Autonomy and assisted suicide: the execution of freedom.
Abstract: Proponents of assisted suicide who base their arguments on autonomy err in ways that are little attended to. In the absence of a substantive theory of the good, in neither a descriptive nor an ascriptive sense can the concept of autonomy distinguish those acts that should be morally prohibited from those that may be permitted. And to impose a particular theory of the good, whether individual liberty or the sanctity of life, violates the autonomy of those who do not share a commitment to that theory.
Subject: Assisted suicide (Ethical aspects)
Autonomy (Philosophy) (Ethical aspects)
Author: Safranek, John P.
Pub Date: 07/01/1998
Publication: Name: The Hastings Center Report Publisher: Hastings Center Audience: Academic; Professional Format: Magazine/Journal Subject: Biological sciences; Health Copyright: COPYRIGHT 1998 Hastings Center ISSN: 0093-0334
Issue: Date: July-August, 1998 Source Volume: v28 Source Issue: n4
Accession Number: 21195465
Full Text: For the last two thousand years, most western civilizations have proscribed assisted suicide and other forms of euthanasia as a violation of innocent human life. The Hippocratic Oath, professed by doctors through the centuries, explicitly condemns administering a deadly drug or even suggesting it. Yet in recent decades scholars have criticized this traditional position for violating personal autonomy[1] and precluding beneficent treatment of the suffering.[2] Their appeals to beneficence have not proven decisive, in part because the traditional position counsels beneficence: it permits physicians to alleviate suffering even when their treatment jeopardizes patients' lives. It is rather the concept of personal autonomy, recently ascendant as a principle of human action, that has transformed the social discourse of assisted suicide.

Opponents of euthanasia have not criticized assisted-suicide claims grounded on autonomy. Instead they resort to arguments based on human dignity,[3] the sacredness of life[4] or slippery slopes[5]--one of which directly addresses the potent autonomy claim. This article challenges the argument for autonomy: it distinguishes the descriptive and ascriptive aspects of autonomy and probes whether either can ground a right to assisted suicide or logically delimit such a right. By differentiating these two aspects of autonomy, this discussion manifests the shortcomings of many autonomy-based moral and legal theories.

Autonomy is a protean concept, but most ethicists employ the term in either a descriptive or an ascriptive sense. The descriptive sense of autonomy encompasses those conditions (for example, a choice of options, competence to act)[6] and characteristics (for example, moral authenticity or integrity)[7] that scholars associate with autonomy. These characteristics and conditions are essential components of self-governed action, but they cannot justify acts of assisted suicide (or most other human acts) because they are bereft of normativity: they merely describe necessary conditions of morality, viz., voluntary agency[8] without specifying the moral character of any particular act. If both the virtuous and the vicious can act autonomously, then the mere possession of autonomy neither specifies an agent's moral character nor justifies his acts.

Although the descriptive sense of autonomy is insufficient to justify--either morally or legally--acts of assisted suicide, many scholars use the term in an ascriptive sense by grounding individual rights on autonomy.[9] Ethicists and legal scholars employ the ascriptive sense of autonomy synonymously with liberty in grounding moral or legal claims. For example, the eminent liberal philosopher Joseph Raz claims that personal autonomy "is essentially about the freedom of persons to choose their own lives,"[10] while John Stuart Mill states that the principle of liberty "requires liberty of tastes and pursuits; of framing the plan of our life to suit our own character."[11]

Jurists also identify autonomy with liberty: the Supreme Court asserts that "choices central to personal dignity and autonomy are central to the liberty protected by the Fourteenth Amendment. At the heart of liberty is the right to define one's own concept of existence ...."[12] And writing in defense of a right to euthanasia, the respected legal scholar Ronald Dworkin claims that individuals' right to autonomy is "a right to make important decisions defining their own lives for themselves."[13] Thus in both ethics and law, autonomy and liberty are similarly concerned with an individual's freedom to make important choices for himself, unfettered by social proscriptions. This ascriptive sense of autonomy grounds the claim for a "right" to assisted suicide by guaranteeing an individual's freedom to enlist assistance in his suicide.

Although this purported right has achieved some measure of public support, its autonomy-based justification faces several formidable challenges. First, proponents of assisted suicide assert that autonomy is a fundamental good that must be protected, yet they advocate an act that extinguishes the basis of autonomy. The same conundrum prompted John Stuart Mill, a stalwart champion of individual liberty, to favor legal proscription of voluntary slavery. Mill claimed that an individual cannot freely renounce his freedom without violating that good.[14] Similarly, autonomous acts of assisted suicide annihilate the basis of autonomy and thereby undermine the very ground of their justification.

A second criticism is that proponents of ascriptive autonomy have not articulated criteria that distinguish opprobrious from acceptable autonomous acts. As noted above, society does not morally or legally sanction acts merely because they are performed autonomously. In fact, immoral and depraved acts are more objectionable if performed autonomously than if coerced.[15] Thus even if assisted-suicide proponents are granted the claim that society must recognize individuals' autonomy to regulate their own lives, they must formulate a method or principle for distinguishing those autonomous acts that are permissible, since moral agents can use their autonomy perniciously.

Most proponents of assisted suicide have not addressed the first criticism, that an autonomous act of assisted suicide destroys the basis of its justification, but many respond to the second criticism by invoking Mill's harm principle to instill normativity into autonomy. The harm principle states that a person's autonomy (or equivalently, his liberty) to perform an act can be circumscribed only if the act harms another.[16] A corollary of the harm principle is that an individual is not subject to others' moral principles if he is not harming them. Thus scholars claim that the harm principle prevents individuals from imposing their view of the good on a person who seeks assisted suicide since he is not harming others.[17]

Yet these scholars themselves ineluctably subject others to a peculiar view of the good when they ascribe autonomy-based rights to acts considered "self-regarding," but not to those perceived as "harmful." To specify an act as harmful is to judge it injurious, but what constitutes injury will depend on one's theory of the good: both the hedonist and the Aristotelian virtuous man would endorse the harm principle, yet they would offer divergent examples of harmful acts. The harm principle is wholly formal insofar as it does not articulate the criteria of a harmful act, or correlatively, which acts should be proscribed. Therefore any scholar who wields the harm principle must turn elsewhere to generate practical moral or legal precepts, namely, to a moral theory. As Joseph Raz notes,

Since "causing harm" entails by its

very meaning that the action is

prima facie wrong, it is a normative

concept acquiring its specific

concrete content from the moral

theory within which it is embedded.

Without such a connection

to a moral theory the harm principle

is a formal principle lacking

specific meaning and leading to

no policy conclusions.[18]

Thus the autonomy proponent can impart normativity to the harm principle--and to autonomy--only by advancing a theory of the good. But in imposing his peculiar theory of the good on adherents of a discrepant theory, the autonomy proponent thereby undermines their autonomy.

This critique of autonomy is more radical than those that criticize autonomy proponents for proposing a view of human nature that is atomistic[19] or self-regarding,[20] for even if ascriptive autonomy does not require an individualistic view of human beings, it entails a profound dilemma. The principle of autonomy or liberty requires a "harm" principle to justify prohibiting certain types of autonomous acts, but whether an act is specified as harmful or harmless will depend on the preferred theory of good. Therefore the normative use of the principle of autonomy is performatively self-refuting: when scholars proscribe certain autonomous acts in the name of harm, or defend other autonomous acts judged harmless, they impose an axiology and subvert autonomy.[21]

Autonomy is inextricably linked to a theory of good, even if the esteemed theory is cast in "thin" terms.[22] Thin theories of the good, which protect only a few basic values such as life or property, must still specify when these goods are inviolable, such as in defense of one's country, life, material goods, etc. (A thin theory that refuses such specification is--like the harm principle--devoid of practical import and therefore unfeasible as a moral or political theory.) Thus even a thin theory of the good must proscribe certain human acts, such as stealing another's property or killing a thief to retain some trivial material object. But any individual who would rather violate these thin proscriptions capitulates his autonomy if he instead acquiesces to the heteronomous legal proscription.

To be sure, autonomy must be circumscribed to maintain the good of society or protect human life, but this appeal to goods violates the autonomy of those who retain a divergent axiology. A society must criminalize certain acts that threaten its survival, but it thereby circumscribes the autonomy of individuals who prefer to perform these threatening acts.

Moreover, this appeal to the good of society or human life underscores the instrumental character of autonomy: it is respected when invoked to protect acts that most people judge as good, for example, the preservation of society or human life, and is usually proscribed when employed in an invidious manner, as in acts of murder or robbery. Hence autonomy is necessary for the existence of a moral act but is insufficient to justify one. The justification of the act will hinge on the end to which autonomy is employed: if for a noble end, then it is upheld; if depraved, then it is proscribed. It is not autonomy per se that vindicates an autonomy claim but the good that autonomy is instrumental in achieving. Therefore an individual cannot invoke autonomy to justify an ethical or legal claim to acts such as assisted suicide; rather he must vindicate the underlying value that the autonomous act endeavors to attain.

The imposition of a theory of the good does not always undermine autonomy, and indeed can procure it. A child or mentally ill individual, incapable as each may be of rational thought, might perhaps subsequently engage in autonomous action if currently prevented from engaging in immediate life-threatening or mind-altering behavior.[23] But since these individuals did not retain sufficient rationality when incurring the imposition, their autonomy could not be violated. In contrast, a rational adult who is denied such choices sacrifices his autonomy at the moment another restricts his choice. Even if circumscription of a rational individual's autonomy is conducive to future autonomous choices, his extant autonomy is nevertheless subverted when he is prevented from enacting a criminal or mind-altering practice that was rationally chosen. Thus an imposed theory of the good violates the autonomy of those who fulfill the descriptive requisites of autonomy; if it did not, it would not be termed an imposition.

This subordination of autonomy to a view of the good generates many prevalent social controversies. Consider the conflict surrounding pornography: free speech advocates favor legalization of pornography because of the harm incurred by curtailing free expression,[24] while many feminists oppose pornography because it demeans and sexually exploits women.[25] Though both sides endorse the harm principle, their divergent moral and legal principles arise from disparate views of the human good, the former upholding the goodness of free expression, the latter the goodness of women viewed as multidimensional beings. Each group attempts to legislate its view of the good, and regardless of the outcome, one group's autonomy to attain its view of the good will be violated. Legislators or jurists must ultimately choose between the alternative views of the good, but they will inevitably harm (at least) one group by constraining its autonomy to realize its theory of the good. Certainly individuals can disapprove of certain acts but refuse to proscribe them legally, but even the refusal is usually grounded on a view of the good, such as the socially deleterious effects of abridging free speech.

Similarly, in the assisted-suicide debate, the disputants seek to legislate conflicting moral views of human life for society. Assisted-suicide proponents endorse a subjective valuation of life's worth[26] or extol an analgesic death.[27] Hence they claim that the individual should be free to terminate his devalued or painful existence. Their arguments are equivalent to the moral claims, "A competent person ought to be assisted in suicide upon request if he does not value his life," or "An individual ought to be free to terminate a painful existence."

Opponents of assisted suicide view innocent human life as intrinsically good[28] or fear a social slide to more dubious types of killing.[29] Their claims reduce to, "It is wrong to intentionally kill innocent human beings, even if they seek to die," or "Society ought to avoid traversing a slippery slope." Both sides of the debate propound moral principles embedded in theories of the good, the former extolling a subjective valuation of life's worth or the avoidance of pain, the latter, innocent human life or extant social conditions.

Thus the debate over assisted suicide is a conflict between competing theories of the good, and not a dispute between proponents of autonomy and the sanctity or dignity of life. Because each side seeks the autonomy to attain its respective theory of the good, neither can invoke autonomy to vindicate its attempt to achieve its good. Therefore a proponent of assisted suicide cannot claim that restrictions violate autonomy by dictating a theory of the good.[30] For to justify assisted suicide by autonomy or to prohibit depraved acts by the harm principle, the proponent himself abrogates the autonomy of those who subscribe to a discrepant set of goods: he a fortiori undermines his own principle of autonomy.

These serious shortcomings of autonomy-based justifications of assisted suicide are nowhere more apparent than in the restrictions placed on the act by proponents of autonomy. Most support the act conditionally, but their stipulations underscore the tension that persists between the liberating character of autonomy and the social necessity of limiting certain autonomous acts.

This tension persists because autonomy justifies assisted suicide for nearly any reason.[31] If a person retains the autonomy to be killed, then it is irrelevant whether he is suffering from a terminal disease or a temporary illness--his autonomy is violated when he is denied death in either condition[.32] Moreover, society cannot coherently limit assisted suicide to those suffering from physical disease; to deny a person liberation from psychological suffering would likewise infringe his autonomy. Hence the Netherlands consistently extended the right to assisted suicide to those facing bleak futures.[33]

Furthermore, the principle of autonomy prevents any distinction on the basis of age: a competent twenty-year-old individual denied assisted suicide--for physical or mental suffering--is denied his autonomy as much as an eighty-year-old. The former's killing seems even more beneficent because he may face a more extended trial of suffering. And if an individual can be assisted in suicide when he autonomously deems his life valueless, then he should also be granted the autonomy to sell his organs for their fair market value or to dispose of his life in any other manner.

Thus proponents of autonomy can restrict the right to assisted suicide only by implicitly or explicitly articulating normative claims, for example, that the young or the physically healthy should not be assisted in suicide because their lives are worthwhile or might later improve. They mandate these normative views of the good for other autonomy supporters who subscribe to a divergent axiology, for example, that the young or healthy can be assisted in suicide if they autonomously so choose. If proponents of assisted suicide-either of limited or unconditional scope--are in fact legislating a particular view of the good, then they must defend their proscription of alternative views of the good, including those that extol the goodness of innocent human life or extant social conditions.

Therefore proponents of assisted suicide face a dilemma in appealing to autonomy, whether of the descriptive or ascriptive stripe. Descriptive autonomy, devoid of normativity, defies restrictions because society thwarts the autonomy of an individual when it denies his "authentic" or "self-creative" choice to die, or any other choice. Ascriptive autonomy retains normativity, but only by capitulating its moral neutrality and violating the autonomy of adherents of conflicting moral theories. Certainly society must limit the harm that autonomous acts could yield, but any limitation implicates a view of the good and undermines opponents' autonomy.

Thus autonomy-based arguments for assisted suicide are self-refuting in two regards: first, acts of assisted suicide committed in the name of autonomy annihilate the very basis of individual autonomy; second, arguments grounded on autonomy ultimately depend on a view of the good that, if socially prescribed, would subvert individuals' autonomy to attain alternative views of the good.

Proponents of assisted suicide justify the act in moral and legal contexts by appealing to the concept of autonomy. However, they have overlooked the problematic nature of this claim, particularly its dependence on a theory of the good. The debate involving assisted suicide, like so many other social disputes, hinges on discrepant views of the good, rather than on autonomy or beneficence. Only by focusing on the conflicting views of the goods at stake--and abandoning or re-formulating the argument for autonomy--will ethicists and legal scholars resolve this controversial social issue.[34]


I would like to thank Louis Safranek, M.D., Ph.D., for his substantive and technical critiques.


[1.] Dan Brock, "Voluntary Active Euthanasia," Hastings Center Report 22, no. 2 (1992): 10-22; Baruch Brody, "Voluntary Euthanasia and the Law." In Beneficent Euthanasia, ed. M. Kohl (Buffalo, N.Y.: Prometheus Books, 1975); "Physician-Assisted Suicide and the Right to Die with Assistance" [Notes], Harvard Law Review 105 (1991): 2021-40.

[2.] Richard L. Risley, A Humane and Dignified Death: A New Law Permitting Physician Aid-in-Dying (Glendale, Calif.: Americans Against Human Suffering, 1987); Sidney Wanzer, "Maintaining Control in Terminal Illness: Assisted Suicide and Euthanasia," Humane Medicine 6, no. 3 (1990): 186-88.

[3.] Leon Kass, "Death with Dignity and the Sanctity of Life," Commentary (March 1990):'33-43; John Mahony, Bioethics and Belief(London: Sheed and Ward, 1984).

[4.] Richard Gula, Euthanasia: Moral and Pastoral Perspectives (New York: Paulist Press, 1994), pp. 24-28; Joseph Boyle, "Sanctity of Life and Suicide: Tensions and Developments Within Common Morality." In Suicide and Euthanasia, ed. Baruch Brody (Boston: Kluwer Academic Publishers, 1989), pp. 221-50; Richard Roach, "Medicine and Killing: The Catholic View," The Journal of Medicine and Philosophy 4, no. 4 (1979): 383-97.

[5.] Richard Fenigsen, "A Case against Dutch Euthanasia," Ethics and Medicine 6, no. 1 (1990): 11-18; Kathleen Foley, "Competent Care for the Dying Instead of Physician-Assisted Suicide," NEJM 336, no. 1 (1997): 54-58.

[6.] Richard H. Fallon, Jr., "Two Senses of Autonomy," Stanford University Law Review 46 (1994): 875-99, at 875.

[7.] Joel Feinberg, Harm to Self (Oxford: Oxford University Press, 1986), ch. 18.

[8.] Aristotle, Nicomachean Ethics, trans. Richard McKeon (New York: Random House, 1941), III, I.

[9.] Jos Welie, "The Medical Exception: Physicians, Euthanasia and the Dutch Criminal Law," The Journal of Medicine and Philosophy 17, no. 4 (1992): 419-37, at 419.

[10.] Joseph Raz, The Morality of Freedom (Oxford: Oxford University Press, 1985), p. 370.

[11.] John Stuart Mill, On Liberty in Utilitarianism, On Liberty, Considerations on Representative Government, ed. H. B. Acton (London: J. M. Dent & Sons, 1972), p. 81.

[12.] Planned Parenthood v. Casey, 112 S.Ct. 2791, 2807 (1992).

[13.] Ronald Dworkin, Life's Dominion (New York: Alfred Knopf, 1993), p. 222.

[14.] Mill, On Liberty, p. 172.

[15.] William Galston, Justice and the Common Good (Chicago: University of Chicago Press, 1980), p. 127.

[16.] Mill, On Liberty, p. 78.

[17.] Max Charlesworth, Bioethics in a Liberal Society (Cambridge: Cambridge University Press, 1993), p. 4.

[18.] Raz, The Morality of Freedom, p. 414.

[19.] Alaisdair Macintyre, After Virtue (Notre Dame, Ind.: University of Notre Dame Press, 1981).

[20.] Charles Taylor, Sources of the Self. The Making of the Modern Identity (Cambridge, Mass.: Harvard University Press, 1989).

[21.] John Safranek and Stephen Safranek, "Can the Right to Autonomy Be Resuscitated after Glucksberg," University of Colorado Law Review 69 (1998): 744-49.

[22.] H.L.A. Hart, Essays in Jurisprudence and Philosophy (Oxford: Clarendon Press, 1983), pp. 80-82.

[23.] Richard Arneson, "Autonomy and Preference Formation." In In Harm's Way: Essays in Honor of Joel Feinberg, ed. Jules Coleman and James Buchanan (Cambridge: Cambridge University Press, 1994), pp. 43-75, at 47.

[24.] Carrie Benson Fischer, "Employee Rights in Sex Work: The Struggle for Dancer's Rights as Employees," Law and Inequality 14 (1996): 521-35.

[25.] Catherine Mackinnon, Feminism Unmodified: Discourses on Life and Law (Cambridge, Mass.: Harvard University Press, 1987), ch. 11-16; Rosemarie Tong, Women, Sex, and the Law (Totowa, N.J.: Rowman and Allenheld Press, 1984), ch. 1.

[26.] Peter Singer, Practical Ethics (Cambridge: Cambridge University Press, 1993), p. 195; Jonathan Glover, Causing Death, Saving Lives (New York: Penguin Books, 1977), p. 192.

[27.] Timothy E. Quill, "Death and Dignity: A Case of Individualized Decision Making," NEJM 329 (1990): 1881-83; James Rachels, The End of Life (New York: Oxford University Press, 1986), pp. 15254; Marcia Angell, "The Supreme Court and Physician-Assisted Suicide--The Ultimate Right," NEJM 336 (1997): 50-53, at 51.

[28.] Byron L. Sherwin, "Jewish Views of Euthanasia." In Beneficent Euthanasia, pp. 3-11.

[29.] Leon Kass, "Neither for Love Nor Money: Why Doctors Must Not Kill," Public Interest 94 (Winter 1989): 25-46; Martin Gunderson and David J. Mayo, "Altruism and Physician Assisted Death," The Journal of Medicine and Philosophy 18, no. 3 (1993): 281-95 at 289-90; Edmund Pellegrino, "Doctors Must Not Kill," The Journal of Clinical Ethics 3, no. 2 (1992): 95-102, at 99-101.

[30.] H. Tristam Engelhardt, Jr., "Death by Free Choice." In Suicide and Euthanasia, ed. Baruch Brody (Boston: Kluwer Academic Publishers, 1989), pp. 251-80.

[31.] Compassion in Dying v. State of Washington, 49 F.3d 586 (9th Cir. 1995).

[32.] Yale Kamisar, "Against Assisted Suicide-Even a Very Limited Form," University of Detroit-Mercy Law Review 72, no. 4 (1995): 735-69, at 740.

[33.] Guus de Haas, "Euthanasia and the Legal Situation in the Netherlands," Nursing Times 91, no. 20 (1995): 30-31.

[34.] The arguments for assisted suicide grounded on dignity and beneficence incur some of the same criticisms as autonomy: what constitutes beneficent or dignified action varies according to each individual's view of the good. Assisted suicide is not considered beneficent or dignified for those who esteem the good of innocent human life or fear a precipitous social decline. Therefore proponents of dignity or beneficence must justify their violation of their opponents' view of dignity and beneficence, a view, that is, of the good. To appeal to such formal concepts as beneficence and dignity in justifying assisted suicide is to beg the question of why this peculiar view of the good should be legislated, particularly if others' dignity or sense of beneficence is subverted in the process. Ultimately, the debate must be engaged in terms of goods rather than formal concepts such as dignity, beneficence, or autonomy.

John P. Safranek, "Autonomy and Assisted Suicide: The Execution of Freedom" Hastings Center Report 28, no. 4 (1998): 32-36.
Gale Copyright: Copyright 1998 Gale, Cengage Learning. All rights reserved.