In the March 27, 1997, issue The New York Review published an amicus curiae brief submitted by six moral and political philosophers to the Supreme Court, urging the Court, in cases then pending, to recognize a limited constitutional right of terminally ill and competent patients to the help of a doctor in ending their lives, in order to avoid further pointless suffering and anguish.1 On June 26, the Court decided these cases, refusing to recognize such a right by an apparently crushing 9-0 vote.2 But though press reports did not make this clear, the unanimity of the vote was deceptive. Five of the six justices who wrote opinions made it plain that they did not reject such a right in principle, suggesting that the Court might well change its mind in a future case when more evidence of the practical impact of any such right was available.
The cases raised, moreover, not only the question of a specific right to assisted suicide, but a more fundamental constitutional issue as well—how to understand and enforce the “due process clause” of the Fourteenth Amendment, which declares that states may not “deprive any person of life, liberty or property, without due process of law.” That clause offers more potential protection to the liberty of individual citizens than any other constitutional provision. Its language is very abstract, however, and the actual protection it offers depends on whether it is read narrowly or expansively. The Court’s assisted suicide decisions contain the fullest and most candid judicial debate of that issue for many decades, and I shall discuss that debate first, not only because the issue is intrinsically important, but in order to understand the great differences in the opinions that made up the deceptively unanimous verdict.
The words “due process of law” might conceivably have been taken to mean, at one extreme, that government may compromise liberty in any way it likes so long as it follows stipulated procedures of lawmaking in doing so. At the other extreme, the clause might be interpreted to say that judges may strike down any law that offends what they themselves deem to be a requirement of pure justice. But almost every lawyer rejects both those extremes. The due process clause, according to the general understanding, condemns all the laws—and only the laws—that curtail liberties that are, in an often-quoted passage from an earlier Supreme Court decision, “deeply rooted in this Nation’s history and tradition.” But that historical standard is also very abstract and lawyers disagree about what it means.
In his majority opinion in the assisted suicide cases, on behalf of himself and four other justices, Chief Justice William Rehnquist defended an historicist view of the “Nation’s history and tradition” test. He insisted that the due process clause protects only those specific liberties that have historically been respected by American states, so that the clause does protect citizens from unwanted and invasive medical treatment, because the common law of most states has for a long time granted that protection. But, according to Rehnquist, the clause permits laws prohibiting a doctor from helping people dying in great pain to die sooner, because almost all states have long prohibited such help.
In an unusually candid statement of this view, Rehnquist acknowledged that it might well produce anomalies of principle, because it might well be that no principled distinction can be drawn between liberties that American states have historically protected and those that they have denied.3 He said that the Supreme Court’s suggestion, in its earlier Cruzan decision,4 that the due process clause gives people a right to have life-saving apparatus removed from their bodies was drawn only from common-law practice, and was
not simply deduced from abstract concepts of personal autonomy…. The decision to commit suicide with the assistance of another may be just as personal and profound as the decision to refuse unwanted medical treatment, but it has never enjoyed similar legal protection.
Justice David Souter, on the other hand, in his separate opinion concurring in but not joining Rehnquist’s majority opinion, offered a much more expansive view of the historical test.5 He said that the nation’s history and traditions include not just the specific rights that have been recognized in the past, but the “basic values” that are revealed when we interpret those rights to see which more general principles of political morality they represent. It may be, he said, that states have not always been wholly faithful to those basic values, and that some of even the oldest legal practices, like the long prohibition on abortion, can now be seen to offend them and so to violate the due process clause.
Judges, he said, must take care in deciding which principles of political morality do underlie the nation’s history, because these values can be identified at varying levels of generality and judges must not state them more broadly than a sound interpretation would justify. He conceded that identifying principles at the right level of generality is not a mechanical matter. “Selecting among such competing characterizations demands reasoned judgment about which broader principle, as exemplified in the concrete privileges and prohibitions embodied in our legal tradition, best fits the particular claim asserted in a particular case.” He drew, from his understanding of the due process clause, very different conclusions about assisted suicide from Rehnquist’s. If we apply reasoned judgment to the assisted suicide issue, Souter argued, we can identify arguments of what he called “increasing forcefulness for recognizing some right to a doctor’s help in suicide.”
The strongest of these arguments, he said, rests on a general principle, embedded in past traditions, that guarantees a
right to medical care and counsel, subject to the limiting conditions of informed, responsible choice when death is imminent…. There can be no stronger claim to a physician’s assistance than at the time when death is imminent, a moral judgment implied by the State’s own recognition of the legitimacy of medical procedures necessarily hastening the moment of impending death [e.g., terminating life support and allowing pain relief that advances death].
So Rehnquist’s and Souter’s views of the due process clause are dramatically different: the first protects individuals only from laws that few states have seen any reason to enact, and offers no protection at all against historically popular invasions of individual freedom. The second holds out the possibility that even longstanding and popular legal rules, like the ban on assisted suicide, might be held unconstitutional when they can be seen to violate more general and established principles of freedom.
It is therefore important to try to gauge the popularity of each of these views in the present Supreme Court. Rehnquist, as I have said, had no difficulty in using his historicist approach to reject any right to assisted suicide out of hand. Four other justices—Justices Anthony Kennedy, Sandra Day O’Connor, Antonin Scalia, and Clarence Thomas—joined Rehnquist’s opinion, and we may safely assume that Scalia and Thomas do in fact embrace the historicist assumptions of that opinion.
O’Connor and Kennedy, however, were two of the three justices—the other was Souter—who wrote a joint opinion in the 1992 Casey abortion decision6 endorsing the interpretive view of due process that Souter defended in this case, and it is therefore puzzling why they joined Rehnquist’s opinion. Perhaps they did so out of institutional courtesy, so that one opinion—Rehnquist’s—could attract five votes and so count as the opinion of the Court, avoiding the inelegant result of a unanimous decision with no majority opinion.
In any case, however, O’Connor wrote a separate opinion which makes it plain that she still does not accept Rehnquist’s historicist understanding of the due process clause. She identified the question posed by the cases as “whether a mentally competent person who is experiencing great suffering has a constitutionally cognizable interest in controlling the circumstances of his or her imminent death.” She said that she saw no need to decide that question because, even if such a patient had such a right, the state laws challenged in the cases did not violate it because they allowed doctors to administer pain-relieving drugs even when these hastened death. She left open, that is, the crucial question—whether patients have some right to control how they die—that the historicist understanding would have answered, negatively and immediately. Kennedy did not write a separate opinion. But it is unlikely that he accepts the historicist account, given not only his opinion in Casey but also his explicitly interpretive reading of the other great clause of the Fourteenth Amendment—the equal protection clause—in the recent Colorado homosexual rights case.7
Justices Breyer, Ginsburg, and Stevens, who declined to join Rehnquist’s opinion, each made it at least as plain as O’Connor did that they were much closer to Souter’s than to Rehnquist’s understanding of the due process clause. Ginsburg said simply that she agreed with O’Connor. Breyer said pointedly that he joined O’Connor’s opinion “except insofar as it joins the majority.” He added that he would formulate the patient’s claim in these cases, not as Rehnquist had, but rather in “words roughly like a ‘right to die with dignity,”‘ and he said that “our legal tradition may provide greater support” for such a right.
Breyer said, however, that he did not have to decide the question whether the due process clause actually does require judges to recognize such a right, because “the avoidance of severe physical pain (connected with death) would have to comprise an essential part of any successful claim,” and he agreed with O’Connor that pain can be avoided because the states do not prohibit even pain-relieving treatment that advances death. He concluded with the important observation that if states did interfere with the “administration of drugs as needed to avoid pain at the end of life,” then “as Justice O’Connor suggests, the Court might have to revisit its conclusions in these cases.”
The remaining justice—Stevens—wrote an eloquent separate opinion to explain that his vote to reverse the lower court decisions was based on procedural rather than substantive grounds. He said that since the patients who were plaintiffs in the cases at hand had all died before the Supreme Court decision, the question before the Court was not whether the anti-assisted suicide laws could constitutionally be applied to patients who were dying when they asked for relief. Instead, he said, once the patients had died, the case required the Court to decide whether the anti-suicide laws could constitutionally be applied to anyone at all, including, for example, a depressed but otherwise healthy person who had expressed a wish to die. Since he thought that a state could properly prevent doctors from aiding some people who wanted to die, he voted to sustain the statutes as not “facially” invalid.
His opinion left little doubt, however, that in what he deemed an appropriate case he would vote to overrule a statute that prevented doctors from helping competent and informed dying patients—not just those whose pain could not otherwise be relieved—to die sooner. He emphasized, as the Philosophers’ Brief had, that different people have different religious and ethical convictions about what kind of death most respects the value of their life, and that individual freedom demands that dying patients be permitted to die according to their own convictions. He ended with the uncompromising statement that
In my judgment…it is clear that [the states’] so-called “unqualified interest in the preservation of human life”…is not itself sufficient to outweigh the interest in liberty that may justify the only possible means of preserving a dying patient’s dignity and alleviating her intolerable suffering.
Stevens’s opinion, though technically a vote against those who challenged the prohibitory statutes, was in fact a vote for all that they asked.
So the narrow, historicist view of the due process clause is now probably confined to a core group of the three most conservative members of the present Court—Rehnquist, Scalia, and Thomas—and that is welcome news for those who favor a principled construction of the individual rights that American citizens enjoy against their government. It is also welcome news that though the Supreme Court refused to recognize a right to assisted suicide in these cases, five justices took care not to foreclose the constitutional debate over such a right for the future.
Stevens’s opinion declared his readiness to recognize that right whenever an appropriate case arose. Souter’s opinion said three times that his vote was only “at this time.” O’Connor and Breyer each said that changed circumstances might cause them to reconsider. And Ginsburg, joining O’Connor’s opinion rather than that of the Court, made it clear that she agreed. It is therefore important to consider why each of these justices, except Stevens, declined to recognize the right “at this time.”
O’Connor, Ginsburg, and Breyer argued, as I have said, that any constitutional right would be limited to relief from pain. But that limitation seems arbitrary, for several reasons. These justices did not explain why what Breyer called a right to die “with dignity” means only dying without pain when, as Souter noted, many people dread a drug-induced stupor just as much, and understandably think it just as offensive to their dignity.8 Though these justices declared themselves satisfied that pain could be prevented in all but a very few cases, moreover, they did not attempt to respond to the substantial evidence, cited by Stevens, to the contrary. And it seems odd that they should be ready to overrule carefully considered decisions of the lower federal courts recognizing a limited right to assisted suicide by appealing to a factual claim vigorously contested in the briefs without any argument for that claim.
Nor did these justices explain why those individuals whose pain concededly cannot be relieved except by rendering them unconscious, no matter how few or many they are, do not have a right to assistance in dying. Breyer did acknowledge that many patients, particularly poor patients, do not receive the palliative treatment, which is often expensive, that would benefit them. But, he said, that is for “institutional reasons or inadequacies or obstacles, which would seem possible to overcome.” It is unsatisfactory, however, to argue that even if dying poor patients have a right not to die in extreme pain, a state may nevertheless forbid them the assistance of a doctor in suicide, even when that means that they actually will die in great pain because the state refuses to pay for the pain relief that would help them if it were provided.
Souter offered a more elaborate and impressive set of reasons for not recognizing any constitutional right at this time. He mentioned the arguments, pressed by opponents of assisted suicide, that it would prove impossible to design any system of regulatory control that would protect people whose death was not in fact imminent, or who did not really wish to die, from being coaxed into suicide by relatives or hospitals who did not want to bear the expense of keeping them alive, or by compassionate doctors who thought them better off dead. He cited, in particular, books and articles purporting to show that the only documented program of assisted suicide and euthanasia—the Dutch one—had failed to prevent many such mistakes. He noted, however, that these analyses of the Dutch case were contradicted by other reports, and that many writers did think that states could develop an effective regulatory scheme that would reduce mistakes to the level that was unavoidable in any grave medical procedure. But, he said, judges should not declare laws enacted by almost all the states unconstitutional on the basis of disputed and controversial judgments of fact, particularly in circumstances when legislatures, who can deploy investigative committees, are in a better position to assess the facts than judges are. So, he concluded, the Court should not declare anti-assisted suicide laws unconstitutional “at this time,” though it might be right to do so later, when and if better evidence is available or more persuasive studies have been made.
That seems reasonable in principle. But, as Souter candidly acknowledged, the Court has assumed that dying patients have a constitutional right to terminate life support even when this means that they will die immediately; and there is as much danger that such patients will be coaxed into a request to die in that way as by requesting lethal pills, particularly since life-support techniques are typically very expensive.9 In any case, the question whether a factual issue is too difficult or uncertain for judges to decide, so that they ought to defer to legislative decisions for that reason, is itself a complex and difficult one, and the courts should answer it only after very careful review of the evidence, particularly when putatively fundamental rights of individual citizens are at stake. Careful review would seem particularly crucial, moreover, in the assisted suicide debate, because many of the social scientists who have com-piled the relevant evidence have strong ethical opinions, including religious convictions, or convictions about proper medical ethics, that might impair their scientific independence.
Many opponents of assisted suicide argue that in the Netherlands—where doctors are not prosecuted if they help patients to die, provided the doctors follow guidelines approved by the courts—people have been, in the words of the psychiatrist Herbert Hendin, “seduced” into death.10 (Hendin is the executive director of an organization called “The Suicide Foundation,” which he describes in his book as “working to prevent suicide.”11 ) Souter cited Hendin’s book, among others, as supporting his own argument that American courts are not now in a position to dismiss the view the books defended.
It is unclear, however, why, even if Hendin and other critics are right about “the Dutch experience,” it follows that American states, which would be free to impose much stricter regulations than the Dutch courts have, could not offer much greater protection to patients who might be pressured into death. It is also unclear, moreover, that a closer look at the books critical of Dutch medical practice would not have revealed serious methodological flaws. In a forthcoming book, three scholars in Groningen University in the Netherlands—John Griffiths, Alex Bood, and Heleen Weyers—offer a balanced and to some degree critical account of Dutch practice, suggesting ways in which enforcement of the court-approved guidelines for euthanasia might be improved.12 But they are highly critical of Hendin’s methodology. They allege that his book contains many mistakes, and that several of his reports of his conversations with Dutch doctors, on which many of his arguments are based, misinterpreted those conversations.13
Souter referred, particularly, to Hendin’s claim that in the Netherlands doctors have ended the life of fully competent patients without a request from, and without consulting, the patients. Rehnquist also referred to the apparently shocking fact that, “The Dutch government’s own study revealed that in 1990, there were… more than 1,000 cases of euthanasia without an explicit request.” In fact, an official report on that study published in English in 1992, which neither Souter nor Rehnquist cited, shows that the cases described as euthanasia “without a patient’s request” were of many different types. They included, for example, cases of patients who could not consent because they were in the final throes of death, and whose death the doctors made easier with drugs that hastened the dying process by minutes, cases of people who had earlier expressed a wish to die but not in terms conforming to the strict Dutch test for a technically explicit request, and cases of just-born infants who would have died within days anyway, and whose prompt death, at their parents’ request, saved those parents agony.14 If American states believe such practices offensive or undesirable, they could prohibit them with straightforward legislative provisions, and it is hard to see how allowing assisted suicide for those who do request it, subject to strict regulations and reporting, would make such practices more likely.
I do not mean to suggest that the Groningen scholars’ criticism of Hendin is justified; I am not able to assess it. But it is regrettable that Souter and the other justices were not in a position to inspect more thoroughly the research techniques of Hendin and the other critics of Dutch practice they cited before deciding that their books raised issues, about the inevitable threat that permitting highly regulated assisted suicide would pose to people reluctant to die, that judges are now incapable of resolving. Perhaps there really is no sound research suggesting that the Dutch experience demonstrates such a threat.
It is hard to resist the conclusion that those justices who, like Souter, accept an interpretivist rather than an historicist understanding of the due process clause, but who were understandably reluctant to invalidate the laws of almost every state, were unable to find convincing arguments to justify their decision not to do so. That makes it even more likely that the constitutional debate will continue.
What effects will the Court’s decisions have in the meantime? The Court made plain that citizens are free to press for permissive reforms in the present law in the ordinary way, through legislation or referendums, and, since the idea that dying patients should have a chance for a doctor’s help in easing their death remains popular, such efforts might well succeed in some further states.
But the more immediate consequences of the Court’s decision may be medical rather than political: the combination of opinions I described strongly suggests that the Court, though perhaps by a divided vote, would be ready to recognize that dying patients do have a right to pain relief even in doses, if necessary to stop pain, that bring death. It is, of course, very difficult to judge whether a doctor and patient who agree on such high doses of painkilling drugs aim at the death that follows or only at pain relief. But thousands of doctors have been prescribing such drugs in lethal doses, for dying patients whose families they know and trust, in order to hasten death.15 The effect of the present decisions may well be not only to confirm but to extend that practice. That is paradoxical, because the administration of very high levels of morphine and other drugs is not regulated in the way any permitted assisted suicide would be, and is therefore much more open to abuse, particularly in the case of the poor and vulnerable patients whose safety has been cited as a main reason for refusing to recognize any right to the latter.
As I said in my introduction to “The Philosophers’ Brief,” a strictly regulated scheme of assisted suicide would demand full information for and informed consent by patients who seek such assistance. It would also require hospitals to satisfy supervising authorities that all options for treatment and palliation had been explained and offered to them. Such a scheme might improve the situation of such patients, and might well extend their lives. A policy that encourages financially stretched hospitals routinely to recommend morphine in potentially lethal doses, for the ostensible purpose of relieving pain, subject to no specific code of regulation at all, may seem less a violation of established medical practices. But it hardly offers less danger to poor patients dying in pain whose relatives or doctors might prefer them dead at once.
—August 28, 1997
Washington et al. v. Glucksberg, 1997 WL 348094; Vacco v. Quill, 1997 WL 348037. ↩
It is worth noticing that he did not defend the historicist approach, as other conservative lawyers have attempted to do, by appealing to the “original intention” of those who wrote and enacted the Fourteenth Amendment. He defended it on the different, and I think now much more popular ground, that weakening the due process clause in this (or indeed any other available) way reduces the power of judges to contradict the wishes and convictions of a majority of a state’s citizens. For an assessment of that argument, see my recent book, Freedom’s Law (Harvard University Press, 1996). ↩
Cruzan v. Director, Missouri Dept. of Health, 497 US 261 (1990). ↩
In an earlier article in The New York Review, I defined and contrasted the historicist and interpretivist views of the due process clause at greater length. See “Sex, Death, and the Courts,” The New York Review, August 8, 1996, pp. 44-50. ↩
Planned Parenthood v. Casey, 505 US 833 (1992). ↩
Of course, all pain can be relieved by total anesthesia. But I assume that the justices meant that in almost all cases it can be relieved while leaving the patient in some, though perhaps a reduced, state of consciousness. Many people would regard living the rest of their lives under total anesthesia as worse than death. ↩
Souter said, in a footnote, that the state has a weaker interest in preventing death following the removal of life support, because then “nature” causes the death. That is not a morally pertinent distinction, as the Philosophers’ Brief pointed out, and in any case not pertinent when the question is only whether a patient’s request is genuine or forced. He also said that because life support is a bodily intrusion, there is less reason to fear that someone requesting its removal, when the result is death, is not fully responsible than there is in the case of someone requesting lethal pills. That seems a non sequitur: people often request that life support be terminated not because they find its presence particularly offensive, but because they want to die, and the dangers of coaxing someone to make that decision seem equally great in the two cases. ↩
Herbert Hendin, Seduced by Death:Doctors, Patients, and the Dutch Cure (Norton, 1997). I should note that the book takes issue with my own writing on its subject. ↩
Hendin, Seduced by Death, p. 223. ↩
John Griffiths, Alex Bood, and Heleen Weyers, Euthanasia and Law in the Netherlands, forthcoming from Amsterdam University Press. ↩
Five doctors, four of whom Hendin describes as “major sources” of his research, wrote a joint letter to the journal that published his initial article, which had the same title as the later book. The letter read, in part, “The following persons interviewed by dr. (sic) Herbert Hendin wish to declare that the texts of the interviews do not contain a truthful description of the interviews. The text contains several errors and flawed interpretations.” They asked that their letter be published with the article. It was not, and though Hendin made some changes in the article before publication, these changes, according to the Groningen scholars and three of the doctors, with whom I spoke on the telephone, did not correct the misinterpretations, which, in their opinion, are perpetuated in Hendin’s later writings. ↩
See P.J. van der Maas, J.J.M. van Delden, and L. Pijnenborg, “Euthanasia and other Medical Decisions Concerning the End of Life,” translated and printed as special supplement 1 and 2 to Volume 22 of Health Policy (Elsevier Science Publishers, 1992). ↩
A recent New York Times article described this practice and quoted a University of California professor of medicine and anesthesiology as saying, “It happens all the time.” Gina Kolata, “When Morphine Fails to Kill,” The New York Times, July 23, 1997. That article reports the opinion of several doctors that patients whose dosage of morphine is slowly increased can tolerate, in the end, very high doses. But that has not been proved to be true for all patients, and it would be difficult to challenge a doctor’s decision that pain could not be controlled soon enough except through doses that, in fact, caused death. For an example of the wide variation of opinions among doctors on these issues, see the series of articles in Volume 12, No. 4, Journal of Palliative Care (1996). ↩