It's hard not to be moved by emotional accounts of how laws prohibiting assisted suicide can drive pain-wracked people to desperate ends. A year ago, in The New Yorker, Andrew Solomon wrote eloquently about how he and his brother helped their mother take sleeping pills to spare her the final agonies of ovarian cancer, an ordeal made even more harrowing by the fear of prosecution. All of the jurors who acquitted Dr. Jack Kevorkian in May similarly said they were influenced by videotapes in which two women suffering from chronic pain described their anguish and pleaded to be allowed to die. (The jurors were unmoved by the fact that the ailments were not fatal.) Confronted with heartbreaking stories like these, it's tempting to conclude that all laws barring competent citizens from dying with dignity should be repealed or invalidated. And, indeed, when two federal appeals courts, during the past few months, struck down state laws in New York and Washington prohibiting physician-assisted suicide, the judges were widely praised for their statesmanship and humanity.
But, in fact, the judicial discovery of a sweeping constitutional right to die is analytically unconvincing and politically indefensible. The recent decisions rely heavily on the similarities between abortion and euthanasia, both of which, to use Ronald Dworkin's typically abstract formulation, involve "choices for death." Both euthanasia and abortion inspire profound religious and moral disputes; both have been, at certain times in American history, prohibited by the states; and the case for each becomes more or less compelling at different points in the life cycle. But creating constitutional rights by analogy is often treacherous, as the Supreme Court discovered in the wake of Roe v. Wade, when it was witheringly criticized for expanding its earlier cases concerning the privacy of the marital bedroom into an apparently unrelated right of doctors to perform first-trimester abortions in hospitals. By blithely repeating the errors of Roe, and expanding the narrow right to refuse unwanted medical treatment into a much broader right to hasten one's own death, the recent assisted suicide decisions show the dangers of constitutional abstraction. What follows is a narrative of how a constitutional right is invented.
Until the judicial decisions earlier this year, America's euthanasia debate was organized around a series of carefully maintained distinctions. The most basic right, now recognized by the Supreme Court, Congress and the state legislatures, is the right to refuse unwanted medical treatment, such as feeding tubes and hydration. Less widely accepted is assisted suicide, still criminal in thirty-four states, which occurs when a doctor prescribes medication that may ease pain or cause death (the double effect), and the patient administers the lethal dose himself. Most controversial is "voluntary passive euthanasia," which takes place when the doctor, rather than the incapacitated patient, administers the lethal dose with the patient's consent.
Last March, in San Francisco, the U.S. Court of Appeals for the Ninth Circuit impatiently swept aside these painstaking distinctions and became the first federal appeals court in the country to recognize a constitutional right to "determine the time and manner of one's own death." Writing for eight members of the Court, Judge Stephen Reinhardt struck down a Washington state law prohibiting physician-assisted suicide, and draped himself in Justice Blackmun's mantle. "In deciding right-to-die cases, we are guided by the Court's approach to the abortion cases," he declared, and proceeded to discover a constitutional right to die in the interstices of the Constitution's due process clause.
Judge Reinhardt's grandiloquent opinion is a marvel of activism in the old style. It brings to mind a Fourth of July oration by a Victorian dignitary: long, strenuous, delivered at full volume, and ostentatiously ornamented with shows of learning. "Like the Court in Roe," he intones, "we begin with ancient attitudes"; he then delivers a leisurely disquisition on views of suicide in classical literature and mythology from Greek and Roman times to our own. ("The first of all literary suicides, that of Oedipus's mother, Jocasta, is made to seem praiseworthy," Reinhardt notes helpfully. "Hundreds of Jews killed themselves at Masada in order to avoid being captured by Roman legions.") In the space of several pages, Reinhardt also manages to quote St. Augustine, Bracton and King Lear. ("Vex not his ghost. O, let him pass! He hates him/That would upon the rack of this tough world/Stretch him out longer.") By the end of the opinion, we know just how Lear feels.
Reinhardt's constitutional analysis, however, is less weighty. First, he invokes the Supreme Court's 1990 Cruzan opinion, in which the parents of Nancy Cruzan, a patient in a persistent vegetative state, sought to remove her feeding tube, even though the patient herself had not left a living will. Chief Justice Rehnquist assumed, for the sake of argument, that competent people have a "constitutionally-protected liberty interest in refusing unwanted medical treatment." But Rehnquist held that a state could require convincing evidence of a patient's wishes before requiring hospitals to pull the plug.
By recognizing a right to refuse medical treatment, Judge Reinhardt suggests expansively, the Supreme Court "necessarily recognize[d] a liberty interest in hastening one's own death." But in fact the Supreme Court did no such thing. The Court emphasized that each state has an "unqualified interest in the preservation of human life," and to illustrate this point, it noted that "the majority of states in this country have laws imposing criminal penalties on one who assists another to commit suicide." Far from questioning these laws, the Supreme Court explicitly appeared to endorse them.
The Supreme Court's distinction between refusing treatment and committing assisted suicide, between killing and letting die, is analytically coherent and nationally accepted. (In 1990, Congress codified the right to refuse medical treatment in the Federal Patient Self-Determination Act.) A long line of cases say that patients have a right to be protected against unwanted physical contact from doctors, which can be seen as a kind of violent battery or a violation of the principle of informed consent. Hooking people up to tubes against their will might also be seen as an unreasonable search or seizure of the body, prohibited by the Fourth Amendment. By contrast, a patient who wants to commit assisted suicide is not seeking to have a feeding tube withdrawn, but to have lethal medication applied. Prohibiting doctors from actively killing can hardly be considered as a bodily seizure or a battery.
Perhaps recognizing the inadequacy of the Cruzan case as a fountainhead for his broad right to die, Reinhardt turns to the abortion cases. But, even on its own terms, Reinhardt's analogy between abortion and assisted suicide is implausible. The cases from which Roe v. Wade derived a constitutional right to abortion involved reproductive autonomy or marital privacy. All have been reconceived, in the years since Roe, as cases involving gender discrimination and the special burdens suffered by women during unwanted pregnancy. But the right to die has no logical connection to sexual autonomy or gender equality.
The only way Reinhardt can sustain his analogy between abortion and euthanasia is by characterizing the sexual privacy cases at the highest possible level of generalization: "A common thread running through these [abortion and contraception] cases is that they involve decisions that are highly personal and intimate, as well as of great importance to the individual. Certainly, few decisions are more personal, intimate, or important than the decision to end one's life, especially when the reason for doing so is to avoid excessive and protracted pain." But this analogy is too slippery to be convincing. For some citizens, no decision is more "personal, intimate, or important" than the decision to engage in homosexual conduct; but this choice, according to the Supreme Court, is not one the Constitution protects. And it's hard to imagine a democracy that would authorize judges to invent fundamental rights guided by nothing more than their own estimation of which decisions are "personal, intimate, or important."
Reinhardt quotes a passage from Planned Parenthood v. Casey, the 1992 abortion case: "At 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." This doesn't help matters. Taken out of context, as Judge Robert Beezer emphasized dryly in dissent, the right to define one's own concept of existence "is so broad and melodramatic as to seem almost comical in its rhetorical flourish." But Reinhardt skirted over the preceding sentences in Casey, which emphasize that "our law affords constitutional protection to personal decisions relating to marriage, procreation, contraception, family relationships, child rearing, and education." Decisions about death are a different matter.
The unbounded quality of Reinhardt's "right to hasten one's own death" is the best reason to reject it. In a series of prescient articles written since 1958, Yale Kamisar of the University of Michigan has argued convincingly that courts should decline to expand the narrow right to refuse unwanted medical treatment into a broader right to control the timing of one's own death, because the broader right has no logical stopping point. And Reinhardt's opinion vindicates Kamisar's most dramatic fears. Isn't a healthy but depressed person's decision to commit suicide just as "personal, intimate, or important" as a decision to commit suicide near the end of an incurable illness? Wouldn't the right to define "one's own concept of existence" protect the right of all citizens to kill themselves, for whatever reasons they please? Reinhardt anxiously denies that he means to endorse a right of active voluntary euthanasia, but his "liberty interest" has no principled limits.
Axmonth after Judge Reinhardt struck down Washington state's law prohibiting assisted suicide, three judges on the U.S. Court of Appeals for the Second Circuit struck down New York state's assisted suicide law. The New York judges went out of their way to repudiate the reasoning of Judge Reinhardt's opinion, which they found open-ended and unconvincing. But they reached the same result for different reasons. The majority opinion, written by Judge Roger Miner, held that New York denies its citizens the equal protection of the laws, because it fails to "treat equally all competent persons who are in the final stages of fatal illness and wish to hasten their deaths." By allowing terminally ill patients on life support to refuse unwanted medical treatment, but prohibiting terminally ill patients who are not on life support to "hasten their deaths" with the aid of their physicians, New York, according to Judge Miner, fails to treat "similarly circumstanced persons alike."
Judge Miner's theory raises the specter of the slippery slope even more alarmingly than Judge Reinhardt's theory. As Yale Kamisar pointed out in testimony before the House Judiciary Committee at the end of April, if people off life support have to be treated as if they are "similarly circumstanced," to use Miner's jargon, to people on life support, why aren't patients who want to kill themselves with prescription drugs "similarly circumstanced" to patients who want to be strangled by their doctors? And how can this right be confined to the terminally ill? If New York has a duty to "treat equally all competent people who are in the final stages of fatal illness and wish to hasten their deaths," to use Miner's words again, isn't Kamisar correct that it would be similarly arbitrary to exclude people who suffer from Alzheimer's, severe depression or even arthritis?
There are good reasons, of course, why it was perfectly rational for the New York legislature to maintain the legal distinction between passively allowing a patient to die (which is permitted) and actively assisting in a patient's death (which is prohibited). As George Fletcher has argued, the distinction between action and omission is supported by the general reluctance, in American law, to hold people liable for their failure to prevent harm. And there's a broad medical, legal and philosophical consensus that stopping treatment should be considered not an action to kill but an omission to cure. Viewed in this light, assisted suicide violates the doctor's duty to avoid killing, while letting a patient die does not.
The New York case concludes with a concurring opinion by Judge Guido Calabresi. Of the three arguments for striking down assisted suicide laws, Calabresi's is the most modest and, on its face, the most plausible. As a Yale law professor in 1981, Calabresi had written an important book suggesting that courts could avoid some of the hazards of judicial activism by asking legislatures to take a "second look" at archaic statutes whose original purposes and popular support seemed to have eroded over time. Drawing extensively on his book, Judge Calabresi argues that the original rationales for New York's assisted suicide laws, like Connecticut's anti-abortion laws, are no longer widely accepted. In the nineteenth century, suicide and attempted suicide were both crimes in New York, and assisting someone to commit the crime of suicide was, accordingly, a crime as well. But, in 1919, the New York legislature repealed the criminal punishments for suicide and attempted suicide, while leaving in place the punishments for assisted suicide.
It's at least arguable, Calabresi suggests, that the assisted suicide ban survived as an oversight; and perhaps it no longer commands the support of New York legislators, now that its original purposes have been lost in the mists of time. By voting to strike down the assisted suicide law, Calabresi suggests, he wants to force the New York legislature to go before the people and to defend the ban openly, rather than relying, by inertia, on archaic premises.
Calabresi's theory is ingenious; but his historical narrative is selective and, ultimately, unconvincing. In fact, the New York legislature, like most other statute legislatures, made a series of deliberate and self-conscious decisions, over many decades, to retain the ban on assisted suicide, long after the decision to decriminalize suicide itself. As the dissenting judge in the California case noted, the reason most states abandoned civil and criminal penalties for suicide during the nineteenth century had nothing to do with a desire to exalt individual autonomy, or to remove social stigma from an act that was still considered reprehensible. Instead, legislatures were increasingly reluctant to punish the suicide's innocent family, and increasingly doubtful that criminal punishment was a very sensible way of deterring an act that came to be seen as a result of mental illness. But no state suggested that there was a "right" to commit suicide, or that people shouldn't be deterred from helping others commit it. On the contrary, after long debates in the 1950s, the Model Penal Code, the main text in criminal law teaching, concluded in 1962 that neither suicide nor attempted suicide should be a crime, but that aiding or soliciting another person to commit suicide should be a felony.
The New York state legislature was obviously influenced by the Model Penal Code in 1965, when it removed the nineteenth-century declaration that suicide was a "grave public wrong" but refused to remove the prohibition on assisted suicide. New York made its intentions even clearer in 1990, when the legislature codified the right to refuse unwanted medical treatment, but added the following caution: "This article is not intended to permit or promote suicide, assisted suicide, or euthanasia." For Judge Calabresi to insist that the New York legislature needs to express its views once again on the question of assisted suicide is like asking for one divine revelation too many.
The most jarring feature of all three right-to-die opinions is their unabashed reliance on opinion polls. In a section of his opinion called "Current Societal Attitudes," Judge Reinhardt cites a 1994 Harris Poll indicating that 73 percent of the respondents favored legalizing physician-assisted suicide under certain conditions. "Polls have repeatedly shown that a large majority of Americans--sometimes nearing 90 percent--fully endorse recent legal changes granting terminally ill patients, and sometimes their families, the prerogative to accelerate their death by refusing or terminating treatment," Reinhardt enthuses. Judges are usually more subtle in their efforts to follow the election returns; but in any event, "current societal attitudes" are considerably more complicated than the various judges suggest. Three states have held referenda on proposals to allow physician-assisted suicide for the terminally ill. Voters in two states--Washington and California--rejected the proposal by margins of about 54 percent to 46 percent. Oregon voters are the only ones to endorse a decriminalization measure, by a narrow margin of 51 percent to 49 percent. Six state legislatures have recently rejected bills that would permit physician-assisted suicide, and no legislature has passed a decriminalization bill.
This popular ambivalence suggests that the abortion analogy, so heavily relied on by right-do-die advocates, is particularly inappropriate. David Garrow argues that in the abortion battles of the 1960s, professional opinion was substantially ahead of popular opinion: the American Medical Association supported liberalization of abortion laws, for example, while popular support for liberalization lagged behind. By contrast, Garrow notes, in the right-to-die battles today, popular opinion seems to be ahead of professional opinion: the American Medical Association opposes liberalization of assisted suicide laws, while Oregon voters and the Kevorkian jurors seem to support it. Of course, judges have no business reading the polls under any circumstances; but the deep national division on the assisted suicide question should make courts all the more hesitant to jump into the fray.
All this is to say that the assisted suicide question, even more obviously than the abortion question, should be decided by the state legislatures and voters rather than by the courts. If left to their own devices, what are the states likely to decide? The policy question seems much harder than the constitutional question. Clearly, the laws prohibiting assisted suicide are rarely enforced: between 1950 and 1993, only eleven doctors in the United States were prosecuted for killing their terminally ill patients, and none went to prison. As one might expect, the infrequency with which the laws are enforced has dramatically reduced their deterrent effect: one in five doctors, according to a survey by the American Society of Internal Medicine, say they have helped their patients to commit suicide, as have one in five nurses, according to a more recent survey by The New England Journal of Medicine. This tacit public policy of refusing to prosecute physician-assisted suicides, except in the most suspicious cases, makes Judge Reinhardt's laments about the "onerous burden" of assisted suicide laws all the more hyperbolic.
In fact, the tension between what the law formally forbids and what society widely tolerates seems to have few social costs: with the exception of showmen like Dr. Kevorkian who openly court prosecution, doctors are relatively free to exercise their professional judgment in close cases. Why is there a burning need to repeal laws prohibiting assisted suicide? In Aging and Old Age, Judge Richard Posner argues that "permitting physician-assisted suicide ... [in] cases of physical incapacity might actually reduce the number of suicides and postpone the suicides that occur." If the only choice is between killing themselves now and suffering later, patients who fear they might some day become so incapacitated that they would be unable to commit suicide on their own will kill themselves now. But, Posner argues, if the choice is between suicide now or physician-assisted suicide later, the aging may well choose suicide later, because there is a chance they will be proved to have been mistaken about their future suffering or incapacity.
If Judge Posner's utility curves are correct, legalizing assisted suicide might save a few lives and alleviate some pain. But the difficulty of reaching a consensus about appropriate regulations for physician-assisted suicide, especially given the hysteria of popular elections, suggests that it may be more sensible to leave matters as they are. The circuit courts in California and New York both agreed that the right to physician-assisted suicide can't be exercised without accompanying safeguards; and both circuits, in effect, ordered their state legislatures to adopt procedural regulations to protect against "errors and abuse." (What an odd constitutional right that judges invent, and then command the state legislatures to restrict!) But the gothic complexity of the model regulations cited by the courts should give a pragmatist pause. Judge Reinhardt, for example, commended the following safeguards to the Washington legislature: "witnesses to ensure voluntariness; reasonable, though short, waiting periods to prevent rash decisions; second medical opinions to confirm a patient's terminal status ... psychological examinations to ensure that the patient is not suffering from momentary or treatable depression; reporting procedures that will aid in the avoidance of abuse." These suggestions may or may not be welcomed by the medical community; but surely doctors themselves are better qualified than state legislatures or judges to draft detailed codes of professional regulation.
It's naive, perhaps, to be surprised at the brazen overconfidence of the New York and California judges, as they substituted their benign moral impulses for the more rigorous judgments of physicians, philosophers and the people's representatives. Maybe, given a chance to play king for a day, most of us would have trouble restraining ourselves, too. But I doubt that advocates of the right to die will thank the judges for this extravagant favor. By rushing to preempt a national debate that has only just begun, and in the most heavy-handed way possible, the California and New York decisions will surely energize the most extreme opponents of assisted suicide and fan the flames of their evangelical crusade. We've been down this road before, of course, but the aftermath of the abortion cases was supposed to have weaned us of the urge to venture down it again.
Jeffrey Rosen is legal affairs editor at The New Republic and president and CEO of the National Constitution Center.