Whose harm? Which metaphysic?

Abstract

Douglas Diekema has argued that it is not the best interest standard, but the harm principle that serves as the moral basis for ethicists, clinicians, and the courts to trigger state intervention to limit parental authority in the clinic. Diekema claims the harm principle is especially effective in justifying state intervention in cases of religiously motivated medical neglect in pediatrics involving Jehovah’s Witnesses and Christian Scientists. I argue that Diekema has not articulated a harm principle that is capable of justifying state intervention in these cases. Where disagreements over appropriate care are tethered to metaphysical disagreements (as they are for Jehovah’s Witnesses and Christian Scientists), it is moral-metaphysical standards, rather than merely moral standards, that are needed to provide substantive guidance. I provide a discussion of Diekema’s harm principle to the broader end of highlighting an inconsistency between the theory and practice of secular bioethics when overriding religiously based medical decisions. In a secular state, ethicists, clinicians, and the courts are purportedly neutral with respect to moral-metaphysical positions, especially regarding those claims considered to be religious. However, the practice of overriding religiously based parental requests requires doffing the mantle of neutrality. In the search for a meaningful standard by which to override religiously based parental requests in pediatrics, bioethicists cannot avoid some minimal metaphysical commitments. To resolve this inconsistency, bioethicists must either begin permitting religiously based requests, even at the cost of children’s lives, or admit that at least some moral-metaphysical disputes can be rationally adjudicated.

Introduction

Although parental authority enjoys wide latitude in medical decision-making for children, this authority is not absolute.¹ The best interest standard has been widely advanced as the proper basis for overriding parental authority with state intervention [2, p. 263; 3; 4, p. 573; 5; 6].² However, Douglas Diekema has argued that it is not the best interest standard, but the harm principle that serves as the moral basis for ethicists, clinicians, and the courts to trigger state intervention to limit parental authority in the clinic [7,8,9].³

Although it focuses on Diekema’s argument for the harm principle, this paper advances a larger thesis about moral standards for constraining decision-making in the clinic.⁴ I argue that where disagreements about what to do in difficult medical-moral cases are tethered to metaphysical disagreements, it is moral-metaphysical standards, and not merely moral standards, that are needed to provide substantive guidance. I aim not to give a comprehensive defense of the harm principle, but only to argue that it requires metaphysical elaboration to be effective. Even were the harm principle to be replaced by a stronger moral standard, my thesis would still obtain. It is not just the harm principle, but any moral principle that aims to provide substantive guidance in metaphysically tethered disputes that requires metaphysical elaboration to be effective.

I proceed in two parts. In the first part, I present Diekema’s harm principle, arguing that one of his main criticisms of the best interest standard—that it is inherently value-laden—also applies to the harm principle that he offers in its place. Others have leveraged this critique against Diekema, but their arguments have not been put as forcefully as they could have been. I conclude that the harm principle is made stronger when it is simply understood as inherently value-laden; Diekema’s claims to the contrary serve only to make the harm principle seem less credible and should be rejected. A more plausible reading of Diekema’s claims here is one in which people are more likely to agree about what constitutes an unacceptable threshold of harm than they are to agree about what is in the best interest of a child. I use this claim to transition into the second part of the paper, which argues that even if Diekema is right, the way he has articulated the harm principle is—despite his claims to the contrary—unable to resolve metaphysically tethered cases of disagreement.

In the second part, I argue that Diekema’s harm principle is not actually effective in the cases where he claims that it is most easily applied—cases in which Jehovah’s Witnesses and Christian Scientists refuse medical treatment for their children. To show this, I argue that a community of Jehovah’s Witnesses could use Diekema’s harm principle as a basis for preventing parents from giving their children blood transfusions. This result is possible because Diekema’s principle is cast as merely moral, whereas I argue that harm must be understood as a moral-metaphysical principle if it is to provide substantive moral guidance in the clinic.

Metaphysical claims are understood in this essay as worldview-level ontological beliefs that are critical to forming the general perspective through which one sees and interprets the world. I am especially focused on worldview-level claims traditionally considered religious, such as the existence of God, sin, salvation, prayer, heaven, thetans, wintis, and so forth.⁵ Although the kinds of claims addressed in this essay are considered metaphysical, they do not exhaust the range of topics to which metaphysics can refer. Metaphysics also includes questions about the nature of identity, necessity and possible worlds, causation, free will, space and time, universals and particulars, the ultimate origin and sustaining of the universe, and the presuppositions one brings to their experience of reality and knowledge [16]. Metaphysical claims about what exists can be distinguished from claims of moral theory, which focus on the right, the good, and moral worth.⁶ My argument is that both moral and metaphysical claims are required to produce principles that offer substantive guidance in the clinic. By substantive guidance, I mean guidance that cannot be adopted by all parties in a debate. Substantive guidance sits in contrast to the guidance of principles whereby, for example, both Jehovah’s Witness parents and secular physicians may claim that their stance on blood transfusions prevents harm to a child. In such cases, merely talking about harm does nothing to resolve the disagreement, since all sides think their position is the less harmful one. Diekema’s harm principle makes no mention of metaphysics and thus fails to offer substantive guidance in the cases to which he claims it applies.

I provide a discussion of Diekema’s harm principle to the broader end of highlighting an inconsistency between the theory and practice of secular bioethics in cases of overriding religiously based medical decisions. In a secular state, ethicists, clinicians, and the courts are purportedly neutral with respect to moral-metaphysical positions, claiming to take no view with respect to matters of moral-metaphysical significance, especially regarding those claims considered to be religious.⁷ However, the practice of overriding religiously based parental requests requires doffing the mantle of neutrality. In the search for a meaningful standard by which to override religiously based parental requests in pediatrics, bioethicists cannot avoid the assumption of minimal metaphysical commitments. To resolve this inconsistency, bioethicists must either begin permitting religiously based requests, even at the cost of children’s lives, or admit that at least some moral-metaphysical disputes can be rationally adjudicated.

Judgments of harm are value-laden

In this section, I elaborate on Diekema’s conception of the harm principle and his argument that it improves upon the best interest standard, which he claims to be inherently value-laden. In response, I argue that the harm principle is also inherently value-laden.

Diekema has not called for rejection of the best interest standard outright, but only for clarification of its proper role. He has argued that the best interest standard has been invoked to serve two very distinct purposes: (1) to offer guidance to parents and (2) to challenge and limit parental authority via state intervention [8].⁸ Diekema believes the best interest standard is useful when examining a child’s medical options and giving recommendations to parents, but he believes it should be confined to that purpose, not used as a standard for state intervention [8, 9]. By conflating these two purposes, Diekema argues that health care professionals may be tempted to seek state intervention when parents merely disagree with them over the proper treatment for their child [8, p. 130]. The best interest standard should be used for making recommendations and even for verbally challenging parents’ decisions where appropriate, but the harm principle should be used for involving state authorities with the power to override parental decisions by force if necessary.

To illustrate his point, Diekema explains that physicians have a strong belief that vaccinations are in a child’s best interests and will attempt to persuade parents to vaccinate their children. In these situations, few physicians would be willing to seek state intervention to compel vaccination. However, if a child has sustained a deep wound that is contaminated with dirt, which is likely to lead to serious harm via tetanus, and the parents still refuse vaccination, then many physicians would seek state intervention. Diekema writes, “What has changed … is not the physician’s belief that vaccination is in the child’s best interest, but the physician’s assessment that the parent’s refusal to vaccinate has placed the child at significant risk of serious harm” [8, p. 132].

Diekema’s central point—that the harm principle, not the best interest standard, is what guides state intervention—is strong. Diekema argues that the best interest standard is not used because intervention is not sought every time parents make a decision that is suboptimal for their children [7, p. 247].⁹ For example, parents are not required to keep their children in peak physical condition, nor are they required to send them to the best schools. On Diekema’s view, the harm principle affords parents a great deal of latitude in decision-making, but state intervention is triggered “where the decision of a parent places the child at substantial risk of serious harm” [7, p. 250].¹⁰ Diekema cites John Stuart Mill’s classic statement of the harm principle: “The only purpose for which power can be rightfully exercised over any member of a civilized community, against his will, is to prevent harm to others. His own good, either physical or moral, is not a sufficient warrant” [19, p. 14]. Diekema further cites Joel Feinberg, who adds that the mode of state intervention must be proven effective at preventing the harm in question, and there must be no other way of preventing the harm that would be less intrusive to individual liberty [20, p. 26].

Diekema argues against the best interest standard, contending that it is too often reduced to “objective medical interests alone,” which both overemphasize the weight of a child’s chance of survival and undervalue other emotional burdens or the effect of treatment on others [7, p. 247].¹¹ To avoid a similarly medically reductionist notion of harm, Diekema appeals to several scholars in order to articulate a wider range of harms that should be considered when establishing harm thresholds sufficient for triggering state intervention. He cites Lainie Ross, who asks that the consideration of best interests be traded in for a consideration of basic needs [22]. Basic interests are satisfied when “abuse, neglect, and exploitation are prohibited, and children must be provided with goods, skills, liberties, and opportunities necessary to become autonomous adults capable of devising and implementing their own life plans” [7, p. 251]. On Ross’s view, parents can sacrifice the best interests of their child for some other familial interest so long as the basic needs of the child are still met [22]. Diekema also summarizes Joel Feinberg, who suggests that serious harm includes interference with interests necessary for one’s “more ultimate goals,” such as “one’s physical health and vigor, the integrity and normal functioning of one’s body, the absence of absorbing pain and suffering or grotesque disfigurement, minimal intellectual acuity, emotional stability,” and so forth [20, p. 37]. Additionally, Diekema references Dworkin, who argues that parents have an obligation to make decisions for children that “preserve their share of what Rawls calls ‘primary goods’; that is, such goods as liberty, health, and opportunity, which any rational person would want to pursue whatever particular life plan he chooses” [23, p. 205].

In addition to critiquing this tendency toward medical reductionism, Diekema also claims that the best interest standard is “inherently a question of values” [7, p. 246]. He writes:

Appealing to a best interest standard does not help the courts decide whose conception of the child’s best interest should prevail. Ultimately, these are not objective “best interest” cases, but involve assessments of which values should carry the most weight. They pit the state’s determination of “best interest” against that of the parents. [7, pp. 246–247]

Elsewhere, Diekema and colleagues write, “if parents disagree with the clinicians, the best interest standard is a shaky foundation for restricting parental choice because it is subjective and value-laden” [9, p. 937].¹² This claim can be interpreted in a strong or weak sense.

The strong interpretation of this claim is that the harm principle improves on the best interest standard because it is not value-laden but objective. Although Diekema and colleagues almost certainly did not mean for their claim to be taken this way, in his own works, Diekema sometimes words this view so as to imply the strong interpretation, as evidenced above. The strong interpretation is untenable because it contradicts itself by claiming that (1) the fact that a standard for state intervention is “inherently a question of values” is a reason to reject said standard and (2) the harm principle should be conceived in a way that avoids being medically reductionist. The contradiction arises because the second claim is based on values that the first claim endeavors to exclude. Either the harm principle is not value-laden (in which case Diekema would need to say much more about what “value-laden” could possibly mean) or it is rife with normative commitments that avoid a medically reductionist view of harm.

The weak interpretation of Diekema and colleagues’ claim is that the harm principle is more straightforward insofar as people are more likely to agree about what constitutes significant imminent harm than they are to agree about what is in a child’s best interest. This interpretation appears to capture the view that Diekema actually holds. In a recent work, Diekema et al. claim that the harm principle does not exhibit “the same amount of ambiguity as the best interest standard” [25, p. 301]. They argue that ambiguity arises in the best interest standard because the state requires some theory of what counts as a good life in order to determine what is in a child’s best interest. Any such theory is bound to court more controversy than are claims of harm, which do not require a theory of the good life [25, p. 301].

However, critics have also questioned the weak interpretation of Diekema’s claim. Giles Birchley has argued that the harm principle is indeterminate, citing examples from rulings in English law where the harm principle has led different judges to produce disparate judgments in similar situations [15, pp. 113–114]. Birchley argues that these disagreements over what constitutes harm show that “harm does not rest on a self-evident basis” [26, p. 123]. Birchley concludes that “all we have done is rename the best interests test while dealing with none of its failings” [15, p. 114].¹³ Johan Bester concedes that the best interest standard requires weighing values, but also argues that “the harm principle does not avoid this, nor is it less vague and indeterminate” [27, p. 300].^14,15 These objections to the weak interpretation are far too brief, leaving the reader with only examples of reasonable disagreements and not an underlying conceptual framework that shows how such disagreements arise—that is, a framework that shows what categories of harm people disagree about.¹⁶

Here I believe Erica Salter’s schema of the normative dimensions involved in various formulations of the best interest standard is instructive, as it sketches an explanation for why so much disagreement arises in the first place [1].¹⁷ Salter categorizes views on the best interest standard in terms of whether they consider physiological, psychological-emotional, relational, or cognitive-developmental interests. Physiological interests include those basic elements “necessary for survival,” such as oxygen, water, and food; psychological interests include “internal affective states,” such as sense of security, depression, and anxiety; relational interests include “those interests that concern interpersonal connections, or relationships,” such as a patient’s desire to postpone major surgery until after a friend’s wedding; finally, cognitive-developmental interests are related to the development of “certain cognitive capacities, knowledge, and skills,” such as the ability to make reasoned decisions [1, p. 187]. Salter also organizes interpretations of the best interest standard according to whether they emphasize present- or future-oriented interests and according to whether they use an individualistic model centered on patient interests or a relational model that considers interests outside those of the individual patient.

Salter shows how six common iterations of the best interest standard differ with respect to the kinds of interests they recognize, their emphasis on present- or future-oriented interests, and their consideration of interests internal or external to the patient alone. Salter concludes:

What constitutes “benefit”? What constitutes “harm”? What range of benefits and harms do you consider: only physiological benefits or psychological, relational and emotional benefits as well? And how are we to assign value to benefits or harms? Depending on one’s particular world view and value set, these questions can be answered in numerous different ways. [1, p. 189]

I agree with Salter that the normative dimensions of the best interest standard inevitably court controversy, and I assert that the same questions arise when making judgments about harm.

It is not my intention here to exhaustively analyze the views of Ross, Dworkin, and Feinberg that Diekema compiles to formulate his harm threshold; nor do I aim to situate these views within the schema given by Salter. Rather, I contend that conceptions of harm, like formulations of best interests, can be divided according to Salter’s schema. Dividing harm in this way precipitates the same problems that arise in an analysis of best interests. Theorists are likely to diverge in their emphases on present-oriented or future-oriented harm, physiological or relational harm, harm to the patient or harm to others affected by a decision, and so forth. Different conceptions of harm are apt to split along these lines because different authors bring different worldviews and values to their considerations of the concept.¹⁸ Applying Salter’s schema to the harm principle offers an explanation for why Birchley and Bester are able to provide so many examples which challenge Diekema’s claim that harm is less indeterminate than considerations of best interest.

Even if Diekema is right about the greater determinacy of the harm principle in general use, not all will agree about what constitutes imminent and significant harm to a child. This discordance is particularly evident in cases involving metaphysically tethered harms, which Diekema questionably presumes are the examples most easily settled through application of the harm principle. Far from demonstrating the effectiveness of the harm principle, such application to metaphysically tethered disputes reveals a significant weakness.

The harm principle needs metaphysical elaboration

In this section, I argue that Diekema’s harm principle does not offer the substantive guidance that he claims it does in cases of Jehovah’s Witnesses and Christian Scientists.¹⁹ This is because Diekema’s formulation fails to incorporate claims regarding what Gregory Bock has called “spiritual harm” [29, p. 91]. Bock’s conception of spiritual harm is ambiguous; he claims it includes both (1) consideration of how a patient might be “ostracized or excommunicated” by their spiritual community and (2) acknowledgment that “there might be eternal consequences that cannot be undone” [29, p. 91]. Here I am concerned with spiritual harm in Bock’s second, metaphysical sense. Whether blood transfusion refusal constitutes harm depends upon the metaphysical beliefs one holds—beliefs about reality at the level of one’s worldview. Because Diekema’s conception of harm has no metaphysical elaboration, whether it can be used to prevent or to force blood transfusions hinges on the metaphysics of the one who wields it.²⁰ A notion of harm without metaphysical elaboration is not the right tool for resolving metaphysically loaded disagreements over what constitutes harm. Whether blood transfusions for children should be permitted depends, in part, upon the plausibility of the claim that transfusions lead to the loss of eternal salvation.

Most harm is ametaphysical harm, meaning it may be readily seen or experienced as harm without necessarily needing a metaphysical explanation to be understood as harmful. Severe full body burns, diabetes, breast cancer, and broken bones exemplify ametaphysical harm. Metaphysical harm is that which may or may not be directly experienced as harmful, but can come to be understood as harm through the acceptance of metaphysical claims. Metaphysical claims have the power to change the meaning of the mundane. As Engelhardt puts it, “belief gives transforming meaning to the ordinary” [34, p. 9]. For instance, what is merely one kind of meat among others to many Western eyes (pork) is, for some Muslims, an unholy object. Metaphysical claims about the unholiness of pork are required to turn the mundane object of pork into something considered harmful. These beliefs carry over into medicine through, for example, the belief of some Muslims that vaccines containing pork gelatin ought to be avoided.²¹ Once it is understood that harm can have a metaphysical dimension, it becomes clear that Diekema’s notion of harm cannot actually adjudicate the very cases where he claims it most easily applies—those of Jehovah’s Witnesses and Christian Scientists.

Diekema seems unaware of the metaphysical dimension of some harm when he claims his harm principle would easily dispatch the spiritual concerns of Jehovah’s Witnesses, Christian Scientists,²² and, one can presume, many other similar views [7, p. 255].²³ To demonstrate why this is so, Diekema offers eight conditions that must be met to determine if a harm threshold has been crossed, claiming that cases involving Jehovah’s Witnesses and Christian Scientists meet all eight.

• 1. By refusing to consent are the parents placing their children at significant risk of serious harm?

• 2. Is the harm imminent, requiring immediate action to prevent it?

• 3. Is the intervention that has been refused necessary to prevent the serious harm?

• 4. Is the intervention that has been refused of proven efficacy, and therefore, likely to prevent harm?

• 5. Does the intervention that has been refused by the parents not also place the child at significant risk of serious harm, and do its projected benefits outweigh its projected burdens significantly more favorably than the option chosen by the parents?

• 6. Would any other option prevent serious harm to the child in a way that is less intrusive to parental autonomy and more acceptable to the parents?

• 7. Can the state intervention be generalized to all other similar situations?

• 8. Would most parents agree that the state intervention was reasonable? [7, p. 252]

In applying these conditions to Jehovah’s Witnesses and Christian Scientists, Diekema writes:

These cases satisfy the eight conditions of the harm principle rather easily. In both situations, withholding treatment (blood or insulin) represents a significant risk of serious harm (i.e., death). Treatment is necessary to prevent harm befalling the child, and treatment is of proven efficacy. Treatment provides great benefit (prevention of death), imparts minimal risk of harm, and represents proportionately greater benefit than harm to the child. State intervention is justified in all similar cases, and is not restricted solely to those cases in which the parents refuse treatment on religious grounds. For example, if a parent refused because of their concern about the potential for tainted blood being introduced into their child, state intervention would still be sought (because the potential for harm to the child does not differ). Finally, the decision to intervene in these situations can be defended in the public forum and will most likely be overwhelmingly supported. [7, p. 255]

Diekema’s claim that Jehovah’s Witnesses and Christian Scientists would violate the harm principle in the cases he presents is not a conclusion that can be drawn from the harm principle as he formulates it, but one already presupposed by the metaphysic he smuggles in. To demonstrate this, a flip test can be applied to Diekema’s eight requirements. Suppose Diekema finds himself as a clinical ethicist in a society where ninety-five percent of people subscribe to the Jehovah’s Witness religion. In such a situation, how might Diekema’s questions fare if a parent requested a blood transfusion for their child? By requesting the intervention (in this case, a blood transfusion), are the parents placing their child at a significant risk of serious harm? Is the harm imminent, requiring immediate action to prevent it? Can the state intervention be generalized to all similar situations? Would most parents agree that the state intervention was reasonable?²⁴ In this hypothetical Jehovah’s Witness society, would not all these questions be answered in the affirmative? Following Diekema’s harm principle, the state would refuse to honor parental requests for blood transfusions for their children.

Diekema’s harm principle thus leads to an opposite result when the metaphysic of those deploying the principle is changed. The substantive guidance of the harm principle comes from the metaphysic of those deploying it, not from the formulation of the harm principle itself. Salter has argued for a standard of decision-making that has external consistency, meaning “it must produce similar outcomes when applied to similar cases” [1, p. 196]. Diekema’s principle fails that test, showing that the harm principle and its eight requirements do not in themselves dictate that state intervention be triggered to deny Jehovah’s Witness refusals. Whether one understands a blood transfusion as harmful depends upon whether one believes there to be a plausible link between refusing blood and eternal salvation. Diekema fixes the conclusion by presuming in advance that considerations of spiritual harm are not to be taken seriously, especially when doing so is likely to lead to a child’s death. Diekema is right that it is easier to override Jehovah’s Witnesses and Christian Scientists because state intervention in such cases is likely to be relatively uncontroversial with the public and there is significant legal precedent, but this is not a good argument that the harm principle he has offered is sufficient for doing so. In deciding what actions cross a harm threshold, one must first ask whose harm, which metaphysic?

In the practice of medicine, this question has already been answered; Jehovah’s Witnesses are not permitted to deny their children low-risk and proven medical procedures that can prevent death or disability.²⁵ Just as Diekema argues that the harm principle seems already to be operative as the actual principle appealed to in cases of state intervention, so too do I argue that some metaphysical commitments are already presumed by the clinicians, ethicists, and courts who argue that there should be limits on what kinds of religious beliefs can motivate pediatric medical decisions.²⁶ The state justifies these limits by claiming to be saving children from death [39, p. 490]; this presumes death is worse than the claims of spiritual harm being made by parents. Forced transfusions are also justified by the claim that parents have a duty to the “health, safety, and wellbeing” of their children; they cannot “make decisions that may permanently harm or otherwise impair their healthy development” [40, p. 715].

But the presumption that the risk of death is worse than the risk of losing eternal salvation through a blood transfusion is a metaphysical claim. Likewise, what can words like health, safety, and wellbeing mean in the context of metaphysically tethered disagreements situated within a system purportedly absent of metaphysical commitments? Metaphysical neutrality is broken in these cases in the name of the platitude “harm.” The state does the right thing, but for insufficient reason. Appeals to harm must come with a moral-metaphysical standard; moral claims must be accompanied by claims about the nature of ultimate reality. Although state intervention is triggered in such cases, it cannot be triggered on the basis of a harm principle purportedly devoid of metaphysical commitments, as such commitments are required to reach the judgment that the death of a child is an unacceptable evil that ought to be avoided by means of force if necessary.

The harm principle does not need to be understood within the context of a comprehensive metaphysical worldview, but only requires commitment to some metaphysical claims. Where contemporary middle-level moral theory is one in which bioethics is asked to do some, but not too much, moral theory, so too should bioethics commit itself to some, but not too much, metaphysics. A middle-level metaphysic is needed to match the middle-level moral theory of bioethics. Examples of middle-level metaphysical claims could include the following: sickness is not a result of spiritual distance from God best remedied through prayer alone; blood transfusions do not bar one from salvation; exorcism is not an effective means of curing autism;²⁷ the ultra-orthodox Jewish practice of metzitzah b’peh (sucking with the mouth) is not a spiritually valid way to practice circumcision.²⁸ Such claims are “middle-level” because they do not require the adoption of a comprehensive worldview, but function as side-constraints upon the religious beliefs that guide pediatric decision-making.

The reader should take notice of the negative form of the examples offered above. These examples show that a claim denying some metaphysical reality is no less a metaphysical claim, a point that is often misunderstood. Timothy Murphy demonstrates this error when he argues for an “irreligious bioethics” (his term for a bioethics based on naturalism) that differs from “religious” approaches in that it does not base its values on presuppositions “accessible only by logically prior commitments to certain theological claims” [42, p. 7]. However, Murphy’s proposition that irreligious bioethics does not rely on logically prior commitments is false, since his irreligious/naturalistic approach is based on the theological/metaphysical claim that the natural world is all there is. Naturalism is not somehow free of its own metaphysical claims; its claims about the ultimate nature of reality need to be argued for and defended like those of any other religion. Therefore, the denial of a positive metaphysical, religious, theological claim (e.g., there is no God, there is no soul, there is no sin) is no less a metaphysical, religious, theological claim.²⁹ The competing claims that (1) blood transfusion leads to loss of eternal salvation and (2) blood transfusion does not lead to loss of eternal salvation are both equally metaphysical. Regardless of the position one takes with respect to the connection between blood transfusions and salvation, one is doing metaphysics, and a moral-metaphysical notion of harm must take a position on such questions to offer substantive guidance.³⁰

The idea that a moral-metaphysical standard of decision-making is needed for clinical ethics receives resistance on several fronts. Secular political pressure, which asserts the doctrine of neutrality with respect to basic values, religion, and metaphysics, is one force of resistance [18, pp. 56–57; 47, p. 209; 48, p. 46; 49]. Philosophical resistance to substantive moral-metaphysical claims arises from the fact that moral theory is in an entrenched state of dissensus regarding the nature of the right, the good, and moral worth [50, 51], and strong arguments have been made that there is no rational way out of these disagreements [34, 52, 53]. These political and philosophical pressures have had a strong influence on clinical ethics, especially with respect to the standardization debate, where theorists have argued over the question of whether ethicists have ethics expertise.³¹

The problem of ethics expertise considers whether clinical ethicists have the ability to make authoritative recommendations about what ought to be done in difficult moral-medical dilemmas. What I will call the negative of ethics expertise is that which denies that ethicists have ethics expertise, full stop. This negative view has been championed in recent years by Tristram Engelhardt. It has motivated an approach to clinical ethics whereby, it is argued, ethicists serve as useful jacks of many trades, who are able to provide services such as quasi-legal advice, risk management, conflict mediation, and intellectual insight without ever claiming to know what should actually be done [55, p. 365; 56, p. 299; 57, p. 141].³² Although Engelhardt has been a vocal advocate of the negative view, it has received strong support from bioethicists since the genesis of the field. Larry Churchill succinctly expresses support for this view when he writes, “The ethicist is qualified to examine, explore, and illumine the complex facets of moral questions, and to clarify their logic; but he stops short of telling people what is right and wrong” [58, p. 14]. Such a clinical ethicist can help individuals clarify their views or avoid logical fallacies, but she cannot—with any sense of authority—inform the content of their views beyond these thin constraints.³³

What makes Engelhardt such an able champion of the negative view is that he offers a strong argument in favor of its adoption. Engelhardt repeatedly argues that moral pluralism is not only a sociological fact, but also a part of our “ontological condition” [59, p. xiii]. By this he means that it is an inescapable fact of reason that is unable to ground any particular moral view. To support such a strong claim, Engelhardt relies on an argument he attributes to the early philosopher Agrippa the Skeptic [60, p. 251].³⁴ This argument states that where moral conflict exists between moral strangers (i.e., those who lack shared basic premises and rules of evidence), disputants argue in a circle, beg the question, or engage in an infinite regression of justification [34, p. ix; 56, p. 364; 58, p. 138; 61, pp. 251–252]. The implication of Engelhardt’s argument is radical; it concludes that no moral disagreements can be rationally resolved. This conclusion applies to the more familiar disagreements over whether justice demands universal health care, but it also applies with equal force to disagreements over whether denying a blood transfusion to a child crosses a harm threshold that warrants state intervention. In fact, the conclusion is so radical, and so contrary to the need for normative guidance that originally gave rise to bioethics, that Engelhardt and others have claimed that both academic and clinical bioethics are in need of foundationally critical reconsideration [62].

The negative view of ethics expertise grounded in an argument for moral pluralism is flatly inconsistent with the facilitation model of ethics consultation put forward by the American Society for Bioethics and Humanities (ASBH). ASBH advances a facilitation model wherein the ethicist helps to guide decision-making toward solutions that exist within the “range of ethically acceptable options” [63, p. 8]. Consistent with the ASBH facilitation model, standards like Diekema’s harm principle represent attempts to establish the boundaries of ethically acceptable options. However, the setting of such boundaries, which establish a line that one should not cross, requires ethics expertise, the positing of which is inconsistent with the negative view.³⁵

Cases of religiously based medical requests in pediatrics are philosophically interesting because they lay bear this inconsistency between the facilitation model and negative views of ethics expertise. Either Ethicists, clinicians, and the courts must concede to the philosophical and political pressures driving the negative view and remain neutral with respect to moral-metaphysical claims, which would require permitting Jehovah’s Witnesses and Christian Scientists to deny their children medical treatment, or they must admit that these moral-metaphysical worldviews are rationally insufficient to warrant risking the deaths of children, which would require conceding that ethicists have some ability to demarcate the boundaries of ethically acceptable options in the clinic. Diekema’s project attempts to defend a standard for drawing some lines of ethical acceptability; my thesis is that no such lines can be drawn in metaphysically tethered disputes without metaphysical supplementation. The harm principle can be saved, but doing so will require bioethicists to accompany their middle-level moral theory with some middle-level metaphysics.

Conclusion

Diekema has argued that cases of state intervention do the right thing but appeal to the wrong moral principle in justifying this action. I have argued that Diekema gives the right moral principle but without sufficient elaboration to do the job. Such disputes involve not only value commitments but also metaphysical claims at the worldview level. Thus, the harm principle, and any principle deployed in metaphysically tethered disagreements, must be considered moral-metaphysical if it is to succeed in offering substantive guidance. Yet positing a moral-metaphysical standard of decision-making is inconsistent with the negative view of ethics expertise. Cases of state intervention for the children of Jehovah’s Witnesses and Christian Scientists force a resolution to this inconsistency. Either ethicists, clinicians, and the courts must begin permitting Jehovah’s Witnesses to deny their children transfusions or it must be admitted that there is an inconsistency such that reason has adjudicated a moral-metaphysical dispute. Admitting that such an inconsistency exists is no small claim amidst the moral pluralism that characterizes the postmodern ethos. If the inconsistency is acknowledged, then ethicists, clinicians, and the courts are implicated in the practice of positing moral-metaphysical standards and implementing them at the state level with a threat of force if necessary. If the inconsistency is acknowledged, then scholars must stop denying that clinical ethicists make moral-metaphysical claims, that they do not tell others what is right or wrong, that they do not take positions on the ultimate nature of reality. If the inconsistency is acknowledged, then transparency requires bioethicists to articulate the minimal metaphysical commitments of the harm principle, or any moral standard to which they appeal. Explicit minimal metaphysical commitments provide the backbone necessary for the harm principle to avoid becoming just another moral standard that offers little meaningful guidance. I have argued here only that explicit metaphysical commitments are needed in order for the harm principle, or any other moral standard, to be able to offer substantive guidance in metaphysically tethered disputes,³⁶ and such a move is inconsistent with negative views of ethics expertise. I have not described or justified any particular approach to determining the specific metaphysical commitments that ought to be admitted. This is the positive project that I leave for future work on this topic.