Full opinion text
Opinion by Judge REINHARDT; Dissent by Judge BEEZER; Dissent by Judge FERNANDEZ; Dissent by Judge KLEINFELD. REINHARDT, Circuit Judge: I. This case raises an extraordinarily important and difficult issue. It compels us to address questions to which there are no easy or simple answers, at law or otherwise. It requires us to confront the most basic of human concerns — the mortality of self and loved ones — and to balance the interest in preserving human life against the desire to die peacefully and with dignity. People of good will can and do passionately disagree about the proper result, perhaps even more intensely than they part ways over the constitutionality of restricting a woman’s right to have an abortion. Heated though the debate may be, we must determine whether and how the United States Constitution applies to the controversy before us, a controversy that may touch more people more profoundly than any other issue the courts will face in the foreseeable future. Today, we are required to decide whether a person who is terminally ill has a constitutionally-protected liberty interest in hastening what might otherwise be a protracted, undignified, and extremely painful death. If such an interest exists, we must next decide whether or not the state of Washington may constitutionally restrict its exercise by banning a form of medical assistance that is frequently requested by terminally ill people who wish to die. We first conclude that there is a constitutionally-protected liberty interest in determining the time and manner of one’s own death, an interest that must be weighed against the state’s legitimate and countervailing interests, especially those that relate to the preservation of human life. After balancing the competing interests, we conclude by answering the narrow question before us: We hold that insofar as the Washington statute prohibits physicians from prescribing life-ending medication for use by terminally ill, competent adults who wish to hasten their own deaths, it violates the Due Process Clause of the Fourteenth Amendment. II. Preliminary Matters and History of the Case This is the first right-to-die case that this court or any other federal court of appeals has ever decided. The plaintiffs are four physicians who treat terminally ill patients, three terminally ill patients, and a Washington non-profit organization called Compassion In Dying. The four physicians — Dr. Harold Glucksberg, Dr. Thomas A. Preston, Dr. Abigail Halperin, and Dr. Peter Shalit— are respected doctors whose expertise is recognized by the state. All declare that they periodically treat terminally ill, competent adults who wish to hasten their deaths with help from their physicians. The doctors state that in their professional judgment they should provide that help but are deterred from doing so by a Washington statute that makes it a felony to knowingly aid another person to commit suicide. Under the Washington statute, aiding a person who wishes to end his life constitutes a criminal act and subjects the aider to the possibility of a lengthy term of imprisonment, even if the recipient of the aid is a terminally ill, competent adult and the aider is a licensed physician who is providing medical assistance at the request of the patient. The Washington statute provides in pertinent part: “A person is guilty of promoting a suicide attempt when he knowingly causes or aids another person to attempt suicide.” RCW 9A.36.060 (emphasis added). A violation of the statute constitutes a felony punishable by imprisonment for a maximum of five years and a fine of up to $10,000. RCW 9A.36.060(2) and 9A.20.020(l)(c). On appeal, the four plaintiff-doctors asserted the rights of terminally ill, competent adult patients who wished to hasten their deaths with the help of their physicians so that they might die peacefully and with dignity. That group included the three patient-plaintiffs. The district court described the patient-plaintiffs, each of whom desired to obtain prescription drugs to hasten his death, as follows: Jane Roe is a 69-year-old retired pediatrician who has suffered since 1988 from cancer which has now metastasized throughout her skeleton. Although she tried and benefitted temporarily from various treatments including chemotherapy and radiation, she is now in the terminal phase of her disease. In November 1993, her doctor referred her to hospice care. Only patients with a life expectancy of less than six months are eligible for such care. Jane Roe has been almost completely bedridden since June of 1993 and experiences constant pain, which becomes especially sharp and severe when she moves. The only medical treatment available to her at this time is medication, which cannot fully alleviate her pain. In addition, she suffers from swollen legs, bed sores, poor appetite, nausea and vomiting, impaired vision, incontinence of bowel, and general weakness. Jane Roe is mentally competent and wishes to hasten her death by taking prescribed drugs with the help of Plaintiff Compassion in Dying. In keeping with the requirements of that organization, she has made three requests for its members to provide her and her family with counseling, emotional support, and any necessary ancillary drug assistance at the time she takes the drugs. John Doe is a 44 — year-old artist dying of AIDS. Since his diagnosis in 1991, he has experienced two bouts of pneumonia, chronic, severe skin and sinus infections, grand mal seizures and extreme fatigue. He has already lost 70% of his vision to cytomegalovirus retinitis, a degenerative disease which will result in blindness and rob him of his ability to paint. His doctor has indicated that he is in the terminal phase of his illness. John Doe is 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 in June of 1991. He also observed his grandfather’s death from diabetes preceded by multiple amputations as well as loss of vision and hearing. Mr. Doe is mentally competent, understands there is no cure for AIDS and wants his physician to prescribe drugs which he can use to hasten his death. James Poe is a 69-year-old retired sales representative who suffers from emphysema, which causes him a constant sensation of suffocating. He is connected to an oxygen tank at all times, and takes morphine regularly to calm the panic reaction associated with his feeling of suffocation. Mr. Poe also suffers from heart failure related to his pulmonary disease which obstructs the flow of blood to his extremities and causes severe leg pain. There are no cures for his pulmonary and cardiac conditions, and he is in the terminal phase of his illness. Mr. Poe is mentally competent and wishes to commit suicide by taking physician-prescribed drugs. Compassion In Dying, 850 F.Supp. at 1456-57. The names of the patients are pseudonymous in order to protect their privacy. All three patients died after the case began. Two had died by the time the District Court issued its decision. See Compassion In Dying v. State of Washington, 850 F.Supp. 1454, 1456 n. 2 (W.D.Wash.1994). The other died prior to the date of the decision by the three-judge panel of this court. Compassion In Dying v. State of Washington, 49 F.3d 586, 588 (9th Cir.1995). Since the District Court properly granted the physicians standing to assert the rights of their terminally ill patients in general, 850 F.Supp. at 1467, it is clear that this case was not rendered moot by the death of the three named patients. The physicians meet both Article III and jurisprudential standing requirements. See Singleton v. Wulff, 428 U.S. 106, 116-17, 96 S.Ct. 2868, 2875-76, 49 L.Ed.2d 826 (1976) (holding that doctors had standing to challenge — on behalf of women patients in general — a Missouri law banning Medicaid reimbursement for abortions that were not medically required). See also Doe v. Bolton, 410 U.S. 179, 188, 93 S.Ct. 739, 745, 35 L.Ed.2d 201 (1973) (holding that physicians, asserting the rights of their patients, have standing to challenge the constitutionality of a criminal abortion statute even though “the record does not disclose that any one of them has been prosecuted, or threatened with prosecution, for violation of the State’s abortion statutes”); Planned Parenthood of Cent. Mo. v. Danforth, 428 U.S. 52, 62, 96 S.Ct. 2831, 2837, 49 L.Ed.2d 788 (1976) (same). Athough there is some ambiguity in Bolton as to whether the physicians were asserting their own rights or the rights of their patients, the Court in Singleton, after discussing Griswold v. Connecticut, 381 U.S. 479, 85 S.Ct. 1678, 14 L.Ed.2d 510 (1965), described Bolton as a case “where the Court also permitted physicians to assert the rights of their patients.” The doctors in Bolton were held to have standing to assert their patients’ rights even though the doctors had never been threatened with prosecution. The doctors here also meet the standing requirements because they run a severe risk of prosecution under the Washington statute, which proscribes the very conduct in which they seek to engage. The state has never indicated that it would not prosecute doctors who violate that law. See Babbitt v. United Farm Workers National Union, 442 U.S. 289, 99 S.Ct. 2301, 60 L.Ed.2d 895 (1979) (holding that plaintiff does not have to risk arrest or prosecution in order to have standing to challenge the constitutionality of a criminal statute). See also Planned Parenthood of Cent. Mo., 428 U.S. at 62, 96 S.Ct. at 2837; Bolton, 410 U.S. at 188, 93 S.Ct. at 745 (saying that the “physician is the one against whom these criminal statutes directly operate” and that the “physician-appellants, therefore, assert a sufficiently direct threat of personal detriment. ... [and] should not be required to await and undergo a criminal prosecution as the sole means of seeking relief’). We need not decide whether, the deaths of the three patient-plaintiffs would negate the ability of their lawyers to continue the challenge that those patients brought while they were still alive. See Southern Pacific Terminal Co. v. ICC, 219 U.S. 498, 515, 31 S.Ct. 279, 283, 55 L.Ed. 310 (1911) (holding a case is not moot when the controversy is capable of repetition yet evading review). We note, however, that in invoking the capable-of-repetition-yet-evading-review doctrine in Roe v. Wade, 410 U.S. 113, 93 S.Ct. 705, 35 L.Ed.2d 147 (1973), the Court specifically relied, in part, on the fact that other women would become pregnant. The Court said; [W]hen, as here, pregnancy is a significant fact in the litigation, the normal 266-day human gestation period is so short that the pregnancy will come to term before the usual appellate procedure is complete. If that termination makes a case moot, pregnancy litigation will seldom survive much beyond the trial stage, and appellate review will be effectively denied. Our law should not be that rigid. Pregnancy often comes more than once to the same woman, and in the general population, if man is to survive, it will always be with us. Roe, 410 U.S. at 125, 93 S.Ct. at 713 (emphasis added). So, too, unfortunately, will terminal illness. The District Court in this case reached only claims asserted by two of the three categories of plaintiffs: the patients’ claims that they had a right to receive medical assistance from their physicians and the claims that the physicians asserted on behalf of their patients. It did not address the claim asserted by Compassion In Dying. Nor, correlatively, did it reach the claim by the terminally ill patients that they had a right to receive assistance from organizations such as Compassion In Dying. Like the District Court, we decide only claims brought by the terminally ill patients and the doctors. We consider those claims to the extent that they relate to the provision of certain medical assistance to terminally ill persons by physicians or persons acting pursuant to their authorization or direction. The claims involving Compassion In Dying are not before us. The district court suggested that it would reach those additional claims at a later stage in the proceedings if Compassion In Dying so desired. We have jurisdiction over this appeal from partial summary judgment because the district court certified the appeal at the request of both parties under Federal Rule of Civil Procedure 54(b). The plaintiffs do not challenge Washington statute RCW 9A.36.060 in its entirety. Specifically they do not object to the portion of the Washington statute that makes it unlawful for a person knowingly to cause another to commit suicide. Rather, they only challenge the statute’s “or aids ” provision. They challenge that provision both on its face and as applied to terminally ill, mentally competent adults who wish to hasten their own deaths with the help of medication prescribed by their doctors. The plaintiffs contend that the provision impermissibly prevents the exercise by terminally ill patients of a constitutionally-protected liberty interest in violation of the Due Process Clause of the Fourteenth Amendment, and also that it im-permissibly distinguishes between similarly situated terminally ill patients in violation of the Equal Protection Clause. In an extremely thoughtful opinion, Chief District Judge Barbara Rothstein held that “a competent, terminally ill adult has a constitutionally guaranteed right under the Fourteenth Amendment to commit physician-assisted suicide.” 850 F.Supp. at 1462. Ruling on cross-motions for summary judgment, the District Court concluded that the Washington statute places an undue burden on the exercise of that constitutionally-protected liberty interest. Id. at 1465. The District Court held that the Washington law also violates the Equal Protection Clause because it impermissibly treats similarly situated groups of terminally ill patients differently. Id. at 1467. Although the scope of the relief the District Judge ordered is not clear, id. at 1456, 1459, 1462-1464, 1467, it appears that she declared the statute invalid only insofar as it applies to the prescription of medication to terminally ill competent adults who wish to hasten their deaths — or, to use the district court’s precise terminology, only insofar as it applies to “physician-assisted suicide,” id. at 1467. On appeal, a three-judge panel of this court voted 2-1 to reverse the district court decision. Compassion In Dying v. State of Washington, 49 F.3d 586 (9th Cir.1995). The majority held that there is no due process liberty interest in physician-assisted suicide. It also concluded that the Washington statute does not violate the Equal Protection Clause. Accordingly, the majority held that the statute is not invalid facially or as applied. Judge Wright dissented and would have held that the statute is invalid as applied to terminally ill, mentally competent adults because it violates their privacy and equal protection rights. Id. at 594, 597 (Wright, J., dissenting). Because of the extraordinary importance of this case, we decided to rehear it en banc. Compassion In Dying v. State of Wash., 62 F.3d 299 (9th Cir.1995). We now affirm the District Court’s decision and clarify the scope of the relief. We hold that the “or aids” provision of Washington statute RCW 9A.36.060, as applied to the prescription of life-ending medication for use by terminally ill, competent adult patients who wish to hasten their deaths, violates the Due Process Clause of the Fourteenth Amendment. Accordingly, we need not resolve the question whether that provision, in conjunction with other Washington laws regulating the treatment of terminally ill patients, also violates the Equal Protection Clause. III. Overview of Legal Analysis: Is There a Due Process Violation? In order to answer the question whether the Washington statute violates the Due Process Clause insofar as it prohibits the provision of certain medical assistance to terminally ill, competent adults who wish to hasten their own deaths, we first determine whether there is a liberty interest in choosing the time and manner of one’s death — a question sometimes phrased in common parlance as: Is there a right to die? Because we hold that there is, we must then determine whether prohibiting physicians from prescribing life-ending medication for use by terminally ill patients who wish to die violates the patients’ due process rights. The mere recognition of a liberty interest does not mean that a state may not prohibit the exercise of that interest in particular circumstances, nor does it mean that a state may not adopt appropriate regulations governing its exercise. Rather, in cases like the one before us, the courts must apply a balancing test under which we weigh the individual’s liberty interests against the relevant state interests in order to determine whether the state’s actions are constitutionally permissible. As Chief Justice Rehnquist, writing for the Court, explained in Cruzan v. Director, Missouri Dept. of Health, 497 U.S. 261, 110 S.Ct. 2841, 111 L.Ed.2d 224 (1990), the only right-to-die case that the Court has heretofore considered: [D]etermining that a person has a “liberty interest” under the Due Process Clause does not end our inquiry; “whether respondent’s constitutional rights have been violated must be determined by balancing his liberty interests against the relevant state interests.” Youngberg v. Romeo, 457 U.S. 307, 321, 102 S.Ct. 2452, 2461, 73 L.Ed.2d 28 (1982); See also Mills v. Rogers, 457 U.S. 291, 299, 102 S.Ct. 2442, 2448, 73 L.Ed.2d 16 (1982). Cruzan, 497 U.S. at 279, 110 S.Ct. at 2851-52 (footnote omitted). The Court has invoked a balancing test in a number of substantive due process eases, not just in the right-to-die context. For example, as the Cruzan Court noted, the Court applied a balancing test in Youngberg and Mills, liberty interest cases involving the right to refuse medical treatment. Youngberg addressed the rights of patients involuntarily committed to state mental institutions. The Court said: “In determining whether a substantive right protected by the Due Process Clause has been violated, it is necessary to balance the liberty of the individual and the demands of organized society.” Youngberg, 457 U.S. at 320, 102 S.Ct. at 2460 (internal citation and quotation omitted). Mills addressed the question of the right of mental patients to refuse treatment with an-tipsychotic drugs. There, the Court stated explicitly that the “state interests” are “to be balanced against an individual’s liberty interests.” 457 U.S. at 304, 102 S.Ct. at 2451. As the Cruzan Court also noted, the use of a balancing test is deeply rooted in our legal traditions. The Court has been applying a balancing test in substantive due process cases at least since 1905, when in Jacobson v. Massachusetts, 197 U.S. 11, 25 S.Ct. 358, 49 L.Ed. 643 (1905), “the Court balanced an individual’s liberty interest in declining an unwanted smallpox vaccine against the State’s interest in preventing disease.” Cruzan, 497 U.S. at 278, 110 S.Ct. at 2851. As Justice O’Connor explained in her concurring opinion in Cruzan, the ultimate question is whether sufficient justification exists for the intrusion by the government into the realm of a person’s “liberty, dignity, and freedom.” Cruzan, 497 U.S. at 287, 289, 110 S.Ct. at 2856, 2857 (O’Connor, J., concurring). If the balance favors the state, then the given statute — whether it regulates the exercise of a due process liberty interest or prohibits that exercise to some degree — is constitutional. If the balance favors the individual, then the statute — whatever its justifications — violates the individual’s due process liberty rights and must be declared unconstitutional, either on its face or as applied. Here, we conclude unhesitatingly that the balance favors the individual’s liberty interest. IY. Is There a Liberty Interest? Before beginning our inquiry into whether a liberty interest exists, we reiterate a few fundamental precepts that guide us. The first lies in the Court’s cautionary note in Roe v. Wade, 410 U.S. 113, 116, 93 S.Ct. 705, 708, 35 L.Ed.2d 147 (1973): We forthwith acknowledge our awareness of the sensitive and emotional nature of the ... controversy, of the vigorous opposing views, even among physicians, and of the deep and seemingly absolute convictions that the subject inspires. One’s philosophy, one’s experiences, one’s exposure to the raw edges of human existence, one’s religious training, one’s attitude toward life and family and their values, and the moral standards one establishes and seeks to observe, are all likely to influence and to color one’s thinking and conclusions.... Like the Roe Court, we endeavor to conduct an objective analysis of a most emotionally-charged of topics. In doing so, we bear in mind the second Justice Harlan’s admonition in his now-vindicated dissent in Poe v. Ullman, 367 U.S. 497, 543, 81 S.Ct. 1752, 1776-77, 6 L.Ed.2d 989 (1961) (Harlan, J., dissenting from dismissal on jurisdictional grounds): [T]he full scope of the liberty guaranteed by the Due Process Clause cannot be found in or limited by the precise terms of the specific guarantees elsewhere in the Constitution. This ‘liberty* is not a series of isolated points pricked out in terms of the taking of property; the freedom of speech, press, and religion; the right to keep and bear arms; the freedom from unreasonable searches and seizures; and so on. It is a rational continuum which, broadly speaking, includes a freedom from all substantial arbitrary impositions and purposeless restraints, ... and which also recognizes, what a reasonable and sensitive judgment must, that certain interests require particularly careful scrutiny of the state needs asserted to justify their abridgment. Applying Justice Harlan’s teaching, we must strive to resist the natural judicial impulse to limit our vision to that which can plainly be observed on the face of the document before us, or even that which we have previously had the wisdom to recognize. Most important, we undertake our difficult task with a profound respect for the noble objectives of the Constitution, as described by Justice Brandéis in the second most famous dissent in American jurisprudence. In Olmstead, v. United States, 277 U.S. 438, 48 S.Ct. 564, 72 L.Ed. 944 (1928), Justice Brandéis wrote, and his words have since been quoted in full in several opinions of the Court and in innumerable appellate court decisions: The makers of our Constitution undertook to secure conditions favorable to the pursuit of happiness. They recognized the significance of man’s spiritual nature, of his feelings and of his intellect. They knew that only a part of the pain, pleasure and satisfaction of life are to be found in material things. They sought to protect Americans in their beliefs, their thoughts, their emotions and their sensations. They conferred, as against the government, the right to be let alone — the most comprehensive of rights, and the right most valued by civilized men. Id. at 478, 48 S.Ct. at 572 (Brandéis, J., dissenting). In examining whether a liberty interest exists in determining the time and manner of one’s death, we begin with the compelling similarities between right-to-die cases and abortion cases. In the former as in the latter, the relative strength of the competing interests changes as physical, medical, or related circumstances vary. In right-to-die cases the outcome of the balancing test may differ at different points along the life cycle as a person’s physical or medical condition deteriorates, just as in abortion cases the permissibility of restrictive state legislation may vary with the progression of the pregnancy. Equally important, both types of eases raise issues of life and death, and both arouse similar religious and moral concerns, Both also present basic questions about an individual’s right of choice. Historical evidence shows that both abortion and assisted suicide were for many years condemned, but that the efforts to prevent people from engaging in the condemned conduct were always at most only partially successful. Even when prohibited, abortions and assisted-suicides flourished in back alleys, in small street-side clinics, and in the privacy of the bedroom. Deprived of the right to medical assistance, many pregnant women and terminally ill adults ultimately took matters into their own hands, often with tragic consequences. Because they present issues of such profound spiritual importance and because they so deeply affect individuals’ right to determine their own destiny, the abortion and right-to-die eases have given rise to a highly emotional and divisive debate. In many respects, the legal arguments on both sides are similar, as are the constitutional principles at issue. In deciding right-to-die eases, we are guided by the Court’s approach to the abortion cases. Casey in particular provides a powerful precedent, for in that ease the Court had the opportunity to evaluate its past decisions and to determine whether to adhere to its original judgment. Although Casey was influenced by the doctrine of stare decisis, the fundamental message of that case lies in its statements regarding the type of issue that confronts us here: “These 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.” Casey, 505 U.S. at 851, 112 S.Ct. at 2807. A. Defining the Liberty Interest and Other Relevant Terms The majority opinion of the three-judge panel that first heard this ease on appeal defined the claimed liberty interest as a “constitutional right to aid in killing oneself.” Compassion In Dying, 49 F.3d at 591 (emphasis added). However, the subject we must initially examine is not nearly so limited. Properly analyzed, the first issue to be resolved is whether there is a liberty interest in determining the time and manner of one’s death. We do not ask simply whether there is a liberty interest in receiving “aid in killing oneself’ because such a narrow interest could not exist in the absence of a broader and more important underlying interest — the right to die. In short, it is the end and not the means that defines the liberty interest. The broader approach we employ in defining the liberty interest is identical to the approach used by the Supreme Court in the abortion cases. In those eases, the Court initially determined whether a general liberty interest existed (an interest in having an abortion), not whether there was an interest in implementing that general liberty interest by a particular means (with medical assistance). Specifically, in Roe v. Wade, 410 U.S. 113, 93 S.Ct. 705, 35 L.Ed.2d 147 (1973), the Court determined that women had a liberty interest in securing an abortion, not that women had a liberty interest in obtaining medical assistance for purpose of an abortion. The Court did so even though the Texas statute at issue did not prohibit a woman from inducing her own abortion; nor did it criminalize a woman’s conduct in securing an abortion. Rather, the Texas statute, like the Washington statute here, prohibited the rendering of assistance; specifically, the Texas statute prohibited only assisting a woman to secure an abortion. Roe, 410 U.S. at 151-52, 93 S.Ct. at 725-26. The Court first determined that a woman had a constitutional right to choose an abortion. Only after it did so, did it proceed to the second step: to determine whether the state’s prohibition on assistance unconstitutionally restricted the exercise of that liberty interest. Similarly, in Planned Parenthood v. Casey, 505 U.S. 833, 112 S.Ct. 2791, 120 L.Ed.2d 674 (1992), the Court first reaffirmed, after extensive analysis, its earlier holding that women have a liberty interest in obtaining an abortion. In determining the existence of that liberty interest, the Court did not address the subject of spousal notification. As in Roe, only after affirming a woman’s right to have an abortion, did the Court proceed to the second step: to examine whether the statutory provision requiring married women to notify their spouses prior to obtaining an abortion posed an undue burden on the exercise of that liberty interest. In this case, our analysis is necessarily the same. First we must determine whether there is a liberty interest in determining the time and manner of one’s death; if so, we must then examine whether Washington’s ban on assisted suicide unconstitutionally restricts the exercise of that liberty interest. While some people refer to the liberty interest implicated in right-to-die cases as a liberty interest in committing suicide, we do not describe it that way. We use the broader and more accurate terms, “the right to die,” “determining the time and manner of one’s death,” and “hastening one’s death” for an important reason. The liberty interest we examine encompasses a whole range of acts that are generally not considered to constitute “suicide.” Included within the liberty interest we examine, is for example, the act of refusing or terminating unwanted medical treatment. As we discuss later at pp. 821-822, a competent adult has a liberty interest in refusing to be connected to a respirator or in being disconnected from one, even if he is terminally ill and cannot live without mechanical assistance. The law does not classify the death of a patient that results from the granting of his wish to decline or discontinue treatment as “suicide.” Nor does the law label the acts of those who help the patient carry out that wish, whether by physically disconnecting the respirator or by removing an intravenous tube, as assistance in suicide. Accordingly, we believe that the broader terms — “the right to die,” “controlling the time and manner of one’s death,” and “hastening one’s death” — more accurately describe the liberty interest at issue here. Moreover, as we discuss later, we have serious doubts that the terms “suicide” and “assisted suicide” are appropriate legal descriptions of the specific conduct at issue here. See infra 824. There is one further definitional matter we should emphasize. Following our determination regarding the existence of a liberty interest in hastening one’s death, we examine whether the Washington statute unconstitutionally infringes on that liberty interest. Throughout that examination, we use the term “physician-assisted suicide,” a term that does not appear in the Washington statute but is frequently employed in legal and medical discussions involving the type of question before us. For purposes of this opinion, we use physician-assisted suicide as it is used by the parties and district court and as it is most frequently used: the prescribing of medication by a physician for the purpose of enabling a patient to end his life. It is only that conduct that the plaintiffs urge be held constitutionally-protected in this case. B. The Legal Standard There is no litmus test for courts to apply when deciding whether or not a liberty interest exists under the Due Process Clause. Our decisions involve difficult judgments regarding the conscience, traditions, and fundamental tenets of our nation. We must sometimes apply those basic principles in light of changing values based on shared experience. Other times we must apply them to new problems arising out of the development and use of new technologies. In all cases, our analysis of the applicability of the protections of the Constitution must be made in light of existing circumstances as well as our historic traditions. Historically, the Court has classified “fundamental rights” as those that are “implicit in the concept of ordered liberty,” Palko v. Connecticut, 302 U.S. 319, 325-26, 58 S.Ct. 149, 151-52, 82 L.Ed. 288 (1937). The Court reasserted this historic standard, along with an alternative description, in its highly controversial Bowers v. Hardwick opinion, 478 U.S. 186, 191-92, 106 S.Ct. 2841, 2844-45, 92 L.Ed.2d 140 (1986): Striving to assure itself and the public that announcing rights not readily identifiable in the Constitution’s text involves much more than the imposition of the Justices’ own choice of values on the States and the Federal Government, the Court has sought to identify the nature of the rights qualifying for heightened judicial protection. In Palko v. Connecticut, 302 U.S. 319, 325, 326, 82 L.Ed. 288, 58 S.Ct. 149 [152] (1937), it was said that this category includes those fundamental liberties that are “implicit in the concept of ordered liberty,” such that “neither liberty nor justice would exist if [they] were sacrificed.” A different description of fundamental liberties appeared in Moore v. East Cleveland, 431 U.S. 494, 503, 97 S.Ct. 1932, 1937, 52 L.Ed.2d 531 (1977) (opinion of POWELL, J.), where they are characterized as those liberties that are “deeply rooted in this Nation’s history and tradition.” Id. at 503, 97 S.Ct. at 1938 (POWELL, J.). In recent years, the Court has spoken more frequently of substantive due process interests than of fundamental due process rights. Compare Thornburgh v. American Coll. of Obst., 476 U.S. 747, 772, 106 S.Ct. 2169, 2184, 90 L.Ed.2d 779 (1986) (describing “fundamental right” to abortion) and Akron v. Akron Center for Reproductive Health, Inc., 462 U.S. 416, 420 n. 1, 103 S.Ct. 2481, 2487 n. 1, 76 L.Ed.2d 687 (1983) (same) with Webster v. Reproductive Health Services, 492 U.S. 490, 520, 109 S.Ct. 3040, 3057, 106 L.Ed.2d 410 (1989) (plurality opinion) (describing women’s entitlement to an abortion as a “liberty interest protected by Due Process Clause”). See also Cruzan, 497 U.S. 261, 110 S.Ct. 2841. The Court has also recently expressed a strong reluctance to find new fundamental rights. Collins v. City of Harker Heights, Tex., 503 U.S. 115, 123, 112 S.Ct. 1061, 1068, 117 L.Ed.2d 261 (1992). The Court’s evolving doctrinal approach to substantive due process claims is consistent with the basic truth enunciated by Justice Harlan and later endorsed by the Court in Casey: “the full scope of the liberty guaranteed by the Due Process Clause is a rational continuum which, broadly speaking, includes a freedom from all substantial arbitrary impositions and purposeless restraints ...” Casey, 505 U.S. at 848, 112 S.Ct. at 2806, citing Poe v. Ullman, 367 U.S. 497, 543, 81 S.Ct. 1752, 1776-77, 6 L.Ed.2d 989 (1961) (Harlan, J., dissenting from dismissal on jurisdictional grounds). As Justice Harlan noted, some liberty interests are weightier than others. Under the Court’s traditional jurisprudence, those classified as fundamental rights cannot be limited except to further a compelling and narrowly tailored state interest. See Collins, 503 U.S. at 123, 112 S.Ct. at 1068. Other important interests, such as the liberty interest in refusing unwanted medical treatment, are subject to a balancing test that is less restrictive, but nonetheless requires the state to overcome a substantial hurdle in justifying any significant impairment. Recent cases, including Cruzan, suggest that the Court may be heading towards the formal adoption of the continuum approach, along with a balancing test, in substantive due process cases generally. If so, there would no longer be a two-tier or three-tier set of tests that depends on the classification of the right or interest as fundamental, important, or marginal. Instead, the more important the individual’s right or interest, the more persuasive the justifications for infringement would have to be. We see the evolution in the Court’s approach more as a recognition of the artificiality of the current classification system than as a fundamental change in the Court’s practical approach to specific issues. So long as the liberty interest is an important one, the state must shoulder the burden of justifying any significant limitations it seeks to impose. However, we need not predict the Court’s future course in order to decide the case before us. Here, as we have said, even under the Court’s traditional mode of analysis, a balancing test is applicable. Nothing in Reno v. Flores, 507 U.S. 292, 113 S.Ct. 1439, 123 L.Ed.2d 1 (1993), the insubstantial reed on which the dissent rests its ease — even though the case was not cited by any of the parties or any of the eleven amici who filed briefs before this court— suggests anything to the contrary. In Flores, the Court simply declined to find a new fundamental right, and repeated its general reluctance to do so. Id. at 302, 113 S.Ct. at 1447. The Court did not, as the dissent implies, purport to establish a new classification system under which all liberty interests other than fundamental rights would be subject to rational basis review. Nor did Flores purport to overrule, or even hint at any desire to modify, the Court’s ninety-year-old practice of using a balancing test in liberty interest cases that raise important issues of the type before us. In fact, Flores did not mention Cruzan, Youngberg, Mills, Jacobson, or any other balancing case. While one might legitimately argue either that the liberty interest at issue here rises to the level of a fundamental right or that it is simply an important liberty interest that is subject to a balancing test, one point is absolutely clear: there can be no legitimate argument that rational basis review is applicable, and nothing in Flores suggests that it is. Although in determining the existence of important rights or liberty interests, the Court examines our history and experience, it has stated on a number of occasions that the limits of the substantive reach of the Due Process Clause are not frozen at any point in time. In Casey, the Court said: “Neither the Bill of Rights nor the specific practices of States at the time of the adoption of the Fourteenth Amendment marks the outer limits of the substantive sphere of liberty which the Fourteenth Amendment protects.” 505 U.S. at 848, 112 S.Ct. at 2805. Justice Frankfurter may have put it best when, writing for the Court in Rochin v. California, 342 U.S. 165, 171-72, 72 S.Ct. 205, 209, 96 L.Ed. 183 (1952), he declared, “To believe that this judicial exercise of judgment could be avoided by freezing ‘due process of law at some fixed stage in time or thought is to suggest that the most important aspect of constitutional adjudication is a function for inanimate machines and not for judges.... ” Certainly, it would be difficult to imagine a more felicitous expression of the dynamism of constitutional interpretation. Thus, while historical analysis plays a useful role in any attempt to determine whether a claimed right or liberty interest exists, earlier legislative or judicial recognition of the right or interest is not a sine qua non. In Casey, the Court made it clear that the fact that we have previously failed to acknowledge the existence of a particular liberty interest or even that we have previously prohibited its exercise is no barrier to recognizing its existence. In discussing a woman’s liberty interest in securing an abortion, the Casey Court stated that pregnancy involves “suffering [that] is too intimate and personal for the State to insist, without more, upon its own vision of the woman’s role, however dominant that vision has been in the course of our history and culture. ” Casey, 505 U.S. at 852, 112 S.Ct. at 2807 (emphasis added). In contrast to Casey, the majority opinion of the three-judge panel in the case now before us erroneously concluded that a historical analysis alone is sufficient basis for rejecting plaintiffs’ claim to a substantive liberty interest or right. Compassion In Dying, 49 F.3d at 591. As explained below, we believe that the panel’s historical account is misguided, but even if it were indisputably correct, historical evidence alone is not a sufficient basis for rejecting a claimed liberty interest. Were history our sole guide, the Virginia anti-miscegenation statute that the Court unanimously overturned in Loving v. Virginia, 388 U.S. 1, 87 S.Ct. 1817, 18 L.Ed.2d 1010 (1967), as violative of substantive due process and the Equal Protection Clause, would still be in force because such anti-miscegenation laws were commonplace both when the United States was founded and when the Fourteenth Amendment was adopted. The Court explicitly acknowledged as much in Casey, 505 U.S. at 847, 112 S.Ct. at 2805, in rejecting the view that substantive due process protects rights or liberties only if they possess a historical pedigree. In Casey, the Court said: It is ... tempting ... to suppose that the Due Process Clause protects only those practices, defined at the most specific level, that were protected against government interference by other rules of law when the Fourteenth Amendment was ratified _ But such a view would be inconsistent with our law. It is a promise of the Constitution that there is a realm of personal liberty which the government may not enter. We have vindicated this principle before. Marriage is mentioned nowhere in the Bill of Rights and interracial marriage was illegal in most States in the 19th century, but the Court was no doubt correct in finding it to be an aspect of liberty protected against state interference by the substantive component of the Due Process Clause in Loving v. Virginia, 388 U.S. 1, 12, 87 S.Ct. 1817, 1824, 18 L.Ed.2d 1010 (1967), (relying, in an opinion for eight Justices, on the Due Process Clause). Similar examples may be found in Turner v. Safley, 482 U.S. 78, 94-99, 107 S.Ct. 2254, 2265-67, 96 L.Ed.2d 64 (1987) [holding that prisoners have a constitutionally protected right to marry a civilian or other inmate]; in Carey v. Population Services International, 431 U.S. 678, 684, 686, 97 S.Ct. 2010, 2015-2017, 52 L.Ed.2d 675 (1977) [holding that the state cannot prohibit the sale of contraceptives to all minors or bar everyone but licensed pharmacists from selling contraceptives to adults]; in Griswold v. Connecticut, 381 U.S. 479, 481-82, 85 S.Ct. 1678, 1680-81, 14 L.Ed.2d 510 (1965) [holding that a Connecticut law forbidding the use of contraceptives uneon-stitutionally intrudes on the right of marital privacy] ... Casey, 505 U.S. at 847-48, 112 S.Ct. at 2805. Indeed, if historical evidence of accepted practices at the time the Fourteenth Amendment was enacted were dispositive, the Court would not only have decided Loving differently, but it would not have held that women have a right to have an abortion. As the dissent pointed out in Roe, more than three-quarters of the existing states (at least 28 out of 37 states), as well as eight territorial legislatures restricted or prohibited abortions in 1868 when the Fourteenth Amendment was ádopted. Roe, 410 U.S. at 175-76 & n. 1, 93 S.Ct. at 737-39 & n. 1 (Rehnquist, J., dissenting). C. Historical Attitudes Toward Suicide The majority opinion of the three-judge panel claimed that “a constitutional right to aid in hilling oneself’ was “unknown to the past.” Compassion In Dying, 49 F.3d at 591. As we have pointed out at p. 803, our inquiry is not so narrow. Nor is our conclusion so facile. The relevant historical record is far more checkered than the majority would have us believe. Like the Court in Roe, we begin with ancient attitudes. In Greek and Roman times, far from being universally prohibited, suicide was often considered commendable in literature, mythology, and practice. The first of all literary suicides, that of Oedipus’ mother, Jocasta, is made to seem praiseworthy, an honorable way out of an insufferable situation. Homer records self-murder without comment, as something natural and heroic. The legends bear him out. Aegeus threw himself into the sea — which therefore bore his name— when he mistakenly thought his son Theseus had been slain by the Minotaur. A. Alvarez, The Background, in Suicide: The Philosophical Issues 18 (M. Pabst Battin and David J. Mayor, eds. 1980). In Athens, as well as the Greek colonies of Marseilles and Ceos, magistrates kept a supply of hemlock for those who wished to end their lives. The magistrates even supplied those who wished to commit suicide with the means to do so. Whoever no longer wishes to live shall state his reasons to the Senate, and after having received permission shall abandon life. If your existence is hateful to you, die; if you are overwhelmed by fate, drink the hemlock. If you are bowed with grief, abandon life. Let the unhappy man recount his misfortune, let the magistrate supply him with the remedy, and his wretchedness will come to an end. While Socrates counseled his disciples against committing suicide, he willingly drank the hemlock as he was condemned to do, and his example inspired others to end their lives. Id. at 19. Plato, Socrates’ most distinguished student, believed suicide was often justifiable. He suggested that if life itself became immoderate, then suicide became a rational, justifiable act. Painful disease, or intolerable constraint were sufficient reasons to depart. And this when religious superstitions faded was philosophic justification enough. Id. Many contemporaries of Plato were even more inclined to find suicide a legitimate and acceptable act. In Roe, while surveying the attitudes of the Greeks toward abortion, the Court stated that “only the Pythagorean school of philosophers frowned on the related act of suicide,” 410 U.S. at 131, 93 S.Ct. at 716; it then noted that the Pythagorean school represented a distinctly minority view. Id. The Stoics glorified suicide as an act of pure rational will. Cato, who killed himself to avoid dishonor when Caesar crushed his military aspirations, was the most celebrated of the many suicides among the Stoics. Montaigne wrote of Cato: “This was a man chosen by nature to show the heights which can be attained by human steadfastness and constancy- Such courage is above philosophy.” Like the Greeks, the Romans often considered suicide to be acceptable or even laudable. To live nobly also meant to die nobly and at the light time. Everything depended on a dominant will and a rational choice. This attitude was reinforced by Roman law.... According to Justinian’s Digest, suicide of a private eitizen was not punishable if it was caused by “impatience of pain or sickness, or by another cause,” or by “weariness of life ... lunacy, or fear of dishonor.” Since this covered every rational cause, all that was left was the utterly irrational suicide “without cause,” and that was punishable on the grounds that “whoever does not spare himself would much less spare another.” In other words, it was punished because irrational, not because it was a crime. Id. at 22-23. The Romans did sometimes punish suicide. Under Roman law, people convicted of crimes forfeited their property to the Emper- or, thereby disinheriting their heirs. Roman law imposed a special penalty on people who were caught committing a crime and then committed suicide prior to conviction to avoid forfeiting the property. To protect the Emperor’s interests, the property of people who committed suicide under such circumstances was forfeited, just as if they had been eon-victed of the crime involved. Marzen at 57-58. The early Christians saw death as an escape from the tribulations of a fallen existence and as the doorway to heaven, “In other words, the more powerfully the Church instilled in believers the idea that this world was a vale of tears and sin and temptation, where they waited uneasily until death released them into eternal glory, the more irresistible the temptation to suicide became.” Id. at 25. The Christian impulse to martyrdom reached its height with the Dona-tists, who were so eager to enter into martyrdom that they were eventually declared heretics. Gibbon, in the Decline and Fall of the Roman Empire, described them this way: They sometimes forced their way into courts- of justice and compelled the aff-righted judge to give orders for their execution. They frequently stopped travellers on the public highways and obliged them to inflict the stroke of martyrdom by promise of a reward, if they consented— and by the threat of instant death, if they refused to grant so singular a favour. St. Augustine said of the Donatists, “to kill themselves out of respect for martyrdom is their daily sport.” Id. at 27. Prompted in large part by the utilitarian concern that the rage for suicide would deplete the ranks of Christians, St. Augustine argued that committing suicide was a “detestable and damnable wickedness” and was able to help turn the tide of public opinion. Id. Even staunch opponents of a constitutional right to suicide acknowledge that “there were many examples of Christian martyrs whose deaths bordered on suicide, and confusion regarding the distinction between suicide and martyrdom existed up until the time of St. Augustine (354-430 A.D.).” In 562 A.D., the Council of Braga denied funeral rites to anyone who killed himself. A little more than a century later, in 693 A.D., the Council of Toledo declared that anyone who attempted suicide should be excommunicated. Id. at 27-28. Once established, the Christian view that suicide was in all cases a sin and crime held sway for 1,000 years until philosophers, poets, and even some clergymen — Montesquieu, Voltaire, Diderot, Francis Bacon, David Hume, John Donne, Sir Thomas More, among others — began to challenge the all-encompassing nature of the dominant ideology. In his book Utopia, Sir Thomas More, who was later canonized by the Roman Catholic Church, strongly supported the right of the terminally ill to commit suicide and also expressed approval of the practice of assisting those who wished to hasten their deaths. Hume argued that a decision by a terminally ill patient to end his life was often laudable. France even enacted a statute legalizing suicide in 1790, primarily as a result of the influence of the nation’s leading philosophers. Suicide was a crime under the English common law, at least in limited circumstances, probably as early as the thirteenth century. Bracton, incorporating Roman Law as set forth in Justinian’s Digest, declared that if someone commits suicide to avoid conviction of a felony, his property escheats to his lords. Bracton said “[i]t ought to be otherwise if he kills himself through madness or unwillingness to endure suffering.” Despite his general fidelity to Roman law, Bracton did introduce a key innovation: “[I]f a man slays himself in weariness of life or because he is unwilling to endure further bodily pain ... he may have a successor, but his movable goods [personal property] are confiscated. He does not lose his inheritance [real property], only his movable goods.” Bracton’s innovation was incorporated into English common law, which has thus treated suicides resulting from the inability to “endure further bodily pain” with compassion and understanding ever since a common law scheme was firmly established. Sir Edward Coke, in his Third Institute published in 1644, held that killing oneself was an offense and that someone who committed suicide should forfeit his movable property. But Coke listed an exception for someone who “by the rage of sickness or infirmity or otherwise,” kills himself “while he is not of compos mentía,” or sound mind. In eighteenth century England, many and perhaps most juries compensated for the perceived unfairness of the law by concluding that anyone who killed himself was necessarily not of sound mind. Thus, although, formally, suicide was long considered a crime under English common law, in practice it was a crime that was punished leniently, if at all, because juries frequently used them power to nullify the law. The traditional English experience was also shaped by the taboos that have long colored our views of suicide and perhaps still do today. English common law reflected the ancient fear that the spirit of someone who ended his own life would return to haunt the living. Accordingly, the traditional practice was to bury the body at a crossroads — either so the suicide could not find his way home or so that the frequency of travelers would keep his spirit from rising. As added insurance, a stake was driven through the body. English attitudes toward suicide, including the tradition of ignominious burial, carried over to America where they subsequently underwent a transformation. By 1798, six of the 13 original colonies had abolished all penalties for suicide either by statute or state constitution. There is no evidence that any court ever imposed a punishment for suicide or attempted suicide under common law in post-revolutionary America. By the time the Fourteenth Amendment was adopted in 1868, suicide was generally not punishable, and in only nine of the 37 states is it clear that there were statutes prohibiting assisting suicide. The majority of states have not criminalized suicide or attempted suicide since the turn of the century. The New Jersey Supreme Court declared in 1901 that since suicide was not punishable it should not be considered a crime. “[A]ll will admit that in some cases it is ethically defensible,” the court said, as when a woman kills herself to escape being raped or “when a man curtails weeks or months of agony of an incurable disease.” Campbell v. Supreme Conclave Improved Order Heptasophs, 66 N.J.L. 274, 49 A. 550, 553 (1901). Today, no state has a statute prohibiting suicide or attempted suicide; nor has any state had such a statute for at least 10 years. A majority of states do, however, still have laws on the books against assisting suicide. D. Current Societal Attitudes Clearly the absence of a criminal sanction alone does not show societal approbation of a practice. Nor is there any evidence that Americans approve of suicide in general. In recent years, however, there has been increasingly widespread support for allowing the terminally ill to hasten their deaths and avoid painful, undignified, and inhumane endings to their lives. Most Americans simply do not appear to view such acts as constituting suicide, and there is much support in reason for that conclusion. See infra at p. 824. Polls have repeatedly shown that a large majority of Americans — sometimes nearing 90% — fully endorse recent legal changes granting terminally ill patients, and sometimes their families, the prerogative to accelerate their death by refusing or terminating treatment Other polls indicate that a majority of Americans favor doctor-assisted suicide for the terminally ill. In April, 1990, the Roper Report found that 64% of Americans believed that the terminally ill should have the right to request and receive physician aid-in-dying. Another national poll, conducted in October 1991, shows that “nearly two out of three Americans favor doctor-assisted suicide and euthanasia for terminally ill patients who request it.” A 1994 Harris poll found 73% of Americans favor legalizing physician-assisted suicide. Three states have held referenda on proposals to allow physicians to help terminally ill, competent adults commit suicide with somewhat mixed results. In Oregon, voters approved the carefully-crafted referendum by a margin of 51 to 49 percent in November of 1994. In Washington and California where the measures contained far fewer practical safeguards, they narrowly faded to pass, each drawing 46 percent of the vote. As such referenda indicate, there is unquestionably growing popular support for permitting doctors to provide assistance to terminally ill patients who wish to hasten their deaths. Just as the mere absence of criminal statutes prohibiting suicide or attempted suicide does not indicate societal approval so the mere presence of statutes criminalizing assisting in a suicide does not necessarily indicate societal disapproval. That is especially true when such laws are seldom, if ever, enforced. There is no reported American case of criminal punishment being meted out to a doctor for helping a patient hasten his own death. The lack of enforcement of statutes prohibiting assisting a mentally competent, terminally ill adult to end his own life would appear to reflect widespread societal disaffection with such laws. Our attitudes toward suicide of the type at issue in this ease are better understood in light of our unwritten history and of technological developments. Running beneath the official history of legal condemnation of physician-assisted suicide is a strong undereur-rent of a time-honored but hidden practice of physicians helping terminally ill patients to hasten their deaths. According to a survey by the American Society of Internal Medicine, one doctor in five said he had assisted in a patient’s suicide. Accounts of doctors who have helped their patients end their lives have appeared both in professional journals and in the daily press. The debate over whether terminally ill patients should have a right to reject medical treatment or to receive aid from their physicians in hastening their deaths has taken on a new prominence as a result of a number of developments. Two hundred years ago when America was founded and more than one hundred years ago when the Fourteenth Amendment was adopted, Americans died from a slew of illness and infirmities that killed their victims quickly but today are almost never fatal in this nation — scarlet fever, cholera, measles, diarrhea, influenza, pneumonia, gastritis, to name a few. Other diseases that have not been conquered can now often be controlled for years, if not decades — diseases such as diabetes, muscular dystrophy, Parkinson’s disease, cardiovascular disease, and certain types of cancer. As a result, Americans are living longer, and when they finally succumb to illness, lingering longer, either in great pain or in a stu-porous, semi-comatose condition that results from the infusion of vast amounts of pain killing medications. Despite the marvels of technology, Americans frequently die with less dignity than they did in the days when ravaging diseases typically ended their lives quickly. AIDS, which often subjects its victims to a horrifying and drawn-out demise, has also contributed to the growing number of terminally ill patients who die protracted and painful deaths. One result has been a growing movement to restore humanity and dignity to the process by which Americans die. The now recognized right to refuse or terminate treatment and the emergent right to receive medical assistance in hastening one’s death are inevitable consequences of changes in the causes of death, advances in medical science, and the development of new technologies. Both the need and the capability to assist individuals end their lives in peace and dignity have increased exponentially. E. Prior Court Decisions Next we examine previous Court decisions that delineate the boundaries of substantive due process. We believe that a careful examination of these decisions demonstrates that there is a strong liberty interest in determining how and when one’s life shall end, and that an explicit recognition of that interest follows naturally, indeed inevitably, from their reasoning. The essence of the substantive component of the Due Process Clause is to limit the ability of the state to intrude into the most important matters of our lives, at least without substantial justification. In a long line of cases, the Court has carved out certain key moments and decisions in individuals’ lives and placed them beyond the general prohibitory authority of the state. The Court has recognized that the Fourteenth Amendment affords constitutional protection to personal decisions relating to marriage, Loving v. Virginia, 388 U.S. 1, 87 S.Ct. 1817, 18 L.Ed.2d 1010 (1967), procreation, Skinner v. Oklahoma, 316 U.S. 535, 62 S.Ct. 1110, 86 L.Ed. 1655 (1942), family relationships, Prince v. Massa