John R. Shook
Oklahoma State University
January 2001
Euthanasia, Unnecessary Suffering, and the Proper Aims of Medicine
It does not require any prolonged exposure to the debates over euthanasia to grasp that a large portion of the effort expended by disputants goes towards arguing over definitions. This is likely a good sign of healthy philosophical argument, so long as this effort contributes to a deeper understanding of the issues, and diminishes the chances that opposed sides are really arguing past each other. My reading of recent work on euthanasia does not give me such confidence, however. Like many others, I have gradually become persuaded that the very term "euthanasia" and its famous species, "active" and "passive," have mostly lost their usefulness. Instead of attempting to better define these terms, this essay will propose to abandon them in favor of newly minted terms created by consulting a Latin dictionary. So that these new terms can be pragmatically tested, an on-going debate generated by James Rachels and Bonnie Steinbock is analyzed.1 Rachels’s two primary conclusions are defended: the act of removing life-support is just as much an act of homicide as giving a lethal injection, and the prevention of unnecessary suffering is not a primary aim of the medical profession. Steinbock’s reply to Rachels do not affect these conclusions; indeed, her views actually support them. Steinbock thus fails to show that the AMA does not inconsistently reject doctors intentionally killing patients while approving the withdrawal of life-support. However, Steinbock’s own primary conclusion is also defended: the AMA’s permission to remove futile treatment is not inconsistent with its rejection of mercy killing. The essay concludes by clarifying the AMA’s position on euthanasia using the new terminology.
I. Withdrawal Euthanasia is Homicide
Before we turn to the terminological issues, one presupposition required by this essay’s treatment of these issues must be defended. This tenet is that an act of withdrawal euthanasia is an act of homicide. Rachels defends the tenet that an act of withdrawal euthanasia is an act of homicide, but his defense proceeds in a very unusual direction. His defense requires us to first agree that the moral qualities of passive euthanasia are not significantly different from those of active euthanasia. Rachels asks to compare the moral evil of Smith, who deliberately drown his cousin, with that of Jones, who stands by and allows his cousin to die by "accident." Taking for granted that the reader should agree with Rachels that Smith and Jones are equally morally evil, Rachels takes this as primary evidence of passive euthanasia’s status as an act of killing. However, I am not so sure that I can easily agree, and when asked, my medical ethics classes are about evenly divided over this question. The root of this disagreement apparently lies, upon questioning, in my students’ appeal to their understanding of the law. If both Smith and Jones were brought to trial, and all the facts of the case were before the jury, in the U.S. legal system it is most likely that only Smith could be convicted and punished for murder. This is so, because it would be difficult for a prosecutor to even claim that Jones committed an act of homicide, since it is highly questionable whether Jones directly or indirectly was a cause of his cousin’s death. I do not presume to resolve such questions, but the cause of reasonable disagreement with Rachels’s own intuitions on the equivalent moral guilt of Smith and Jones is obvious.
Rachels’s efforts to assimilate the moral qualities of passive euthanasia to those of active euthanasia are thus trebly confusing. First, it is highly doubtful whether our moral intuitions about Smith and Jones must coincide with Rachels’s. Second, the case of Jones is not sufficiently analogous to the withdrawal of life-support, the concern of the AMA, but rather to the withholding of life-support. Withholding euthanasia is a type of euthanasia which neither Rachels nor the AMA nor Steinbock really consider to be as significant a moral question as active or withdrawal euthanasia. Third, if Rachels wants to show that the AMA inconsistently decries killing while countenancing withdrawal of life-support, he should not appeal to their dubious moral equivalence, but rather to their common status as simple homicides. A homicide, following criminal law in the U.S., is that act of a person which is a significant and proximate cause, directly or indirectly, of the death of another person. The nature of this person’s intent behind doing this act is irrelevant to the question of whether it should be categorized as homicide, although it is quite relevant to whether the act is also murder.
It will be objected that since withdrawal euthanasia is one type of passive euthanasia, and that passive euthanasia must by definition be not really an action at all but only a non-action, then withdrawal euthanasia could not be an act of homicide – it is not even an act at all! This objection seems to be specifically saying, for example, that when the artificial respiration machine is switched off to halt the flow of oxygen through an air tube that goes into the patient’s trachea, or a feeding tube is pulled from a patient’s body, whoever is responsible for these events performs an action to a machine or a tube, but does not perform any action to the patient, much less an act of homicide. The appropriate response to this objection is to ask what primarily caused the lack of oxygen in the patient’s lungs, or caused the lack of nutrition in the patient’s stomach, if not the person who performed the switching or pulling? I cannot follow a line of argument inferring that such a person could not be the primary cause of the absence of oxygen in the patient’s lungs from the admitted premise that this person was the only cause of the machine halting the flow of oxygen through the air tube. Not can I follow a line of argument inferring that such a person could not be a primary cause of the patient’s death from the admitted premise that this person was the primary cause of the absence of oxygen in the patient’s lungs. But perhaps this objection is instead claiming that a person could be a primary cause of the patient’s death yet does not perform an act of killing, since an act of killing must be the only cause of death. Where the flow of oxygen is halted, or nutrition is halted, such halting is but one of the causes of death; there are many others, such as the body’s vital need for oxygen or for nutrition. After all, if the patient didn’t need oxygen so badly, the halting of oxygen through the air tube would not have been able to contribute to the patient’s death! However, it is simply not the case that an action cannot be a killing unless it is the only cause of death; if it were so, then a pyromaniac could justifiably claim that his action of setting a house on fire could not be the killing of the child inside, since the house contained oxygen to support the fire.
The withdrawal of life-support is an act of homicide because the withdrawal is a significant cause of the patient’s hastened death. It need not be the only cause, of course, but neither is it an insignificant cause. The opposed view, that the only significant cause of death in the case of the withdrawal of life-support is the disease/injury which created the need for life-support, is unreasonable. To prove that this is so, let us examine a hypothetical case in which Smith withdraws the life-support of another person. Suppose Smith is a new member of a criminal gang, and must now prove his worthiness by passing a test: killing Doe, a leader of a rival gang, within the next 24 hours. Since Doe is on life-support in a hospital, making a slow recovery from a mysterious disease, Smith sneaks into Doe’s room and aims his pistol at the unconscious Doe. Just then Smith notices all of the tubes and wires attaching Doe’s body to large machines at the bedside, and has a flash of inspiration: instead of making more noise and leaving bullets at the scene of the crime, he could use his gloved hand to detach all of tubes and wires! Smith does so, and he watches as the last breaths heave out of the dying Doe. Satisfied that Doe is dead after five minutes, he then executes his getaway plan by jumping out of the hospital room window. Alas for Smith, he forgot that the room was on the third floor; security easily catches him as he tries to limp away from the hospital with a broken leg. Smith’s trial is now at hand. All the pertinent facts are known, but Smith’s defense attorney is confident that he can get his client off on the lesser charge of breaking and entering. The attorney will simply explain to the judge that Smith cannot be prosecuted for murder, much less homicide, because Smith was not a significant cause of Doe’s death – only the Doe’s disease killed Doe! Granted, Smith pulled all the tubes and wires from Doe’s body, but the cause of death according to the coroner was the disease, not Smith.
I hope, and expect, that this attorney’s interesting defense would never succeed, and that Smith’s trial would proceed. Of course Smith was a significant cause of death: if Smith had not withdrawn Doe’s life-support, Doe’ disease would not have been able to contribute to Doe’s death on that fateful day. Smith committed an act of homicide, and this is the case whether or not Doe’s disease would have otherwise killed him later on, and even if it was certain that Doe would have actually died later that same day. The Smith example proves that an act of withdrawal of life-support is a homicide, just as if Smith had instead shot Doe to death. When a doctor withdraws life-support from patient which hastens the patient’s death, from reasoning by analogy, that doctor performs an act of homicide on that patient. Objections to drawing such an analogy fall into four possible categories. First, the doctor’s intentions behind withdrawal are presumably quite dissimilar from Smith’s intentions. Second, the doctor’s action may be very praiseworthy while Smith’s action is only reprehensible. Third, it could be claimed that doctors cannot by definition kill patients. Fourth, in the Smith case Doe would likely have survived his hospital visit if not for Smith, since Doe was on the road to full recovery, but a doctor who withdraws futile life-support should do so only when the patient will soon die anyway.
The first objection is irrelevant, since as explained above, the nature of the intent behind an act does not affect the question of whether it is a homicide. A person can commit a homicide for all sorts of reasons besides trying to profit from another’s death. Negligent homicides happen all the time, and prison executioners surely perform homicides in the performance of their duties. The intent of the doctor bears only on the question of the moral status of withdrawal, not on the bare fact of whether a homicide occurs. This reply to the first objection also satisfies the second objection. As for the third objection, it asks us to define "doctor" so that a real doctor would never under any circumstances kill a patient. This notion is defeated by the unfortunate incidents of gross medical malpractice. The fourth objection is defeated by considering the hypothetical Smith further. Even if it were instead certain that Doe would have actually died later that same day from D, Smith’s defense attorney could not, and should not, be able to persuade a judge that Doe’s inevitable death precludes the possibility that Smith was a significant cause of Doe’s actual death. The attorney’s defense amounts to the claim that in some real sense, at the time Smith was pulling out Doe’s tubes and wires, Doe was already dead from the disease, preventing Smith from being a cause of Doe’s death. Aside from completely lacking any proper legal standing, this defense asks us to unreasonably confound the common sense and legal understanding of the "time of death."
One oft-mentioned version of objection four is worded as follows: the withdrawal of futile life-support is not homicide, because the doctor only removes an artificial medical barrier to the patient’s natural death. This objection could be understand in four different ways. In its first sense, this objection portrays the doctor’s action to withdraw life-support as no cause of anything at all, and certainly not a cause of the patient’s death. This sense has already been refuted above. In its second sense, this objection portrays withdrawal as an action performed on the patient, not as a cause of the patient’s death, but rather as a cause contributing to nature’s ability to cause the patient’s death. This portrayal of withdrawal also portends a major and unreasonable violation of common sense and legal theory, as shown by the Smith case. The fact that a coroner would not list the doctor’s withdrawal is but professional courtesy. In its third sense, this objection portrays withdrawal as an action performed on the patient, not as a killing of the patient, but rather as the hastening of the patient’s death. This sense likewise violates common sense and legal theory, since the definition of homicide does not recognize any valid difference between killing and hastening death. An alleged murderer is not released when it is discovered that his victim would have died later that day, or later that hour. In its fourth sense, this objection portrays withdrawal as an action performed on the patient, not as the cause of the hastening of the patient’s death, but only as causing the patient to die sooner than the patient would have, had the life-support not been withdrawn. This distinction eludes me completely; I should think that these descriptions were pretty much synonymous.
II. Rachels’s Dispute with the AMA
Since withdrawal euthanasia is an act of homicide, Rachels is right to accuse the AMA of both denouncing and defending the killing of a patient by a doctor. Let us be reminded of the AMA’s 1973 statement to which Rachels refers.
The intentional termination of the life of one human being by another – mercy killing – is contrary to that for which the medical profession stands and is contrary to the policy of the American Medical Association. The cessation of the employment of extraordinary means to prolong the life of the body when there is irrefutable evidence that biological death is immanent is the decision of the patient and/or his immediate family. The advice and judgment of the physician should be freely available to the patient and/or his immediate family.2
Aside from the fact that the AMA offers a poor definition of mercy killing (Smith terminated Doe’s life but that act was hardly motivated by mercy), the AMA does appear to presume in this statement that the withdrawal of medical treatment is not the termination of a human life by another. That is to say, the AMA presumes that withdrawal is not homicide. This presumption is false, as I have argued. However, Rachels uses this point to step up to a larger claim: that the withdrawal of futile medical treatment is mercy killing. This larger claim is rightly challenged by Steinbock, but she does not successfully find a counter-example case of withdrawal.
Unfortunately, the manner in which Steinbock goes about her argument portrays the issue as one of whether the withdrawal of life-support is even a matter of euthanasia at all. She tries to defend the AMA, not by directly arguing that withdrawal is not killing (good thing, too) but rather by arguing that the AMA’s approval of the withdrawal of futile life-support is not designed to approve a doctor’s use of withdrawal as the means of preventing unnecessary suffering. While Steinbock is able to show that Rachels wrongly supposed that the AMA’s statement must be understand as giving such approval, Steinbock tacitly gives the reader the misleading impression that euthanasia must be defined as the use of the patient’s hastened death as the means of preventing unnecessary suffering. Such a definition is very unhelpful indeed, as we will see below. Still, Steinbock cannot be entirely blamed for this confusion, since Rachels says nothing to dispel the notion that this definition is Rachels’s own, and says much to support this notion. Rachels is most interested of course to expose the AMA’s statement as ample evidence of the AMA’s failure to place patient comfort ahead of prolonging life, caused by the medical profession’s cowardly fear of prosecution for homicide and murder. Rachels successfully exposes this failure by pointing out that if the medical profession were really motivated to prevent unnecessary suffering in cases where further treatment could but temporarily prolong life, then only the AMA’s false belief that withdrawal is not homicide would stand in the way of its approval of active euthanasia over withdrawal euthanasia. If the AMA were to realize that both active and withdrawal euthanasias were homicides, then it could proceed to the next question: which sort of homicide is in the best interests of the patient? Rachels’s plea for active euthanasia, as often being the most effective way to prevent unnecessary suffering, only makes sense at this stage. By portraying active euthanasia as a means to accomplish such prevention, however, Rachels leads the reader to similarly see withdrawal euthanasia as the use of the patient’s death to prevent unnecessary suffering. From Rachels, then, we are drawn to the view that mercy killing is always an act of causing the patient’s death in order to prevent unnecessary suffrage. It is precisely on this point that Steinbock aims her reply. Is it really true, she asks, that a doctor’s withdrawal of futile life-support is always a matter of the doctor using the patient’s death as a means to prevent unnecessary suffering?
Steinbock, like the AMA, apparently is unwilling to countenance the notion that withdrawal is homicide, that withdrawal is killing, but instead of trying to argue that point directly, she instead argues that the AMA actually approves of cases where the motivation is not one of merciful killing. There are at least two kinds of cases, Steinbock says, where a doctor may permissibly withdraw medical treatment without performing an act of merciful killing (and hence an act of euthanasia). In the first, the doctor withdraws treatment only because that is the patient’s wish. In the second, the doctor withdraws treatment only because continued treatment prolongs the patient’s life for no sufficiently worthy gain. In the first case, Steinbock (successfully) argues that it could be the case that the doctor does not withdraw treatment because the patient is suffering unnecessarily, but rather because the doctor believes that she has the duty to respect this patient’s wish to have treatment withdrawn. Even if that may be the case, Steinbock cannot prove that this case is not mercy killing, because she only argues that the motivation can be other than one of mercy. In this case, Steinbock does not argue that this withdrawal is not killing. As will be explained in the third section, Steinbock does actually have here a counter-example to Rachel’s view that a doctor’s withdrawal of futile life-support is a case of mercy killing, but she fails to use it carefully.
In the second case, where anti-cancer treatment is withdrawn, the motivation of the doctor is indeed merciful. However, Steinbock simply declares, without detailed argument, that although the act of withdrawal does hasten the patient’s death, the act is not a homicide. Her own wording must be consulted:
The decision here to cease anti-cancer treatment cannot be construed as a decision that the patient die, or as the intentional termination of life. It is a decision to provide the most appropriate treatment for the patient at that time. Rachels suggests that the point of the cessation of treatment is the intentional termination of life. But here the point of discontinued treatment is not to bring about the patient’s death but to avoid treatment that will cause more discomfort than the cancer and has little hope of benefiting the patient.3
Steinbock is claiming, without argument, that an act which hastens a person’s death is not also an act of intentionally terminating that person’s life. Granted, the very terms themselves suggestively inspire agreement with Steinbock. The word "termination" hardly seems fit to describe the consequences to a cancellation of cancer treatment. If so, then the proper perspective on Steinbock’s proposed counter-example is not to agree with Steinbock that withdrawals are not homicides, but to get past terminological fog and go to root of the question: is the doctor’s act of cessation of life-prolonging treatment an act of homicide: an act of directly or indirectly causing the hastening of the patient’s death? The answer, in light of the Smith case above, is surely yes, since if the doctor had not done this act, in the absence of other factors, the patient would not have died sooner. Now, I am well aware that counterfactuals deserve their notorious reputation for starting arguments rather than ending them. Nevertheless, we have exposed one of the essential questions that must be faced in the euthanasia debate, no thanks to either Rachels or Steinbock. Lacking an argument to support her claim that the withdrawal of futile treatment is not the intentional hastening of the patient’s death, Steinbock’s second case ends up much as her first case did, remaining a case of merciful killing.
Because neither of Steinbock’s cases successfully refute Rachels’s claim that the withdrawal of futile medical treatment is mercy killing, she fails to defend the AMA’s position that it completely repudiates euthanasia in any form. Therefore, she is powerless to defend the AMA against Rachels’s moral condemnation of the medical profession’s failure to place the prevention of unnecessary suffering as the priority over the mere prolongation of life for it own sake. Indeed, her first case can only impact Rachels’s argument if it construes the doctor’s motivation as something other than the desire to prevent unnecessary suffering. The overall impression of the collision between these two articles is that despite appearances, they actually do not manage to intersect much at all. The problem is largely terminological, and the remedy must likewise be terminological.
III. Clementhanasia
The linguistic heritage of "clementhanasia" represents the effort to return to the fundamental basis of euthanasia: that euthanasia is, aside from all other connotations, an act of mercy killing. Giving us our English word "clemency," the Latin term "clement" means roughly "mercy." Joining this term with "thanasia," meaning "death," creates "clementhanasia": an act of "merciful killing." Person Q performs an act of clementhanasia A on person P if and only if all three criteria below are met. No one can perform an act of clementhanasia on themselves; suicide and assisted suicide are not considered in this paper.
Criterion one offers the traditional distinctions because they do reflect moral distinctions that many feel ought to be made. The order of their listing is not designed to show any moral preference. "Active" and "passive" are replaced with "infusio" and "withdrawal," respectively. Purists might instead refer to withdrawal clementhanasia as "recipio-clementhanasia," since "recipio" means to "take back." The notion of a "lethal agent" is supposed to be as broad as possible, covering anything introduced into the patient’s body which seriously interferes with its natural functions to cause death. An injection of potassium chloride is a paradigm example, as is the cruder but no less deadly bullet fired from a pistol at close range. Criterion one rules out two other conceivable kinds of euthanasia. First, it rules out what has been traditionally called "withholding" euthanasia, in which Q refrains from doing some action that would have prolonged P’s life. This type of euthanasia is problematic for many reasons. It is difficult to classify as an action at all. Furthermore, it typically involves the rationale of "allowing nature to take its course." This rationale portrays people as under an obligation to not interfere with human destiny, which is profoundly opposed to the spirit of clementhanasia: that there are some circumstances where people should undertake responsibility for the manner of another’s death. Perhaps "withholding euthanasia" is a misnomer altogether. Refusing to interfere with nature’s course should instead be called "cosmothanasia": "death by nature." Second, criterion one rules out those in positions of influence or authority from performing clementhanasia. Person Q may suggest or persuade or command or give permission to another person R, who then performs clementhanasia, but if R does so, Q does not therefore also perform clementhanasia.
Criterion two requires that P’s death must be a consequence of action A. P cannot survive an act of clementhanasia performed on her, and there cannot be some other intervening circumstances that are the real cause of P’s hastened death. Criterion two does not require that Q knows that A will hasten P’s death, since that stronger requirement amounts to adding that Q must justifiably believe that A will hasten P’s death. This is too strong a requirement. Q might act on a false belief about P’s medical condition yet still manage to hasten P’s death by doing action A because, unbeknownst to Q, P’s real medical condition made him very susceptible to A. In such a case, Q still managed to perform clementhanasia on P. Another interesting aspect of criterion two is that only "hastening death" is required, not "killing." Why, in light of my strenuous efforts to categorize hastening death as killing, do I side-step this issue? The careful choice of words here is motivated by two considerations. First, my strenuous efforts may well be insufficient. Although killing is evidently a matter of hastening death, some readers may still balk at the reverse claim. Second, "hastening" death is a matter of degree (for how much time was the patient’s life shortened?), while "killing" is not. Since the degree of hastening is quite likely a valid and important moral factor, the definition of clementhanasia should not obscure it. One further fact about the wording of criterion two should be highlighted: it does not require that Q intends to commit an act of homicide on P, nor does it require that Q’s purpose in doing A must be to hasten P’s death. Far too much purely verbal wrangling over "intending" has considerably clouded euthanasia debates. One side argues that a doctor, knowing the fatal consequences of withdrawal, must thereby intend the patient’s death, while the other side argues that merely foreseeing the fatal consequences of withdrawal cannot imply that a doctor intends those consequences. Far too much moral weight is placed on ordinary vague English words here. One side uses "Q intends to do A" to mean that "Q knows that a description of his actions could include A." The other side uses "Q intends to do A" to mean that "Q by his actions primarily desires A to happen." There is just no right answer here, since in ordinary language the term "intends" and its synonyms easily cover both usages. The pragmatic answer to this problem is to define clementhanasia so that the essential nature of euthanasia is unobscured: the way that a doctor goes about preventing unnecessary suffering requires the trade-off of shortened life.
Criterion three is designed to also avoid the prevalent wrangling over the question of the exact nature of Q’s intentions while performing A. It instead casts the issue as a question of how Q would sincerely account for doing A. It is largely this criterion which applies to Steinbock’s attempt to carve out two cases of hastening a patient’s death that fail to be acts of euthanasia. The first case supposes that the doctor performs action A, a withdrawal of treatment, which hastens P’s death, because the doctor believes that she is under an overriding obligation to respect P’s wish to have that treatment removed. On Steinbock’s telling of this story, a crucial matter has been left out: why does the doctor believe that she is under this overriding obligation? Perhaps this doctor believes that a patient’s wish to not be treated always overrides any other considerations. We are indeed left to read this case in this way; on this reading the doctor does not perform an act of clementhanasia. Since the doctor is not motivated by mercy, but by the command of the patient, the doctor’s action should instead be called "imperothanasia": a "commanded death." However, it is quite possible that this doctor instead believes that this obligation is overriding in this particular case only because of this patient’s present condition. For example, suppose the doctor believes that this particular patient has the right to refuse treatment because she believes that any person ought to have the right to refuse treatment in circumstances where continued treatment would only prolong suffering. If this were indeed the case, then criterion three applies, and the doctor would be performing an act of clementhanasia.
The second case offered by Steinbock describes the doctor’s motivation behind withdrawing treatment as a desire to prevent unnecessary suffering, not as a desire to hasten the patient’s death in order to prevent unnecessary suffering. Steinbock understands Rachels, as explained above, to be defining euthanasia in such a way as to require that the doctor intends to use the patient’s death as the means of preventing unnecessary suffering. My definition of clementhanasia contains no such requirement. Steinbock’s doctor correctly believes that the withdrawal of treatment will hasten the patient’s death; thus criterion two is satisfied. Criterion three is satisfied because the doctor’s reason for withdrawing treatment essentially involves the belief that unnecessary suffering for P should be prevented. Thus Steinbock’s second case is an act of withdrawal clementhanasia.
In a similar way my definition of clementhanasia easily handles the oft-discussed case of the doctor who gives accelerated doses of morphine to a terminally ill patient in order to "keep him comfortable." Here, the doctor gives repeated heavy doses of morphine in the full knowledge that in addition to preventing pain, the morphine will hasten P’s death by suppressing the patient’s breathing. While this doctor can sincerely claim that the morphine doses were only intended to secure the patient’s comfort, and not to directly cause the patient’s death, this doctor’s action easily meets all three criteria for active clementhanasia. It does not meet the typical criteria for euthanasia, since euthanasia is widely understood to require both that the doctor use the patient’s death to prevent suffering, and that the doctor primarily intends by the action to kill the patient. In this morphine case, the doctor is using the morphine to relieve the suffering, not to kill the patient. While morphine also acts as a lethal agent, killing the patient hours later, the doctor can sincerely claim that this death was but a foreseen but unintended consequence of the morphine doses. The definition of clementhanasia reveals some essential moral questions which euthanasia obscures: are there any circumstances in which a shortened life is a morally acceptable trade-off for less suffering, and if so, when in those circumstances does another person has a moral duty to shorten someone’s life to prevent her suffering?
IV. Understanding the AMA’s Stand on Euthanasia
In this paper, many sorts of actions that have at one time or another been taken to be acts of euthanasia are re-sorted into three categories: clementhanasia, cosmothanasia, and imperothanasia. For convenience, let us set down their definitions together.
Person Q performs an act of clementhanasia A on person P if and only if
Person Q allows nature to perform cosmothanasia on person P if and only if
Person Q performs an act of imperothanasia A on person P if and only if
The AMA’s stand on euthanasia can now be rephrased using these terms. The AMA rejects infusio-clementhanasia, while apparently approving in some cases withdrawal clementhanasia, cosmothanasia, and imperothanasia. I say "apparently," because by the AMA’s statement policy, these approved cases must concern only a patient whose biological death is immanent. In order for the AMA to avoid Rachels’s argument that by permitting only withdrawal it fails to consistently show adequate concern for patient suffering, it must simply point out that Rachels has falsely assumed that it is in any case primarily motivated by patient suffering. Steinbock fails to show that the AMA does sometimes encourage concern for patient suffering, because her second case rashly assumes (falsely) that such withdrawal is not a homicide. Without that case, Steinbock can only defend the AMA from Rachels’s charge of inconsistency by raising a case of imperothanasia, but in such a case the doctor is not asked to be primarily motivated by a concern for the patient’s suffering, but only by a respect for the patient’s legal right to refuse medical treatment. Nor can the AMA’s approval of cosmothanasia be construed as a primary concern for patient suffering, since such a concern would morally justify the immediate relief of infusio-clementhanasia, not the lingering suffering of cosmothanasia. While the AMA does not have an inconsistent position on the priority of preventing unnecessary suffering, it does contradict itself by authorizing homicide by withdrawal while forbidding homicide by more active means, like the lethal injection. The AMA’s only logical recourse, if it wishes to forbid doctors to perform homicides, is to forbid withdrawal as well. Such an explicit stand against both clementhanasia and imperothanasia would relieve the AMA from logical contradiction, but at the heavy price of exposing its fundamental doctrine that any amount of suffering is always medically justified by extended life.
In conclusion then, the AMA does not believe that the medical profession should place the efficient relief of patient suffering ahead of the priority of extending life. It does permit imperothanasia only because it has been legally required to do so, as the result of recent legal battles over patient’s rights. If the medical profession wants to preserve its 20th century obsession with postponing death at all costs, and if its definition of medical care cannot include the efficient prevention of suffering when amelioration or cure is no longer possible, then perhaps the euthanasia debate should now shift to a different question: under what circumstances should a patient be released from medical care to the care of professionals trained to perform clementhanasia?
NOTES
1. James Rachels, "Active and Passive Euthanasia," New England Journal of Medicine 292 (1975): 78-80. References are to its reprinting in Killing and Letting Die, 2nd ed., edited by Bonnie Steinbock and Alastair Norcross (New York: Fordham University Press, 1994), pp. 112-119. Bonnie Steinbock, "The Intentional Termination of Life," Ethics in Science and Medicine 6 (1979): 59-64. References are to its reprinting in Killing and Letting Die, 2nd ed., pp. 120-130.
2. Rachels, pp. 112-113.
3. Steinbock, p. 124.