1 I have chosen the term “infanticide” because this word (like “homicide“) connotes an act about which one ought to have serious reservations in almost all circumstances, but which may sometimes be considered justifiable.

2 See M. Tooley, Abortion and Infanticide 315-22 (1983); R. Weir, Selective Nontreatment of Handicapped Newborns 5-16 (1984); Rosenblum Budde, Historical and Cultural Considerations of Infanticide, in Infanticide and the Handicapped Newborn 1, 1 (D., Horan M., Delahoyde eds. 1982)Google Scholar [hereinafter cited as Infanticide]. Infanticide has not disappeared in all parts of the world. See, e.g., Chicago Sun-Times, Feb. 19, 1985 (5-star ed.), at 6 (reporting that female infanticide is widely practiced as a result of state population-control policies in Communist China).

3 Obviously, economic pressures can also lead to infanticide in less necessitous situations. For example, parents who calculate that preserving a handicapped child is not worthwhile (perhaps, because it will reduce the chance that the family will be able to send other children to college) are responding to economic pressures.

4 See R., Weir, supra note 2, at 18–20;Google Scholar Rosenblum Budde, supra note 2, at 1-3.

5 Many congenital defects can now be diagnosed in utero, and the birth of children suffering from these problems can be prevented by means of abortion. See Horan, Valentine, , The Doctor's Dilemma: Euthanasia, Wrongful Life, and the Handicapped Newborn, in Infanticide, supra note 2, at 33.Google Scholar Whether this is properly characterized as a form of “birth control“ and whether it is compatible with traditional medical ethics and general moral principles are debatable questions, but the law currently treats this practice as a legitimate one. See generally L., Kass, Perfect Babies: Prenatal Diagnosis and the Equal Right to Life in Toward a More Natural Science: Biology and Human Affirs 80 (1985).Google Scholar

6 Traditional forms of infanticide sometimes involved more or less active measures, for example, suffocation and strangulation. Often, however, people have sought to avoid such direct participation in the child's death; this apparently accounts for the prevalence of more “passive” means of inducing death, such as, exposure to the elements, “accidental” suffocation that occurred when the child and parent shared a bed at night, and arrangements whereby third parties (often servants or midwives) tacitly agreed to dispose of the unwanted child. The distinctions between “active” and “passive” infanticide are not compelling, since the intent and the result are the same in both—indeed, the attempt to disguise the practice by using “passive“ or “indirect” devices is itself probative evidence suggesting that it is regarded as morally wrong. For further criticism of the distinction between “active” and “passive” euthanasia, see Capron, , Borrowed Lessons: The Role of Ethical Distinctions in Framing Law on Life-Sustaining Treatment, 1984 Ariz. St. L.J. 647, 650;Google Scholar Rachels, , Active and Passive Euthanasia, 292 New Eng.J. Med. 78 (1975).Google Scholar In this Article, I will assume that deliberately withholding the necessities of life is fundamentally equivalent to more violent means of inducing death.

7 It is estimated that in the United States, about 100 infants are killed each year. P. Hoffer N. Hall, Murdering Mothers 161 (1981). By comparison, there were over 22,000 homicides in the United States during 1981. Federal Bureau of Investigation, U.S. Dep't of Justice, Uniform Crime Reports 6 (1981).

8 Poverty in the traditional sense, for example, has almost disappeared from the United States, where the government provides food, shelter, and basic medical care to virtually everyone who is willing to accept it. This, however, has raised to new prominence the phenomenon of “relative poverty” (i.e. inequality in the possession of wealth), a condition that many people seem to find more intolerable as they become objectively more wealthy.

9 Child Abuse Amendments of 1984, Pub. L. No. 98-457, 98 Stat. 1749-55 (1984) (to be codified at 42 U.S.C. § § 5101-05).

10 Child Abuse Prevention and Treatment Act, Pub. L. No. 93-247, 88 Stat. 4 (1974) (codified as amended at 42 U.S.C. § 5101 (1983 Supp. 1985)).

11 Child Abuse Prevention and Treatment and Adoption Reform Act, Pub. L. No. 95-266, 92 Stat. 205 (1978) (current version at 42 U.S.C. § 5101 (1982)); Omnibus Budget Reconciliation Act of 1981, Pub. L. No. 97-35, 95 Stat. 357, 488-89 (1981).

12 Appropriations have ranged from $4,500,000 to $22,928,000; authorizations have been considerably higher, as is typical with federal programs. See S. REP. NO. 246, 98th Cong., 1st Sess. 3 (Table 1) (1983).

13 Rehabilitation Act of 1973, Pub. L. No. 93-112, 87 Stat. 355 (codified as amended at 29 U.S.C. §§701-794(1982)).

14 § 504 (29 U.S.C. § 794 (1982)) provides: No otherwise qualified handicapped individual in the United States … shall, solely by reason of his handicap, be excluded from the participation in, be denied the benefits of, or be subjected to discrimination under any program or activity receiving Federal financial assistance.

15 United States v. University Hospital, 729 F.2d 144 (2d Cir. 1984).

16 410 U.S. 113 (1973).

17 Id. at 162-63.

18 See, e.g., P. Ramsey, Ethics at the Edges of Life 190 (1978); Ramsey, , Reference Points in Deciding about Abortion, in the Morality of Abortion 60, 79–86Google Scholar (J. Noonan ed. 1970); M., Tooley, supra note 2, at 323-24;Google Scholar Tooley, , A Defense of Abortion and Infanticide, in The Problem of Abortion 51 (J., Feinberg ed. 1973).Google Scholar

19 Roe v. Wade, 410 U.S. at 152-54.

20 Traditionally, medicine has been devoted to the health of the patient and therefore to preventing and curing disease and injury. The use of medical tools to interfere with normal biological processes (whether overtly for the patient's convenience, as when a woman has a normal breast removed because it interferes with her golf swing, or under a specious “mental health” rationale) requires considerable stretching of that traditional notion. For a thorough analysis of this problem, see Kass, , Regarding the End of Medicine and the Pursuit of Health, 40 Pub. Interest 11 (1975)Google Scholar (Kass discusses the example of the golfer with the inconvenient breast at 13). See also Kass, , Ethical Dilemmas in the Care of the III, 244 J.A.M.A. 1811, 1814-15 (1980).Google Scholar

21 During the 1984 presidential debates, for example, Walter Mondale used the phrase “the most personal decisions in American life,” apparently as a euphemism for “abortion.“ N.Y. Times, Oct. 8, 1984, at B5 (transcript of televised presidential campaign debate of Oct. 7, 1984). See also id. (Mr. Mondale referring to abortion as something that concerns “the private lives of the American people;” referring to abortion regulations as “[government… reaching into your living rooms;” and claiming that “[i]n America, on basic moral questions we have always let the people decide in their personal lives“).

22 Cf. infra notes 56-64 and accompanying text.

23 See, e.g., Roe v. Wade, 410 U.S. at 153 (discussing “the problem of bringing a child into a family already unable,psychologically and otherwise, to care for it” (emphasis added)); id. at 163 (discussing the importance of a “meaningful life“). In Justice Blackmun's opinion, as in the debates over the new infanticide, there is considerable ambiguity about whose welfare (the child's or its parents’) is being considered.

24 This issue is discussed infra at notes 56-85 and accompanying text.

25 Duff, Campbell, , Moral and Ethical Dilemmas in the Special-Care Nursery, 289 New Eno. J. Med. 890 (1973).Google Scholar

26 Id. at 892, 894.

27 Immediately after the article appeared, two other physicians from Duff and Campbell's own newborn special care unit publicly dissociated themselves from the nontreatment policy, which they characterized as a “therapy of nihilism.” Venes Huttenlocher, Letter to the Editor, 290 New Eng. J. Med. 518 (1974).

28 The literature on the subject is now enormous. For a brief summary of the principal contending positions, see R., Weir, supra note 2, at 60–90.Google Scholar

29 Several recent surveys indicate that a majority of pediatricians would withhold treatment in cases similar to the Bloomington “Baby Doe” case, discussed infra Part VI, that provoked widespread outrage in the public and Congress. See Child Abuse Prevention and Treatment and Adoption Reform Act Amendments of 1983: Hearings Before the Subcommittee on Family and Human Services of the Senate Committee on Labor and Human Resources, 98th Cong., lstSess. 6, 8 (1983) (testimony of the Surgeon General of the United States) [hereinafter cited as Senate Hearings﹜. Older survey data suggest that the practice of withholding treatment from handicapped newborns began long before it came to public attention. See Ellis, , Letting Defective Babies Die: Who Decides? 7 Am. J.L. Med. 393, 399 Be n. 16–19 (1982).Google ScholarPubMed

30 Note, Life or Death in the Intensive Care Nursery: Who Should Decide?, 11 Wm. Mitchell L. Rev. 127, 128 n.2 (1985).

31 Karen Quinlan was 22 years old when she became comatose and was being kept alive only (or so her doctors thought) with the help of a respirator. Her father, invoking several putative federal constitutional rights, asked the New Jersey courts to appoint him her legal guardian with express authority to terminate use of the respirator; this request was opposed by Karen's doctors, the hospital, the county prosecutor, the State of New Jersey, and her guardian ad litem. The New Jersey Supreme Court ruled that Karen enjoyed a constitutional right to “privacy” that might permit termination of treatment. The court appointed her father her guardian, set up certain procedures to be followed by the physicians and hospital before the respirator could be disconnected, and immunized all participants from possible legal liability so long as the specified procedures were followed. The standard specified for determining the course of treatment was one of “substituted judgment,” that is, an analysis of what Karen would probably have chosen had she been capable of making a decision. In re Quinlan, 70 N.J. 10, 355 A.2d 647, cert, denied, 429 U.S. 922 (1976). When the doctors disconnected the respirator, Karen Quinlan failed to die as she had been expected to do. In fact, she lived until June 11, 1985. N.Y. Times, June 12, 1985, at 12.

32 Joseph Saikewicz was 67 years old and suffering from terminal cancer. Because he was extremely retarded (with a mental age of less than 3 years), he could not make an informed decision about whether to undergo painful chemotherapy treatments. The superintendent of the state institution that had custody of him petitioned the Massachusetts courts to appoint a guardian with authority to make the necessary decisions. The Supreme Judicial Court cited Quinlan for the proposition that a constitutional right to privacy could be asserted on behalf of an incompetent patient, but criticized the Quinlan court for leaving too much discretion to physicians. The court performed its own “substituted judgment” analysis and decided that the best interests of the patient required that chemotherapy treatment be withheld. Superintendant v. Saikewicz, 378 Mass. 728, 370 N.E.2d 417 (1977).

33 See, e.g., P., Ramsey, supra note 18, at 190;Google Scholar Ramsey, , Reference Points in Deciding about Abortion in the Morality of Abortion 60 (J., Noonan ed. 1970);Google Scholar Tooley, , A Defense of Abortion and Infanticide, in the Problem of Abortion 51 (J., Feinberg ed. 1973).Google Scholar

34 In Roe v. Wade, the Supreme Court seized upon the concept of fetal “viability” (i.e. ability to survive outside the mother's womb) in an effort to draw a principled line between constitutionally protected abortions and those that the states could regulate. 410 U.S. at 163. The Court failed to explain why this should be a legally meaningful distinction, except to note that the fetus is then “presumably” capable of “meaningful life” outside the mother's body. Id. Moreover, the Court seemed to assume that “viability” occurred around the beginning of the third trimester and that it could certainly never occur during the first trimester, id., an assumption that now seems quite unwarranted. See Akron, v. Akron Center for Reproductive Health, 462 U.S. 416, 457 (1983)Google Scholar (O'Connor, J. dissenting) (“[F]etal viability in the first trimester of pregnancy may be possible in the not too distant future.“). Thus, the Court's rule is inconsistent with the rationale that purports to underlie it. See id. at 457 (the Roe framework “is clearly on a collision course with itself“). Recent developments with in vitro fertilization and artificial placentas (i.e. “test-tube” babies) suggest that fetuses may eventually be “viable” in the Roe sense virtually from the moment of conception. See generally Annas, Elias, , In Vitro Fertilization and Embryo Transfer: Medicolegal Aspects of New Techniques to Create a Family, 17 Fam. L.Q. 199, 203-06, 207-10 (1983);Google Scholar Grobstein, Flower Mendeloff, External Human Fertilization: An Evaluation of Policy, 222 Science 127 (1983); Kass, , “Making Babies” Revisited, 54 Pub. Interest 32, 35–40 (1979)Google Scholar (discussing the meaning of “viability” in relation torn vitro fertilization). Once this occurs, the entire Roe framework will completely disintegrate.

It is not surprising that modern science has failed to develop a meaningful distinction between fetuses and young children. While birth itself is a significant event in the life of the individual, it can occur at any time during a fairly extended period of continuous development. The particular stage of development at which birth typically occurs varies widely from species to species, and evolutionary biology suggests that the “embryonic” stage in humans properly extends through about the first year of postnatal life. See S., Gould, Human Babies as Embryos, in Ever Since Darwin 70 (1977).Google Scholar

35 The Oath provides in relevant part: “I will neither give a deadly drug to anybody if asked for it nor will I make a suggestion to this effect. Similarly, I will not give to a woman an abortive remedy.“

36 For a detailed explication of the coherence of the Oath and an explanation of why it has been so durable, see L., Kass, IS There a Medical Ethic1?: The Hippocratic Oath and the Sources of Ethical Medicine, in Toward a More Natural Science: Biology and Human Affirs 224 (1985).Google Scholar

37 Neither Roe v. Wade nor any of its progeny compels any physician to perform an abortion. In this sense, there is no constitutional right to an abortion. See Maher v. Roe, 432 U.S. 464, 470-71 (1977). The Supreme Court's abortion decisions therefore do not compel the medical profession to abandon the Hippocratic Oath. In light of the internal problems in the Court's abortion jurisprudence, the persuasive value of these decisions is also quite limited. Cf. Brief for the United States as Amicus Curiae, Thornburgh v. American College of Obstetricians and Gynecologists (U.S. Nos. 84-495 and 84-1379); supra note 34.

38 The Oath provides: “I will apply dietetic measures for the benefit of the sick according to my ability and judgment…. Into whatever houses I may enter, I will come for the benefit of the sick ….“

To claim that physicians have traditionally aspired to benefit their patients is not to imply that it has always been crystal clear what the patient's good entailed. As the Quinlan and Saikewicz cases, discussed supra notes 31-32, so vividly remind us, the patient himself has almost always been permitted to choose to forego painful or “heroic” treatments.

A somewhat different problem arises when the physician is confronted with multiple patients whose interests conflict. This dilemma has produced the general and traditional rule permitting an abortion if medically necessary to save the life of the mother. It has also produced the triage system discussed infra note 77.

39 See infra notes 57-85 and accompanying text.

40 Physicians, especially at university hospitals, may sometimes be tempted to apply treatments that have an extremely small chance of success, in hopes of thereby increasing medical knowledge for the benefit of future patients (and perhaps also for the benefit of the physician's own prestige and professional advancement). See Goichman, , A Critique of Louisiana's Approach to Withholding Medical Treatment from Defective Newborns, 9 S.U.L. Rev. 157, 160 (1983).Google Scholar The problems associated with the use of subjects who have not consented to certain treatments in medical experiments present an interesting topic in their own right, but will not be pursued in this Article.

41 There may, however, be a hidden demand for the right to eliminate children who have these characteristics. In other times and places, infanticide of females and bastards has been especially common. See R., Weir, supra note 2, at 18–21.Google Scholar In our own culture, there is anecdotal evidence indicating that abortion is frequently used for just these reasons, though statistics would be extremely difficult to compile since women do not have to disclose their reasons for demanding the procedure. Sec L., Kass, supra note 5, at 95;Google Scholar see also Newsweek, Jan. 14, 1985 at 29 (reporting that women are said to have had abortions because their fetuses were of the wrong sex). We do know, however, that in 1979 approximately 30% of pregnant women chose abortion. Henshaw, Forrest, Sullian & Tietze, Abortion in the United States, 1978-1979, 13 Family Planning Perspectives 6 Jan.-Feb. 1981. Thus, if infanticide were granted the same legal status as abortion, we might see it used—like abortion—as part of a kind of private eugenics programs. It is extremely unlikely that the medical profession would resist such a development. One survey suggests that, even under the present laws, 8% of the surgeons sampled would acquiesce in a parent's decision to refuse an otherwise healthy child treatment for a simple intestinal blockage, though such operations are virtually always successful and the patient lives a completely normal life afterwards. Koop, Ethical and Surgical Considerations in the Care of the Newborn with Congenital Abnormalities, in Infanticide and the Handicapped Newborn 98 (D. Horan M. Delahoyde eds. 1982).

42 See, e.g., Milligan, Shennan, , Perinatal Intensive Care: Where and How to Draw the Line, 148 Obstetrics and Gynecology 499–503 (1984).Google Scholar

43 Low weight at birth is in some ways generally a better predictor of medical problems than time spent in utero. The distinction is not important for my discussion except insofar as one recognizes that neither predictor is perfect. See infra notes 44-46 and accompanying text.

44 For babies weighing 500-800 grams, for example, about 20% survive and 15% of the survivors suffer from major neurological handicaps. See Fitzhardinge, Bennet-Britton, , Is Intensive Care Justified for Infants Less Than 800 Grams at Birth?, 14 Pediatrics Res. 590 (1980).Google Scholar Vigorous treatment, however, can affect the mortality rate considerably. See Strong, The Tiniest Newborns, Hastings Center Rep. Feb. 1983, at 14, 15-16 (citing study indicating that, for infants weighing 600-799 grams at birth, replacing “conservative” treatment with “aggressive“ treatment raised survival rates from 5% to 52%). Costs of caring for these children can be as much as five times greater than for larger premature babies. See McCarthy, , Koop, , Honeyfield, Butterfield, , Who Pays the Bill for Neonatal Intensive Care?, 95 J. Pediatrics 755, 758 (1979).Google Scholar

45 Cf. Meyer, , Protecting the Best Interests of the Child: Is the State the Necessary Blunt Instrument?, 1984 Ariz. St. L J. 627, 634Google Scholar (suggesting that devices such as the “Baby Doe Hotline“ “may be expected to result in defensive medical practice that may, however, unintended, prolong futile and painful treatments“).

46 See R., Weir, supra note 2, at 44–45;Google Scholar Paris, McCormick, , Saving Defective Infant's Options for Life and Death, 1983 Am. 313, 313;Google Scholar Senate Hearings, supra note 29, at 144.

47 See survey data cited supra note 29; Todres, infra note 50.Google Scholar Note also that victims of Down's Syndrome are a principal target of prenatal diagnosis (mainly through amniocentesis) and abortion; cf. supra note 5.

48 See, e.g., Hatcher, , Economic Grounds No Criteria, 5 BRit. Med. J. 285, 285 (1973)Google Scholar (“I have always sensed that attempts to preserve life by surgery are more likely to be asked for [by parents] if the prospects for the child's intelligence are good.“).

49 This was true in the Baby Doe case discussed infra notes 88-93 and accompanying text. There is presently a waiting list for the adoption of children with Down's Syndrome. Tedeschi, , Infanticide and its Apologists, 1984 Commentary 31, 31.Google Scholar

50 See, e.g., Todres, , Pediatricians’ Attitudes Affecting Decision-Making in Defective Newborns, 60 Pediatrics 197 (1977).Google Scholar

51 See text accompanying notes 68-82 infra for a discussion of the theory characterized by the phrase “quality of life.“

52 For a thorough, relatively nontechnical discussion of spina bifida, see Swinyard, , Chaube, Nishimura, , Spina Bifida as a Prototype Defect for Decision Making: Nature of the Defect, in Decision Making and the Defective Newborn 17 (C., Swinyard ed. 1978).Google Scholar

53 If all patients were treated vigorously, at least half would survive. See Lorber, , Results of Treatment of Myelomeningocele, 13 Dev. Med. Child Neurology 279 (1971).Google Scholar

54 As in the case of Down's Syndrome, it is not possible to predict accurately at birth the degree of a specific patient's handicaps; nor is it possible to predict the extent to which medical treatment early in life will prevent later medical problems. See Soare, Raimondi, , Quality of Survival in Treated Myelomeningocele Children, in Decision Making and the Defective Newborn 68 (C., Swinyard ed. 1978).Google Scholar

55 Sec Discussion, 5 Brit. Med. J., 286, 286 (1973) mortality rate for untreated children with myelomeningocele is 50% at 6 months and 90% at 1 year). Physicians have since achieved higher mortality rates through more sophisticated selection techniques, but they still have not reached 100%. See Lorber, , Selective Treatment of Myelomeningocele: To Treat or Not to Treat, 53 Pediatrics 307, 308 (1974).Google ScholarPubMed

56 The draftsman of the new law may have been aware of these differences, and may have consciously decided that the best way to take them into account was to use broad and vague language. In any event, they did erect a rather vague legal standard. See the analysis of the law, infra Part VII, and particularly the definition of “withholding of medically indicated treatment” quoted infra note 110, which at the crucial points uses the phrase “physician's and physicians’ reasonable medical judgment.“

For a critique of the new law, arguing that it is so vague that it may turn out to be unconstitutional as applied, see Vitiello, , The Baby Jane Doe Litigation and Section 504: An Exercise in Raw Executive Power, 17 Conn. L. Rev. 95, 154-59 (1984).Google Scholar

57 The homicide laws, for example, pretty clearly prohibit the use of lethal drugs to hasten the death of children who have been selected for nontreatment. Even here, however, physicians seem to have found ways of evading legal proscriptions: there are hints in the literature of cases of overdoses of sedatives and analgesics and of stratagems such as heavy sedation accompanied by a policy of feeding only “on demand.” See, e.g., Freeman, , To Treat or Not to Treat: Ethical Dilemmas of Treating the Infant with a Myleomeningocele, 20 Clinical Neurosurgery 134, 141 (1972).Google Scholar Compare the discussion of prosecutorial discretion, infra note 114. See also infra note 78.

58 For further discussion of this point, see infra notes 59, 63-64.

59 Cf. Ellis, , supra note 29, at 414Google Scholar (in the perinatal period, “[p]arent-child bonding may not yet be complete, and the parental love assumed by society to exist in other contexts may not yet have developed“).

60 The special child abuse laws of the present time merely supplement the common law doctrine that imposed a duty of care on the parents of minor children. Every state has statutes that specifically impose upon parents the duty to provide necessary medical care. See Robertson, , Involuntary Euthanasia of Defective Newborns, 27 Stan. L. Rev. 213, 218 (1975).Google Scholar

61 Something akin to self-interest is perhaps involved. To the extent that physicians' training and professional outlook encourage them to pursue the health or biological fitness of their patients, the severely handicapped stymie and frustrate their professional goals. A kind of resentment towards this revelation of professional impotence may account in part for doctors’ zeal in encouraging genetic screening and eugenic abortion. Cf. Horan, Valentine, supra note 5;Google Scholar infra notes 62, 81-84 and accompanying text. On the other hand, the desire to exercise all possible professional skill to defeat the disease may also emerge.

62 Dr. Leon Kass has argued that there is a clear set of ethical principles flowing from the nature of medicine itself and has suggested that for a long time this offered an appropriate substitute for detailed legal regulation of physicians’ conduct. See L., Kass, supra note 36;Google Scholar Kass, , Professing Ethically, 249 J.A.M.A. 1305 (1983).Google Scholar Note, however, that Kass acknowledges that medicine, like all human pursuits, is and must be subject to the control of the political authorities. His principal concern seems to be with helping the profession provide itself with a set of principles that will minimize the need for direct government interference in the details of medical practice.

Some commentators have suggested relying on the decisions of hospital “ethics committees.“ See, e.g., Note, Withholding Treatment from Birth-Defective Newborns: The Search for an Elusive Standard, 31 WAYNE ST. L. Rev. 187 (1984). There is little reason, however, to suppose that such committees would have any special moral competence beyond that of the individual physicians who would compose them:

Fost, Baby Doe: Problems and Solutions, 1984 Ariz. St. L.J. 637, 641; see also Longino, , Withholding Treatment from Defective Newborns: Who Decides, and on What Criteria?, 31 Kan. L. Rev. 377, 403 (1983)Google Scholar (“[Ethics committees] can easily become ‘bureaucratized’ and may render decisions of no greater quality than those of the physician and family.“).

63 But see Angell, , Handicapped Children: Baby Doe and Uncle Sam, 309 N. Eng. J. Med. 659, 660 (1983)Google Scholar (because any set of general rules would be “insensitive and vague when applied to a particular patient,” government interference with physicians’ discretion is “both arrogant and foolish“). Dr. Angell is not the first to observe that general rules will not do justice in all the particular cases to which they apply. See, e.g., Aristotle, Nicomachean Ethics 1137 b 10-25. Aristotle, however, did not draw the conclusion that legal rules should be abolished.

One commentator has offered an even more radical assertion to justify government nonintervention in treatment decisions: “[H]ow can parents in such situations give the wrong answer since there is no way of knowing the right answer?” Goldstein, , Medical Care for the Child at Risk: On State Supervention of Parental Autonomy, 86 Yale L.J. 645, 655 (1977).Google Scholar Taken seriously, this suggestion, like Dr. Angell's, would undermine the rule of law itself.

64 Physicians tend, naturally enough, to prefer self-regulation to government “interference“ in their lives. So, probably, does everyone else, but that is not a sufficient reason for allowing self-regulation. Physicians have also worried, with some justification, that increased legal regulation will have bad side-effects, especially if courts and bureaucrats begin making medical decisions under the guise of enforcing the ethical principles of the law. This same concern, however, is also appropriate in many other fields in which the law seeks to curtail the misuse of power by experts who justifiably assume that courts and bureaucrats tend to have an insufficient appreciation of the nature of the day-to-day problems they face in their work. Conspicuous examples include policemen, prison wardens, and businessmen in their relations with their labor force. The natural clumsiness of the law is a factor that should be considered when deciding how to regulate and how much to regulate, but it is not by itself a reason to give unfettered discretion to the possessors of technical expertise.

65 It may not be superfluous to note that while the law has always taken a position and must take one today, it is also reasonable for the law to take a position. Among the strongest justifications for favoring republican government in general, and the American system of constitutional democracy in particular, is its ability to protect the rights of minorities (or the weak generally) against the contrary interests of the majority (or those who can manipulate and direct the majority) without thereby converting the protected minority itself into an oligarchy. See The Federalist No. 10 (J. Madison). This is what accounts for our remarkable panoply of protective civil rights and civil liberties laws; it would be strange to suggest that handicapped children are not the sort of weak and vulnerable minority that should be entitled to legal protection for their rights, however those rights might finally be defined.

66 For brief presentations of Koop's views, see Koop, , Ethical and Surgical Considerations in the Care of the Newborn with Congenital Abnormalities, in Infanticide and the Handicapped Newborn 89 (D., Horan M., Delahoyde eds. 1982);Google Scholar Senate Hearings, supra note 29, at 6-21 (statement of Everett Koop).

67 It should be noted, however, that it is easy to overstate the inevitability of the suffering that defective children impose upon their families. Adoption, as in the Bloomington case, is often available. Furthermore, most states have procedures by which parents can divest themselves of their legal obligations to their offspring, even when there are no offers of adoptions. See Robertson, , Dilemma in Danville, 11 Hastings Center Rep. 515 (1981).Google Scholar

68 There is now a large literature on this subject. Some of it consists of attempts to develop and refine criteria for deciding which children will be selected for nontreatment; this literature usually ignores the underlying questions about the propriety of developing such criteria and is mainly technical in nature. Another large portion of the literature is purely negative, and consists of attacks on the “sanctity of life” position (or some caricature thereof), usually by describing extremely pitiful patients in a way that implies that the infants are no good to themselves and a positive harm to others; the conclusion (either expressed or implied) is that some alternative must be found, but the exact nature of that alternative is often left extremely vague or uncertain.

69 In this Article, I will focus on what seem to be the most moderate and well thought out versions of the quality of life analysis. It should be noted, however, that some prominent commentators are much more radical:

Singer, , Sanctity of Life or Quality of Life, 72 Pediatrics 128, 129 (1983);Google ScholarPubMed see also The Hardest Question, Newsweek, Jan. 14, 1985, at 29 (reporting that Prof. Virginia Abernathy of Vanderbilt Medical School believes that until the age of three or four, infants are “nonpersons” whose interests are always subordinate to the rights of “persons“).

Even more moderate advocates of the “quality of life” position seem to have trouble formulating reasonable general positions. John Lorber, for example, suggests that one precondition of a meaningful life is the ability to earn one's “own living in competitive employment and be self-supporting with a secure, independent place in society.” Shaw, , Spina Bifida Cystica: Results of Treatment of 270 Consecutive Cases with Criteria for Selection for the Future, 47 Archives of Disease in Childhood 854 (1972).Google Scholar If this criterion were applied to the adult population, it would seem to suggest the need for a large scale euthanasia program.

70 See, e.g., P. Stinson R. Stinson, The Long Dying of Baby Andrew (1983); Lorber, , Dilemmas of “Informed Consent” in Children, 289 New Eng. J. Med. 885, 887-88 (1973);Google Scholar Portela, , The Elin Daniels Case: An Examination of the Legal, Medical, and Ethical Considerations Posed When Parents and Doctors Disagree on Whether to Treat a Defective Newborn, 18 Forum 709 (1983)Google Scholar (describing a case in which medical treatment was performed, pursuant to court order, over parents’ objections; it turned out that the child was not, as had been feared, mentally retarded).

71 See, e.g., Baer, , Should Imperfect Infants Survivel?, 35 Nat'l Rev. 1069, 1092 (1983)Google Scholar (“One boy, for example, whose indignant parents refused Dr. Duff's fnontreatment] prescription when he was born is now an A student. Dr. Duff had said the child would be a ‘vegetable.'“).

72 See, e.g., Freeman, , supra note 57, at 143-44Google Scholar (“[T]he family has rights, just as the affected child has rights … the ‘best’ for everyone concerned may be the death of the child“).

73 See, e.g., Senate Hearings, supra note 29, at 138 (testimony of Mrs. Karen Green-McGowan, R.N.) (“The physician often comes in wanting to rescue the family, saying, if I were in your place, gee, I wouldn't want a child like that.“).

74 See, e.g., Fost and Cranford, Hospital Ethics Committees: Administrative Aspects, 253 J.A.M.A. 2687-91 (1985).

75 Shaw, , Defining the Quality of Life, Hastings Center Rep., Oct. 1977, at 11.Google Scholar

76 Shaw himself disclaims any intent to use the formula to calculate “the numerical value of human life.” Indeed, he seems to recognize that there may be something problematic about the formula when he points out that for a physically normal child born to an unwed drugaddict mother, “H” and “S” could both be zero, so that the baby's “QL” would be the same as that of an extremely defective newborn, namely zero. Shaw, however, declines to draw any inference from this apparent parallel except to note that a person's quality of life “may be determined to a significant degree by factors physicians frequently fail to consider.” Id.

77 Shaw's formula should be contrasted with the use of the medical “triage system,” to which it has a certain misleading similarity. The triage system is used by doctors in emergency situations where there is a shortage of medical resources, for example in battle or after major natural disasters such as earthquakes. Under that system, patients are divided into three groups: those who will probably survive even without immediate attention, those who will probably not survive even with immediate attention, and those for whom immediate attention will probably make the difference between life and death. First priority is given to patients in the last group, as a combination of common sense and the principle of the sanctity of life quite clearly dictate. Even in this situation, where reasonably objective (though of course fallible) medical judgment is exercised, doctors have typically felt that they were being compelled only by necessity to make decisions that were in sharp tension with their moral and professional inclinations; they have been worried by the prospect of making errors in classifying the patients and have sought (for example, by demanding that additional medical resources be supplied as quickly as possible) to avoid being compelled to make such decisions.

The use of Shaw's formula, or “quality of life” criteria in general, is very different from the triage system. It requires the use of far less objective criteria, which is troubling enough in itself. More important, however, it is not imposed on doctors as the least bad alternative in the face of a scarcity of resources. Rather, it requires that they embrace the opportunity to allow some individuals to die according to criteria selected by the physician (or by whomever is charged with calculating the patient's expected “quality of life“).

78 The New Jersey Supreme Court, which decided the Quinlan case, see supra note 31, recently adopted a test that seems to incorporate a “quality of life” criterion in this sense. At least for elderly patients with severe mental and physical disabilities, life-prolonging treatment may be withheld, even without evidence that the patient himself would have refused the treatment, if “the net burdens of the patient's life with the treatment [w]ould ‘clearly and markedly outweigh’ the benefits that the patient derives from life.” Matter of Conroy, 98 N.J. 321, 486 A.2d 1209 (1985).

It should be noted that if one believes that it is sometimes in the patient's interest to be allowed to die because his life will be filled with more pain than he would wish to endure, one should find it hard to resist the conclusion that the process of dying should be made as swift and painless as possible. The law, however, clearly prohibits active euthanasia (e.g. by lethal injection); hence, a commitment to this position should require one to favor changes in the legal status quo. Cf. Harris, , Ethical Problems in the Management of Some Severely Handicapped Children, 7 J. Med. Ethics 117 (1981)Google Scholar (suggesting that active euthanasia is preferable to selective nontreatment); Freeman, , supra note 57, at 144Google Scholar (“[U]ntil active euthanasia, with whatever appropriate safeguards, becomes acceptable to-society.“); Parness, Stevenson, , Let Live and Let Die: Disabled Newborns and Contemporary Law, 37 U. Miam. L. Rev. 43, 54–60 (1982)Google Scholar (suggesting that a constitutional “right to die” be established on the basis of the Supreme Court's privacy decisions); Smith, , Life and Death Decisions in the Nursery: Standards and Procedures for Withholding Lifesaving Treatment from Infants, 27 N.Y.U. L. Rev. 1125, 1166-68 (1982)Google Scholar (advocating active euthanasia in some circumstances).

79 Shaw seems to recognize this when he distinguishes contributions to the individual by the home and society from contributions to the home and society by the individual. Indeed, one of Shaw's main points seems to be to discourage confusion between the “QL” of the individual and the “QL” of his family and society. Shaw is utterly ambiguous, however, on the question of whether it is proper for a physician to consider the quality of life of persons other than his patient, and he certainly does not propose a means by which decision makers could in practice convert “H” and “S” into truly independent variables. Shaw, , supra, note 75.Google Scholar

80 See supra notes 58-59 and accompanying text.

81 See supra notes 35-39 and accompanying text.

82 See, e.g., Eckstein, , The Problems of Selection, 5 Brit. Med. J., 284, 284 (1973)Google Scholar (“The parents’ opinion should be considered, but in practice very few parents express a strong opinion which is different from those of the medical attendants.“); Rickham, , The Ethics of Surgery on Newborn Infants, 8 Clinical Pediatrics 251, 252 (1969)Google Scholar (“At the end it is usually the doctor who has to decide the issue.“).

83 Eugenics programs can be divided into several types, but they all share the goal of improving the biological fitness of a population. Some are “positive” (programs to promote reproduction by individuals with superior genes) while others are “negative” (programs to prevent reproduction by individuals having inferior or defective genes). Some programs focus on eliminating certain genotypes (e.g. by genetic screening of prospective parents) while others require direct destruction of the unwanted phenotypes (e.g. the Nazis’ extermination programs at German mental hospitals).

The new infanticide is somewhat unusual in that most of its promoters are primarily and consciously concerned only with the improvement of very small populations (individual families) rather than with society at large. The aggregate effect, however, is the same as it would be if the specific intent to improve the overall gene pool were present. In any case, the crucial moral distinction among eugenics programs is between those that aim only to eliminate certain undesirable biological traits (e.g. congenital diseases) and those that destroy undesired individuals (e.g. those who suffer from congenital diseases). The new infanticide is in the latter category.

84 Sterilization programs have been carried on at various times in this country and have received the outspoken support of at least one highly respected American jurist. See Buck v. Bell, 274 U.S. 200, 207 (1927) (Holmes, J.) (“It is better for all the world, if instead of waiting to execute degenerate offspring for crime, or to let them starve for their imbecility, society can prevent those who are manifestly unfit from continuing their kind…. Three generations of imbeciles are enough.“). Sterilization, however, is significantly less drastic than the destruction of unwanted individuals.

85 See Alexander, , Medical Science under Dictatorship, 241 New Eng. J. Med. 39, 39, 44 (1949).Google Scholar

86 In its simplest version, the argumentum ad hitlerem takes the following form: Policy X was adopted by the Nazi regime; the Nazi regime was evil incarnate; we should therefore not adopt Policy X. The fallacy in this argument is obvious enough that it should not require explication. The argument sometimes takes a more sophisticated form: Policy X was adopted by the Nazis and we should reject the policy in order to avoid taking what could be the first step toward Nazism. In this form, the argumentum ad hitlerem is dangerous, but not completely invalid: it is always prudent to be wary of taking steps that could lead to undesirable unintended consequences, though one should also be wary of overestimating such dangers and of losing sight of the more immediate virtues of the proposed policy

A sober version of the slippery slope argument was used by Abraham Lincoln to demonstrate the untenability of the proslavery position:

II Collected Works of Abraham Linxoln 222 (R. Basler ed. 1953) (emphasis in original). Lincoln's argument applies, mutatis mutandis, with equal force to the claim that supposedly defective or inferior people should be killed.

87 The advocates of the quality of life argument for infanticide have themselves fell the need to worry about the possibility of Nazi-type consequences. See, e.g., Duff, Campbell, , supra note 25, at 892Google Scholar (mentioning Nazi Germany); id. at 894 (mentioning the need to avoid the “extreme excesses of ‘Hegelian rational utility’ under dictatorships“).

It should also be noted that the danger of a slippery slope is taken very seriously by some thinkers who are very far removed from the religious and politically conservative influences that manifestly affect many opponents of infanticide, l'rominant libertarian Nat Hentoff, for example, has asked: “If fetuses have no rights, handicapped infants have no rights, can the aged and infirm be far behind?” The Hardest Question, Newsweek, Jan. 14, 1985, at 29.

88 See supra notes 16-42 and accompanying text.

89 The case described below was not the first to become publicly known; there was a small flurry of publicity about similar cases in the early 1970's. See Ellis, , supra note 29, at 399–401.Google Scholar

90 Paris 8c McCormick, , supra note 46, at 313.Google Scholar

91 The decisions of the Indiana courts do not seem to have been published. My summary of the facts is based on the following sources: J. Lyon, Playing god in the Nursery 21-39 (1985); Senate Hearings, supra note 29, at 1-2 (statement of Sen. Denton); S. Rep. NO. 246, 98th Cong, 1st Sess. 5-6 (1983); Lyon, , The Death of Baby Doe, Chicago Tribune, Feb. 10, 1985,Google Scholar (Magazine), at 10; Tedeschi, , Infanticide and its Apologists, supra note 49, at 31;Google Scholar Comment, Withholding Treatment from Seriously III and Handicapped Infants: Who Should Make the Decision and How?—An Analysis of the Government's Response, 33 De Paul L. Rev. 495, 499-501 (1984); Note, Withholding Treatment from Defective Infants: “Infant Doe” Postmortem, 59 Notre Dam. L. Rev. 224, 234-35 (1983).

Ironically, Indiana was the only state in the Union at the time that had a statute specifically providing for criminal and civil liability where newborn children have been denied needed medical treatment. Ind. Code Ann. § 35-1-58.5-7 (West Supp. 1984-85). At least three other states have now responded to the problem with new legislation. See Kuzman, , The Legislative Response to Infant Doe, 59 Ind. L.J. 366, 400-05 (1984).Google Scholar

92 Doe v. Bloomington Hospital 464 U.S. 961 (1983).

93 Although the “Baby Doe” problem has received considerable attention in the media, there is some evidence that the coverage has been inaccurate, euphemistic, and generally biased in favor of parents and physicians who withhold food or medical treatment from handicapped children. See Tedeschi, , supra note 49.Google Scholar Despite whatever suppression of the truth has occurred, enough emerged to provoke both the President and the Congress to act.

94 Victims of Down's Syndrome suffer mental retardation that can be quite severe but is often rather mild. The degree of retardation cannot be predicted at birth. Paris McCormick, supra note 46, at 313.

95 The legislative history of the new law focuses heavily on the Bloomington case, which was clearly the catalyst for congressional action. See, e.g., Senate, Hearings, supra note 29, at 1–2Google Scholar (statement of Sen. Denton); S. Rep. No. 246, 98th Cong., 1st Sess. 5-6 (1983).

96 48 Fed. Reg. 9630(1983) (to be codified at 45 C.F.R. pt 84), (proposed Mar. 7, 1983).

97 American Academy of Pediatrics v. Heckler, 561 F. Supp. 395, 396 (D.D.C. 1983). Reviewing courts are given the authority, under the Administrative Procedure Act, to hold unlawful and set aside agency action that is found to be arbitrary, capricious, or an abuse of discretion. 5 U.S.C. § 706(2)(a) (1971).

98 45 C.F.R. § 84.61 (1982). For a detailed discussion of the regulatory activity preceding the Second Circuit's decision, discussed infra, see Lawton, , Carder, Weisman, , Recent Governmental Action Regarding the Treatment of Seriously 111 Newborns, 11 J. Coll. Univ. L. 405, 408-09 (1985).Google Scholar

99 United States v. University Hospital, 729 F.2d 144 (2nd Cir. 1984). The Court quoted. Gesell as follows: “The legislative history … [on this subject] focuses on discrimination against adults and other children and denial of access to federal programs. As far as can be determined, no congressional committee or member of the House or Senate ever even suggested that Section 504 would be used to monitor medical treatment of defective newborn infants or establish standards for preserving a particular quality of life.” Id. at 158 (quoting American Academy, 561 F. Supp. at 401). The majority opinion is somewhat ambiguous about exactly what kinds of situations are not covered by § 504. A strong dissent suggested that the majority's opinion employed such a tendentious reading of the legislative history that it constituted an “outright disagreement with Congress'judgment and an unconstitutional act in itself.” Id. at 163 (Winter J., dissenting). The Supreme Court will take up the issue presented in this case during the next term. See Heckler v. American Hospital Association, No. 84-1529 cert, granted, 105 S. Ct. 3475 (1985).

For a detailed history of the administration's attempt to address the Baby Doe problem through the regulatory process, see Note, The HHS’ Final Rule on Health Care for Handicapped Infants: Equal Protection Not Guaranteed, 11 J. Legis. 269 (1984).

100 Child Abuse Amendments of 1983, H.R. 1904, 98th Cong., 1st Sess. (1983).

101 Child Abuse Prevention and Treatment Act, 42 U.S.C. § 5101(a) (1974) (amended 1983 Supp. 1985). This law is discussed supra notes 10-12 and accompanying text.

102 The crucial provision of the Child Abuse Amendments of 1984, Pub L. No. 98-457, 98 Stat. 1749 (1984), as finally enacted, is § 122, which adds a clause ((4)(b)(2)(k)) to the existing Act. It requires that states receiving funds under the federal child abuse program:

103 To Amend the Child Abuse Prevention and Treatment and Adoption Reform Act of 1978: Hearing Before the Subcommittee on Select Education of the Committee on Education and Labor of the House of Representatives, 98th Cong., 1st Sess. 16, 38, 187-89, 336-43, 359-71 (1983).

104 H.R. Rep. NO. 159, 98th Cong., 1st Sess. (1983).

105 130 Cong. Rec. H423 (daily ed. Feb. 2, 1984).

106 S. 1003, 98th Cong., 1st Sess. (1983).

107 This provision was apparently intended to give HHS the authority that the Reagan administration had attempted to exercise under § 504 of the Rehabilitation Act. Though this did not survive to become part of the bill as enacted, the new law contains a “savings” clause (§ 127 (a)) that provides: No provision of this Act or any amendment made by this Act is intended to affect any right or protection under section 504 of the Rehabilitation Act of 1973.

108 See Senate Hearings, supra note 29, at 47-60, 111-17.

109 Tiiecrucjal provision (§ 122) of t h e bill as Finally enacted is quoted supra note 102.

110 The bill as finally enacted defined the prohibited “withholding of medically indicated treatment” as follows:

§ 121(3) (adding § 3(3) to the Act).

111 Implementation of the new Act's crucial provision (quoted supra note 102) may be made easier than it would otherwise be by the availability of a model state statute developed by the ABA's Developmental Disabilities State Legislative Project. The model law provides that physicians who deny “necessary medical care and treatment” would be subject to loss of their medical licenses (§ 11) and that physicians would be immune for good faith provision of care without parental consent (§ 5(2)), if delay in providing the care would jeopardize the infant's life (§ 5(1)). Unlike the new federal law, the model statute defines the protected group of children (those born alive and less than two years old) (§ 3(1)). See B. Sales, D. Powell R. Duizend, Disabled Persons and the Law: Stat. Legislative Issues 88-89 (1982).

112 For a detailed analysis, see R. Dahl, Pluralist Democracy in the United States (1967).

113 Pub. L. No. 93-247, 88 Stat. 4 (1974). The new legislation consists of amendments to this law. See supra notes 99-107 and accompanying text.

114 There have apparently not been any successful murder prosecutions in this area. See Ellis, , supra note 29, at 401 n.32.Google Scholar In one case, parents and physicians were charged with conspiring to commit murder when food was withheld from newborn Siamese twins, but the charges were dismissed for lack of evidence. See Robertson, , supra, note 67, at 5.Google Scholar Prosecutors and juries, however, often consider factors that are foreign to the applicable legal concepts— lynch mobs, for example, were immune from liability for murder in certain sections of the country until fairly recently. Factors that may discourage prosecution in Baby Doe cases include sympathy with the “afflicted” parents, deference to the judgment of physicians (who are rightly assumed not to be acting out of vicious motives), and a natural tendency not to regard extremely defective infants as fully human.

Note, however, that the ordinary criminal law—when unmodified by prosecutorial discretion—quite clearly prohibits parents and physicians from withholding medical treatment. See Horan, , Euthanasia, Medical Treatment and the Mongoloid Child: Death as a Treatment of Choice, 27 Baylor L. Rev. 76, 78 (1976);Google Scholar Robertson, , supra, note 60, at 217-44;Google Scholar Paulson, , The Legal Framework for Child Protection, 66 Colum. L. Rev. 679, 680-93 (1966).Google Scholar

115 It is by no means true that all observers are disturbed by cases like the Bloomington incident. Cf. Sorenson, , The Rationalizing of Reproduction and Parenthood: Some Societal Developments, in Decision Making and the Defective Newborn 261, 275-76Google Scholar (C. Swinyard ed. 1978) (discussing survey data suggesting significant levels of approval for euthanasia among the general population).

116 The Am. presented its own scheme for dealing with the issue, which was defeated in Congress. Cong. Q. Weekly, Feb. 2, 1984.

117 See supra note 108. Note also that the new law contains a provision (Section 127(b)) that indicates an intent to retain the traditional deference towards physicians:

Child Abuse Amendments of 1984, Pub. L. 98-457, § 127(b), 98 Stat. 1749, 1754 (1984).

118 For a suggestion that the lack of public interest is in part caused by a lack of publicunderstanding, which in turn has been caused by a campaign of distorted and tendentious journalism, see Tedeschi, supra note 49, at 31.

119 See supra note 106 and accompanying text.

120 For a discussion of the argument that “we should muddle through and live with a striking dichotomy between the law on the books and the law in action rather than developing more precise standards or adopting new procedures,” see Mnookin, , Two Puzzles, 1984 Ariz. St. L. J. 667, 683.Google Scholar