self-perceived burden is rarely the only reason someone chooses to hasten death. It may be regrettable if every other consideration leads a person to want to prolong life but she chooses not to only because she perceives herself to be a burden. It does seem deeply concerning for a patient who otherwise wants nothing other than to live to choose to die because she perceives she is trouble for her loved ones. But that is a very atypical situation, and having the kind of situation in the forefront of our minds can distort our intuitions about burdenbased decisions in general. Much more commonly, a patient who feels herself to be a

18 When Is Self-perceived Burden an Acceptable Reason to Hasten Death?

319

burden is also experiencing numerous other hardships that can give her reasons not to prolong life (Chochinov et al. 2002; Chochinov et al. 2007; Ganzini et al. 2002; Geng et al. 2017; McPherson et al. 2010; Sullivan et al. 2001; Wilson et al. 2005). This correlation between self-perceived burden and other hardships is hardly surprising. Many of the reasons a patient perceives herself to be a burden are inextricably linked to physical hardship, to the inability to engage in activities that are meaningful, to assaults on her sense of self. In the mind of such a patient, the thought of relieving the burden on her loved ones does not pull in the opposite direction of her other desires. Her desire not to be a burden and many other of her strong desires may be in concord. At times, moreover, it may be misleading to speak of self-perceived burden and those other considerations as entirely distinct reasons. In the mind of a patient, considerations may be bound up with each other in a manner that makes “not wanting to be a burden to others” a motivation that defies the altruistic/self-oriented distinction. Consider, for instance, a patient’s desire for her loved ones’ memories to be mainly of her as vibrant, engaged, loving, and caring, rather than for their memories to be dominated by thoughts of her going through an agonizing dying process in a significantly compromised condition. That desire could be described as self-oriented and unrelated to burden: she does not want people to think a certain way about her. Or it could be described as burden-based and altruistic: she does not want her loved ones to carry burdensome memories. But probably the more accurate thing to say is that the distinction between self-oriented and burden-based is in this case inapt. It might be equally inapt to draw that distinction with regard to the patient’s desire that her loved ones not have to attend to some of her bodily functions. In the mind of the patient, the other-oriented wish not to be a burden and the self-oriented wish not to undergo certain unpleasant experiences may be one and the same, even if she might be categorized in a research context as being motivated by self-perceived burden (see Mak & Elwyn 2005; Malpas et al. 2012). The end of life can involve a sense of passivity and helplessness that can make dying seem meaningless. Actions to benefit others can counteract that passivity. They can make one’s death feel more meaningful. Such actions can include: writing wills and getting one’s affairs in order, publicizing one’s condition to enable others to take preparatory or preventative measures, repairing relationships that will benefit loved ones after one’s death. Dying individuals are praised for the benefits such actions can have for others. It is also generally understood that the agents themselves can gain something from these actions—that far from being merely a sacrifice, such conduct can be valuable for the dying individuals themselves. In light of this, to insist that not wanting to be a burden is an essentially unacceptable reason seems arbitrary, the result of a simplistic psychological view of human motivation. That’s the initial sketch of acceptable burden-based decisions. What might lead us to think that almost actual burden-based decisions are regrettable nonetheless? In the next two sections we’ll look at two possible reasons: mistaken beliefs and mistaken value-judgments.

320

M. B. Gill

18.3 Mistaken Beliefs Terminal patients who make burden-based decisions are influenced by the belief that their continued care is a burden to their loved ones. A decision of that sort would be regrettable if the belief underlying it were mistaken. Are there reasons to think those beliefs typically are mistaken? “Is the sense of being a burden to others mainly a mental construct, perhaps subject to cognitive distortion, or is it grounded in social reality?” (Wilson et al. 2005, 121). A number of researchers have expressed the worry that burden-based beliefs are mistaken. Rodriguez-Prat et al. (2019, 418) maintain, “There is no evidence … that family members actually shared the patient’s point of view that the wish to die was a way of sparing others from suffering. A person who perceives him or herself as not worthy of living may project this feeling on to others and believe that they have the same view.” McPherson et al. (2007a, 141) suggest that better communication may succeed in “alleviating patients’ sense of having become a burden to others” because they will come to realize how their caregivers really feel. Rehmann-Sutter (2015, 168) writes, “The person who thinks they will be an unacceptable burden to others presumes these others’ perspectives in some way. The assumption is that existence is a burden to them. The real others would possibly say that this is not the case. Their perception of being burdened need not correspond to how the patient imagines them to be burdened.” If the point these writers are making is that individual patients can misperceive how much of a burden they are to their loved ones, they are correct. Patients are not always completely accurate in their self-perception of burden. Some do overestimate the costs to loved ones of their continued care, and, perhaps more significantly, underestimate the benefits their loved ones gain from continuing to care for them (Gudat et al. 2019, 426). To prevent misperception, it is important to promote communication between patients and caregivers. Greater communication may very well reduce some patients’ self-perceived burden. But the evidence of caregiver burden, and of correlations between caregiver burden and patients’ perceptions, speaks against the idea that drastic misperception plays a decisive role in all or most burden-based decisions. A plethora of studies have found that people who provide care for loved ones with terminal illness bear significant physical, social, financial, and emotional burdens (Cagle et al. 2016; Diwan et al. 2004; Galvin et al. 2016; Given et al. 2001; Glajchen 2004; Grunfeld et al. 2004; Hoefman et al. 2013; Laizner et al. 1993; Oldenkamp et al. 2017; Schneider et al. 1999; Scholte op Reimer et al. 1998; Song et al. 2012; Stetz 1987; Torti et al. 2004; Weitzner et al. 1999). Physical burdens arise from the exertion of moving and cleaning someone who is incapacitated, from lack of sleep, and from the loss of exercise and other healthy activities. Social burdens arise from decreased time and attention for children and other family members, sacrifice of connection with friends and others who provide personal support, and loss of opportunity for entertainment, travel, and other social activities. Financial burdens arise from loss of income and occupational advancement as a result of extended

18 When Is Self-perceived Burden an Acceptable Reason to Hasten Death?

321

periods of care, and from savings-depleting medical expenses. Emotional burdens arise from the difficult responsibility of providing round-the-clock care, the agony of watching a spouse or parent suffer painful decline, the transformation of a home into a hospital-like environment, and the experience of a prolonged dying process that can obscure memories of the patient from earlier times (Cahill et al. 2009; Grunfeld et al. 2004; Leroy et al. 2016; Malpas et al. 2012; Simmons 2007). Recent studies have also found that that there is generally moderate correspondence between patients’ self-perception of burden and actual caregiver burden. Chio et al. (2005, 1782) found that among patients with ALS, “awareness of their impact on caregivers, measured with SPBS [self-perceived burden scale], was high.” McPherson et al. (2010, 200) found that among stroke patients “care recipients’ sense of being a burden to others was correlated with disrupted schedule and health problems on the caregivers’ CRA [care reaction assessment] as well as with most subscales of the caregivers’ quality of life measure. This suggests that SPB may be grounded to some extent in the actual burden borne by the caregiver in providing their care.” Kowal et al. (2012, 1740) found that the “significant association between SPB [self-perceived burden] and caregiver reports of feeling burdened is consistent with past research on ALS and stroke. It is also in line with end-stage renal disease patients’ perceptions of being a burden to unpaid caregivers. This suggests that patients’ sense of burdening others is founded, at least in part, on an accurate perception of caregiver burden.” Leroy et al. (2016, 1282) found that “patients had quite a good perception of their caregivers’ burden.” McPherson et al. (2007b) found less of a correlation between caregiver burden and self-perceived burden among advanced cancer patients than had been found in ALS patients. Even in their study, however, there was a significant correlation between patients’ perceptions and certain aspects of caregiver burden. McPherson et al. (2007b, 135) write, “The family members of patients who reported that SPB was a significant problem had higher scores on the caregiver burden scale than family members of other patients, although the overall correlation was modest” (see also Gauthier et al. 2007; Kowal et al. 2012; McPherson et al. 2007a, 2010; Morita et al. 2004). Patients are not entirely accurate in their assessment of caregiver burden, but neither are they wildly inaccurate. The evidence does not comport with the idea that burden-based decisions are virtually all based on beliefs that are drastically mistaken. Perhaps, though, there’s a worry here about something other than just the possibility of patients’ overestimating the physical, financial, social, or emotional strain on their loved ones. We can articulate this worry by distinguishing two different meanings of ‘burden.’ On meaning 1, to say that a terminally ill patient is a burden to her loved ones is to say that caring for her has significant physical, social, or financial costs—that, for instance, as a result of caring for her, loved ones experience disrupted sleep for extended periods of time, or drastically curtail their social life, or take extended leave from their jobs. On meaning 2, to say that a terminally ill patient is a burden is to say that her loved ones would prefer that she die sooner rather than later so that they no longer have to care for her. These two meanings are distinct. Someone can think a patient is a burden in the first sense, but not a burden in the second sense. The evidence suggests that patients generally aren’t drastically

322

M. B. Gill

inaccurate in their assessment of whether they are a burden in sense 1. The worry is that they are inaccurate in their assessment of whether they are a burden in sense 2. The worry is that their desire not to prolong life is influenced by the false belief that their loved ones would prefer that they die sooner rather than later—that their decision-making is distorted by a misperception of their loved ones’ desires.This worry about burden in the sense of meaning 2 may be what Rodriguez-Prat et al. (2019, 418) have in mind when they say that there “is no evidence … that family members actually shared the patient’s point of view.” The same may be true of Jahn et al. (2011, 215), who speak of perceived burdensomeness as involving “a sense of self-hate. Perceived burdensomeness arises when a person believes he or she is so flawed as to create a liability on others. This perception leads to affectively-laden cognitions of self-hatred.” Some may be unduly influenced by a mistaken belief that others would prefer they were no longer living. But there’s no reason to think this mistaken belief underlies most burden-based decisions of terminal patients. A terminally ill patient and her loved ones may be on the same page in thinking that caring for her involves significant physical, social, or financial costs, while holding different attitudes toward the prolongation of her life. A patient can prefer non-prolongation to imposing those costs, even while understanding that her loved ones would prefer her prolongation. She can prefer an earlier death because she thinks she is a burden in sense 1 without also thinking that she is a burden in sense 2. Indeed, it is not uncommon for a terminally ill patient drawn toward a burdenbased decision to believe she is a burden in sense 1 without thinking she is a burden in sense 2. As Seidlein et al. (2019) have found, many caregivers prefer that their loved ones lives are prolonged despite the costs involved, while at the same time the caregivers would prefer not to impose those costs on their loved ones themselves. Seidlein et al. (2019, 453) write, “[T]he informal caregivers interviewed find themselves in an asymmetric situation: they uncompromisingly accept undertaking care work in the family and the consequences it has on their own lives, but, on the other hand, they would not want to be care recipients themselves, both because of the feeling of being a burden to their next of kin and the fear of not being treated in an appropriate way.” As a woman caring for her husband put it, “I wouldn’t really want [him to do] that. I would also not want my daughter to do it for me. I would not want my daughter to care for me, to clean my backside, and I wouldn’t want my husband to do that either” (Seidlein et al., 453). Many nurses and other professional caregivers also exhibit this preference not to be a burden to their loved ones. “I do not want to be cared for by my child. I know that he should live his own life,” said one professional caregiver in Seidlein et al.’s study (453). Said another, “No, I don’t want my family to have to care for me some day. Concerning my parents, I would never put them in a nursing home, but I would not want it the other way round” (Seidlein et al., 453). I expect many of us can identify with the asymmetrical preferences of this caregiver. These asymmetrical preferences do not result from false beliefs about burdening. This caregiver understands the burdens her loved ones would bear by caring for her; she herself is experiencing those burdens right now. She understands that her children may want to continue to care for her; she herself wants to continue to

18 When Is Self-perceived Burden an Acceptable Reason to Hasten Death?

323

care for her parents. She doesn’t want her children to bear those burdens nonetheless. Recall Example 1 from the previous section. It’s precisely because the person in that example appreciates the burdens involved and realizes that her children may choose to continue to bear those burdens that she puts in her advance directive instructions not to pursue life-extending treatment. I have talked about two senses in which caring for a terminally ill family member can be a burden: the sense of its having significant physical, social, or financial costs; and the sense of preferring that the patient’s life is no longer prolonged. It’s easier to write about this difference on the page than it is to speak about it in actual situations. In actual situations, for one to say that caring for a terminally ill family member is a burden in the first sense might seem to invite the impression that one thinks it’s a burden in the second sense. The lack of acknowledgement of the distinction may partly explain the taboo against admitting, to others and to oneself, that caring for a terminally ill family member is a burden. (Compare to the case of caring for small children: many parents readily acknowledge that caring for small children involves burdens in the first sense without worrying that they will be misunderstood as asserting that it is a burden in the second sense). The lack of acknowledgment of this distinction may thus explain some of the difficulties of communication in end-of-life situations. Clarifying the distinction between the two senses may make communication easier. Undoubtedly some patients’ thinking is distorted by misperceptions about the burdens their loved ones are bearing. It’s important to try to prevent those misperceptions and the decisions they would lead to. It seems reasonable to think that communication about such matters will be facilitated by diminishing the taboo against considering being a burden to be a legitimate consideration. But there’s no warrant for thinking that the elimination of false beliefs would reduce the incidence of burden-based decisions to anywhere near zero.1

18.4 Mistaken Value-Judgments Burden-based decisions could still be regrettable even if patients have the facts right. For burden-based decisions could still involve mistaken value-judgments. Let’s turn now to the possibility of that kind of mistake. The mistaken value-judgment cannot be simply that the patient has decided against prolonging her life. A terminal patient’s decision to hasten death is not in and of itself a regrettable failure. That was the point of Example 2: the decision of the patient of Example 2 is a (sad but) acceptable response to her situation, not an indication of a moral problem. If there is a moral problem in burden-based decisions, it resides in the burden-based reason itself, not simply in the non-prolongation of life. Or maybe we would want to say that there is always some reason in favor of prolonging life, but 1

For discussion of improving communication about burden at the end of life, see Gudat et al. (2019, 426–7).

324

M. B. Gill

that that reason can be overridden by other reasons. Example 2 is an example of the reason in favor of prolonging the life of a competent patient is overridden by other reasons. The question then becomes: why isn’t the wish to unburden loved ones a reason that can acceptably override the reason in favor of prolonging life? This is where Example 1 is relevant. Example 1 shows that the wish to unburden loved ones (at least when one is considering a future situation in which one is unconscious or significantly cognitively impaired) can be an acceptable reason against prolonging life, a reasonable response rather than a moral problem.

18.4.1 Overvaluing Autonomy? A number of commentators maintain, however, that burden-based decisions are their own special moral problem nonetheless. Gunderson (2004, 38) argues that one has a duty to “allow oneself to be a burden” at the end of life. When we choose not to prolong life because we don’t want to be a burden to our loved ones, we “betray” our relationship with them (Gunderson 2004, 38). A true commitment to our relationships with our loved ones, according to Gunderson (2004, 40), “provides a reason for each person to accept the other’s care when care is needed.” Caring relationships “generate reasons to accept being a burden” (Gunderson 2004, 39). Meilaender makes a similar point, arguing that being a burden to each other is what “it means to belong to a family” (Meilaender 2010). Meilaender criticizes any “last-ditch attempt” at the end of life “to bypass the interdependence of human life, by which we simply do and should constitute a burden to those who love us” (Meilaender 2010). In the same vein, Corby (2019) contends that the inclination not to burden others violates what is distinctive about our humanity. He writes, “Is there something about being a burden that is essential to being human? Yes, the anthropological vision that allows a person to burden loved ones is the same that awakens him or her to the true nature of humanity. In other words, the realization of being a burden not only recognizes the interconnectedness of our human relations, but more deeply touches on our existential dependence as created beings. To deny this is to deny ourselves” (Corby 2019, 372). Gunderson, Meilaender, and Corby all believe that incorrect valuing underlies the frame of mind that leads to burden-based decisions. The culprit most often identified is an incorrect overvaluing of self-reliance, autonomy, or independence. We have an irrational fear of dependency, an unreasonable obsession with personal autonomy and control. Fraser expresses the idea when he writes, “I do want to be a burden on my loved ones just as I want them to be a burden on me—it’s called looking after each other. Obviously, I know people are terrified of the indignity of dying and of being ill generally. Having someone wipe our bums, clean up our mess, put up with our incoherent ramblings and mood swings is a threat to our cherished sense of personal autonomy. But this is where the liberal model of individual self-determination breaks down” (Fraser). We err in thinking

18 When Is Self-perceived Burden an Acceptable Reason to Hasten Death?

325

that dependence is the highest form of indignity. Once we correct that mistake, Gunderson, Meilaender, Corby, and Fraser believe, we will lose our reluctance to allow ourselves to burden others. Kittay has developed one of the richest accounts of the view suggested by Gunderson, Meilaender, Corby, and Fraser. According to Kittay (2013), our society’s overemphasis on individual autonomy has prevented us from acknowledging that dependency is an essential feature of human life. It has prevented us from realizing that we have a “moral responsibility to receive with grace care that is offered in good will and with the requisite competence” (81). Overemphasis on individual autonomy threatens to “unfit” us for the “virtues of acknowledged dependence” that ought to be part of human life (Kittay 2011, 56).2 Many who write on self-perceived burden at the end of life agree with Kittay that negative feelings about being a burden result from a lack of acknowledgment of the essential dependency of human being. Malpas et al. (2012, 16) write, “Although dependence on others is an essential feature of human development, and an essential condition of what it is to be a human person some participants seemed to view dependency as deficiency; almost as though the loss of independence was a failure on their part.” Metselaar and Widdershoven (2019, 435) believe we ought to approach issues of self-perceived burden with an ethics of care that starts from the “fundamental assumption that vulnerability is intrinsic to the human condition, and acknowledges that we realize our life goals while we are interdependent and immersed in a web of caring relationships with others.” Rodriguez-Prat et al. (2019, 419) believe that the failing behind burden-based decisions is an unnatural “aversion to caring for the elderly or the ill.” This mistake in valuing is often attributed to Western culture’s obsession with autonomy and control. According to this way of thinking, Western culture fetishizes self-determination. It pegs dignity and self-worth to independence. It views reliance on others as a deficiency or failure. These ideas lead to an inability to accept our frailty and dependence at the end of life. As a result, when we perceive we are a burden, we judge that our lives are no longer worth living. We are incapable of seeing the worth in our continued existence once the independence that had given our lives value has gone. This is based on a faulty notion of what we human beings are. As other cultures have long realized, dependence is intrinsic to human life. To disvalue being a burden is to deny our essential nature. Truly to understand humanity is to realize that we are always dependent on others. Being a burden is not a deviant condition that must be eliminated. It is essential to what it means to be human. Once we embrace this truth about ourselves, we will no longer judge being a burden as a reason not to prolong life. As Rodriguz-Prat et al. (2019, 419) put it, we’ll come to a view of human life

2

Kittay is drawing on Baier’s claim that morality that emphasizes autonomy, “if unsupplemented, may unfit people to be anything other than what its justifying theories suppose them to be, ones who have no interest in each others’ interests” (Baier 1994, 29).

326

M. B. Gill

that makes it “impossible to ever feel like a burden to others, as is the case in some cultures.” But evidence does not support the idea that concern about burdening loved ones at the end of life is due to uniquely Western values. The concern is salient not only in North America, Europe, and New Zealand, but also in Japan, South Korea, Taiwan, and Hong Kong. In a study of Japanese patients, Akazawa et al. (2010, 230) found that the levels of self-perceived burden are “generally close to those reported by patient and family studies conducted in Western countries,” and that not being a burden is “an important factor in achieving a good death in terminally ill cancer patients in both Western and Eastern cultures” (Akazawa et al. 2010, 225). Kuo et al. (2018, 2) explicitly considered the question of whether “factors associated with high SPB [self-perceived burden] may differ between Asian and Western countries due to different cultural considerations.” Kuo et al. found that Taiwanese patients’ self-perceived burden was in certain respects comparable to, and in other respects greater than, self-perceived burden of Western patients. Chan et al.’s (2009) study of patients in Hong Kong had similar results. Chan et al. (2009, 526) suggest that concerns about being a burden at the end of life might actually be more pronounced in Chinese culture because of the dominant value placed on “family-style collectivism” and “interdependence”(see also Hirai et al. 2006; Steinhauser et al. 2000). These results suggest the opposite of the claim that concerns about self-perceived burden are due to distinctively Western values of independence, self-reliance, or autonomy (see Akazawa et al. 2010; Steinhauser et al. 2000; Hirai et al. 2006. Kuo et al. 2018: Chan et al. 2009). Nor in Western studies is there clear evidence of demographic differences that would suggest that concerns about burden are unique to the values of any particular aspect of Western culture. In a study of self-perceived burden among stroke patients, McPherson et al. (2010, 198) found, “There were no significant correlations between care recipient SPBS scores and any demographic variables, including age, sex, education, or time elapsed since the stroke.” Wilson et al. (2005, 119) found among advanced cancer patients “no differences between [low and high-burden] groups in the basic demographic characteristics of age, gender or education.” Leroy et al. (2016, 1278) found that “SPB affects a large proportion of patients at the end of their life (one to three terminally ill cancer patients in four according to studies), regardless of their culture.” Suri Rita et al. (2011, 2320) “found no associations among perceived caregiver burden with demographic factors, including age, sex, race and level of education.”3 Be that as it may, we still need to address the claim that burden-based decisions are regrettable. For if burden-based decisions in one culture exemplify an unreasonable or irrational obsession with personal autonomy, such decisions will be problematic regardless of whether other cultures accept burden-based decisions for different 3

In a qualitative study of US patients, Cahill et al. 2009 found that the white patients they interviewed were somewhat more likely than black patients to make decisions based on concern about being a burden, and that the women they interviewed were somewhat more likely to make such decisions than the men. But the numbers were small, and the differences were not extreme enough to imply that burden-based decisions are almost always the result of the values of a certain demographic.

18 When Is Self-perceived Burden an Acceptable Reason to Hasten Death?

327

reasons. And the decisions of other cultures could be unreasonable and irrational even if they are based on reasons other than autonomy and control. Many who contend that burden-based decisions are regrettable claim that such decisions are based on values that fail to cohere with the natural or essential interdependence of human beings. There is a general and a specific problem with this claim. The general problem is the difficulty of basing substantive and controversial judgments about how to live on ideas about what is natural or essential to human beings. If the concept of what is natural or essential to human beings is scientific or descriptive, there will be significant questions about how to draw from it any interesting and robust prescriptive conclusions. If the concept is moral or normative, there will be significant questions about how to formulate it in a non-question-begging way that is determinate enough to adjudicate contested practical issues. The specific problem is the difficulty of bringing ideas about what is natural or essential into contact with medical decisions facing twenty-first century patients with terminal disease. Many patients facing these end-of-life situations suffer from illnesses that would have already killed them were it not for recent advances of technologically sophisticated interventions. They are not only living longer than humans with similar conditions would have lived in the past. They are living with effects of disease and invasive treatment that have not been part of past human experience. Many of the challenges facing them on a day-to-day basis—and many of the challenges facing their caregivers—were not faced by most people throughout human history. It is hard to see the principled reason for claiming that choosing an extremely technologically advanced life-extending treatment in order to live for an additional six months is more natural than allowing a disease to take its course. Kittay (2005, 112) writes, “No other creatures devote themselves to the care of the ill, the disabled and the frail elderly that humans do as a matter of course… Our caring for … those who are ill, impaired or frail exhibits a moral capacity as characteristically human and as worthy of human dignity as moral autonomy.” Kittay makes a profound point. But it does not follow that there is something “characteristically human” about administering long-term round-the-clock modern medical care to a patient undergoing chemotherapy—or that doing so comes naturally to us. This is important to note because it may help allay caregivers’ unwarranted feelings of guilt or inadequacy. That a caregiver finds it extraordinarily trying to administer twentyfirst century medical care to a patient with a late-stage disease that would have taken the life of someone from any other period of human history should not make the caregiver feel that she is lacking in care or concern that is natural to human beings. Nor should she think herself in any way deficient in the milk of human kindness if she perceives that her loved one perceives that this twenty-first century effort is extraordinarily difficult for her.

328

M. B. Gill

This distinction between traditional caring and aspects involved in the medical care of a patient undergoing twenty-first century life-extending treatment points to another problem with this objection to accepting burden-based decisions. Valuing interdependence rather than independence is not an all-or-nothing affair. It may be fundamentally mistaken for a human being to aspire to complete independence. To value absolute self-reliance—to disvalue help of any kind—may very well be to deny something one ought not deny. But it may be sensible nonetheless to form judgments about the desirability of being more or less dependent in particular situations. There are some kinds of help I ought to accept from family and friends, some kinds of help the refusal of which would be not only self-destructive but also disrespectful and hurtful. But it does not follow that I ought never turn down any assistance, that I ought to accept all the help that is ever offered. There may be some kinds of assistance I might understandably choose not to ask for, even if I believe a loved one would give it without question. There may be some offers of assistance it may be wise for me to refuse, because I think it is too much for the person to offer or too much or me to accept. That a patient does not wish to place certain burdens on others should not be equated with a blanket denial of dependency. Her wish may coexist with her accepting help—affirming her dependency—in many other respects. A terminal patient may accept the care of her loved ones for months or years. She may affirm wholeheartedly the interdependence of human life, and her position as dependent on others. At that the same time, she may come to think that there is a certain point beyond which she does not wish certain kinds of care to extend. That is consistent with her having previously accepted help graciously and with love, and with continuing to accept with grace and love palliative care and other kinds of assistance as the end approaches. To characterize such a person as denying her essential human dependency is simply inaccurate. That making a burden-based decision is compatible with accepting a wealth and depth of care and concern also has an important implication for caregivers: giving profound care and concern to a loved one is compatible with accepting her burdenbased decision. Accepting a loved one’s burden-based decision does not entail ceasing to care for her with love. As a result of a loved one’s making a burden-based decision, the quantity of time spent with her will diminish. But the quality need not. In the time she has remaining, she will still have needs to attend to, emotional and spiritual matters to engage with. Her decision will not eliminate the opportunity to show her care, concern, love. Worries about burden-based decisions sometimes seem animated by a vision of meaningful end-of-life interactions in other eras and cultures. But those interactions do not require the use of all available high-tech life-extending treatments.

18.4.2 Values that Make for a Worse Human Life? The claim that burden-based decisions conflict in some important way with what is essential or natural to being human is also construed as the claim that the values

18 When Is Self-perceived Burden an Acceptable Reason to Hasten Death?

329

opposed to burden-based decisions make for a better human life. According to this line of thought, if there are two patients in the same external circumstances—both have debilitating terminal disease, both are being cared for by their loved ones, and the loved ones of both are experiencing the same level of caregiver burden that studies have found to be very common—the patient who does not consider a burden-based decision will have the values that we should generally promote because those are the values that will make us better off. There are three problems with this view. The first problem is that there is no one set of values that characterizes all those who make burden-based decisions, and no one set of values that characterizes all those who do not. People fall into both categories for a wide variety of reasons. This objection to burden-based decisions seems to turn on the idea that we can identify one (unfortunate) common value-system at the root of all burden-based decisions. But the evidence from multiple cultures and demographic groups gives us reason to doubt there is that kind of uniformity. One could claim, I suppose, that any set of values that leads to a decision not to prolong life is self-evidently worse than any set of values that leads to a decision to prolong life. But that claim would be far from uncontroversial. DNRs and the discontinuation of life-extending treatment are not generally considered indications of mistaken values. The patients I described in examples 1 and 2 exhibit kinds of end-of-life decision-making that are not universally considered indicative of regrettable attitudes toward human life. The second problem is that a patient’s burden-based decision may be downstream of values that are not going to change. Even if it is true in some cases that the patient would be better off if she had held values incompatible with a burden-based decision, a burden-based decision might be better for her given the values fixed in her character. My father’s suffering in his last year may have been exacerbated by certain entrenched ideas and tendencies of his. His last year might have been significantly improved had he held different views about life’s purpose and meaning. But those entrenched ideas and tendencies were not going to be dislodged in his final months. That he might have been happier if he had held values incompatible with a burden-based decision does not alter the fact that a burden-based decision may have been the right thing for him given the values that until the end he actually did hold. Some people undergo transformations at the end of life. Epiphanies do occur. But not always. We might not do a person any favors by resisting her burden-based decision because we think she would be better off if she came to a new, profound realization about what it is to be a human being. Accepting a burden-based decision may be inseparable from accepting who a person is. The third problem with the rejection of values underlying burden-based decisions is that it presupposes substantive answers to the question of what makes for a thriving human life. But it is far from clear anyone has the right answers to that question, or that there even is a fact of the matter as to what the right answers are. A recognition of the imponderability of this issue is what underlies the most common view of appropriate decisions for patients at the end of life. That common view is that the appropriate decision is the one that accords with the values the patient has affirmed and conducted herself by throughout her life. This view is neutral between a valuesystem that emphasizes independence, autonomy, or self-reliance, and a value-system

330

M. B. Gill

that emphasizes dependence, or connection, or care. It does not grant priority to values of individualism or to values of community. It holds that whatever mattered to the way the person lived is what we ought to honor in treating the person as she dies. Some people may have values that are compatible with burden-based decisions, while others have values that are incompatible. Rather than endorsing one approach over the other, the common view accepts decision-making that is of a piece with the patient’s life as a whole.4

18.4.3 Values Distorted by Depression? But what about depression? Terminally ill patients who perceive themselves to be a burden may be depressed (see Kowal et al. 2012; Van Orden et al. 2012; Chio et al., 2005; McPherson et al. 2007a; Wilson et al. 2005). But, someone who objects to burden-based decisions might contend, depression distorts decision-making. Terminally ill patients may thus not be making decisions in accord with their own values after all. We shouldn’t take their burden-based decisions at face value any more than we would take at face value the suicidal thoughts of a physically healthy twenty-yearold who is undergoing a depressive episode. In both cases, the appropriate response is to treat the depression. Treatment can lead a patient away from her wish to hasten death. The result will be that the patient gains valuable continued life that she would have lost if we had accepted her earlier decision at face value. To evaluate this worry, let’s start by distinguishing between those terminal patients who are depressed, and those who are not. It is undoubtedly appropriate to try to treat the depression of those in the first category. It will be a good result if the patient responds quickly to the treatment and then chooses a life-extending course of action. But what of the patient who does not respond quickly? What attitude ought we bear to her burden-based decision? Obviously relevant is that the patient is terminal. Her time is limited. This makes her situation different from that of a physically healthy twenty-year-old considering suicide. Even if the twenty-year-old has not responded well to months of treatment, it might be reasonable to continue to urge measures that may lead her to change her mind. But it could be best to accept the state of mind of a terminal patient sooner than that. Resisting her decision because we think that were she not depressed she would come to a different decision might succeed only in making us unaccepting of something about her that is not going to change—and in extending her misery. Moreover, it’s problematic to equate the depression of a twenty-year-old to the state of mind of a terminal patient considering a burden-based decision. A terminal patient may meet criteria of being depressed because she is responding to her situation in accord with her settled ideas about what makes life worthwhile. What is deemed a depressive state of mind may be downstream of her long-term concerns. If we mean 4

Dworkin (1993, 207) writes, “Making someone die in a way that others approve, but he believes a horrifying contradiction of his life, is a devastating, odious form of tyranny.”

18 When Is Self-perceived Burden an Acceptable Reason to Hasten Death?

331

by ‘depression’ a cognitively distorting condition, it would be incorrect to label her depressed. If we mean by ‘depression’ a certain psychological that includes the wish not to prolong life, then she will be depressed, but such a label will tell us nothing about whether the reasons guiding her decisions ought to be deemed problems to be solved or values to accept. And then there is the category of terminal patients considering burden-based decisions who are not depressed. Van Orden et al. (2012, 856) found that the connection between self-perceived burden and the wish not to prolong life can “persist above and beyond the contribution of depressive symptoms to meaning in life… [T]he relationship between perceived burdensomeness and meaning in life is not solely of depression.”5 It is reasonable to worry that a patient’s burden-based decision results not from her enduring values but from psychological distortions of depression. But it is a worry we should recognize can be allayed.

18.4.4 Values that Harm Vulnerable Populations I’ve been discussing burden-based decisions in relation to the mental states of terminal patients. But there is an objection to accepting burden-based decisions that concerns how such attitudes bear on treatment of other groups. According to this objection, terminal patients who make burden-based decisions are motivated by aversion to dependency. They act on the idea that a life dependent on the continuous care of others is not worth living. To accept their burden-based decisions is to affirm that attitude toward dependency. But another group of people also require continuous care—people with disabilities. To accept the attitude toward dependency that underlies burden-based decisions is thus to affirm the idea that the lives of people with disability are less valuable. Acceptance of burden-based decisions thus promotes the ableist prejudice that independence is essential for dignity and worth. In response, I refer first to my earlier point that attitudes toward dependency are not all-or-nothing. A terminal patient who has decided not to prolong life may embrace her dependency in many ways. That she has reached a point in her disease that makes continued life-extending treatment undesirable to her is entirely compatible with her accepting all sorts of other care with grace and love. It is a mistake to conflate her decision with a fetish for self-reliance or an aversion to dependency. It is equally mistaken to conflate loved ones’ acceptance of her decision with that fetish. Her loved ones can recognize her reasons for not prolonging life while continuing to provide her compassionate care. They do not have to resist her burden-based decision in order to express and act on their love for her. Accepting an individual’s decision not to prolong life is compatible with cherishing her existence. It need not embody any devaluation of her worth.

5

Van Orden et al. (2012) are not contending that we should accept burden-based decisions. They are in favor of interventions to try to prevent burden-based decisions.

332

M. B. Gill

Will acceptance of burden-based decisions have a wider societal effect of feeding ableist prejudice? Here is a reason to think it will not. Underlying that acceptance may be not aversion to dependency but acceptance of terminal patient’s making endof-life decisions in accord with their own values. Such acceptance does not privilege autonomy or self-determination above care and connection. It prioritizes honoring whatever concerns are most important to patients themselves. As we have seen, moreover, autonomy-prioritizing value-systems do not have a monopoly on burdenbased decisions. Such decisions are also evident in cultures typically contrasted with the West and its prioritization of autonomy. Between burden-based decisions and the fetishization of self-reliance there is no necessary connection. At this point I am also inclined to say that empirical evidence does not support the worry that acceptance of burden-based decisions will have deleterious effect on societal attitudes toward people with disability. This worry seems to me similar to the worries that were raised in the nineteen-eighties and nineties that legalizing physician-assisted suicide would have deleterious effects on palliative care. Those worries about the effects of legalizing physician-assisted suicide have been refuted by the empirical evidence that has been collected over the last twenty-five years. I do not see compelling reason to believe empirical evidence will do any more to vindicate the worry that acceptance of burden-based decisions will have deleterious effects on attitudes toward disability. I am aware, however, that holding out for empirical evidence on this matter may seem inappropriate—because of the gravity of the potential harms, because the concerns at issue may not be reducible to quantifiable correlations, and because the lived experiences of people with disabilities give their judgments more authority on this matter than armchair arguments about the separability of various attitudes toward burden and disability. So I will close by saying only that I hope to have articulated a position on burden-based decisions that is reasonable and humane, albeit with awareness that the path societal attitudes travel is often far from the most reasonable and humane one.

References Akazawa, Terukazu, Tatsuo Akechi, Tatsuya Morita, Satoru Tsuneto, Yasuo Shima, and Toshiaki A. Furukawa. 2010. Self-perceived burden in terminally ill cancer patients. Journal of Pain and Symptom Management 40 (2): 224–234. Akechi, T., T. Okuyama, Y. Sugawara, T. Nakano, Y. Shima, and Y. Uchitomi. 2004. Suicidality in terminally ill Japanese patients with cancer. Cancer 100: 183–191. Ashby M, C. Op’t Hoog, A. Kellehear, et al. 2005. Renal dialysis abatement: Lessons from a social study. Palliative Medicine 19: 389–396. Back, A.L., J.I. Wallace, H.E. Starks, and R.A. Pearlman. 1996. Physician-assisted suicide and euthanasia in Washington State: Patient requests and physician responses. Journal of the American Medical Association 275: 919–925. Baier, Annette. 1994. Moral Prejudices: Essays on Ethics. Cambridge, MA: Harvard University Press.

18 When Is Self-perceived Burden an Acceptable Reason to Hasten Death?

333

Bausewein, Claudia, Natalie Calanzani, Barbara A. Daveson, et al. 2013. ‘Burden to others’ as a public concern in advanced cancer: A comparative survey in seven European countries. BMC Cancer 13 (105). https://doi.org/10.1186/1471-2407-13-105 Berkman, C.S., P.F. Cavallo, W.C. Chesnut, and N.J. Holland. 1999. Attitudes toward physicianassisted suicide among persons with multiple sclerosis. Journal of Palliative Medicine 2: 51–63. Cagle, John G., Dawn C. Carr, Seokho Hong, and Sheryl Zimmerman. 2016. Financial burden among US households affected by cancer at the end of life. Psycho-Oncology 25: 919–926. Cahill, Eileen, Lisa M. Lewis, Frances K. Barg, and Hillary R. Bognar. 2009. ’You don’t want to burden them’: Older adults’ views on family involvement in care. Journal of Family Nursing 15 (3): 295–317. Chan, Wallace C.H., Irwin Epstein, Dona Reese, and Cecilia L.W. Chan. 2009. Family predictors of psychosocial outcomes among Hong Kong Chinese cancer patients in palliative care. Social Work in Health Care 48: 519–532. Chio, A., A. Gauthier, A. Calvo, P. Ghiglione, and R. Mutani. 2005. Caregiver burden and patients’ perceptions of being a burden in ALS. Neurology 64: 1780–1782. Chochinov, Harvey Max, Linda J. Kristjanson, Thomas F. Thomas, Susan McClement. Hassard, and Michael Harlos. 2002. Burden to others and the terminally ill. Journal of Pain and Symptom Management 34 (5): 463–471. Chochinov, Harvey Max, Thomas Hack, Thomas Hassard, Linda J. Kristjanson, Susan McClement, and Mike Harlos. 2005. Understanding the will to live in patients nearing death. Psychosomatics 46: 7–10. Chochinov, Harvey Max, Thomas Hack, Thomas Hassard, Linda J. Kristjanson, Susan McClement, and Mike Harlos. 2007. Burden to others and the terminally ill. Journal of Pain and Symptom Management 34 (5): 463–471. Cohen, S. Robin, and Anne Leis. 2002. What determines the quality of life of terminally ill cancer patients from their own perspective? Journal of Palliative Care 18 (1): 48–58. Cohen-Mansfied, J., J.A. Droge, and N. Billig. 1992. Factors influencing hospital patients’ preferences in the utilization of life-sustaining treatments. The Gerontologist 32 (1): 89–95. Corby, Paschal H. 2019. The fear of being a burden on others. The National Catholic Bioethics Quarterly 19 (3): 369–376. Cousineau, Natalie, Ian McDowell, Steve Hotz, and Paul Hebert. 2003. Measuring chronic patients’ feelings of being a burden to their caregivers: Development and preliminary validation of a scale. Medical Care 41 (1): 110–118. De Faye, Barbara J., Keith G. Wilson, Susan Charter, Raymond A. Viola, and Pippa Hall. 2006. Stress and coping with advanced cancer. Palliative Support Care 4 (3): 239–249. Diwan, S., G.W. Hougham, and G.A. Sachs. 2004. Strain experienced by caregivers of dementia patients receiving palliative care. Journal of Palliative Medicine 7 (6): 797–807. Dworkin, Ronald. 1993. Life’s Dominion. London: HarperCollins. Dyeson, T.B. 2000. Burden self-image: A mediating variable of depressive symptoms among chronically il care recipients. Journal of Gerontological Social Work 33 (1): 17–33. Eitel, P., L. Hatchett, R. Friend, K.W. Griffin, and N.K. Wadhwa. 1995. Burden of self-care in seriously ill patients: Impact on adjustment. Health Psychology 14 (5): 457–463. Filiberti A, C. Ripamonti, A. Totis, et al. 2001. Characteristics of terminal cancer patients who committed suicide during a home palliative care program. Journal of Pain and Symptom Management 22: 544–553. Foley, G., V. Timonen, and O. Hardiman. 2016. ‘I hate being a burden’: The patient perspective on carer burden in amyotrophic lateral sclerosis. Amyotrophic Lateral Sclerosis and Frontotemporal Degeneration 17 (5–6): 351–357. Fraser, Giles. 2013. I want to be a burden on my family as I die, and for them to be a burden on me. Guardian 3 May 2013. Accessed on January 12, 2021: https://www.theguardian.com/commentis free/belief/2013/may/03/burden-loved-ones-dying-euthanasia Galvin, M., B. Corr, C. Madden, et al. 2016. Caregiving in ALS—A mixed methods approach to the study of Burden. BMC Palliative Care 15 (1): 81.

334

M. B. Gill

Ganzini, L., W.S. Johnston, B.H. McFarland, S.W. Tolle, and M.A. Lee. 1998. Attitudes of patients with amyotrophic lateral sclerosis and their care givers toward assisted suicide. The New England Journal of Medicine 339 (14): 967–973. Ganzini, L., H.D. Nelson, T.A. Schmidt, D.F. Kraemer, M.A. Delorit, and M.A. Lee. 2000. Physicians’ experiences with the Oregon death with dignity act. New England Journal of Medicine 342 (8): 557–563. Ganzini, Linda, Maria J. Silvera, and Wendy S. Johnston. 2002. Predictors and correlates of interest in assisted suicide in the month of life among ALS patients in Oregon and Washington. Journal of Pain and Symptom Management 24 (3): 312–317. Ganzini, L., T.M. Beer, and M. Brouns. 2006. Views on physician-assisted suicide among family members of Oregon cancer patients. Journal of Pain and Symptom Management 32 (3): 230–236. Gauthier, A., A. Vignola, A. Calvo, et al. 2007. A longitudinal study on quality of life and depression in ALS patient–caregiver couples. Neurology 68: 923–926. Geng, D., R. Ou, X. Miao, L. Zhao, Q. Wei, X. Chen, et al. 2017. Patients’ self-perceived burden, caregivers’ burden and quality of life for amyotrophic lateral sclerosis patients: A cross-sectional study. Journal of Clinical Nursing 26 (19–20): 3188–3199. Given B.A., C.W. Given, and S. Kozachik. 2001. Family support in advanced cancer. CA: A Cancer Journal for Clinicians 51 (4): 213–231. Glajchen, M. 2004. The emerging role and needs of family caregivers in cancer care. The Journal of Supportive Oncology 2 (2): 145–155. Grunfeld, E., D. Coyle, T. Whelan, J., et al. 2004. Family caregiver burden: results of a longitudinal study of breast cancer patients and their principal caregivers. Canadian Medical Association Journal 170(12): 1795–1801. Gudat, Heike, Katherin Ohnsorge, Nina Streeck, and Christoph Rehmann-Sutter. 2019. How patients’ feelings of being a burden to others can motivate a wish to die. Moral challenges in clinics and families. Bioethics 33: 421–430. Gunderson, Martin. 2004. Being a burden: Reflections on refusing medical medical care. Hastings Center Report 34 (5): 37–43. Hirai K, M. Miyashita, T. Morita, M. Sanjo, and Y. Uchitomi. 2006. Good death in Japanese cancer care: a qualitative study. Journal of Pain and Symptom Management 31. Hoefman, R.J., J. van Exel, and W.B. Brouwer. 2013. Measuring the impact of caregiving on informal carers. Health and Quality of Life Outcomes 11: 173. Jahn, D.R., K.C. Cukrowicz, K. Linton, and F. Prabhu. 2011. The mediating effect of perceived burdensomeness on the relation between depressive symptoms and suicide ideation in a community sample of older adults. Aging & Mental Health 15 (2): 214–220. Kittay, E.F. 2011. The ethics of care, dependence, and disability. Ratio Juris 24 (1): 49–58. Kittay, E.F. 2013. Caring for the long haul: Long-term care needs and the (moral) failure to acknowledge them. The International Journal of Feminist Approaches to Bioethics 6: 66–88. Kittay, E.F. 2005. Equality, dignity, and disability. In Perspectives on Equality, eds. Mary Ann Lyons and Fionnuala Waldron, 93–119. Dublin: The Liffey Press. Kowal, J., K.G. Wilson, L.A. McWilliams, K. Péloquin, and D. Duong. 2012. Self-perceived burden in chronic pain: Relevance, prevalence, and predictors. Pain 153 (8): 1735–1741. Kuo, Su-Chian, Wen-Chi Chou, Ming-Mo Hou, Chiao-En Wu, Wen-Chi Shen, Fur-Hsing Wen, and Siew-Tzuy Tang. 2018. Changes in and modifiable patient- and family caregiver-related factors associated with cancer patients’ high self-perceived burden to others at the end of life. European Journal of Cancer Care 27. Laizner, A.M., L.M. Yost, F.K. Rag, and R. McCorkle. 1993. Needs of family care-givers of persons with cancer: A review. Seminars in Oncology Nursing 9: 114–120. Lee, J., D.W. Shin, J. Cho, et al. 2015. Caregiver burden, patients’ self-perceived burden, and preference for palliative care among cancer patients and caregivers. Psycho-Oncology 24 (11)L: 1545–1551.

18 When Is Self-perceived Burden an Acceptable Reason to Hasten Death?

335

Leroy, Tanguy, Emmanuelle Fournier, Nicolas Penel, and Veronique Christophe. 2016. Crossed views of burden and emotional distress of cancer patients and family caregivers during palliative care. Psycho-Oncology 25: 1278–1285. Lulé, D., S. Nonnenmacher, S. Sorg, et al. 2014. Live and let die: Existential decision processes in a fatal disease. Journal of Neurology 261 (3): 518–525. Mak, Y.Y.W., and G. Elwyn. 2005. Voices of the terminally ill: Uncovering the meaning of desire for euthanasia. Palliative Medicine 19 (1): 343–350. Malpas, Phillipa J., Kay Mitchell, and Malcolm H. Johnson. 2012. ‘I wouldn’t want to become a nuisance under any circumstances’—A qualitative study of the reasons some healthy older individuals support medical practices that hasten death. The New Zealand Medical Journal 125: 9–19. McGlade, K.S., L. Slaney, B.P. Bunting, and A.G. Gallagher. 2000. Voluntary euthanasia in Northern Ireland: General practitioners’ beliefs, experiences, and actions. British Journal of General Practice 50: 794–797. McPherson, Christine J., Keith G. Wilson, Michelle M. Lobchuk, and Susan Brajtman. 2007a. Selfperceived burden to others: Patient and family caregiver correlates. Journal of Palliative Care 23 (3): 135–142. McPherson, Christine J., Keith G. Wilson, and M.A. Wilson. 2007b. Feeling like a burden to others: A systematic review focusing on the end of life. Palliative Medicine 21: 115–128. McPherson, Christine J., Keith G. Wilson, Livia Chyurlia, and Charles Leclerc. 2010. The balance of give and take in caregiver-partner relationships: An examination of self-perceived burden, relationship equity and quality of life from the perspective of care recipients following stroke. Rehabilitation Psychology 55 (2): 194–203. Meilaender, Gilbert. 2010. I want to burden my loved ones. First Things March 2010. https://www. firstthings.com/article/2010/03/i-want-to-burden-my-loved-ones Metselaar, S., and G. Widdershoven. 2019. Moral dilemmas in (not) treating patients who feel they are a burden. Bioethics 33: 431–438. Morita, T., Y. Sakaguchi, K. Hirai, S. Tsuneto, and Y. Shima. 2004. Desire for death and requests to hasten death of Japanese terminally ill cancer patients receiving specialized inpatient palliative care. Journal of Pain and Symptom Management 27: 44–52. Morita, T., J. Tsunoda, S. Inoue, and S. Chihara. 2000. An exploratory factor analysis of existential suffering in Japanese terminally ill cancer patients. Psycho-Oncology 9: 164–168. Murray, M.A., A.M. O’Connor, V. Fiset, and R. Viola. 2003. Women’s decision-making needs regarding place of care at end of life. Journal of Palliative Care 19: 176–184. Oldenkamp, M., M. Hagedoorn, R. Wittek, R. Stolk, and N. Smidt. 2017. The impact of older person’s frailty on the care-related quality of life of their informal caregiver over time. Quality of Life Research 26 (10): 2705–2716. Patrick, D.L., R.A. Pearlman, H.E. Starks, K.C. Cain, W.G. Cole, and R.F. Uhlman. 1997. Validation of preferences for life-sustaining treatment: Implications for advance care planning. Annals of Internal Medicine 127: 509–517. Pearlman, R.A., K.C. Cain, D.L. Patrick, M. Appelbaum-Maizel, H.E. Starts, N.S. Jecker, and R.F. Uhlmann. 1993. Insights into patient assessments of states worse than death. Journal of Clinical Ethics 4 (1): 33–41. Rehmann-Sutter, Christoph. 2015. End-of-life ethics from the perspectives of patients’ wishes. In The Patient’s Wish to Die: Research, Ethics, and Palliative Care, ed. Christoph Rehmann-Sutter, Heike Gudat, and Katherin Ohnsorge, 161–171. Oxford University Press. Rodriguez-Prat, A., A. Balaguer, I. Crespo, and C. Monforte-Roye. 2019. Feeling like a burden to others and the wish to hasten death in pateints with advanced illness: A systematic review. Bioethics 33 (4): 411–420. Schneider, J., J. Murray, S. Banerjee, and A. Mann. 1999. EUROCARE: A cross-national study of co-resident spouse carers for people with Alzheimer’s disease. International Journal of Geriatric Psychiatry 14 (8): 651–661.

336

M. B. Gill

Scholte op Reimer, W.J., R.J. de Haan, P.T. Rijnders, and G.A. van den Bos. 1998. The burden of caregiving in partners of long-term stroke survivors. Stroke 29: 1605–1611. Seidlein, Anna-Henrikje., Ines Budhholz, Maresa Buchholz, and Sabine Salloch. 2019. Relationships and burden: An empirical-ethical investigation of lived experience in home nursing arrangements. Bioethics 33 (4): 448–456. Simmons, A.L. 2007. Self-perceived burden in cancer patients: Balidation of the self-perceived burden scale. Cancer Nursing 30 (5): 405–411. Singer, P.A., D.K. Martin, and M. Kelnor. 1999. Quality end-of-life care: Patient’s perspectives. Journal of American Medical Association 218 (2): 163–168. Song, J.I., D.W. Shin, J.Y. Choi, et al. 2012. Quality of life and mental health in the bereaved family members of patients with terminal cancer. Psycho-Oncology 21 (11): 1158–1166. Steinhauser, K.E., N.A. Christakis, E.C. Clipp, et al. 2000. Factors considered important at the end of life by patients, family, physicians, and other care providers. Journal of the American Medical Association 214 (19): 2476–2482. Stetz, K.M. 1987. Caregiving demands during advanced cancer: The spouse’s needs. Cancer Nursing 10: 260–268. Street, A., and D.W. Kissane. 2000. Dispensing death, desiring death: An exploration of medical roles and patient motivation during the period of legalized euthanasia in Australia. Omega (westport) 1999–2000 (40): 231–248. Sullivan, A.D., K. Hedberg, K., and D. Hopkins. 2001. Legalized physician-assisted suicide in Oregon, 1998–2000. New England Journal of Medicine 344 (8): 605–607. Suri Rita, S., Brett Larive, Amit X. Garg, et al. 2011. Burden on caregivers as perceived by hemodialysis patients in the frequent hemodialysis network (FHN) trials. Nephrology Dialysis Transplantation 26 (7): 2316–2322. Torti, F.M., Jr., L.P. Gwyther, S.D. Reed, J.Y. Friedman, and K.A. Schulman. 2004. A multinational review of recent trends and reports in dementia caregiver burden. Alzheimer’s Disease and Associated Disorders 18: 99–109. Van Orden, K.A., P.M. Bamonti, D.A. King, and P.R. Duberstein. 2012. Does perceived burdensomeness erode meaning in life among older adults? Aging & Mental Health 16 (7): 855–860. Vig, E.K., and R.A. Pearlman. 2004. Quality of life while dying: A qualitative study of terminally ill older men. Journal of American Geriatric Society 51: 1595–1601. Virik, K., and P. Glare. 2002. Requests for euthanasia made to a tertiary referral teaching hospital in Sydney, Australia in the year 2000. Supportive Care in Cancer 10: 309–313. Weitzner, M.A., P.B. Jacobsen, H. Wagner Jr., J. Friedland, and C. Cox. 1999. The caregiver quality of life index–cancer (CQOLC) scale. Quality of Life Research 8: 55–63. Wilson, Keith G., Dorothyann Curran, and Christine J. McPherson. 2005. A burden to others: A common source of distress for the terminally ill. Cognitive Behaviour Therapy 34 (2): 115–123. Wilson, K.G., L.A. McWilliams, and K. Péloquin. 2013. Chronic pain and the interpersonal theory of suicide. Rehabilitation Psychology 58 (1): 111–115. Zweibel, N.R., and C.K. Cassel. 1989. Treatment choices at the end-of-life: A comparison of decisions by older patients and their physician-selected proxies. The Gerontologist 29: 615–621.

Michael B. Gill is Professor of Philosophy at the University of Edinburgh. He has published numerous articles on organ donation and physician-assisted death. He also writes on meta-ethics and the history of ethics, and is the author of Humean Moral Pluralism (Oxford University Press, 2014) and A Philosophy of Beauty (Princeton University Press, 2022).