Search

Lawmakers are stewards of social resources. A current debate--over screening newborns for genetic disorders--illuminates dilemmas of that stewardship that have particularly plagued bioethics. Recently in the Report, Mary Ann Baily and Thomas Murray told the story of little Ben Haygood. He died from MCADD, a genetic disorder that can make long fasting fatal. (1) Screening at birth would have let doctors alert Ben's parents. "After Ben died," Baily and Murray wrote, "his father became a passionate advocate for expanding Mississippi's newborn screening program to add MCADD and other disorders." Soon, the Ben Haygood Comprehensive Newborn Screening Act increased the number of genetic disorders covered by the state's program from five to forty and provided teams to follow up positive tests. Is this a prudent and productive way to make public policy? Certainly it is understandable. Mr. Haywood knew that parents suffered and children died in ways newborn screening could prevent. He acted on his knowledge. The legislature knew that suffering and death could be prevented. It acted on its knowledge. Was it supposed to let Tiny Tim die? Who could say yes? But was the legislature a good steward? In its first year, as Baily and Murray reported, Mississippi's expanded screening identified three cases of MCADD "along with twelve cases of other new disorders, out of a total of 116 newborn screening diagnoses." But "to help pay for the expansion, the state doubled the [screening] fee to seventy dollars. This meant that a substantial share of the resources for expansion came from Mississippi's Medicaid funds, since Medicaid covers more than half of Mississippi births." Around this time, Mississippi's infant mortality rate had been increasing, "Medicaid eligibility requirements were tightened, and some programs were cut." One county's infant mortality rate, however, "fell sharply after a private charity began providing intensive in-home visits using local women as counselors and busing pregnant black women to pre- and postnatal classes." If resources spent screening had instead been spent on such a program, would more lives have been saved and more suffering averted? When legislators are confronted with the tragedy and passion of a parent like Mr. Haygood, such questions are painful and seem pointless. But if they are not asked, several anonymous Tiny Tims may die to save one whose story lawmakers hear. The American College of Medical Genetics proffers a somewhat more systematic way to make screening policy. (2) Its "expert panel identified 29 conditions for which screening should be mandated. An additional 25 conditions were identified because they are part of the differential diagnosis of a condition in the core panel or are clinically significant and revealed by the screening technology but lack an efficacious treatment ... or because there are incidental findings for which there is potential clinical significance." The ACMG report is a stride toward more orderly stewardship, and most states have implemented a version of the report's proposals. But like Mr. Haygood, the ACMG has purposes and passions that lead it to see a world in which screening is indispensable. Its report does address the issue of cost and benefit, but in such a cautiously qualified way that it's hard to embrace its conclusions. Furthermore, the report enthusiastically says that "screening is more than testing. It is a coordinated and comprehensive system consisting of education, screening, follow-up, diagnosis, treatment and management, and program evaluation." How would a "comprehensive system" affect costs and benefits? Finally, the report does not ask whether there are yet more rewarding ends to which resources could be put. In a recent white paper, the President's Council on Bioethics sees screening from yet another perspective. (3) It finds the cost question harder than the ACMG. Few infants actually benefit: Of about four million babies screened annually, "about 5,000 are identified as having serious heritable disorders, most of which are, in varying degrees, amenable to treatment. …

[I]t is as craftsmen that we get our satisfactions and our pay. --Learned Hand The Bill of Rights Oliver Wendell Holmes--a great judge--said that "the command of the public force is intrusted to the judges in certain cases, and the whole power of the state will be put forth, if necessary, to carry out their judgments and decrees." Appellate courts command that force in ways that principle and practicalities leave little fettered. Judges must fetter themselves, not least by honoring the judicial duty of craftsmanship. That duty obliges courts to respect procedural rules, for they keep courts within their bounds and promote fair and sound decisions. That duty obliges courts to analyze legal authority scrupulously, since judicial legitimacy depends on that authority. That duty obliges courts to explain their decision and reasoning lucidly, since only then can the litigants and we the governed tell what is required of us and see that the court acted lawfully. No court, no judge, no person always achieves high standards of craft. But a court that betrays its duty of craftsmanship is a court that has lost its moral authority and wrongfully asserted its legal power. (1) Such a court is the court that decided Grimes v. Kennedy Krieger Institute. (2) Grimes is the case (actually two cases consolidated in a single appeal) in which parents whose homes had been part of KKI's research into ways to abate the dangers of lead paint in old buildings sued KKI on the grounds that KKI's negligence had harmed their children. They centrally claimed (the court seems to say) that KKI had failed to warn them properly about dangers to their children arising from the lead paint that had once been used in their homes. The Maryland Court of Appeals (Maryland's supreme court) eagerly commanded the public force. But its justification for its judgments and decrees fell intolerably short of the requirements of judicial craftsmanship in area after area. The most primitive requirement of the judge's craft is writing well enough to be understood. Even this task overwhelms the Grimes court. Its opinion is swamped with errors in syntax, grammar, punctuation, and even proofreading. It is awash in sentences that cannot be parsed. It is flooded with passages so bungled that they stupefy more than clarify. For example: "If consent agreements contain such provisions, and the trial court did not find otherwise, and we hold from our own examination of the record that such provisions were so contained, mutual assent, offer, acceptance, and consideration existed, all of which created contractual relationships imposing duties by reason of the consent agreement themselves (as well, as we discuss elsewhere, by the very nature of such relationships)." Or: "Though not expressly recognized in the Maryland Code or in our prior cases as a type of relationship which creates a duty of care, evidence in the record suggests that such a relationship involving a duty or duties would ordinarily exist, and certainly could exist, based on the facts and circumstances of each of these individual cases." This is not how competent law students, much less competent judges, write. All the students in my last brief-writing course wrote more correctly and comprehensibly. Second, a court's duty of explanation requires it to set out a case's facts, its procedural history, and its legal issues. Grimes spends eight of its twenty-eight thousand words on the facts, but its account is so jumbled and disjointed that even a heroic reader retires defeated. Ordinarily, appeals go from trial courts to intermediate appellate courts, but the reader never learns how Maryland's highest court got the case. Statements of issues litter the opinion, but the reader can only wonder whether those are the issues the litigants actually posed. Third, a court's duty of justification obliges it to identify the governing law and then show how it applies to the issues at hand. …

If you attempt to shave too closely in money matters--grabbing when a patient ... [is] so low that it is no longer decent to take fees; or hungrily holding watches, jewelry, or other articles as security for payment of your fees; or compelling their wounded or half-dead owners to pawn or sell them for your benefit; or [being] ... too vigorous in your efforts to collect fees from persons in narrow circumstances--[this] would not only be brutal barbarity, but would be very apt to ... create a widespread community feeling of hostility against you... . It is almost better to be defrauded than to collect your fees by such method ... --Daniel Webster Cathell The Physician Himself Courts and legislatures have labored for decades to protect patients' choice of medical treatments, even though patients seize that gift less eagerly than lawmakers expect. Yet while courts have rushed to build the whited sepulchre of informed consent, they have fled from a related problem that patients actually yearn to solve and that actually can be ameliorated--the plight of patients who perforce agree to a treatment before they know its costs and who receive a bill both unrelated to the treatment's value and several times what an insured patient would pay. Increasingly, patients must be consumers in the medical marketplace. This frightens patients, and should. Medical bills can be as alarming and baffling as medical ailments. The costs of illness--particularly medical bills--contribute to more than half the personal bankruptcies in the United States. Even insured patients may find themselves paying for uncovered services that can be both numerous and dear. Furthermore, managed care's effort to subdue health costs by inducing doctors to save money has been badly battered, and the new cry is to give patients such inducements instead. "Consumer-directed health care" makes patients consumers by asking them to purchase insurance programs shrewdly and by using high deductibles (one to five thousand dollars) and health savings accounts to make them purchase specific treatments. In short, individually we must increasingly worry about buying care wisely, and nationally we must worry about our latest adventure in financing health care. So why should courts protect patients as consumers? You, dear consumer, should protect yourself (caveat emptor) by evaluating what you buy, and the market should protect you by disciplining vendors who compete for your business. Evangelists of consumer-directed health care fondly imagine just such a market. Really? You arrive at the doctor's office or the hospital and are told to sign a contract. Like this: In consideration of hospital services rendered to the patient, I jointly or severally, do hereby agree to pay Athens Regional Medical Center any and every account presented to me, or us jointly or severally, for said service or services in accordance with the rates and terms of the hospital. (1) In other words, "Do you want help? Sign a blank check. We'll fill it in later. As we wish." You do want help, so you sign. (And even if you don't, accepting services binds you to pay for them.) Consumer! You must shop for capable care at palatable prices. True, doctors don't advertise prices. So you telephone: "I'm a good consumer trying to direct my health care. How expensive is Dr. Jones?" Even if a human being answers, even if the human being thinks this question tolerable, what answer can you fairly expect? Do you know how long your visit will be? What services you will need? Which of the myriad insurance policies you have? If you don't know--and who does?--what doctor's staff can predict the charges? (And finding out fees is a dream compared to ascertaining the quality of the doctor's work.) Or you ask the clerk about the hospital's prices. The clerk has no idea and resents your impudence. You press up the ladder. …

When law regulates a profession, where does it get its standards? Largely from the profession. Members of professions acquire esoteric and abstract knowledge through formal education and the experience of practice. They use professional judgment in applying this knowledge to each case. Because legislatures and courts lack this expertise, they adopt the standards of the experts. Thus in a malpractice suit, juries are instructed to determine whether the doctor met medicine's standard of care. Furthermore, physicians must be called as expert witnesses to guide juries in that work. Even when lawmakers contemplated intensifying their regulation of medicine by creating the duty of informed consent, they could consult a literature to which doctors and medical ethicists contributed crucially. In some jurisdictions, even the scope of the duty is determined by using the medical standard of disclosure (although in other jurisdictions the standard is the degree of disclosure sufficient to permit the ordinary patient to make a sound decision). Nor have lawmakers striven to extend the reach of informed consent beyond the norms of medicine. They might have done so in two ways. First, they might have broadened the legal standard of disclosure. This seems to have happened only sporadically and tentatively. Second, fact-finders (juries and trial-court judges) might have interpreted the legal standard as demanding elaborate or unusual disclosures. This too has apparently not much happened. Indeed, plaintiffs rarely bring informed consent actions (except as appendages to malpractice suits), rarely win them, and rarely obtain large verdicts. In short, lawmakers have essentially established rules intended to hold medicine to its own standards and then mostly left the system to work unmolested. What lawmakers have not noticed, however, is that the status of informed consent in the medical literature has become parlous. Two developments particularly matter. First, a torrent of empirical evidence now suggests that informed consent does not work as intended: Doctors generally tell patients too little and patients generally understand too little for patients to make the choices that lawmakers had imagined. Second (and relatedly?), the literature seems to be deserting the term "informed consent." And what instead? A comet shower of novel terms. "[T]here is now a profusion of competing models that attempt to convey subtle differences in the sharing of information and power between clinician and patient." (1) A smattering of the latest models: "evidence-based patient choice," "informed decision-making," "informed medical decision-making," "informed treatment decision-making," "physician as perfect agent," "shared decision-making," "shared clinical decision-making," "shared medical decision-making," and "shared treatment decision-making." From this welter of multiplying, mystifying distinctions, one term has emerged most stoutly--"shared decision-making." So, what is shared decision-making? Would that I knew. Or that anyone did. Makoul and Clayman heroically slogged through the literature and concluded that "there is no shared definition of shared decision making." They "identified 31 separate concepts used to explicate SDM, only two of which appeared in more than half of the conceptual definitions." In fact, "60% of articles that purport to focus on shared decision-making failed to include any conceptual definition at all." (2) As this suggests, many proponents of shared decision-making seem to regard its meaning as self-evident. And no doubt most people suppose that they understand the term. "Sharing" does the real rhetorical work here. Who doesn't understand sharing? Who could oppose it? Yet the skimpiest reflection reveals that the slogan is ambiguous unto incoherence. For example, what makes a decision "shared"? It is logical, plausible, reasonable to say, as many advocates of shared decision-making seem to say, that anyone who helps shape a decision, helps give the decision meaning, helps give it effect, "shares" in it. …

In America today, public policy governing an increasing number of social issues is made through the judicial enforcement of constitutional rights. Cruzan raised the question whether policy regarding the withdrawal of medical care from incompetent patients is to be handled similarly. This essay argues that privacy-rights doctrine provides a poor basis for constructing public policy in this area. It suggests that the Court has been unable to articulate a convincing basis for privacy rights and that the basis the Court seems to assume poorly fits many circumstances to which it is applied. The essay further contends that the Court has been unable to describe a persuasive method of analyzing the state interests that are at stake in privacy cases and to accommodate those interests in a reasonable way. The essay concludes that these are issues as to which policy is better formed through a political, not judicial, process.

And why beholdest thou the mote that is in thy brother's eye, but considerest not the beam that is in thine own eye? ... Thou hypocrite, first cast out the beam out of thine own eye; and then shalt thou see clearly to cast out the mote out of thy brother's eye. --MATTHEW 7:3-5 The law of bioethics has been the law of cases. Interpreting the common law and the Constitution, judges have written the law of informed consent, abortion, and assisted suicide. Reacting to causes celebres, legislatures have written the law of advance directives and end of life decisions. The long, sad death of Terri Schiavo eclipsed even the long, sad deaths of Karen Ann Quinlan and Nancy Beth Cruzan in the duration and strength of the attention and passions it evoked. What are Schiavo's lessons? Hard cases, lawyers say, make bad law. Why? First, hard cases are atypical cases. They present abnormal situations that normal rules do not anticipate and cannot handle. Schiavo is egregiously atypical. Persistent vegetative states are unusual. Few families fight so implacably and repellently. Courts--and even less legislatures, governors, and presidents--rarely encounter cases about withdrawing treatments. Decisions usually take days, not decades. Second, hard cases make bad law because hard cases force tragic choices--choices where walking toward one blessing means walking away from another. The interests of the Schiavos and the Schindlers were irreconcilable. Ms. Schiavo's interests in living and dying clashed. The menace of hard cases is that they invite response. But no rule can resolve tragic conflicts well. Even could such a rule be imagined, the law that can manage the hard case mishandles ordinary ones. Thus the best response to hard cases is usually no response. No system of rules ever produces uniformly good results; the goal of legislation is therefore to devise the system that yields the fewest bad results. The course of wisdom is to learn when bad results must be tolerated. But who thinks Schiavo a hard case? It has primarily provoked righteous vindication: Had you but listened to me, all would be well. The conventional bioethical and medical wisdom is that the internecine strife that ignited Schiavo would have been extinguished had Ms. Schiavo done her duty and prepared an advance directive. This is a triumph of faith over reason: First, as Rebecca Dresser has noted, a court long ago granted Mr. Schiavo the authority to act for his wife. Second, early on a court found "clear and convincing evidence" that Ms. Schiavo would have wanted to forego food and water in her circumstances. Both rulings did what advance directives would presumably have done, but neither ended her travail. But suppose Ms. Schiavo had--despite her youth--written a living will. Suppose--heroically--that it accurately and lucidly described a careful decision to refuse food and water in a persistent vegetative state. The Schindlers would still have presented medical evidence that Ms. Schiavo was not in that condition. For that matter, bitter and obdurate litigants don't need a good case to protract litigation endlessly. Indeed, as Ms. Schiavo's guardian ad litem notes in this issue, the Schindlers "stated that even if Theresa had executed a formal, written living will, they would have fought to have it voided because they did not believe it was consistent with their and her beliefs." And they would have had a point. Living wills call for greater powers to predict circumstances and preferences than most of us can muster. Ms. Schiavo would not only have had to imagine the unimaginable about her physical circumstances, she would also have had to anticipate the ways her social circumstances would affect her medical choices. For instance, had Ms. Schiavo anticipated her parents' inconsolable distress at her death, would she have been willing to perdure? Far from demonstrating the merit of living wills, Schiavo demonstrates their danger. …

Marshall Klavan headed the Obstetrics and Gynecology Department of the Crozer-Chester Medical Center. (1) He deeply feared strokes, perhaps because his father had been savaged by one. In 1993, Dr. Klavan wrote an advance directive which said that (as a court later put it) "he 'absolutely did not want any extraordinary care measures utilized by health care providers.'" On April 29, 1997, Dr. Klavan tried to kill himself. He left suicide notes and a note refusing resuscitation. The next morning, medical center employees found him unconscious and took him to the emergency room, where he was resuscitated. By May 2, Dr. Klavan had fallen into a persistent vegetative state. His family and his lawyer told the center about his notes and his advance directive. On May 4, the center "agreed to provide care in accordance with" the advance directive, but on May 5 Dr. Klavan's condition worsened and the center again resuscitated him. He "then suffered a stroke that rendered him mentally and physically incompetent." A suit was brought on Dr. Klavan's behalf in federal court. It claimed that his fourteenth amendment right to refuse medical treatment was being denied. The fourteenth amendment, however, restricts only governments, not private institutions. Had the government so deeply implicated itself in the center's activities that it had effectively become a governmental institution? The court thought not, leaving Dr. Klavan's representative only with state-law claims. The federal court had discretion to take jurisdiction of those claims, but it refused: "While Dr. Klavan's situation cries out for prompt and definitive judicial resolution, we nevertheless decline to exercise our discretion ... precisely because of the gravity of his case." How did this sad case reach this unsettling result? First, the story I have told may be false. (2) Klavan was decided before a trial could determine what had actually happened. To decide whether it had jurisdiction, the court accepted as true everything Dr. Klavan's representative had alleged. The court concluded that even if all the allegations were true, it should not grant the relief Dr. Klavan's representative sought. Therefore we have no idea what the medical center's defense was or how any factual disputes would have been resolved in a trial. (And even had there been a trial, we could not be sure of its conclusions.) More broadly, however, the Klavan court's reluctance to address the dispute was typical of the judicial reaction to the (uncommon) attempts to enforce living wills. Such attempts are of two kinds. First, people locked in a dispute about treating an incompetent patient can ask a court to settle the disagreement. In doing so, the court can look at the living will and any other relevant evidence. Second, after a patient dies, the family might sue to recover the damages the patient suffered became the living will was disobeyed. Why have courts not embraced either kind of litigation? In part, courts' hesitation to enforce living wills reflects a judicial aversion to disputes about treating the dying. Such disputes demand a speed courts can rarely attain. The patient has often died by the time lawyers have been consulted, non-legal solutions have been exhausted, a decision to litigate has been reached, and legal documents and evidence have been prepared. Just as bad is the "standards" problem: These decisions raise issues the law addresses too obscurely to guide courts. And given the complexity and perplexity of end-of-life decisions, more illuminating rules probably are impossible. So courts have acquiesced--tacitly and sometimes expressly--in relegating these decisions to an informal process in which doctors and families work toward a consensus that the time has come to let the patient die. Courts may be right to do so: Evidence is spotty, but it hints that patients' families are generally satisfied with this process and that their dissatisfactions would not be soothed by litigation. …

The amount of law is relatively small which a modern legislature can successfully impose. The reason for this is that unless the enforcement of the law is taken in hand by the citizenry, the officials as such are quite helpless.... For what gives law reality is not that it is commanded by the sovereign but that it brings the organized force of the state to the aid of those citizens who believe in the law. --Walter Lippmann A Preface to Morals In my last contribution to this column (HCR, July-August 2000), I argued that the law of bioethics has repeatedly failed to achieve the hopes cherished for it. I presented evidence, for example, that most doctors breach the duty of informed consent, that advance directives do not direct patients' care, and that repeated legal attempts to increase organ donation have failed to find the success predicted for them. I closed that column by promising to try to explain this chastening experience. It would, of course, take a lifetime of columns to capture all the reasons the law of bioethics has so often disappointed.[1] Here I want to discuss only one, albeit a crucial one: Legal regulation of human behavior is insistently difficult because human behavior and social institutions are bafflingly complex. It is maddeningly hard to mold that behavior and those institutions because they are shaped by many potent forces besides the law and because lawmakers so often cannot accurately identify all those forces and devise reliable methods of altering enough of them in sufficiently precise and predictable ways to achieve the result intended. It is even hard for law to rule in its own house--for example, to shape litigation in useful ways. The law of bioethics illustrates both the general problem (influencing social behavior) and its particular instantiation (influencing legal institutions). Let me once again adduce empirical evidence to show how. Living wills exemplify the first problem. Living wills seemed an obvious solution to the perplexity of making end of life decisions for incompetent patients, but they have betrayed the expectations faithfully nurtured for them. To see why, consider the chain of circumstances necessary for living wills to be well and widely used. First, people must want a living will. Some people say they do not, many of them because they think satisfactory decisions will be made for them without one. Many people believe they want a living will, but very many of these have not signed one even though they know about advance directives. Programs to persuade people to sign living wills have not been conspicuously successful. Is this, perhaps, because people do not actually want them, or do not actually want them enough to overcome ambivalence about them? Second, people must know what treatment they would want should they become incompetent. This requirement has several components. To begin with, people must (a) obtain accurate information about what their choices would be and (b) understand that information. But patients will encounter all the problems acquiring and analyzing information that have become notorious through studies of informed consent. And empirical investigation suggests that doctors are neither anxious to have conversations about living wills nor adept at conducting them.[2] Furthermore, unless people are assisted with exceptional care, they must decipher the advance directive itself. This is no small undertaking. Living wills are often execrably drafted. And even if they were drafted by angels, "[i]n the largest study of functional health literacy in the United States.... 42% of ... [English-speaking patients] were unable to comprehend directions for taking medication on an empty stomach, 26% could not understand information on an appointment slip, and 60% could not understand a standard consent form."[3] In addition, people preparing advance directives not only confront all the perplexities of medical decisions; they also face the special problems of making decisions for a hypothetical future. …

Since the days of Tocqueville, foreign observers have seen America as both a pattern and a problem. They still do, and in ways that illuminate the way law deals with bioethical issues both here and abroad. America was long exceptional in having a written constitution, in allowing its courts the power of judicial review, and in letting courts exercise that power to develop and enforce principles of human rights. Today, that pattern looks markedly less exceptional. After the Second World War, Germany and Japan were persuaded to adopt constitutions that included human rights provisions and that endowed courts with the power to interpret them. Since that time, a number of other countries--Canada, for example--have also moved closer to the American constitutional pattern. Many countries, however, have not been content to borrow American constitutional principles and practices. Their courts have also asserted their authority to develop and enforce principles of human rights in two other ways. First, courts in many countries have assumed jurisdiction over questions involving those rights by virtue of their duty to interpret treaties their countries have signed. Second, and strikingly, courts in many countries have come to see themselves as joint participants in the work of building an international body of human rights law. As my colleague Christopher McCrudden writes in his fascinating study of this development, "It is now commonplace for courts in one jurisdiction to refer extensively to the decision of other courts in interpreting human-rights guarantees."[1] This is where America has come to seem a problem to our foreign observers. For the United States has been irritatingly reluctant to sign these treaties, and American courts have been irritatingly unwilling to consult the decisions of foreign courts. Why? The reasons are--obviously--various. Foreign observers are not infrequently pleased to believe that Americans are irredeemably provincial. (What do you call someone who speaks two languages? Bilingual. What do you call someone who speaks one language? American.) And it is perfectly true that, while lawyers in many countries are likely to have had some instruction in a foreign law (and even to have studied in a foreign country--often the United States), American lawyers have generally not been so blessed. To be sure, Americans lawyers have less need. They work in the world's dominant economic power, one so vast that much more internal trade occurs within American boundaries than within any other industrialized country. American lawyers, to put the point differently, are more narrowly trained because--more than the lawyers in other countries--they can afford to be. Foreign observers also relish suggesting that America has resisted signing some international human rights treaties because it is hopelessly arrogant and cannot believe it has anything to learn about human rights from abroad. It is surely true that American foreign policy has in recent decades often seemed loftily confident of its mission to bring the wretched heathen to the human rights light. But the reluctance to sign such treaties has other, less evident, roots as well. For one thing, a number of international human rights agreements have grown out of regional (and particularly European) efforts at economic and political integration, efforts in which the United States has been, if anything, a competitor. More significantly, however, the government of the United States is--more than almost any foreign country's--federal. We confide many issues with human rights implications to the states, not the federal government. In addition, some Americans have opposed these treaties because they take them seriously enough to be uneasy about where judicial interpretation will take their provisions. That the American executive and legislative branches have not leapt to endorse human rights treaties may help explain why American judges have not felt encouraged to be guided by the developing international law of human rights that judges in many countries are creating. …

In June 1997 a sixteen-year-old girl named Shannon Nixon began to feel ill. Her parents belonged to the Faith Tabernacle Church, one of a number of American sects which believe that illness should be treated spiritually rather than medically. Accordingly, the Nixons prayed for Shannon and took her to be anointed at their church. Shannon reported that she felt better and that the spiritual treatment had gained her her victory--her recovery. Before long, however, Shannon again felt ill. She became weaker and weaker and then fell into a coma. A few hours later she died. An autopsy revealed that she had succumbed to diabetic acidosis. Shannon's parents were charged with involuntary manslaughter and endangering the welfare of a child. A jury found them guilty, and the judge sentenced them to prison for two and half to five years and fined them $1,000.[1] The Nixons' case merits attention not because it sets an astonishing new precedent, but because in its ordinariness it represents several continuing developments in a problem with which the law has had a prolonged and perturbed history. It is standard doctrine that parents have a duty to provide their children the medical care they need. Parents who breach this duty may ordinarily be judged to have "abused or neglected" their child, they may be subject to criminal penalties, and the state may step in to protect the child, even to the extent of terminating the parents' legal relationship with the child. States have long intervened to secure medical care for children who are seriously ill but whose parents refuse to provide them medical care, even if the parents are motivated by their religious beliefs. The first amendment to the Constitution, of course, protects the "free exercise" of religion. Doesn't this mean that parents who deny their children medical care for religious reasons can prevent the state from overriding their decision to be more concerned for their children's immortal souls than for their children's temporal bodies? The conventional answer to that question is no, not least because of the Supreme Court's decision in Prince v. Massachusetts.[2] In an often-quoted passage, it said, "Parents may be free to become martyrs themselves. But it does not follow they are free, in identical circumstances, to make martyrs of their children before they have reached the age of full and legal discretion when they can make that choice for themselves." Suppose, however, that--as easily happens--the state does not find out that a child is being denied treatment until the child has died. The usual rule here is that if a child dies because the parents have withheld medical care, the parents may be held criminally liable for the death. Nevertheless, where parents have refused medical treatment for their children for religions reasons, the concerns that animate the first amendment might lead us not to invoke the criminal law against them. Furthermore, prosecutions of these parents is notoriously difficult. The parents' failure to provide medical care must be identified as the cause of death, a prosecutor must decide to prosecute, a jury must decide to convict, a judge must decide to sentence, and an appellate court must decline to use any of the numerous devices at its disposal to reverse the conviction. It is thus not so surprising that for many years there were no reported appellate court opinions upholding homicide convictions of parents whose children had died because they had been given spiritual rather than medical care (although some parents were convicted of charges less than homicide). In recent years, however, this tacit compromise has been disturbed by two conflicting developments. On one hand, most states have adopted statutes that exempt religiously motivated parents from some of the effects of child protection laws. On the other hand, the number of homicide prosecutions of parents whose children have died after their parents treated them spiritually appears to have increased, and opinions upholding homicide convictions in these situations have begun to appear. …

As I write, the Supreme A Court has just agreed to hear Compassion in Dying v. Washington and Quill v. Vacco, the two cases in which United States circuit courts of appeals held that a state may not constitutionally prohibit physicians from helping a terminally ill person who wishes to commit suicide to do so. These cases have already received lavish comment and criticism, and no doubt the Supreme Court's opinion will garner even more. Reasonably enough, most of this analysis addresses the merits of physician-assisted suicide as social policy. I, here, want to talk about how setting bioethical policy through constitutional adjudication actually works and how its usefulness is diminished by some practical deficiencies. Courts are often thought well-equipped to resolve social issues because they are guided by the thought and wisdom that animated the Constitution. Sometimes this argument has some truth. But bioethical questions are generally resolved under a provision (the Fourteenth Amendment) that says only that no state shall "deprive any person of life, liberty, or property without due process of law." These words, noble though they are, give courts little insight and little direction. The Supreme Court has been able to find in them such a suspiciously wide range of lessons--from the unconstitutionality of statutes limiting a woman's right to an abortion to the unconstitutionality of statutes limiting the hours an employee can be required to work--that one may begin to wonder just what the justices have been consulting. The vagueness of the Fourteenth Amendment throws judges back on their own resources. Because judges vary in experience and talent, these resources vary. But all judges are lawyers. A lawyer's training is centrally about analyzing legal documents, particularly judicial opinions. It does little, for example, to equip lawyers to understand and respect empirical evidence about social behavior and does too much to convince them that the best evidence about how the world works is the particular facts of a litigated case. The practice of law may remedy some of these ills, but today most sophisticated legal practice is too specialized to expose lawyers broadly to social issues. Yet once appointed, judges must become the broadest kinds of generalists. In short, judges are lawyers who--narrowly trained and narrowly experienced--are unlikely to come to the extraordinary range of problems they face with expertise or understanding. What is worse, judges are less well situated to learn about these problems than one might expect. True, the trial-court judge presides over proceedings in which evidence is introduced. But a systematic flaw of constitutional adjudication is that it promulgates grand principles on the basis of a single case's facts. Courts tend to assume that the case before them typifies the social problem at issue and that the parties before them represent all relevant points of view. And since the parties, not the judge, decide what evidence will be introduced, the judge does not hear what will not benefit the parties, and the quality of the information introduced crucially depends on the wit and learning of the attorneys the Quinlans or the Cruzans happen to hire and of the too often modestly gifted political appointees and civil servants who represent state governments. To be sure, interested groups may file "friend of the court" briefs. But in my experience, appellate judges (at least) rarely read them. Suppose, though, that this haphazard process leaves the federal trial judge well informed. That judge's decision will only affect the people within that judge's district. If several states are to be reached, the case must be presented to a circuit court of appeals. If the whole country is to be reached, the case must be taken to the Supreme Court. Ironically, though, as the case's reach widens, the judge's contact with the evidence shrinks. The parties normally reprint only a few significant documents from the trial record for the appellate judges. …

In trying to understand the SUPPORT study, it may be useful to think of contemporary bioethics reform in terms of the principles of consumer protection. The central tendency of that reform (particularly in my own field--the law) has been to employ the model of consumer choice. That model sets as its purpose to allow consumers to choose the kinds of products they prefer. It seeks to accomplish that purpose primarily by supplying consumers the information they need to make choices and by insisting that they are given what they chose. Thus, for example, merchants may be required to reveal the actual terms under which they sell goods on credit, so that customers may decide whether they wish to pay the price. And once the customer has chosen a purchase, the merchant is held to the terms originally agreed upon. In short, the consumer choice model seeks to allow customers to make successful choices by providing them with a market that works in the economist's sense, efficiently. Bioethical reform may be understood along these lines. The doctrine of informed consent is, plainly, intended to provide patients the information they need to make wise choices that express their preferences. The patient is thus to become a consumer well-enough informed to make sensible choices in the market for medical care. The patient's choices are then given more binding effect by, for example, various provisions for advance directives. The consumer choice model is hardly startling and surely makes a good deal of sense. The free-market prescription was, for instance, quite successful when it was employed during the New Deal to improve the operation of the securities market. But of course the consumer choice model rests on some consequential assumptions. It assumes, for one thing, that human preferences differ so much that it is best to let people make purchasing decisions for themselves. It assumes an engaged and energetic purchaser. Finally, it assumes that people's preferences are clear and strong enough to drive them to act on the information they receive to make choices and follow up on them. Thus the New Deal reforms of the securities markets were plausible because many of the people who trade in that market know their way around it, have ready access to the information that is provided, have commanding incentives to use it, and know how to do so. The SUPPORT study does not speak directly to all these assumptions. Nor could any single study definitively establish whether such assumptions arc true. Nevertheless, this ambitious and impressive research raises some fruitful questions about the consumer choice model as it has been applied in medicine. The SUPPORT enterprise followed that model in trying to increase the communication between doctor and patient so that both could make better decisions. But what is striking about the study is, despite a commitment of resources (and, one suspects, enthusiasm) unlikely to be even matched, much less sustained, in the dullness of daily life, the SUPPORT instantiation of the model quite failed to produce statistically significant benefits. What are we to make of this unsettling negative? To put the point simply, a strong reading of the study raises questions about the usefulness of continuing to emphasize the consumer choice model. In saying this, I am far from suggesting that efforts to implement the consumer choice model in bioethics have been wasted. I suspect that that model has helped spur changes in medical attitudes which were necessary to make doctors' behavior more tolerable to contemporary American mores. My point, rather, is that--strongly read--the SUPPORT study provokes doubts about the profit of sustaining the consumer choice model at its present preeminence. But how strongly should we read the SUPPORT data? Our answer will depend on many things. But one reason to read it strongly is that SUPPORT seems to confirm what we already know about the effects of some other bioethical reforms in the consumer choice model--that those reforms often seem to have fewer consequences than the consumer choice model seems to call for. …

Law provides a rich language for thinking about bioethical issues and is a tool for action as well as talk. But the language of the law, often inapt, regularly fails to achieve its desired effect.

Book reviewed in this article: Rights Talk: The Impoverishment of Political Discourse. By Mary Ann Glendon