Assisted Suicide: The Philosophers’ Brief

Later this year the Supreme Court will decide two cases posing the question whether dying patients have a right to choose death rather than continued pain and suffering.¹ We print here the brief filed as amicus curiae in these cases by the group of six moral philosophers listed above, with an introduction by Ronald Dworkin.
—the Editors

Introduction

We cannot be sure, until the Supreme Court decides the assisted suicide cases and its decision is published, how far the justices might have accepted or rejected the arguments of the brief published below.² In this introduction I shall describe the oral argument before them last January, and offer some suggestions about how, if they decide against the brief’s position, as many commentators now think they will, they might do the least damage to constitutional law.
The laws of all but one American state now forbid doctors to prescribe lethal pills for patients who want to kill themselves.³ These cases began when groups of dying patients and their doctors in Washington State and New York each sued asking that these prohibitions be declared unconstitutional so that the patients could be given, when and if they asked for it, medicine to hasten their death. The pleadings described the agony in which the patient plaintiffs were dying, and two federal Circuit Courts of Appeal—the Ninth Circuit in the Washington case and the Second Circuit in the New York case—agreed with the plaintiffs that the Constitution forbids the government from flatly prohibiting doctors to help end such desperate and pointless suffering.⁴
Washington State and New York appealed these decisions to the Supreme Court, and a total of sixty amicus briefs were filed, including briefs on behalf of the American Medical Association and the United States Catholic Conference urging the Court to reverse the circuit court decisions, and on behalf of the AmericanMedical Students Association and the Gay Men’s Health Crisis urging it to affirm them. The justices’ comments during oral argument persuaded many observers that the Court would reverse the decisions, probably by a lopsided majority. The justices repeatedly cited two versions—one theoretical, the other practical—of the “slippery slope” argument: that it would be impossible to limit a right to assisted suicide in an acceptable way, once that right was recognized.
The theoretical version of the argument denies that any principled line can be drawn between cases in which proponents say a right of assisted suicide is appropriate and those in which they concede that it is not. The circuit courts recognized only a right for competent patients already dying in great physical pain to have pills prescribed that they could take themselves. Several justices asked on what grounds the right once granted could be so severely limited. Why should it be denied to dying patients who are so feeble or paralyzed that they cannot take pills themselves and who beg a doctor to inject a lethal drug into them? Or to patients who are not dying but face years of intolerable physical or emotional pain, or crippling paralysis or dependence? But if the right were extended that far, on what ground could it be denied to anyone who had formed a desire to die—to a sixteen-year-old suffering from a severe case of unrequited love, for example?

The philosophers’ brief answers these questions in two steps. First, it defines a very general moral and constitutional principle—that every competent person has the right to make momentous personal decisions which invoke fundamental religious or philosophical convictions about life’s value for himself. Second, it recognizes that people may make such momentous decisions impulsively or out of emotional depression, when their act does not reflect their enduring convictions; and it therefore allows that in some circumstances a state has the constitutional power to override that right in order to protect citizens from mistaken but irrevocable acts of self-destruction. States may be allowed to prevent assisted suicide by people who—it is plausible to think—would later be grateful if they were prevented from dying.

That two-step argument would justify a state’s protecting a disappointed adolescent from himself. It would equally plainly not justify forcing a competent dying patient to live in agony a few weeks longer. People will of course disagree about the cases in between these extremes, and if the Court adopted this argument, the federal courts would no doubt be faced with a succession of cases in years to come testing whether, for example, it is plausible to assume that a desperately crippled patient in constant pain but with years to live, who has formed a settled and repeatedly stated wish to die, would one day be glad he was forced to stay alive. But though two justices dwelled, during the oral argument, on the unappealing prospect of a series of such cases coming before the courts, it seems better that the courts do assume that burden, which they could perhaps mitigate through careful rulings, than that they be relieved of it at the cost of such terrible suffering.
The practical version of the slippery slope argument is more complex. If assisted suicide were permitted in principle, every state would presumably adopt regulations to insure that a patient’s decision for suicide is informed, competent, and free. But many people fear that such regulations could not be adequately enforced, and that particularly vulnerable patients—poor patients dying in overcrowded hospitals that had scarce resources, for example—might be pressured or hustled into a decision for death they would not otherwise make. The evidence suggests, however, that such patients might be better rather than less well protected if assisted suicide were legalized with appropriate safeguards.
More of them could then benefit from relief that is already available—illegally—to more fortunate people who have established relationships with doctors willing to run the risks of helping them to die. The current two-tier system—a chosen death and an end of pain outside the law for those with connections and stony refusals for most other people—is one of the greatest scandals of contemporary medical practice. The sense many middle-class people have that if necessary their own doctor “will know what to do” helps to explain why the political pressure is not stronger for a fairer and more open system in which the law acknowledges for everyone what influential people now expect for themselves.
For example, in a recent study in the state of Washington, which guaranteed respondents anonymity, 26 percent of doctors surveyed said they had received explicit requests for help in dying, and had provided, overall, lethal prescriptions to 24 percent of patients requesting them.⁵ In other studies, 40 percent of Michigan oncologists surveyed reported that patients had initiated requests for death, 18 percent said they had participated in assisted suicide, and 4 percent in “active euthanasia”—injecting lethal drugs themselves. In San Francisco, 53 percent of the 1,995 responding physicians said they had granted an AIDS patient’s request for suicide assistance at least once.⁶ These statistics approach the rates at which doctors help patients die in Holland, where assisted suicide is in effect legal.

The most important benefit of legalized assisted suicide for poor patients however, might be better care while they live. For though the medical experts cited in various briefs disagreed sharply about the percentage of terminal cases in which pain can be made tolerable through advanced and expensive palliative techniques, they did not disagree that a great many patients do not receive the relief they could have. The Solicitor General who urged the Court to reverse the lower court judgments conceded in the oral argument that 25 percent of terminally ill patients actually do die in pain. That appalling figure is the result of several factors, including medical ignorance and fear of liability, inadequate hospital funding, and (as the Solicitor General suggested)the failure of insurers and health care programs to cover the cost of special hospice care. Better training in palliative medicine, and legislation requiring such coverage, would obviously improve the situation, but it seems perverse to argue that the patients who would be helped were better pain management available must die horribly because it is not; and, as Justice Breyer pointed out, the number of patients in that situation might well increase as medical costs continue to escalate.

According to several briefs, moreover, patients whose pain is either uncontrollable or uncontrolled are often “terminally sedated”—intravenous drugs (usually barbiturates or benzodiazepenes) are injected to induce a pharmacologic coma during which the patient is given neither water nor nutrition and dies sooner than he otherwise would.⁷ Terminal sedation is widely accepted as legal, though it advances death.⁸ But it is not subject to regulations nearly as stringent as those that a state forced to allow assisted suicide would enact, because such regulations would presumably include a requirement that hospitals, before accepting any request for assistance in suicide, must demonstrate that effective medical care including state-of-the-art pain management had been offered. The guidelines recently published by a network of ethics committees in the Bay Area of California, for example, among other stringent safeguards, provide that a primary care physician who receives a request for suicide must make an initial referral to a hospice program or to a physician experienced in palliative care, and certify in a formal report filed in a state registry, signed by an independent second physician with expertise in such care, that the best available pain relief has been offered to the patient.⁹
Doctors and hospitals anxious to avoid expense would have very little incentive to begin a process that would focus attention on their palliative care practices. They would be more likely to continue the widespread practice of relatively inexpensive terminal care which is supplemented, perhaps, with terminal sedation. It is at least possible, however, that patients’ knowledge of the possibility of assisted suicide would make it more difficult for such doctors to continue as before. That is the view of the Coalition of Hospice Professionals, who said, in their own amicus brief, “Indeed, removing legal bans on suicide assistance will enhance the opportunity for advanced hospice care for all patients because regulation of physician-assisted suicide would mandate that all palliative measures be exhausted as a condition precedent to assisted suicide.”
So neither version of the slippery slope argument seems very strong. It is nevertheless understandable that Supreme Court justices are reluctant, particularly given how little experience we have so far with legalized assisted suicide, to declare that all but one of the states must change their laws to allow a practice many citizens think abominable and sacrilegious. But as the philosophers’ brief that follows emphasizes, the Court is in an unusually difficult position. If it closes the door to a constitutional right to assisted suicide it will do substantial damage to constitutional practice and precedent, as well as to thousands of people in great suffering. It would face a dilemma in justifying any such decision, because it would be forced to choose between the two unappealing strategies that the brief describes.
The first strategy—declaring that terminally ill patients in great pain do not have a constitutional right to control their own deaths, even in principle—seems alien to our constitutional system, as the Solicitor General himself insisted in the oral argument. It would also undermine a variety of the Court’s own past decisions, including the carefully constructed position on abortion set out in its 1993 decision in Casey. Indeed some amicus briefs took the occasion of the assisted suicide cases to criticize the abortion decisions—a brief filed on behalf of Senator Orrin Hatch of Utah and Representatives Henry Hyde of Illinois and Charles Canady of Florida, for example, declared that the abortion decisions were “of questionable legitimacy and even more questionable prudence.” Protecting the abortion rulings was presumably one of the aims of the Clinton administration in arguing, through the Solicitor General, for the second strategy instead.
The first strategy would create an even more evident inconsistency within the practice of terminal medicine itself. Since the Cruzan decision discussed in the brief, lawyers have generally assumed that the Court would protect the right of any competent patient to have life sustaining equipment removed from his body even though he would then die. In the oral argument, several justices suggested a “common-sense” distinction between the moral significance of acts, on the one hand, and omissions, on the other. This distinction, they suggested, would justify a constitutional distinction between prescribing lethal pills and removing life support; for, in their view, removing support is only a matter of “letting nature take its course,” while prescribing pills is an active intervention that brings death sooner than natural processes would.
The discussion of this issue in the philosophers’ brief is therefore particularly significant. The brief insists that such suggestions wholly misunderstand the “common-sense” distinction, which is not between acts and omissions, but between acts or omissions that are designed to cause death and those that are not. One justice suggested that a patient who insists that life support be disconnected is not committing suicide. That is wrong:he is committing suicide if he aims at death, as most such patients do, just as someone whose wrist is cut in an accident is committing suicide if he refuses to try to stop the bleeding. The distinction between acts that aim at death and those that do not cannot justify a constitutional distinction between assisting in suicide and terminating life support. Some doctors, who stop life support only because the patient so demands, do not aim at death. But neither do doctors who prescribe lethal pills only for the same reason, and hope that the patient does not take them. And many doctors who terminate life support obviously do aim at death, including those who deny nutrition during terminal sedation, because denying nutrition is designed to hasten death, not to relieve pain.

There are equally serious objections, however, to the second strategy the philosophers’ brief discusses. This strategy concedes a general right to assisted suicide but holds that states have the power to judge that the risks of allowing any exercise of that right are too great. It is obviously dangerous for the Court to allow a state to deny a constitutional right on the ground that the state lacks the will or resource to enforce safeguards if it is exercised, particularly when the case for the practical version of the “slippery slope” objection seems so weak and has been little examined. As Justice Rehnquist, who perhaps favors the first strategy, observed in the oral argument, “[I]f we assume a liberty interest but nevertheless say that, even assuming a liberty interest, a state can prohibit it entirely, that would be rather a conundrum.”¹⁰

If the justices believe that they cannot now accept the lower court decisions, in spite of the powerful defense offered by the respondents and the various briefs supporting them, then they should consider a third strategy—postponement. They might declare that both precedent and principle offer strong grounds for a constitutional right to manage one’s own death, but that there is as yet too little experience with legally permitted assisted suicide for the Court to rule that states lack the constitutional power to follow their traditional practice of outlawing it.
That third strategy, unlike the first two, would in effect commit the Court to considering a new challenge in the future when a much more substantial record of experience is available—from Oregon and any other American state that follows its lead in permitting assisted suicide by legislation, and from the Netherlands, Switzerland, the Northern Territories of Australia, and any other jurisdiction whose legislature takes the same course. In the meantime, the public would have had an opportunity to participate more fully in the argument about principle; and, when circumstances make it possible, wide public discussion is a desirable and democratic preliminary to a final Supreme Court adjudication.¹¹ Postponement is not what the philosophers’ brief urges. But it would be the most statesmanlike way in which the Court could make the wrong decision.

Ronald Dworkin
—February 27, 1997

THE BRIEF OF THE AMICI CURIAE

Interest of the Amici Curiae

Amici are six moral and political philosophers who differ on many issues of public morality and policy. They are united, however, in their conviction that respect for fundamental principles of liberty and justice, as well as for the American constitutional tradition, requires that the decisions of the Courts of Appeals be affirmed.

Introduction and Summary of Argument

These cases do not invite or require the Court to make moral, ethical, or religious judgments about how people should approach or confront their death or about when it is ethically appropriate to hasten one’s own death or to ask others for help in doing so. On the contrary, they ask the Court to recognize that individuals have a constitutionally protected interest in making those grave judgments for themselves, free from the imposition of any religious or philosophical orthodoxy by court or legislature. States have a constitutionally legitimate interest in protecting individuals from irrational, ill-informed, pressured, or unstable decisions to hasten their own death. To that end, states may regulate and limit the assistance that doctors may give individuals who express a wish to die. But states may not deny people in the position of the patient-plaintiffs in these cases the opportunity to demonstrate, through whatever reasonable procedures the state might institute—even procedures that err on the side of caution—that their decision to die is indeed informed, stable, and fully free. Denying that opportunity to terminally ill patients who are in agonizing pain or otherwise doomed to an existence they regard as intolerable could only be justified on the basis of a religious or ethical conviction about the value or meaning of life itself. Our Constitution forbids government to impose such convictions on its citizens.
Petitioners [i.e., the state authorities of Washington and New York] and the amici who support them offer two contradictory arguments. Some deny that the patient-plaintiffs have any constitutionally protected liberty interest in hastening their own deaths. But that liberty interest flows directly from this Court’s previous decisions. It flows from the right of people to make their own decisions about matters “involving the most intimate and personal choices a person may make in a lifetime, choices central to personal dignity and autonomy.” Planned Parenthoodv. Casey, 505 U.S. 833, 851 (1992).
The Solicitor General, urging reversal in support of Petitioners, recognizes that the patient-plaintiffs do have a constitutional liberty interest at stake in these cases. See Brief for the United States as Amicus Curiae Supporting Petitioners at 12, Washingtonv. Vacco [hereinafter Brief for the United States] (“The term ‘liberty’ in the Due Process Clause…is broad enough to encompass an interest on the part of terminally ill, mentally competent adults in obtaining relief from the kind of suffering experienced by the plaintiffs in this case, which includes not only severe physical pain, but also the despair and distress that comes from physical deterioration and the inability to control basic bodily functions.”); see also id. at 13 (“Cruzan…supports the conclusion that a liberty interest is at stake in this case.”).
The Solicitor General nevertheless argues that Washington and New York properly ignored this profound interest when they required the patient-plaintiffs to live on in circumstances they found intolerable. He argues that a state may simply declare that it is unable to devise a regulatory scheme that would adequately protect pa-tients whose desire to die might be ill-informed or unstable or foolish or not fully free, and that a state may therefore fall back on a blanket prohibition. This Court has never accepted that patently dangerous rationale for denying protection altogether to a conceded fundamental constitutional interest. It would be a serious mistake to do so now. If that rationale were accepted, an interest acknowledged to be constitutionally protected would be rendered empty.

Argument

I. The Liberty Interest Asserted Here is Protected by the Due

Process Clause
The Due Process Clause of the Fourteenth Amendment protects the liberty interest asserted by the patient-plaintiffs here.
Certain decisions are momentous in their impact on the character of a person’s life—decisions about religious faith, political and moral allegiance, marriage, procreation, and death, for example. Such deeply personal decisions pose controversial questions about how and why human life has value. In a free society, individuals must be allowed to make those decisions for themselves, out of their own faith, conscience, and convictions. This Court has insisted, in a variety of contexts and circumstances, that this great freedom is among those protected by the Due Process Clause as essential to a community of “ordered liberty.” Palko v. Connecticut, 302 U.S. 319, 325 (1937). In its recent decision in Planned Parenthood v. Casey, 505 U.S. 833, 851 (1992), the Court offered a paradigmatic statement of that principle:

matters involving the most intimate and personal choices a person may make in a lifetime, choices central to a person’s dignity and autonomy, are central to the liberty protected by the Fourteenth Amendment.
That declaration reflects an idea underlying many of our basic constitutional protections.¹² As the Court explained in West Virginia State Board of Education v. Barnette, 319 U.S. 624, 642 (1943):
If there is any fixed star in our constitutional constellation, it is that no official…can prescribe what shall be orthodox in politics, nationalism, religion, or other matters of opinion or force citizens to confess by word or act their faith therein.
A person’s interest in following his own convictions at the end of life is so central a part of the more general right to make “intimate and personal choices” for himself that a failure to protect that particular interest would undermine the general right altogether. Death is, for each of us, among the most significant events of life. As the Chief Justice said in Cruzanv. Missouri, 497 U.S. 261, 281(1990), “[t]he choice between life and death is a deeply personal decision of obvious and overwhelming finality.” Most of us see death—whatever we think will follow it—as the final act of life’s drama, and we want that last act to reflect our own convictions, those we have tried to live by, not the convictions of others forced on us in our most vulnerable moment.
Different people, of different religious and ethical beliefs, embrace very different convictions about which way of dying confirms and which contradicts the value of their lives. Some fight against death with every weapon their doctors can devise. Others will do nothing to hasten death even if they pray it will come soon. Still others, including the patient-plaintiffs in these cases, want to end their lives when they think that living on, in the only way they can, would disfigure rather than enhance the lives they had created. Some people make the latter choice not just to escape pain. Even if it were possible to eliminate all pain for a dying patient—and frequently that is not possible—that would not end or even much alleviate the anguish some would feel at remaining alive, but intubated, helpless, and often sedated near oblivion.
None of these dramatically different attitudes about the meaning of death can be dismissed as irrational. None should be imposed, either by the pressure of doctors or relatives or by the fiat of government, on people who reject it. Just as it would be intolerable for government to dictate that doctors never be permitted to try to keep someone alive as long as possible. when that is what the patient wishes, so it is intolerable for government to dictate that doctors may never, under any circumstances, help someone to die who believes that further life means only degradation. The Constitution insists that people must be free to make these deeply personal decisions for themselves and must not be forced to end their lives in a way that appalls them, just because that is what some majority thinks proper.

II. This Court’s Decisions in Casey and Cruzan Compel

Recognition of a Liberty Interest Here
A. Casey Supports the Liberty Interest Asserted Here
In Casey, this Court, in holding that a state cannot constitutionally proscribe abortion in all cases, reiterated that the Constitution protects a sphere of autonomy in which individuals must be permitted to make certain decisions for themselves. The Court began its analysis by pointing out that “[a]t the heart of liberty is the right to define one’s own concept of existence, of meaning, of the universe, and of the mystery of human life.” 505 U.S. at 851. Choices flowing out of these conceptions, on matters “involving the most intimate and personal choices a person may make in a lifetime, choices central to personal dignity and autonomy, are central to the liberty protected by the Fourteenth Amendment.” Id. “Beliefs about these matters,” the Court continued, “could not define the attributes of personhood were they formed under compulsion of the State.” Id.
In language pertinent to the liberty interest asserted here, the Court explained why decisions about abortion fall within this category of “personal and intimate” decisions. A decision whether or not to have an abortion, “originat[ing] within the zone of conscience and belief,” involves conduct in which “the liberty of the woman is at stake in a sense unique to the human condition and so unique to the law.” Id. at 852. As such, the decision necessarily involves the very “destiny of the woman” and is inevitably “shaped to a large extent on her own conception of her spiritual imperatives and her place in society.” Id. Precisely because of these characteristics of the decision, “the State is [not] entitled to proscribe [abortion] in all instances.” Id. Rather, to allow a total prohibition on abortion would be to permit a state to impose one conception of the meaning and value of human existence on all individuals. This the Constitution forbids.
The Solicitor General nevertheless argues that the right to abortion could be supported on grounds other than this autonomy principle, grounds that would not apply here. He argues, for example, that the abortion right might flow from the great burden an unwanted child imposes on its mother’s life. Brief for the United States at 14-15. But whether or not abortion rights could be defended on such grounds, they were not the grounds on which this Court in fact relied. To the contrary, the Court explained at length that the right flows from the constitutional protection accorded all individuals to “define one’s own concept of existence, of meaning, of the universe, and of the mystery of human life.” Casey, 505 U.S. at 851.
The analysis in Casey compels the conclusion that the patient-plaintiffs have a liberty interest in this case that a state cannot burden with a blanket prohibition. Like a woman’s decision whether to have an abortion, a decision to die involves one’s very “destiny” and inevitably will be “shaped to a large extent on [one’s] own conception of [one’s] spiritual imperatives and [one’s] place in society.” Id. at 852. Just as a blanket prohibition on abortion would involve the improper imposition of one conception of the meaning and value of human existence on all individuals, so too would a blanket prohibition on assisted suicide. The liberty interest asserted here cannot be rejected without undermining the rationale of Casey. Indeed, the lower court opinions in the Washington case expressly recognized the parallel between the liberty interest in Casey and the interest asserted here. See Compassion in Dying v. Washington, 79 F.3d 790, 801 (9th Cir. 1996) (en banc) (“In deciding right-to-die cases, we are guided by the Court’s approach to the abortion cases. Casey in particular provides a powerful precedent, for in that case the Court had the opportunity to evaluate its past decisions and to determine whether to adhere to its original judgment.”), aff’g, 850 F. Supp. 1454, 1459 (W.D. Wash. 1994) (“[T]he reasoning in Casey [is] highly instructive and almost prescriptive…”). This Court should do the same.
B. Cruzan Supports the Liberty Interest Asserted Here
We agree with the Solicitor Gen-eral that this Court’s decision in “Cruzan…supports the conclusion that a liberty interest is at stake in this case.” Brief for the United States at 8. Petitioners, however, insist that the present cases can be distinguished because the right at issue in Cruzan was limited to a right to reject an unwanted invasion of one’s body.¹³ But this Court repeatedly has held that in appropriate circumstances a state may require individuals to accept unwanted invasions of the body. See, e.g., Schmerber v. California, 384 U.S. 757 (1966) (extraction of blood sample from individual suspected of driving while intoxicated, notwithstanding defendant’s objection, does not violate privilege against self-incrimination or other constitutional rights); jacobson v. Massachusetts, 197 U.S. 11 (1905) (upholding compulsory vaccination for smallpox as reasonable regulation for protection of public health).
The liberty interest at stake in Cruzan was a more profound one. If a competent patient has a constitutional right to refuse life-sustaining treatment, then, the Court implied, the state could not override that right. The regulations upheld in Cruzan were designed only to ensure that the individual’s wishes were ascertained correctly. Thus, if Cruzan implies a right of competent patients to refuse life-sustaining treatment, that implication must be understood as resting not simply on a right to refuse bodily invasions but on the more profound right to refuse medical intervention when what is at stake is a momentous personal decision, such as the timing and manner of one’s death. In her concurrence, Justice O’Connor expressly recognized that the right at issue involved a “deeply personal decision” that is “inextricably intertwined” with our notion of “self-determination.” 497 U.S. at 287-89.
Cruzan also supports the proposition that a state may not burden a terminally ill patient’s liberty interest in determining the time and manner of his death by prohibiting doctors from terminating life support. Seeking to distinguish Cruzan, Petitioners insist that a state may nevertheless burden that right in a different way by forbidding doctors to assist in the suicide of patients who are not on life-support machinery. They argue that doctors who remove life support are only allowing a natural process to end in death whereas doctors who prescribe lethal drugs are intervening to cause death. So, according to this argument, a state has an independent justification for forbidding doctors to assist in suicide that it does not have for forbidding them to remove life support. In the former case though not the latter, it is said, the state forbids an act of killing that is morally much more problematic than merely letting a patient die.
This argument is based on a misunderstanding of the pertinent moral principles. It is certainly true that when a patient does not wish to die, different acts, each of which foreseeably results in his death, nevertheless have very different moral status. When several patients need organ transplants and organs are scarce, for example, it is morally permissible for a doctor to deny an organ to one patient, even though he will die without it, in order to give it to another. But it is certainly not permissible for a doctor to kill one patient in order to use his organs to save another. The morally significant difference between those two acts is not, however, that killing is a positive act and not providing an organ is a mere omission, or that killing someone is worse than merely allowing a “natural” process to result in death. It would be equally impermissible for a doctor to let an injured patient bleed to death, or to refuse antibiotics to a patient with pneumonia—in each case the doctor would have allowed death to result from a “natural” process—in order to make his organs available for transplant to others. A doctor violates his patient’s rights whether the doctor acts or refrains from acting, against the patient’s wishes, in a way that is designed to cause death.
When a competent patient does want to die, the moral situation is obviously different, because then it makes no sense to appeal to the patient’s right not to be killed as a reason why an act designed to cause his death is impermissible. From the patient’s point of view, there is no morally pertinent difference between a doctor’s terminating treatment that keeps him alive, if that is what he wishes, and a doctor’s helping him to end his own life by providing lethal pills he may take himself, when ready, if that is what he wishes—except that the latter may be quicker and more humane. Nor is that a pertinent difference from the doctor’s point of view. If and when it is permissible for him to act with death in view, it does not matter which of those two means he and his patient choose. If it is permissible for a doctor deliberately to withdraw medical treatment in order to allow death to result from a natural process, then it is equally permissible for him to help his patient hasten his own death more actively, if that is the patient’s express wish.
It is true that some doctors asked to terminate life support are reluctant and do so only in deference to a patient’s right to compel them to remove unwanted invasions of his body. But other doctors, who believe that their most fundamental professional duty is to act in the patient’s interests and that, in certain circumstances, it is in their patient’s best interests to die, participate willingly in such decisions: they terminate life support to cause death because they know that is what their patient wants. Cruzan implied that a state may not absolutely prohibit a doctor from deliberately causing death, at the patient’s request, in that way and for that reason. If so, then a state may not prohibit doctors from deliberately using more direct and often more humane means to the same end when that is what a patient prefers. The fact that failing to provide life-sustaining treatment may be regarded as “only letting nature take its course” is no more morally significant in this context, when the patient wishes to die, than in the other, when he wishes to live. Whether a doctor turns off a respirator in accordance with the patient’s request or prescribes pills that a patient may take when he is ready to kill himself, the doctor acts with the same intention: to help the patient die.
The two situations do differ in one important respect. Since patients have a right not to have life-support machinery attached to their bodies, they have, in principle, a right to compel its removal. But that is not true in the case of assisted suicide: patients in certain circumstances have a right that the state not forbid doctors to assist in their deaths, but they have no right to compel a doctor to assist them. The right in question, that is, is only a right to the help of a willing doctor.

III. State Interests do not Justify a Categorical Prohibition on all

Assisted Suicide
The Solicitor General concedes that “a competent, terminally ill adult has a constitutionally cognizable liberty interest in avoiding the kind of suffering experienced by the plaintiffs in this case.” Brief for the United States at 8. He agrees that this interest extends not only to avoiding pain, but to avoiding an existence the patient believes to be one of intolerable indignity or incapacity as well. Id. at 12. The Solicitor General argues, however, that states nevertheless have the right to “override” this liberty interest altogether, because a state could reasonably conclude that allowing doctors to assist in suicide, even under the most stringent regulations and procedures that could be devised, would unreasonably endanger the lives of a number of patients who might ask for death in circumstances when it is plainly not in their interests to die or when their consent has been improperly obtained.
This argument is unpersuasive, however, for at least three reasons. First, in Cruzan, this Court noted that its various decisions supported the recognition of a general liberty interest in refusing medical treatment, even when such refusal could result in death. 497 U.S. at 278-79. The various risks described by the Solicitor General apply equally to those situations. For instance, a patient kept alive only by an elaborate and disabling life-support system might well become depressed, and doctors might be equally uncertain whether the depression is curable: such a patient might decide for death only because he has been advised that he will die soon anyway or that he will never live free of the burdensome apparatus, and either diagnosis might conceivably be mistaken. Relatives or doctors might subtly or crudely influence that decision, and state provision for the decision may (to the same degree in this case as if it allowed assisted suicide) be thought to encourage it.
Yet there has been no suggestion that states are incapable of addressing such dangers through regulation. In fact, quite the opposite is true. In McKay v. Bergstedt, 106 Nev. 808, 801 P.2d 617 (1990), for example, the Nevada Supreme Court held that “competent adult patients desiring to refuse or discontinue medical treatment” must be examined by two non-attending physicians to determine whether the patient is mentally competent, understands his prognosis and treatment options, and appears free of coercion or pressure in making his decision. Id. at 827-28, 801 P.2d at 630. See also: id. (in the case of terminally-ill patients with natural life expectancy of less than six months, [a] patient’s right of self-determination shall be deemed to prevail over state interests, whereas [a] non-terminal patient’s decision to terminate life-support systems must first be weighed against relevant state interests by trial judge); [and] In re Farrell, 108 N.J. 335, 354, 529 A.2d 404, 413 (1987) ([which held that a] terminally-ill patient requesting termination of life-support must be determined to be competent and properly informed about [his] prognosis, available treatment options and risks, and to have made decision voluntarily and without coercion). Those protocols served to guard against precisely the dangers that the Solicitor General raises. The case law contains no suggestion that such protocols are inevitably insufficient to prevent deaths that should have been prevented.
Indeed, the risks of mistake are overall greater in the case of terminating life support. Cruzan implied that a state must allow individuals to make such decisions through an advance directive stipulating either that life support be terminated (or not initiated) in described circumstances when the individual was no longer competent to make such a decision himself, or that a designated proxy be allowed to make that decision. All the risks just described are present when the decision is made through or pursuant to such an advance directive, and a grave further risk is added: that the directive, though still in force, no longer represents the wishes of the patient. The patient might have changed his mind before he became incompetent, though he did not change the directive, or his proxy may make a decision that the patient would not have made himself if still competent. In Cruzan, this Court held that a state may limit these risks through reasonable regulation. It did not hold—or even suggest—that a state may avoid them through a blanket prohibition that, in effect, denies the liberty interest altogether.
Second, nothing in the record supports the [Solicitor General’s] conclusion that no system of rules and regulations could adequately reduce the risk of mistake. As discussed above, the experience of states in adjudicating requests to have life-sustaining treatment removed indicates the opposite.¹⁴ The Solicitor General has provided no persuasive reason why the same sort of procedures could not be applied effectively in the case of a competent individual’s request for physician-assisted suicide.
Indeed, several very detailed schemes for regulating physician-assisted suicide have been submitted to the voters of some states¹⁵ and one has been enacted.¹⁶ In addition, concerned groups, including a group of distinguished professors of law and other professionals, have drafted and defended such schemes. See, e.g., Charles H. Baron, et. al., A Model State Act to Authorize and Regulate Physician-Assisted Suicide, 33 Harv. J. Legis. 1 (1996). Such draft statutes propose a variety of protections and review procedures designed to insure against mistakes, and neither Washington nor New York attempted to show that such schemes would be porous or ineffective. Nor does the Solicitor General’s brief: it relies instead mainly on flat and conclusory statements. It cites a New York Task Force report, written before the proposals just described were drafted, whose findings have been widely disputed and were implicitly rejected in the opinion of the Second Circuit below. See generally Quill v. Vacco, 80 F.3d 716 (2d Cir. 1996). The weakness of the Solicitor General’s argument is signaled by his strong reliance on the experience in the Netherlands which, in effect, allows assisted suicide pursuant to published guidelines. Brief for the United States at 23-24. The Dutch guidelines are more permissive than the proposed and model American statutes, however. The Solicitor General deems the Dutch practice of ending the lives of people like neo-nates who cannot consent particularly noteworthy, for example, but that practice could easily and effectively be made illegal by any state regulatory scheme without violating the Constitution.
The Solicitor General’s argument would perhaps have more force if the question before the Court were simply whether a state has any rational basis for an absolute prohibition; if that were the question, then it might be enough to call attention to risks a state might well deem not worth running. But as the Solicitor General concedes, the question here is a very different one: whether a state has interests sufficiently compelling to allow it to take the extraordinary step of altogether refusing the exercise of a liberty interest of constitutional dimension. In those circumstances, the burden is plainly on the state to demonstrate that the risk of mistakes is very high, and that no alternative to complete prohibition would adequately and effectively reduce those risks. Neither of the Petitioners has made such a showing.
Nor could they. The burden of proof on any state attempting to show this would be very high. Consider, for example, the burden a state would have to meet to show that it was entitled altogether to ban public speeches in favor of unpopular causes because it could not guarantee, either by regulations short of an outright ban or by increased police protection, that such speeches would not provoke a riot that would result in serious injury or death to an innocent party. Or that it was entitled to deny those accused of crime the procedural rights that the Constitution guarantees, such as the right to a jury trial, because the security risk those rights would impose on the community would be too great. One can posit extreme circumstances in which some such argument would succeed. See, e.g., Korematsu v. United States, 323 U.S. 214 (1944) (permitting United States to detain individuals of Japanese ancestry during wartime). But these circumstances would be extreme indeed, and the Korematsu ruling has been widely and severely criticized.
Third, it is doubtful whether the risks the Solicitor General cites are even of the right character to serve as justification for an absolute prohibition on the exercise of an important liberty interest. The risks fall into two groups. The first is the risk of medical mistake, including a misdiagnosis of competence or terminal illness. To be sure, no scheme of regulation, no matter how rigorous, can altogether guarantee that medical mistakes will not be made. But the Constitution does not allow a state to deny patients a great variety of important choices, for which informed consent is properly deemed necessary, just because the information on which the consent is given may, in spite of the most strenuous efforts to avoid mistake, be wrong. Again, these identical risks are present in decisions to terminate life support, yet they do not justify an absolute prohibition on the exercise of the right.
The second group consists of risks that a patient will be unduly influenced by considerations that the state might deem it not in his best interests to be swayed by, for example, the feelings and views of close family members. Brief for the United States at 20. But what a patient regards as proper grounds for such a decision normally reflects exactly the judgments of personal ethics—of why his life is important and what affects its value—that patients have a crucial liberty interest in deciding for themselves. Even people who are dying have a right to hear and, if they wish, act on what others might wish to tell or suggest or even hint to them, and it would be dangerous to suppose that a state may prevent this on the ground that it knows better than its citizens when they should be moved by or yield to particular advice or suggestion in the exercise of their right to make fateful personal decisions for themselves. It is not a good reply that some people may not decide as they really wish—as they would decide, for example, if free from the “pressure” of others. That possibility could hardly justify the most serious pressure of all—the criminal law which tells them that they may not decide for death if they need the help of a doctor in dying, no matter how firmly they wish it.
There is a fundamental infirmity in the Solicitor General’s argument. He asserts that a state may reasonably judge that the risk of “mistake” to some persons justifies a prohibition that not only risks but insures and even aims at what would undoubtedly be a vastly greater number of “mistakes” of the opposite kind—preventing many thousands of competent people who think that it disfigures their lives to continue living, in the only way left to them, from escaping that—to them—terrible injury. A state grievously and irreversibly harms such people when it prohibits that escape. The Solicitor General’s argument may seem plausible to those who do not agree that individuals are harmed by being forced to live on in pain and what they regard as indignity. But many other people plainly do think that such individuals are harmed, and a state may not take one side in that essentially ethical or religious controversy as its justification for denying a crucial liberty.
Of course, a state has important interests that justify regulating physician-assisted suicide. It may be legitimate for a state to deny an opportunity for assisted suicide when it acts in what it reasonably judges to be the best interests of the potential suicide, and when its judgment on that issue does not rest on contested judgments about “matters involving the most intimate and personal choices a person may make in a lifetime, choices central to personal dignity and autonomy.” Casey, 505 U.S. at 851. A state might assert, for example, that people who are not terminally ill, but who have formed a desire to die, are, as a group, very likely later to be grateful if they are prevented from taking their own lives. It might then claim that it is legitimate, out of concern for such people, to deny any of them a doctor’s assistance [in taking their own lives].
This Court need not decide now the extent to which such paternalistic interests might override an individual’s liberty interest. No one can plausibly claim, however—and it is noteworthy that neither Petitioners nor the Solicitor General does claim—that any such prohibition could serve the interests of any significant number of terminally ill patients. On the contrary, any paternalistic justification for an absolute prohibition of assistance to such patients would of necessity appeal to a widely contested religious or ethical conviction many of them, including the patient-plaintiffs, reject. Allowing that justification to prevail would vitiate the liberty interest.
Even in the case of terminally ill patients, a state has a right to take all reasonable measures to insure that a patient requesting such assistance has made an informed, competent, stable and uncoerced decision. It is plainly legitimate for a state to establish procedures through which professional and administrative judgments can be made about these matters, and to forbid doctors to assist in suicide when its reasonable procedures have not been satisfied. States may be permitted considerable leeway in designing such procedures. They may be permitted, within reason, to err on what they take to be the side of caution. But they may not use the bare possibility of error as justification for refusing to establish any procedures at all and relying instead on a flat prohibition.

Conclusion

Each individual has a right to make the “most intimate and personal choices central to personal dignity and autonomy.” That right encompasses the right to exercise some control over the time and manner of one’s death.
The patient-plaintiffs in these cases were all mentally competent individuals in the final phase of terminal illness and died within months of filing their claims.
Jane Doe described how her advanced cancer made even the most basic bodily functions such as swallowing, coughing, and yawning extremely painful and that it was “not possible for [her] to reduce [her] pain to an acceptable level of comfort and to retain an alert state.” Faced with such circumstances, she sought to be able to “discuss freely with [her] treating physician [her] intention of hastening [her] death through the consumption of drugs prescribed for that purpose.” Quill v. Vacco, 80 F.2d 716, 720 (2d Cir. 1996) (quoting declaration of Jane Doe).
George A. Kingsley, in advanced stages of AIDS which included, among other hardships, the attachment of a tube to an artery in his chest which made even routine functions burdensome and the development of lesions on his brain, sought advice from his doctors regarding prescriptions which could hasten his impending death. Id.
Jane Roe, suffering from cancer since 1988, had been almost completely bedridden since 1993 and experienced constant pain which could not be alleviated by medication. After undergoing counseling for herself and her family, she desired to hasten her death by taking prescription drugs. Compassion in Dying v. Washington, 850 F. Supp. 1454, 1456 (1994).
John Doe, who had experienced numerous AIDS-related ailments since 1991, was “especially cognizant of the suffering imposed by a lingering terminal illness because he was the primary caregiver for his long-term companion who died of AIDS” and sought prescription drugs from his physician to hasten his own death after entering the terminal phase of AIDS. Id. at 1456-57.
James Poe suffered from emphysema which caused him “a constant sensation of suffocating” as well as a cardiac condition which caused severe leg pain. Connected to an oxygen tank at all times but unable to calm the panic reaction associated with his feeling of suffocation even with regular doses of morphine, Mr. Poe sought physician-assisted suicide. Id. at 1457.
A state may not deny the liberty claimed by the patient-plaintiffs in these cases without providing them an opportunity to demonstrate, in whatever way the state might reasonably think wise and necessary, that the conviction they expressed for an early death is competent, rational, informed, stable, and uncoerced.
Affirming the decisions by the Courts of Appeals would establish nothing more than that there is such a constitutionally protected right in principle. It would establish only that some individuals, whose decisions for suicide plainly cannot be dismissed as irrational or foolish or premature, must be accorded a reasonable opportunity to show that their decision for death is informed and free. It is not necessary to decide precisely which patients are entitled to that opportunity. If, on the other hand, this Court reverses the decisions below, its decision could only be justified by the momentous proposition—a proposition flatly in conflict with the spirit and letter of the Court’s past decisions—that an American citizen does not, after all, have the right, even in principle, to live and die in the light of his own religious and ethical beliefs, his own convictions about why his life is valuable and where its value lies.