TFN no.5754.14 337-364


On the Treatment of the Terminally Ill


A Jewish couple is providing care to two relatives with end-stage neurological disease.

Naomi, the couple’s 16-month-old daughter, has Canavan’s Disease, a rare progressive brain disease similar to Tay-Sachs, though the potential exists for the child to survive into teenage. She recognizes her family, smiles and laughs, but she cannot roll over, grasp objects, or hold up her head. Her vision is worsening towards blindness. She is not gaining weight. Her parents are concerned for her “quality of life”: how much discomfort and pain should she suffer, enduring the medical procedures that might be introduced, for the kind of future that she will inevitably have?

Esther is the husband’s 95-year-old grandmother and a patient at a nursing home. She has had Alzheimer’s Disease for over ten years. She can feed herself and roll over in bed but needs help with everything else. She does not recognize her family but smiles with some activities. Based upon her previously stated wishes, life-prolonging medical care will be withheld, including antibiotics, hospitalization, and tube feedings.

What approach to medical care is most appropriate for Naomi? Is it justifiable to treat her in the same manner as a 95-year-old with Alzheimer’s and withhold life-prolonging measures? Does her current “happiness” mandate some or all efforts to extend her life as long as possible? Does her future quantity or quality of life justify painful medical interventions? What differentiates these two cases? (Rabbi Norman M. Cohen, Hopkins, MN)


I. On Euthanasia and Assisted Suicide. It is undeniably difficult to speak to the situation that confronts these parents. We know that whatever counsel we can offer will be inadequate in the face of the heartbreak they endure as they watch their daughter’s deterioration toward an early, inevitable death. Yet they ask us to explore the resources of the Jewish tradition and derive from them a response to her illness that is both compassionate and ethical: compassionate in that it spares her unnecessary pain, ethical in that it meets our moral duty toward a human life that is infinitely precious in God’s sight. How might they best achieve both these goals, striking a proper balance between them?

The parents do not suggest euthanasia, or “mercy killing,” as an option for either Naomi or Esther. For that reason alone, we might be justified in ignoring the subject altogether. Yet it cannot be ignored; the issue of euthanasia and assisted suicide has lately become a hotly-debated one within our culture. In particular, some prominent Reform rabbis have proposed that we rethink our long-standing opposition to euthanasia.[1] Therefore, though it is not directly relevant to this case, we believe it to be our responsibility to examine that question, if briefly, as an essential first step in the consideration of the broader question of the treatment of the terminally ill.

Jewish tradition, as is well known, prohibits suicide, if by “suicide” we mean a rational, premeditated act of self-killing.[2] The prohibition flows from the tradition’s affirmation of the sanctity, the inviolability of human life. This affirmation, in turn, assumes the doctrine that life belongs to God, Who has the final say in its disposal.[3] This implies that the individual has no right of “ownership” over his/her life, no authority to bring that life to an illegitimately premature end. For this reason, the court may not execute a criminal on the strength of his own confession, “for the human life is not the property of man but of God.. one’s confession cannot be accepted with respect to a matter that does not lie within his power…(for) one is not entitled to commit suicide.”[4] Similarly, Jewish law prohibits euthanasia, or mercy killing. Inasmuch as human life remains sacred and inviolable until the final moment of its existence, the sources uniformly reject any distinction in this regard between the dying person (the goses) and any other. “The dying person is like a living person in all respects” (Semachot 1:1). Though he or she lies in a moribund state in which death is imminent,[5] a person is still a person, a human being created in the image of God. This life is to be treasured and protected; even though the prognosis is hopeless, he or she deserves all appropriate care. Just as the laws prohibiting work on Shabbat may be violated in order to save life (pikuach nefesh), so do we violate them on behalf of the goses. We set aside the Shabbat in order to treat this person, despite the fact that this is a life we cannot “save.”[6] The one who kills the goses is guilty of murder.[7] The dying person is compared to a flickering flame: the slightest touch will extinguish his life. It is forbidden to take any action that hastens the death of the goses; “whoever touches him commits bloodshed,”[8] even though this act is taken out of compassion, in order to relieve him of terrible pain and suffering. If such is the case with the goses, then it is surely so concerning patients such as Naomi and Esther who, though incurably ill, have not yet reached the very end of life.[9]

On the other hand, the prohibition against suicide is not absolute. One major exception to the rule is the case of martyrdom: a Jew is obligated to accept death rather than to transgress the Toraitic commandments against murder, idolatry, and sexual immorality. Thus does one fulfill the commandment of kiddush hashem, of sanctifying the Divine Name through one’s decision to die.[10] Some suggest that this requirement extends to active suicide: one who fears that the persecutors will, through excruciating torture, coerce him into violating the Torah may take his life in order to avoid committing that act.[11] Of special interest to our question is the fact that the sources look with considerable understanding upon individuals who commit suicide at moments of personal extremis. A classic case is that of King Saul, who falls upon his sword rather than suffer a degrading death at the hands of the enemy (I Samuel 31:4). Saul, writes one commentator, “committed no sin in taking his own life” under these circumstances.[12] Some authorities derive from Saul’s example that, while such an act is not “permitted,” one who takes his or her life out of a desire to escape terrible pain and degradation (`oni uvizayon) is not in fact a “suicide.” Such a person has in effect been coerced by overpowering circumstances into this most extreme measure; that action is not the “rational, premeditated act of self-killing” forbidden by halakhah.[13] This judgment coheres with the rabbinic tendency to exploit every available pretext in order to declare that a person, though he has died at his own hand, may receive all the customary rites of mourning normally denied to the suicide: “it is most unlikely that a person of sound mind would take such a horrible step.”[14]

If the tradition responds with compassion and empathy to those who commit suicide, some contemporary observers go farther. Citing as a prooftext the story of the death of Saul, as well as the talmudic narratives surrounding the deaths of Rabbi Chanina ben Teradyon and Rabbi Yehudah Hanasi, they argue that Judaism actually permits suicide and mercy killing for those who face the pain and agony of terminal illness. In doing so, however, they face an interesting problem in interpretation. On the one hand, it is certainly true that these stories might plausibly be read so as to support the option of active euthanasia. On the other hand, through the long history of the Jewish study of the Bible and the Talmud, the texts in question have not been understood in this way (see Excursus). This is a point of no little importance to our discussion. We wish to know, after all, whether the “Jewish tradition” offers evidence in support of active euthanasia. It is for this reason that advocates of mercy killing cite these stories in the first place. Yet we find that the very tradition of learning which created these passages and which has studied them for fifteen centuries and more as sources of moral meaning declares consistently and unequivocally against euthanasia. Indeed, the message which emerges from traditional halakhic thought on this subject is quite clear and uniform: we do almost anything to relieve the suffering of the terminally-ill, but we do not kill them and we do not help them kill themselves. It is always possible to read these texts differently than they have ever been read by the Jewish religious community, to discover in them levels of meaning that generations of rabbis and talmidey chakhamim may have missed. Still, the unequivocal voice of the halakhic literature renders it is most difficult to sustain an argument, based upon the citation of a few stories from the Bible and the Talmud, that the “Jewish tradition” permits euthanasia.[15]

As Reform Jews, of course, we consider ourselves free to ascribe “new” Jewish meanings to our texts, to depart from tradition when we think it necessary to secure an essential religious or moral value. In this case, though, we fail to see why we should do so.[16] We see no good reason, first of all, to abandon the traditional Jewish teaching concerning the inestimable value of human life. If the doctrine of life’s essential holiness means anything at all, it means that we must stand in reverence before the very fact of life, the gift of God that renders us human. And this reverence does not diminish as human strength declines, for the dying person still possesses life, a life stamped indelibly with the image of God until the moment of death. It is an awesome and awful responsibility we take upon ourselves when we determine to kill a human being, even when our intentions are good and merciful. Such an action is the ultimate arrogance, for it declares that we are masters over the one thing–life itself–that our faith has always taught must be protected against our all-too-human tendency to manipulate, to mutilate, and to destroy.

Second, we do not believe that the existence of pain and suffering constitutes a sufficient Jewish justification for killing a human being in the name of compassion. It is true that none of us wants to endure a state of physical or psychological agony, and none of us wants this for our loved ones. We have every right to administer treatment to relieve pain. In addition, we are under no obligation to take every conceivable measure to prolong a life of suffering; on all this, see below. It remains a fact, however, that pain and suffering are part and parcel of the human condition. We do not cease to be human, that is, when we experience suffering, even that of a terminal illness. The choice we face when we are ill is essentially the same choice we confront at every other moment of our lives: to determine what we, human beings in covenant with God, propose to do with the time and the strength available to us on this earth. All of life, its end no less than its beginning and its middle, is the arena in which we act out our humanity. Judaism, for its part, bids us to respond to the challenges of life by choosing life, to praise God whether that life brings us joy or sorrow.[17] Even in debilitating illness, when our freedom of action is severely limited, we yet sanctify the divine name by living our relationship with God, by striving toward nobility of conduct and of purpose, by confronting our suffering with courage. To say this is not to ignore the agony of the dying but to recognize a fundamental truth: that even when we are dying we have the power to choose how we shall live. We can kill ourselves, thereby accepting the counsel of despair, or we can choose life, declaring through our actions that despite everything life–all of it–is blessed with the promise of ultimate meaning and fulfillment.

Third, we are uncomfortable with arguments for assisted suicide that proceed from judgments concerning the “quality of life.” While this standard may be persuasive to many, the quality of life by its nature is virtually impossible to determine. That is to say, the decision that “my life is no longer worth living” is an inescapably subjective one; it cannot be quantified, verified, or tested against any principle other than the conviction that one’s suffering is no longer tolerable. For example, it is often suggested that the life of a patient in a protracted coma or persistent vegetative state lacks a minimal element of “quality” and that the patient is therefore justified in giving advance authorization for his or her euthanized death. Yet there is nothing inherent in such a condition that demands suicide or euthanasia; the sole “objective” warrant for mercy killing in this instance is the patient’s stated desire to die. If so, on what grounds can we deny the “right to suicide” to other persons who state that desire, to persons who are paralyzed, severely depressed, aged and infirm? Are they not entitled to decide that their lives lack “quality”? Once we have adopted “quality of life” as our standard, we have no principled reason to oppose the suicide of any person (with the possible exception of children and the insane, who by definition cannot make a “responsible” choice), no matter how flimsy the justification, whether undertaken in response to terminal illness, or to chronic illness, or to psychological or emotional distress. So long as a person concludes that “I do not want to live like this,” we would have no right to oppose that decision.

Indeed, what of persons such as the psychotic, the senile, the defective newborn, who have not or cannot make their own decisions but about whom we can say with confidence that “no one should have to live like this”? Shall we declare for involuntary euthanasia on their behalf, in service of their “human dignity”? The experience of the Netherlands in this regard is both instructive and frightening. In that country, where euthanasia and physician- assisted suicide are officially tolerated, government figures record at least 1000 cases annually of active involuntary euthanasia, defined as “deliberate action to terminate life without the patient’s consent.” Private observers believe that the real figure is much higher, that it should include the approximately 5000 persons killed per year by lethal dose of morphine but whose deaths are currently classified as “pain relief.” All this has occurred despite the fact that euthanasia in Holland is supposed to be “voluntary,” authorized by the patient. In its study of the Dutch experience, the Board of Trustees of the American Medical Association warns that such nonobservance of the rules is inevitable once physicians, the guardians of life, become dispensers of death. Predicting similar (and worse) numbers should these practices become legal in the United States, the Board warns that “meaningful control by a society of (the practice of euthanasia) is illusionary once the physician-patient relationship has been so changed that death becomes an accepted prescription for pain and suffering.”[18] Indeed, the move from voluntary to involuntary euthanasia is a natural one; for once we have convinced ourselves that the absence of an identifiable standard of quality of life justifies the destruction of that life, why should we hold ourselves back from acting upon our belief?

Our duty to the sick is to heal them or, when this is no longer possible, to care for them; it is not to kill them. The sick, the terminally-ill, have a right to expect compassion from us, for such flows from the respect we ought to display to ourselves and to others as children of God. But they are not entitled to ask that we take their lives, and should they make that request, we are not entitled to grant it. For when we define “compassion” so as to include the killing of human beings, we have transgressed the most elemental of Jewish moral standards and the most basic teachings of Jewish tradition as we understand it. We believe that compassion toward the dying is a moral responsibility. But we also believe that this responsibility can and must be discharged without resort to assisted suicide and active euthanasia.

II. The Cessation of Medical Treatment for Terminal Patients. Jewish tradition teaches that we achieve this compassion through two means: measures aimed at the relief of pain, and the cessation of unnecessary medical treatment for the terminally ill. For example, the same tradition which rejects suicide and euthanasia also bids us to strive to alleviate the suffering of the sick and the dying. Patients may undergo risky surgery to relieve pain, even though the surgery may hasten their death; such surgery is, after all, legitimate medicine.[19] Physicians may administer powerful anti-pain medications such as morphine to dying patients, even though such a course of treatment may shorten the patients’ lives, for pain itself is a disease and its relief is a proper medical objective.[20]

In addition to permitting such active measures, the halakhah also supports the withdrawal of medical treatment under some circumstances from terminal patients. The classic source for the discussion of this issue is the comment of R. Moshe Isserles in Shulchan Arukh Yore De`ah 339:1. Drawing upon material from the 13th-century Sefer Chasidim,[21] Isserles rules that while it is forbidden to take any measure that would hasten the death of the goses (e.g., by moving him or by moving the pillow or mattress from beneath him), “if there exists any factor which prevents the soul from departing, such as the sound of a woodcutter near the house or salt on the patient’s tongue…it is permitted to remove that factor. This is not considered a positive act (ma`aseh) but merely the removal of an impediment.” While the realia mentioned in this passage hardly resemble what we recognize as science, Isserles and Chasidim deal here with an issue familiar to all students of contemporary medical ethics. They distinguish between “active euthanasia”, defined as the application of any factor such as physical contact which would hasten the patient’s death, and “letting nature take its course,” the removal of any existing factor which serves only to impede the patient’s otherwise imminent death. The former is forbidden; the latter is permitted. Should we draw the analogy between the technologies of the Middle Ages–the birdfeathers, woodcutters, and salt–and those of our own day, we would discover traditional support for the discontinuation of medical treatment (“turning off the machines”) when that treatment can be viewed as “useless”, an impediment to death.

A problem with this analogy is that the line which separates active euthanasia from the removal of an impediment to death is not always clear. Indeed, Isserles apparently contradicts himself on this point. On the one hand, he forbids the removal of the mattress from beneath the patient, an action taken on the grounds that “some people say that certain birdfeathers have the property of delaying a patient’s death,” because to do so involves physical contact with the patient which hastens his death. On the other hand, he allows the removal of salt from the tongue, which also involves physical contact with the patient and thus presumably hastens his death, because this is merely the “removal of an impediment.” What is the difference between the two? Why may we remove the salt but not the mattress? Halakhic authorities have addressed this contradiction in various ways. Some, opting for extreme caution, declare Isserles wrong and prohibit the removal of the salt altogether.[22] Others allow the removal of the salt as but an “insignificant” contact with the patient.[23] A third approach is provided by R. Yehoshua Boaz b. Barukh, the 16th-century author of the Shiltey Giborim commentary to Alfasi.[24] He notes that while it is forbidden to hasten the death of the goses it is likewise forbidden to take any action that unnecessarily impedes it.[25] Salt, which cannot bring healing but only impede the patient’s death, should never have been put on his tongue. Whoever put it there has acted improperly; thus, its removal, even though it involves physical contact, is permitted as the restoration of the correct status quo ante.

The advantage of the Shilteyi Giborim’s analysis is that it turns our attention away from blurry distinctions between “active” and “passive” measures and toward the nature and purpose of those actions. The essential issue is the medical efficacy of the factor we seek to remove. Certain measures must never be applied to the goses because they lack any trace of therapeutic value. Offering no hope of cure or successful treatment, they serve only to delay his or her otherwise imminent death. Since it is forbidden to do this, to unnecessarily prolong the death of the dying person, these measures may be discontinued even if we must touch the patient’s body in order to do so.[26]

This theory helps to translate the medieval language of the texts into a usable contemporary vernacular. Does there not come a point in a patient’s condition when, despite their obvious life-saving powers, the sophisticated technologies of modern medicine–the mechanical respirator, for example, or the heart-lung machine–become nothing more than mere “salt on the tongue,” mechanisms which maintain the patient’s vital signs long after all hope of recovery has vanished? Answering “yes” to this question, some contemporary poskim allow the respirator to be disconnected when a patient is clearly and irrevocably unable to sustain independent heartbeat and respiration. Even though the machine is considered part of routine medical therapy (for patients are as a matter of course connected to it during emergency-room and surgical procedures), it has at this juncture ceased to serve any therapeutic function. They can no longer aid in the preservation or prolongation of life.[27] Once their therapeutic function is exhausted, the machines “merely prolong in an artificial way the process of dying. We must disconnect the patient from the machines, leaving him in his natural state until the soul departs.”[28]

III. The Duty to Heal. Other authorities, it is true, reject the comparison of modern medical technologies to those mentioned by Isserles. Birdfeathers, woodchoppers, and salt on the tongue fall into the category of segulah, something mystical or metaphysical in nature, whose properties are not subject to scientific verification.[29] Moreover, even if we accept the designation of the respirator and other end-stage technologies and therapies as “impediments to death,” Isserles describes a situation at the last moments of life, a point at which we are certain that “the soul is struggling to depart the body,” when death is imminent and would occur almost instantaneously should the impediment be removed.[30] Even if it is possible to determine precisely when a patient has reached this final extremity (and we are well aware that medicine is not a precise science), the patients who concern us here clearly have not reached it. A goses, particularly one at the very last instant of life, is not the same thing as a terminally-ill patient, who may have weeks, months, or even years to live. Neither Naomi nor Esther is a goseset. The medical treatments they are receiving are, to be sure, keeping them alive, but since neither lies at the very doorstep of death, these treatments do not qualify as “impediments” to imminent death as the tradition understands that concept. If we view their situation according to the criteria of Yore De`ah 339, therefore, we must conclude that we are not justified in withholding these treatments.

Jewish tradition, however, offers another conceptual framework for thinking about the terminally-ill patient whose death is not yet imminent. Under this framework, we consider not only the patient’s specific prognosis but also (and primarily) the nature of the practice of medicine itself. The questions addressed are: what is the Toraitic source of the commandment to heal? How does Jewish law understand “medicine” as an ethical obligation? And, most importantly for our case, does the obligation to provide medical care change or cease altogether when the patient’s illness enters a terminal stage and when hope for successful treatment has vanished? Should the answer to this latter question be “yes”, it might follow that some types of medical care may be withdrawn from a patient even before he or she has arrived at the very last moments of life.

The commandment to heal the sick is never stated explicitly in the Torah and is addressed but obliquely in rabbinic literature. A midrash on Exodus 21:19, which speaks of the tortfeasor’s obligation to pay the medical expenses for the person whom he has injured, declares that “from here we derive that a physician has permission (reshut) to practice medicine.”[31] It is Nachmanides who raises this “reshut”, a term that implies a voluntary act, to the level of a mitzvah, a religious and moral obligation. He bases this deduction upon a logical inference: since we rely upon the physician’s diagnosis to determine whether and when to set aside the laws of Shabbat and Yom Kippur on behalf of the sick, it is obvious that medicine is an integral part of the commandment to save life (pikuach nefesh).[32] Maimonides, on the other hand, also believes that medicine is a mitzvah, but he derives the commandment from Deuteronomy 22:2, a verse which, according to the Talmud, imposes upon us a positive duty to rescue a person from mortal danger.[33] Both approaches see the obligation to practice medicine as a subset of the more general commandment to save life. It follows that the obligation to heal should be understood according to the definition of that more general mitzvah. And, say the rabbis, fundamental to that definition is the element of ability. That is to say, one is required to take action to save life (and, conversely, one is liable if one does not take such action) only when the action has a reasonable chance of success. Thus, as Maimonides puts it, “whoever is able to save another (kol hayakhol lehatsil) and does not do so has violated the commandment: you shall not stand idly by the blood of your neighbor.”[34] One is under no obligation to undertake useless actions, actions which clearly do not contribute to the rescue of another person, for such measures are not to be defined as “the saving of human life.”

The same principle would apply to the practice of medicine: As with lifesaving in general, the obligation to practice medicine holds only when “one sees another in danger and one is able to save that person (veyakhol lehatsilo)” by medical means.[35] Put differently, the point and the essence of medicine is to heal. It is for this reason, and only for this reason, that we are permitted to administer harsh drugs and invasive surgical procedures which, under non-therapeutic conditions, would be strictly prohibited as chabalah, the causing of unnecessary physical harm to the human body.[36] This would imply that once a medical treatment ceases to be effective and beneficial it ceases to be “medicine” as that practice is conceived by Jewish tradition. A physician is obligated to administer those measures which in the judgment of the profession are therapeutic: i.e., they are regarded in medical opinion as contributing to the successful treatment of the disease. On the other hand, treatments which do not effect “healing” are not medicine and thus are not required. While we may be entitled to administer such treatments we are not commanded to do so, inasmuch as they do not partake in the saving of life.

This distinction is of great practical significance in the halakhic discussion of some familiar problems of medical ethics. As we have seen, halakhists rule that a terminally-ill patient may be given powerful pain-killing medications such as morphine, even though these drugs may actually hasten the patient’s death, because the treatment of pain is a legitimate goal of medical practice.[37] Then there is the question whether a person has the right to refuse medical treatment. On the one hand, since medicine is viewed as a mitzvah and suicide is prohibited, it stands to reason that a person is obligated to accept medical treatment for illness. One leading authority deals with the case of an individual who refuses treatment on the apparently admirable ground that the preparation of the medicine would involve a violation of the laws of Shabbat. Such a person, he responds, “is a pious fool (chasid shoteh). This is not an act of piety but of suicide. He is therefore compelled to do what the physicians prescribe.”[38] That is, to accept proper therapeutic treatment for disease is an act of pikuach nefesh; it is a commandment, to which one has no right to say “no.” This standard would apply, however, only to medical procedures classified as r’fu’ah vada’it or bedukah, tested and proven remedies which offer a reasonably certain prospect of successful treatment. On the other hand, should a particular remedy be experimental in nature, if its therapeutic effect upon the disease is uncertain at best, then the patient is not required to accept it.[39] Under such circumstances, the treatment is no longer classified as “life-saving” and is therefore no longer obligatory. While physician and patient have the permission (reshut) to utilize it, there is no moral requirement (mitzvah) that they do so.[40]

On this basis, halakhists can permit the cessation of medical treatment for end-stage patients who have not arrived at brain death or to the point of gesisah, the very last moments of life. R. Moshe Feinstein rules that “when the physicians see that a person cannot recover from his illness but can only continue to live in a state of suffering; and when the treatment they prescribe serves only to prolong his life as it is now, filled with suffering; they must not administer the treatments but leave him alone.” To support his decision, he cites the story of the last days of R. Yehudah HaNasi, or Rabbi, discussed above. Rabbi’s maidservant, to use modern terminology, “pulled the plug” on prayers which had lost their therapeutic value.[41] Thus, we learn that there is a significant difference between healing (r’fu’ah) and medical procedures that needlessly prolong a patient’s suffering; the former is obligatory, the latter is not.[42] R. Immanuel Jakobovits, too, rules that “there is no obligation to prolong the life and the suffering of a clearly terminal patient.” He permits a diabetic who develops terminal, inoperable cancer to cease taking insulin. Although the insulin is a successful treatment for the diabetes, it can now only prolong his suffering and delay his death. This is true “even though he is not yet a goses; since the whole point of medicine is to restore a person’s health, (the insulin) is no longer obligatory but merely voluntary.”[43]

The standard of therapeutic effectiveness, as a tool by which to make judgments concerning medical treatment, allows us to draw some conclusions with moral confidence. Under the heading “therapeutic” and “successful” treatments we would certainly include all medical and surgical procedures, such as antibiotics and routine surgeries, which physicians expect will lead to a cure for the illness in question. These treatments are “obligatory” under the traditional Jewish conception of medicine. Other therapies, though they do not produce a cure, would nonetheless fall under this category because they are able to control the disease and allow the patient a reasonable degree of function. Included here are such therapies as insulin for diabetes (so long as the patient has not developed another, terminal illness; see above) and dialysis for chronic renal disease. These procedures can be unpleasant, true, and they do not offer a cure, but they do offer life; they are to be considered as pikuach nefesh. When, however, a patient has entered the final stages of terminal disease, medical treatments and procedures which serve only to maintain this state of existence are not required. A cancer patient, for example, would accept radiation and/or chemotherapy so long as according to informed medical judgment these offer a reasonable prospect of curing, reversing, or controlling the cancer. Once this prospect has disappeared and the therapies can serve only to increase suffering by prolonging the patient’s inevitable death from the disease, they are no longer to be regarded as medicineand may therefore be withdrawn.

While this standard is useful in helping to direct our thinking, it is by no means free of difficulty. Terms such as “therapeutic” and “successful treatment” are inherently vague and impossible to define with precision. In many situations it will be problematic if not impossible to determine when or even if the prescribed regime of therapy has lost its medical value. Yet the decision to continue or to cease the treatment must nonetheless be made, and those who must make it will confront an element of doubt and uncertainty that cannot be entirely resolved. Every such decision is inherently a matter of choice, a choice between two or more alternatives when none is the obviously correct one. This kind of uncertainty is disturbing to many, who believe (as do we all) that fundamental issues of life and death must be handled with an attitude of reverence and caution. Yet their laudable search for moral certainty has led some authorities toward an extremist position, rejecting the very possibility that treatment can ever be withdrawn from a dying patient. Says one: “every person is obligated in every case to seek out medical treatment, even though he believes that the treatment will not heal him but only prolong his suffering; for we must hope for and await God’s deliverance to the very last moment of our lives.”[44] This conviction is based upon the reasoning that, inasmuch as medicine is not a precise science, even the most definitive medical prognosis is a matter of safek, of doubt. We must work to preserve life until the very end, for while it can never be established with certainty that a patient has absolutely no hope for recovery, it is indeed certain that, should we withdraw medical care, the patient will die.[45]

To this argument we would simply ask: is this truly “medicine” as we conceive it? Our answer, as liberal Jews who seek guidance from our tradition in facing the moral dilemmas of our age, is “no”. We do not adopt the simplistic approach, advocated by some, which holds that Jewish sacred texts have nothing to say to the challenges posed by contemporary medical reality. But we cannot and do not believe that those texts, which bid us to heal the sick and to preserve life, demand that in fulfilling these duties we apply in indiscriminate fashion every available technological device to prolong the death of a dying person. Medical science has made immeasurable advances during recent times, and we are thankful for that fact. Doctors today are able to prevent and to cure disease, to offer hope to the sick and disabled to an extent that past generations could scarcely imagine. Yet there comes a point in time when all the technologies, the chemicals, the surgeries, and the machines which comprise the lifesaving arsenal of modern medicine, become counterproductive, a point when all that medical science can effectively do for a patient is to indefinitely delay his inevitable death. This is not pikuach nefesh; this is not medicine; this is not what physicians, as agents of healing, are supposed to do. There is neither meaning nor purpose in maintaining these treatments. They are salt on the tongue and the sound of a woodchopper. They are not r’fu’ah; no commandments are fulfilled thereby. Yes, life is a precious thing, and every moment of it should be regarded as God’s gift. But we are not required under any reading of the tradition that makes sense to us to buy additional moments of life by undertaking useless and pointless medical treatment.

If this conviction leaves us in doubt as to the “right” answer for particular patients, then it is well to remember that moral, religious, and halakhic truth can never be a matter of absolute certainty. There will always be more than one plausibly correct answer, more than one possible application of our texts and our values to the case at hand. Our task is to determine the best answer, the one that most closely corresponds to our understanding of the tradition as a whole. That search must be conducted by means of analysis, interpretation, and argument. Its outcome will never enjoy the finality of the solution to a mathematical equation; its conclusions will be subject to challenge and critique. Yet this is no reason to shrink from moral argument; it means rather that we have no choice but to enter the fray, to confront difficult cases, and to do the best we can. We may never be absolutely sure that we are “right”; but if we are thorough in our thinking, if we read the texts, consider the case, and conduct our argument carefully and prayerfully, then we canbe sure that we have done our job.

IV. The Cases Before Us. We begin with Naomi. Canavan’s Disease, a “spongy degeneration” of the central nervous system, usually occurs in infants of East European Jewish ancestry. It is characterized by progressive mental deterioration, spasticity, and blindness. Due to diminished chest muscle function, the child will often develop respiratory tract disease. Death will occur in most cases before the age of five. No means presently exist to cure or control the illness; “treatment is symptomatic and supportive.”[46]

Based upon the analysis developed in the foregoing sections of this teshuvah, we turn our attention to the nature and extent of this “symptomatic and supportive” treatment. Since Naomi’s disease is a progressive one and can neither be reversed nor arrested, any measure that might be adduced to prolong her life is essentially an artificial and improper delay of her death and has no therapeutic value. For example, should she develop a respiratory tract disease, the goal of treatment need not be to “cure” that disease, since it is an integral part of a terminal illness which cannot be cured. She should receive treatments directed at relief of physical pain and suffering, so long as those treatments are not themselves so invasive as to increase her suffering. Naomi’s parents are under no Jewish moral obligation to resort to any measures whose purpose it is to lengthen her life.

As for Esther, whose terminal illness has brought her to the very end of her life, it is clear that the family has no duty to administer “life-prolonging measures.” By this we certainly mean “painful medical interventions,” which offer no hope of arresting or controlling her Alzheimer’s Disease. The issue of antibiotics is somewhat more difficult. On the one hand, these drugs do not affect the course of the Alzheimer’s, which is causing her death; on the other hand, they would be considered “successful” treatments for the particular infections which afflict her. In this particular case, we would take the position advocated by Rabbi Jakobovits[47] and counsel that the antibiotics not be administered. We should remember that when we practice medicine, we are treating the patient and not this or that disease. The successful treatment of a particular infection in a terminal patient does not change the fact that the patient remains terminal and that death is inevitable. Antibiotics may be justified in cases where the patient’s death is not imminent or when those drugs offer the prospect of restoring the patient to a reasonable degree of function. This latter judgment must, again, be measured not by the drug’s effectiveness in controlling a specific, identifiable syndrome but rather in the context of the patient’s total medical situation. Esther’s advanced age and medical condition offer convincing evidence both that she is “terminal” and that death is relatively near. For such a patient, the antibiotics serve no reasonable therapeutic function; they are but pointless hindrances to her death.

V. On Artificial Nutrition and Hydration. The conclusion that medical treatment may be withdrawn or withheld raises a difficult question with respect to artificial nutrition and hydration. A terminal patient may be kept alive partially or even primarily by means of food and water supplied through tubes inserted into the veins, nose, or stomach. May we discontinue the supply of nutrients or disconnect the tubes altogether on the grounds that, as all hope for recovery or satisfactory control of the illness has vanished, this feeding serves only to prolong the patient’s death?

The answer to this question depends upon whether we regard artificial nutrition and hydration as a “medical treatment.” As we have seen, Jewish tradition offers strong support for the cessation of medical treatments for the terminally-ill when these treatments have lost their therapeutic effectiveness. We are not commanded to do medicine when our actions are not medicine, when they do not heal. We violate no moral obligation if we refuse to offer a patient drugs or technologies that are medically useless. By contrast, we do violate such an obligation under normal circumstances when we withhold food and water: we have starved that person to death.[48] Though we might respond that a dying patient fed through a tube hardly constitutes a “normal circumstance,” artificial feeding differs from other hospital procedures in one crucial aspect: it can be argued that the feeding tube has nothing to do with “medicine” at all. Its function is not to treat the disease but to provide essential nutrients to the patient, and so long as the patient is capable of digesting these nutrients, the tube is successfully performing its task. In this analysis, artificial nutrition and hydration are not medical treatments, do not lose any “therapeutic” effectiveness, and therefore may not be withdrawn.

One could argue that artificial feeding devices are indeed “medical”, a response to disease. They are utilized precisely because a patient is unable to ingest nutrients in the “normal” manner. As such they are medical interventions and can be withdrawn when the intervention is no longer medically justified. There is no reason to distinguish between feeding tubes and other, indisputable “medical” procedures such as cardiopulmonary resuscitation: both keep the terminal patient alive, and the withholding of either will result in death from the very disease which warranted its introduction in the first place. On the other hand, unlike sophisticated medical procedures, food and water are universal human needs. All of us, whether sick or well, require food and water in order to survive. Moreover, the fact that these nutrients are supplied by a machine does not transform them into exotic medical substances; we all receive our food at the end of a long chain of production, transportation, and distribution technologies. A real and desirable distinction can therefore be made between artificial feeding and medical treatment.

Opinions on this question are deeply divided. A broad coalition that includes medical ethicists,[49] the American Medical Association,[50] and the United States Supreme Court[51] supports the definition of artificial nutrition and hydration as a medical procedure which may be withdrawn from terminal patients. On the other hand, this “emerging medical, ethical and legal consensus”[52] has been challenged by some ethicists who argue that the withdrawal of nutrition resembles killing more than it does the cessation of purely “medical” treatment.[53] The dispute among halakhic scholars is the mirror image of that among ethicists. Most authorities prohibit the withdrawal of food and water; “the reason, quite simply, is that eating is a normal physiological process, required to sustain life, necessary for all, including those who are healthy.”[54] Food and water are not, therefore, medicine; their presence cannot be defined as medically illegitimate.[55] At the same time, some halakhists have suggested the opposite, that artificial nutrition is a medical procedure and may be withdrawn.[56] Reform halakhic opinion is also split: one responsum opposes the removal of the feeding tube,[57] though several others permit it.[58]

Given this division of opinion, we cannot claim that Jewish tradition categorically prohibits the withdrawal of food and water from dying patients. It can be plausibly argued that artificial nutrition and hydration are medical interventions which, on the Judaic grounds that we have cited in the previous two sections of this teshuvah, may be discontinued upon a competent finding that they no longer provide therapeutic benefit to the patient. At the same time, we stress the plausibility of the opposing argument. Food and water, no matter how they are delivered, are the very staff of life (lechem chuki) for the human being. They sustain us at every moment of our lives, in health as well as in illness. It is therefore not at all obvious that we should look upon these substances as “medicine” merely because they come to us in the form of a tube inserted by medical professionals.[59] Moreover, the moral stakes in removing the feeding tube are considerable. As one authority who rules permissively admits, “there is something which is, minimally, highly unaesthetic” about withholding food and water from terminal patients.[60] We agree. Indeed, some of us would use stronger adjectives, for–let us neither mince words nor hide behind comforting euphemisms–we cannot overlook the fact that by removing them we are starving these human beings to death.[61]

We would therefore caution at the very least that the removal of artificial nutrition and hydration should never become a routine procedure. It is preferable that artificial feeding of terminal patients be maintained so that, when death comes, it will not have come because we have caused it by starvation. Nonetheless, because we cannot declare that cessation of articial nutrition and hydration is categorically forbidden by Jewish moral thought, the patient and the family must ultimately let their conscience guide them in the choice between these two alternatives.


As previously noted, supporters of euthanasia will cite various biblical and talmudic passages as evidence that the Jewish tradition supports a permissive stance toward mercy killing or, as it is called these days, physician-assisted suicide. We indicated in brief that the “tradition” does not so interpret those texts. That is to say, however Jews have read and understood the narratives of the deaths of King Saul, R. Chaninah ben T’radyon, and R. Yehudah HaNasi (Rabbi), these episodes have not tended to serve as “proof” that a terminally-ill patient may take active steps to end his or her life to avoid the sufferings of illness or that others may do so on behalf of the patient. We now want to examine this issue in some detail.

A. The Death of King Saul. While many commentators go to great lengths to justify Saul’s suicide, they do so in a way which makes it difficult to use his case as a model for today’s terminally-ill patient. Some write that Saul, as king of Israel, was a unique case, subject by the nature of his office to special ethical and political obligations that make it difficult to draw an analogy from his situation to any other. There was the concern, for example, that should Saul be captured alive the Israelites would have felt bound to attempt to rescue him, an attempt that would have entailed a severe loss of life and a further weakening of Israel’s already-precarious military situation.[62] Others suggest that Saul took his life out of fear that his captors would torture him into committing the sin of idolatry; his death was thus an act of martyrdom and not simply an attempt to avoid suffering.[63] Still others reject the whole tradition which exonerates Saul: he committed suicide, and bears the guilt of sin.[64]

B. The Death of R. Chaninah b. T’radyon. R. Chanina b. T’radyon died a martyr’s death at the hands of the Romans during the Hadrianic persecutions of the second century C.E.[65] According to the story, the Romans burned R. Chanina at the stake, wrapped in the parchment of a sefer Torah, and they place wet woolen rags around him in order to retard the flames and to prolong his agony. He nonetheless refuses to open his mouth and let the fire enter, in order to hasten his death; “it is better that the One who gives life take it away than for a person to bring harm upon himself.” Yet when a Roman guard asks: “Rabbi, if I increase the flame and remove the rags, will you guarantee me life in the world to come?”, R. Chanina answers “yes,” and the guard did so.

Since the guard’s action hastens R. Chanina’s death, it is sometimes suggested that this narrative proves that we are permitted to do the same for the terminally ill. Yet if this is the case, if one may ask another to speed one’s inevitable end, why does Chanina himself not commit suicide? We cannot argue that an individual is forbidden to kill himself but may request others to do so, for such a conclusion runs counter to the most fundamental conceptions of moral responsibility. If I am forbidden to kill myself, I am not entitled to appoint another to kill me.[66] How then may R. Chanina empower the guard to take this lethal action? Traditional commentators resolve this contradiction by reminding us that the story of R. Chanina is a case of martyrdom, for which, as we have seen, special rules apply.[67]The guard is not R. Chanina’s agent but his executioner; the rabbi cannot “appoint” the guard to do anything. The guard is the agent of the Roman authorities, who have the discretion to kill R. Chanina according to their law or by any means they desire. The manner of R. Chaninah’s death is not the “will of Heaven,” “natural law,” or any such thing; it is not up to him or subject to his decision. If the executioner decides to kill him more quickly, that is entirely the executioner’s choice. What is up to R. Chaninah is the decision to participate directly in the hastening of his death, either by his own hand or through an agent; and this he does not do. This set of facts radically distinguishes the case of a martyr from that of the terminally-ill patient.

C. The Death of R. Yehudah Hanasi (Rabbi). When Rabbi is near death, his students gather to pray for his recovery.[68] His maidservant, who at first is sympathetic to their efforts, soon realizes that Rabbi is beyond the point of healing; the time of his death is nigh, and the prayers serve only to prolong his suffering. She therefore prays that Rabbi die quickly. When she sees that the students will not cease their own prayers, she casts a glass vessel from the attic of the house to the ground; the startling sound interrupts the prayers, and Rabbi dies.

Again, some read the story as an example of mercy killing; therefore, they argue that this story provides support for the practice of euthanasia for those like Rabbi suffering the end-stage of terminal disease. The rabbinic tradition, as we have seen,[69] adopts a different interpretation: the maidservant did not kill Rabbi but rather removed an inappropriate impediment to his death. Advocates of mercy killing reject this distinction as excessively formalistic. They claim there is no significant moral difference between taking action to hasten a person’s death and withdrawing treatment so as to allow death to occur; both of these are positive actions which speed the death of the patient. We would reject this claim on two grounds. First, Jewish ethical thought does see a significant difference between action that directly kills a person and inaction which allows him to die.[70] And second, as we have argued above, the removal of an impediment is not an act of killing at all, even passive killing, but in fact a corrective measure taken against a situation which we have wrongly allowed to occur. For while Jewish tradition forbids us to kill a terminal patient it also forbids us to delay her death unnecessarily. It is therefore permitted to remove factors which contribute to that delay. To permit mercy killing, however, would be to permit the taking of a life even in the absence of “impediments” (machines, medications, etc.) that serve no therapeutic effect other than to delay death. Such killing is qualitatively distinct from the removal of an impediment.


1. See the essays by Rabbis Leonard Kravitz and Peter Knobel in Walter Jacob and Moshe Zemer, eds., Death and Euthanasia in Jewish Law (Pittsburgh and Tel Aviv, 1995).

2. This is evident from the Hebrew term for “suicide”: hame’abed atsmo l’da`at; see SA YD 345:2-3 and commentaries ad loc.

3. See Yad, Rotseach 1:4.

4. See the commentary of R. David ibn Zimra to Yad, Sanhedrin 18:6, printed editions. For the prohibition of suicide, see B, Baba Kama 91b (the midrash on Gen. 9:5, and see Rashi, ad loc.) and Yad, Rotseach 2:2-3.

5. For that matter, one is not entitled to subject his/her body to physical damage (chavalah), a principle so well established that the permissibility of cosmetic surgery is a matter of no little controversy within the legal literature. B. Baba Kama 90b; Yad, Chovel 5:1; Sh. A., CM 420:31. On this issue, see the responsum of this Committee no. 5752.7, “On the Permissibility of Cosmetic Surgery,” Teshuvot for the Nineties, pp. 283-288,

6. See Rambam, Commentary to M. Arakhin 1:3. SA, YD 329:4. This ruling seems odd at first glance because the permit to violate the laws of Shabbat is based upon Leviticus 18:5: “these are the laws…which a person shall perform and live by them…” (B. Yoma 85b). Surely the goses, who stands at the brink of death, cannot “live” by these laws for more than a few hours or, at the very most, days. Yet see Mishnah B’rurah, Be’ur Halakhah, OC 329, s.v. ela l’fi sha`ah: we violate Shabbat for the goses because the Torah places inestimable value upon even the briefest span of human life.

7. B. Sanhedrin 78a; Yad, Rotseach 2:7.

8. M. Semachot 1:4; Yad, Avel 4:5; Alfasi, Mo`ed Katan, fol. 16b; Sh. A., YD 339:1.

9. Some authorities point to the t’reifah, the person who is terminally ill and should die within one year, as a major exception to this rule: according to Jewish law, the one who kills a t’reifah is not punished by death as is the killer of the goses. However we try to resolve this apparent contradiction, we should not forget that while the killer of the t’reifah is exempt from execution by the earthly court, he has still committed a serious moral offense that will presumably bring heavenly retribution in its wake. It is arguably possible to rank the t’reifah lower than the healthy person on a scale of priorities in pikuach nefesh; i.e., I may be entitled to save the healthy person before I aid the t’reifah. Still, it is not permitted to kill the t’reifah, a point of no little significance in the debate over euthanasia. See B. Sanhedrin 78a and Yad, Rotseach 2:7-9.

10. B. Sanhedrin 74a-b and parallels; Yad, Hilkhot Yesodey HaTorah 5:1-4; Sh. A., YD 157:1. The sources add that during a time of religious persecution a Jew must accept martyrdom rather than transgress even the smallest detail of customary Jewish observance. Some authorities permit an individual to accept martyrdom even in situations where the law does not require one to do so; see Hil. HaRosh, Avodah Zarah 2:9.

11. Rabbeinu Tam in Tosafot, Avodah Zarah 18a, s.v. ve`al, citing the story of the captive children in B. Gittin 57b (and see Tosafot Gittin 57b, s.v. kaftsu). By the early fourteenth century, such acts of suicide had been approved by leading rabbinic authorities; see Chiddushey HaRitva, Avodah Zarah 18a. See, in general, R. Yosef Karo’s Bedek HaBayit to Tur YD 157.

12. R. David Kimchi ad loc., citing B’reshit Rabah 34:13, which exempts cases “like that of King Saul” from the prohibition against suicide derived from Genesis 9:5 (see note 4, above).

13. Sh. A., YD 345:3. See also R. Shalom Schwadron, Resp. Maharsham, v. 6, YD, # 123, who explicitly rejects the opposing view of the Hatam Sofer (Responsa, YD, no. 326). A similar position is taken by two of Sofer’s early 19th-century contemporaries: R. Efraim Margoliot, Resp. Beit Efraim, YD, # 76, and R. Mordekhai Benet, Resp. Parashat Mordekhai, YD, # 25-26.

14. Arukh HaShulchan, YD 345, # 5: Saul’s suicide was a case of emotional coercion (ones) and not a rational choice. See in general SA, YD 345:3 and Pitchey Teshuvah, # 3.

15. Those who advocate euthanasia sometimes attempt to distinguish between an “agadic tradition” which tells stories that support mercy killing and a “halakhic tradition” which opposes it. We think this attempt is fruitless. Both literary genres are the product of the same religious culture; the rabbis who tell the stories are the same rabbis who read them to learn the law. And it is those rabbis who prohibit euthanasia.

16. Our own Reform responsa tradition, it should be noted, has consistently rejected euthanasia as a morally acceptable response to terminal illness. See the teshuvot of R. Israel Bettan (ARR, #78, ), R. Solomon B. Freehof (ARR, # 77, ); Reform Responsa, # 27, pp. 117- 122; Modern Reform Responsa, # 34,, #35, ), R. Walter Jacob (ARR, no. 79, CARR, # 81,, #83, pp. 138-140; Questions and Reform Jewish Answers, nos. 145 and 157,, R. Moshe Zemer (Halakhah Sh’fuyah, pp. 295-298).

17. Hence, the blessing dayan ha’emet, “blessed be…the True Judge,” traditionally recited at the death of a relative or when one receives evil tidings. See M. B’rakhot 9:5: the commandment to love God “with all your strength” (Deut. 6:5) implies that we are to give expression to this love come what may. See Yad, B’rakhot 10:3, and Sh. A., OC 222:2. This does not mean that we must accept sorrow and tragedy in passivity; the numerous stories of Jewish heroes who “argue with God” against the evil in the world are sufficient proof of that. It does suggest, however, that the experience of evil does not bring an end to the very relationship with God in which a meaningful argument can take place.

18. Report of the Board of Trustees of the American Medical Association in Issues in Law and Medicine 10:1 (Summer, 1994), pp. 89ff, at p. 91. See also at p. 81: “The Board of Trustees recommends that the American Medical Association reject euthanasia and physician-assisted suicide as being incompatible with the nature and purposes of the healing arts.” While Dutch observers report that there is some controversy over these statistics, that controversy in no way lessens the moral gravity of the situation. On the case of the Netherlands, see the article by Joop Al in the forthcoming Jewish Law Annual, v. 12.

19. R. Ya`akov Emden (18th century), Mor uKetsi`ah, ch. 328.

20. R. Eliezer Waldenberg, Resp. Tsits Eliezer, v. 13, # 87. Waldenberg cautions that the intent of the procedure must be to relieve pain and not to hasten the patient’s death, a point made as well by Emden, note 19, above.

21. Ch. 723 (= ch. 315, Wistinetzki-Freimann ed.)

22. Turey Zahav, YD ad loc. He notes that Karo’s ruling, to which Isserles does not object, prohibits even the closing of the patient’s eyelids at the moment of death. The slightest amount of contact, therefore, must be seen as the hastening of death.

23. Nekudot HaKesef ad loc.

24. Shiltey Giborim to Rif, Mo`ed Katan, fol. 16. It should be noted that R. Yehoshua Boaz, an older contemporary of Isserles, never saw the latter’s ruling. He is addressing the same contradiction as it appears in Sefer Chasidim. The thrust of his comment is applied to Isserles by Beit Lechem Yehudah, YD 339.

25. Sefer Chasidim loc. cit.

26. For this reason, too, the Shiltey Giborim supports the prohibition against moving the mattress. Even though the feathers may serve to impede the patient’s death, the mattress itself is supposed to be there; it plays a legitimate role in the care of the goses, and is thus not solely an impediment that must be removed.

27. See R. Eliezer Yehudah Waldenberg, Resp. Tsits Eliezer, vol. 13, # 89. Waldenberg conditions this permit upon the performance of extensive test which show that the patient cannot recover independent respiration. See also R. Chaim David Halevy, Aseh Lekha Rav, v. 5, # 29.

28. Rabbi B. Rabinovits, Sefer Asya, 1976, pp. 197-198.

29. Dr. Ya`akov Levy, No`am, vol. 16, 1973, pp. 53ff.

30. Thus, Waldenberg (see above, n. 27) stresses that his permission to turn off the respirator applies only at the very end of life (gemer k’lot hanefesh).

31. B. Bava Kama 85a. “Permission” is necessary, according to Nachmanides (see next note), for two reasons: in order to protect the physician from claims of liability should he cause injury to the patient, and in order to allow the practice of healing even when it seems to contradict the Divine will (“if God smites a person, who am I to heal him?”).

32. Ramban, Torat Ha’Adam, Chavel ed., pp. 41-42. His discussion forms the basis of the Tur, YD 336, on the laws of medicine. On pikuach nefesh see Lev. 18:5 and B. Yoma 85b.

33. Rambam, Commentary to the Mishnah, Nedarim 4:4. The verse declares that we are obligated to restore a lost object to its rightful owner; the midrash cites a linguistic peculiarity in the verse to extend this duty to the “restoration/rescue of a person’s life” (hashavat gufo; B. Sanhedrin 73a). See the discussion in our responsum 5754.18, Teshuvot for the Nineties, pp. 373-380,

34. Yad, Rotseach 1:14. On “not standing idly by…” see Lev. 19:16 and B. Sanhedrin 73a.

35. Rambam, Commentary to the Mishnah ad loc.

36. M. Baba Kama 8:5; B. Baba Kama 91a-b; Yad, Hilkhot Chovel 5:1; Sh. A., CM 420:31. See, in general, our responsum 5752.7, “On the Permissibility of Cosmetic Surgery,” note 5, above.

37. See at note 20, above.

38. Resp. R. David ibn Zimra (16th-c. Egypt), v. 1, # 1139.

39. R. Ya`kov Emden, Mor uKetsi`ah, 328; R. Moshe Raziel, “Kefi’at choleh lekabel tipul refu’i,” Techumin 2 (1981), pp. 335-336.

40. See R. Ya`akov b. Shmuel (Prussia, 17th century), Resp. Beit Yaakov, # 59. R. Ya`akov Reischer (Germany, 18th century) in his Resp. Sh’vut Yaakov, OC # 13, disputes the Beit Yaakov, but only in that he regards it permissible (mutar) for a physician to administer medications that delay death. He does not claim that it is obligatory to do so.

41. R. Nissim Gerondi, Commentary, B. Nedarim 40a, also cited this story as an halakhic precedent, but in a more restrictive sense: since the maidservant, seeing Rabbi in excruciating pain, had previously prayed for his speedy death, R. Nissim learns that there are times when one is permitted to pray for the death of a suffering patient. Feinstein goes farther, suggesting that the story serves as a precedent for the cessation of medical treatment.

42. Resp. Igrot Moshe, CM, v. 2, # 73-74.

43. Rabbi I. Jakobovits, in HaPardes 31 (1957), # 3, pp. 18-19.

44. R. Natan Zvi Friedman, Resp. Netser Mata`i, # 30. See also R. Eliezer Yehudah Waldenberg, Resp. Tsits Eliezer, vol. 5, Ramat Rachel, # 28, and J. D. Bleich, “The Quinlan Case: A Jewish Perspective,” in Fred Rosner and J. David Bleich, eds., Jewish Bioethics (New York, 1979), pp. 266-276.

45. See R. Avraham Yitschak Hakohen Kook, Resp. Da`at Kohen, # 142; Resp. Chatam Sofer, YD, # 158.

46. J.G. Theone, Physicians’ Guide to Rare Diseases (Montvale, NJ: 1992), pp. 373-374. See also J.B. Wyngaarden and C.H. Smith, Jr., eds., The Cecil Textbook of Medicine, 18th edition (Philadelphia, 1988), p. 2216.

47. At note 43, above.

48. B. Sanhedrin 77a and Rashi ad loc.; Yad, Hilkhot Rotseach 3:9-10.

49. President’s Commission for the Study of Ethical Problems in Medicine and Biomedical and Behavioral Research, Withholding or Withdrawing Life-Prolonging Treatment (Washington, DC: 1983), p. 88; Tom L. Beauchamp and James F. Childress, Principles of Biomedical Ethics (New York, 1989), pp. 163-169; Hastings Center, Guidelines on the Termination of Life-Sustaining Treatment (Briarcliff Manor, NY: 1987); and numerous others.

50. Council on Ethical and Judicial Affairs of the American Medical Association, Opinions (Chicago: 1986), Opinion 2:20.

51. Cruzan v. Director, Missouri Department of Health, 497 U.S. 261 (1990).

52. The term is that of Robert Steinbrook and Bernard Lo, “Artificial Feeding–Solid Ground, Not a Slippery Slope,” New England Journal of Medicine 319 (1988), p. 288.

53. See Patrick G. Derr, “Why Food and Fluids Can Never Be Denied,” Hastings Center Report 16 (February, 1986), pp. 28-30; Gilbert Mailaender, “On Removing Food and Water: Against the Stream,” Hastings Center Report 14 (December, 1984), pp. 11-13; Daniel Callahan, “On Feeding the Dying,” Hastings Center Report 13 (October, 1983), p. 22.

54. R. Moshe Feinstein, Resp. Igrot Moshe, CM, v. 2, # 74, sec. 3.

55. See Nishmat Avraham, YD 339, pp. 245-246; Avraham Steinberg in Sefer Asya 3 (1983), p. 448; R. Immanuel Jakobovits in HaPardes 31:3 (1957), pp. 18-19. Among Conservative thinkers see R. David Feldman, Health and Medicine in the Jewish Tradition (New York, 1986), p. 95, and R. Avram Reisner in Conservative Judaism 43:3 (1991), pp. 52 ff.

56. For Orthodox opinion, see R. Zev Schostak, “Jewish Ethical Guidelines for Resuscitation and Artificial Nutrition and Hydration of the Dying Elderly,” Journal of Medical Ethics 20:2 (June, 1994), p. 98, and R. Zalman Goldberg in Emek Halakhah (Asya) (Jerusalem, 1986), p. 78 (but see note 13 ad loc.). An important Conservative responsum in this vein is authored by R. Elliot N. Dorff in Conservative Judaism 43:3 (1991), pp. 36-39.

57. See “Hospital Patient Beyond Recovery,” Teshuvot for the Nineties no. 5750.5, pp. 365-369,

58. R. Walter Jacob, Questions and Reform Jewish Answers, # 159, pp. 263-269; R. Mark N. Staitman in Jacob and Zemer, Death and Euthanasia, pp. 1-10. See also R. Solomon Freehof, in ARR,# 77, pp. 257-260,, who permits the physician to refrain from connecting or refilling the nutrition apparatus of a dying patient.

59. The argument, advanced by some, that we are not obligated to provide artificial nutrition because it is essentially different from “eating” in the normal sense is hard to understand. In either case, whether we prevent a healthy person from eating or withdraw artificial nutrition from a patient, we are withholding nutrients that are necessary for survival and starving that person to death.

60. Dorff, note 56, above, at p. 38.

61. The Cruzan case (note 51, above) is instructive here. Though Ms. Cruzan’s parents sought to discontinue artificial feeding, they did not ask for the removal of the feeding tube, which they wanted left in place so that medications might be administered to reduce seizures as their daughter died. That is to say, they did not ask to discontinue her medical treatment so that “nature” might take its course; they asked that food and water be withheld so as to cause her death by starvation.

62. See R. Shelomo Luria, Yam shel Shelomo, Baba Kama, ch. 8, # 59.

63. Chiddushei HaRitva, Avodah Zarah 18a. See above at notes 10 and 11.

64. See R. Yosef Karo, Bedek HaBayit, Tur YD 157.

65. B. Avodah Zarah 18a.

66. The operative principle is sh’lucho shel adam k’moto, “one’s agent is the legal equivalent of oneself”. A person’s legal representative, who carries that person’s “power of attorney”, is endowed with only those rights enjoyed by the one who appointed him or her. A corollary is the statement ein shaliach l’devar aveirah: “an agent cannot legally perform a transgression.” Should I instruct my agent to do something prohibited by the Torah, those instructions are null and void; B. Kiddushin 41b-42b and parallels.

67. See above at notes 10 and 11.

68. B. Ketubot 104a.

69. See above, at notes 41-42.

70. The classic example is the case in B. Sanhedrin 74a, where one has been told: kill so-and-so; if you refuse, we will kill you. The ruling in that instance: let yourself be killed. How do you know that your blood is redder than his? Perhaps his blood is redder. That is, the lives of both are equally precious to God. One life will be destroyed in either event; do not compound the tragedy by committing the sin of murder (see Rashi ad loc.). The only proper moral stance is shev ve`al ta`aseh, inaction, for positive action can but make the situation worse.