Balancing in Ethical Deliberation: Superior to Specification and Casuistry

Abstract

Approaches to clinical ethics dilemmas that rely on basic principles or rules are difficult to apply because of vagueness and conflict among basic values. In response, casuistry rejects the use of basic values, and specification produces a large set of specified rules that are presumably easily applicable. Balancing is a method employed to weigh the relative importance of different and conflicting values in application. We argue against casuistry and specification, claiming that balancing is superior partly because it most clearly exhibits the reasoning behind moral decision-making. Hence, balancing may be most effective in teaching bioethics to medical professionals.

Keywords:

• balancing
• casuistry
• education
• ethics consultation
• specification

I. INTRODUCTION

Theories employing multiple principles and rules face two major problems:

1. It is often unclear how they should be applied, and

2. in application norms may conflict.

Resolving these problems is the foremost obstacle facing bioethical theories such as Tom L. Beauchamp and James F. Childress' “principlism.” In the fourth and fifth editions of their text they respond by supporting Henry Richardson's specification. Carson Strong recently countered by claiming that casuistry is superior to specification. Despite support for these methods, we argue that balancing is a better approach and that both specification and casuistry do not adequately support moral reasoning in bioethics.

After offering brief critical accounts of specification and Strong's casuistry, we examine a case that shows the weaknesses of specification and casuistry and the strengths of balancing. The issues we explore ultimately concern effective moral reasoning in bioethics. Even if specification and casuistry can accomplish what they promise — something we doubt — we explain why they would not enhance moral reasoning. Instead, balancing is the correct way to handle conflicting norms, both in complex cases and in teaching bioethics.

II SPECIFICATION

Specification has been praised for its ability to make abstract principles and rules applicable in bioethics. ¹ Specification does this by making “general norms more specific for a particular context or range of cases” (Beauchamp & Childress, 2001, p. 15). Also, “Specification is a process of reducing the indeterminateness of abstract norms and providing them with action-guiding content” (Beauchamp & Childress, 2001, p. 16). Principles, such as “Do no harm,” are too sparse to be applied in complex situations and thus require specification. However, Beauchamp and Childress caution that specification, also used to eliminate contingent conflict, “may be arbitrary, lack impartiality, or fail for other reasons.” They claim that specification does not always eliminate the need for balancing (Beauchamp & Childress, 2001, p. 16).

In their initial statement endorsing specification, Beauchamp and Childress refer to its more technical presentation by Henry S. Richardson. He argues that specification is superior to balancing, which he finds too subjective or intuitive to support collaborative moral problem solving. He rejects “justified balancing of plural considerations as a recommended way of bringing … plural principles to bear on practice” (Richardson, 2000, p. 298). We, instead, strongly support such balancing.

When norms conflict, balancing is the attempt to determine which rules or principles take priority in a particular case. This resolves the conflict by deciding which considerations are the most compelling in the case at hand. In contrast, specification attempts to make principles or rules more specific by adding features. Specification results in a new norm, albeit one that is related to the specified original norm. The new norm typically dissolves the moral conflict in the case at hand and is applicable to all cases similar in basic circumstances (Richardson, 2000, p. 288). ²

For example, when facing conflict between parental decision-making and harm to a child due to a Jehovah's Witness's decision to reject needed blood products for his or her child, Beauchamp specifies the rule supporting parental decision-making by adding a clause relating to children's well-being: “It is morally prohibited to disrespect a parental refusal of treatment, unless the refusal constitutes child abuse, child neglect, or violates a right of the child” (Beauchamp, 2003, p. 270). ³ Because the Jehovah's Witness' refusal of needed treatment is considered some form of neglect, this new moral rule attempts to dissolve moral conflict. That is, with the specification, there is only one applicable norm; in this case the new norm does not conflict with other norms.

Richardson makes the idea of specification more precise by claiming that a genuine specification must involve extensional narrowing. Basically, this means that fewer types of actions are covered by the specified norm. More exactly, extensional narrowing means that actions satisfying the specified norm must also satisfy the original norm. An example may help to understand this. Richardson critiques a specification offered by Beauchamp and Childress that may violate this requirement. They say: “A typical example of a rule that specifies the principle of respect for autonomy by giving it more content is, ‘Follow a patient's advance directive whenever it is clear and relevant’” (Beauchamp & Childress, 2001, p. 39). ⁴ By giving more content, Beauchamp and Childress indicate that it narrows the focus of the principle for application in a particular circumstance. But do actions that satisfy the specified norm also satisfy the original norm? That is, do all actions that follow an advance directive, as a written document, also respect a person's autonomy? It is easy to think of circumstances in which this is not so: for example, when a person changes his or her mind before becoming incompetent without changing the advance directive. Given this, the rule about following an advance directive is probably not a specification of the autonomy principle. Richardson says: “The connection that Beauchamp and Childress make out between respecting autonomy and following advance directives is a theoretical achievement that takes a relatively complex argument to set out” (2000, p. 290).

Specifications are the conclusions of sustained theoretical argumentation, which, of course, are not included in the resulting norms. Whenever a specification is questioned, the specification itself does not give adequate guidance, but instead the guidance may come by way of the original, but perhaps unavailable, argumentation supporting the new norm.

Specification is both a process and a result. Richardson gives few details about the process except by way of examples. The process of producing a specified principle or rule is complex, ironically often using balancing while appealing to other norms. Richardson provides a detailed example of a person deciding whether it is justifiable to use disposable diapers despite their environmental impact. The reasoning is complex, showing how balancing is justifiably used, in Richardson's view, to help determine the properly applied principles (Richardson, 1990, pp. 305 – 308). In general, he gives more instruction on specifications as results than on the deliberative process. As a process, specification helps an individual decision-maker to avoid moral conflicts. As a result, a specification gives a new norm. Presumably, once the new norm is in place, there is no longer need for a repetition of the deliberative process when similar circumstances arise.

Richardson carefully defines what an achieved specification is and indicates that it either stands alongside or replaces the original specified norm (1990, pp. 292 – 295). These new principles are action-guiding. He notes that “an adequate set of action-guiding principles — at least in a fast-changing context such as bioethics, if not in human life in general — requires the progressive collaboration of many practitioners and theorists, each building on the work of others” (2000, p. 286). He also claims that, “What allows the idea of specification to offer a third way of reflectively coping with conflicts among principles is the fact that it offers a change in the set of norms” (2000, p. 298). Thus Richardson seems to support the development of a rich set of specified norms, albeit changing and developing, that can be used to provide an applicable ethical theory in medical ethics. Keep in mind that this set would only contain specified and/or original principles or rules, and not the reasoning that supports them. ⁵

Thus, specification is an attempt to populate our moral universe with a wide variety of additional moral rules. If established, these rules are apparently intended, at least in typical circumstances, to serve as a moral rulebook: look up the appropriate rule and then the proper decision is at hand. Yet the rule itself may be questioned or rejected. Due to the controversial nature of specifications as the product of complex argumentation, such a moral rulebook is unlikely to be realized. Furthermore, to be an adequate rulebook, there would need to be many specifications making it unwieldy and difficult to apply, subverting the intended purpose, to foster application.

Even if we had an agreed-upon moral rulebook, it would be improper to use. The results of specification tend to obscure the importance of some of the values involved in moral decision-making. ⁶ This is especially important when conflicts are difficult to resolve. Because specifications are new principles or rules that have presumably already resolved or dissolved conflict, all the originally conflicting norms may no longer be apparent. As we show in a case more fully covered in the next section involving parental decision-making and the well-being of an severely ill infant, a good moral decision should balance, among other things, the moral authority of the parents, the severity of the deformity, the likelihood of effective intervention, the expected quality of life of the infant, and the uncertainty involved with each factor. These factors are crucial in reaching a well-reasoned moral judgment. Moral decision-makers may disagree. By simply appealing to one of the set of specifications that Richardson speaks of, the deliberation is about acceptance of this or that norm and thus, as a result covers up the role each factor played in moral decision-making and in deciding on a specification. In contrast, when the factors involved are clearly expressed, as in the case of proper balancing, moral deliberation is enhanced.

III CASUISTRY

Casuistry relies on concrete cases as moral guidelines; these are used to indicate how norms are to be applied, especially in cases in which they conflict. Suppose one is faced with a difficult issue involving a parent's desire to withdraw treatment from a newborn. If one can find two cases involving similar circumstances, one permitting the withdrawal of treatment and the other requiring treatment, then the given case might be resolved by considering which of the two is closer in circumstances. If the given case is more like the one requiring treatment, then the case at hand should be decided by requiring treatment against the parent's wishes.

Specification involves the development of a rich set of norms; likewise, casuistry requires many cases, maybe thousands, to cover the multiplicity of concerns in clinical bioethics. It is as difficult to assemble good cases as it is to assemble good specified rules. But if we had a good set of cases, would it be morally advisable for health care professionals to solve moral problems by locating and comparing cases?

To help answer this question, we examine a case first presented by Richardson and then analyzed by Strong. The case centers on whether to withhold nutrition and hydration from a severely malformed newborn on the request of the child's mother and father. ⁷ Richardson begins by locating conflicting norms: one is against deliberate killing; another favors respecting the reasonable choices of parents; and the third is about the duty of medical personnel to benefit persons in their care. These requirements are vague; he offers a relatively complex specification of the norm against deliberate killing with the hope that it will dissolve the conflict:

It is generally wrong directly to kill innocent human beings who have attained self-consciousness, and generally wrong directly to kill human beings with the potential to develop self-consciousness who would not be better off dead, but it is not generally wrong directly to kill human beings who meet neither of these criteria. (Richardson, 1990, p. 304)

The reasoning Richardson offers in support of the specified principle is virtually nonexistent, and is put in the voice of the mother who simply views it as a specification of respect for the newborn. We suspect that balancing relevant considerations would be involved in reaching this complex rule. However, once the rule is established, conflicting considerations are no longer considered. Since no reasoning is offered, one may wonder why this specification is offered. For example, we may question whether an exception to killing should be made due to a lack of potential for self-consciousness. Also, is the specification clear enough for application? Note that it includes a clause about being better off dead. Unclear are the circumstances under which a person is better off dead; this involves balancing. ⁸

Strong has similar doubts about specification. He presents casuistry as a way to counter specification yet provide for specific guidance in moral decision-making. Since Strong objects to Richardson's use of a very sketchy case, one lacking in ethically relevant detail, he presents a more detailed problematic case: An infant is diagnosed with trisomy 18 syndrome. She has serious heart defects and an esophagus that does not extend to her stomach (esophageal atresia). The infant has only a 10% chance of surviving the first year, and then with severe mental deficits. The parents do not want continuation of nutrition and hydration or surgery to correct the infant's esophageal atresia.

Strong shows that applying Richardson's specifications to the more complex problematic case is inconclusive. In the briefer case, Richardson believes that the infant might not be better off dead, but understands that this is a vague notion. Instead of facing that issue, Richardson offers a specification of respect for parental decision-making that dissolves the conflict: One should respect the “reasonable choices of parents regarding their children so long as they respect the children's rights” (Richardson, 1990, p. 305). Strong rightly raises questions about the presumed rights of the child. The rights of the neonates are often unclear. Does the child have a right to the surgery? Also, many, perhaps all, rights are not absolute. Richardson's specification makes the mistake of holding that any presumed right of the child overcomes the norm to respect parental decision-making. A choice between the rights of children and parents involves balancing. The price of discarding balancing is that genuine moral conflicts are hidden by an a priori decision to give a dominating weight to one end of the conflict.

Strong might reject the idea that balancing is required because it appears to play no role in the casuistry he supports. In casuistry, one identifies the main values in the problematic case and the foreseeable alternative actions that might be taken. Then one considers the morally relevant ways cases of this sort might differ. The next step is to find similar paradigm cases taking into account morally relevant features that would justify each alternative action.

Following his plan, Strong addresses the problematic case by presenting two additional cases. The first case supports overriding the parent's decision not to treat an infant's esophageal atresia. The infant suffers from Klinefelter's syndrome and is expected to have relatively mild physical abnormalities and an IQ that is 10 to 15 points lower than otherwise would be expected. Also, the child will probably face behavioral problems, including immaturity (Strong, 2000, p. 332). The features supporting treatment are: high likelihood of survival, acceptable cognitive awareness, and so on (Strong, 2000, p. 333). ⁹

In ways this case is similar to the problematic case. In other ways the case is dissimilar, mainly because of low probability of survival and of extreme cognitive deficit in the problematic case.

The second case involves an anencephalic infant with esophageal atresia. In this case, circumstances suggest that it is morally appropriate to follow the wishes of the parent by stopping treatment intended to prolong life. This case is also unhelpful. Disabilities are more extreme than in the problematic case. Prognosis is more certain, and the outcome is more tragic.

Despite the fact that neither paradigmatic case is adequately analogous, in our opinion, to the problematic case, Strong contends that a resolution depends on deciding which case is closer. He claims: “When the case at hand is closer to one paradigm than to the others, the course of action justifiable in that paradigm would also be justifiable in the case at hand” (2000, p. 324). ¹⁰ He argues that the problematic case is closer to the one involving the anencephalic infant. “Argues” is a key term. Strong relies on general considerations including the likelihood of long-term survival and the extent of cognitive deficit. These are used to show that the case is closer to the one justifying withholding aggressive treatment. However, his analysis gives the reasons in favor of rejecting treatment without reference to the paradigm case. Despite Strong's view, it is not the similarity with the paradigmatic case that matters; rather consideration of the complex “weight” of the medical circumstances of the child does the real moral work. All three cases involve different medical circumstances; determining whether or not treatment should proceed should depend on the seriousness of these circumstances, as it does in the two paradigmatic cases. Thus, in his argumentation, balancing such circumstances is key.

In order to use casuistry to resolve problematic cases, as Strong explains it, access to a very large number of paradigmatic cases would be required. Counterfactually, let us assume that we have a large number of cases acceptable to the relevant community of inquirers. ¹¹ Practitioners and bioethicists would have to locate two cases for each alternative action, one of which justifies the action while the other does not. Then for each alternative action, one would have to argue that the problematic case at hand is closer to one or the other for each feature. This would be a daunting task.

If the point is simply to find the closer case, as it seems to be, then the method is harmful for moral thinking. The two cases Strong presents are different from the problematic case, so finding the closer one does no deep moral work. As the saying goes, the devil is in the details; by relying on the closer case we ignore the details that separate the problematic case from the supposedly easier to decide paradigmatic cases. On the other hand, the moral work may be done by the argumentation suggested by the paradigmatic cases. This seems to run counter to the case method, but paradoxically is involved in it. If the moral argumentation bears the burden, then the cases are not necessary, except perhaps as instructive examples. Either way, emphasis only on cases hinders rather than helps moral inquiry.

It may be helpful to compare casuistry to use of court cases in the common law tradition where cases are precedents by fiat. Often cases are decided by a 5 to 4 judicial vote, yet they are binding in a relatively strong way. This is not the way it is in moral reasoning. A case that is closely decided has a weak moral force, if any. A case that has a significant minority in opposition is also weak. In the law, there are a vast number and variety of cases. Lawyers on each side, hopefully skillful, present arguments that their case is more like past cases that support their client. The opposing side presents similar arguments. That competition is intended to control bias, on both sides, in case selection. Judges then declare a verdict, often involving detailed argumentation. This system “works” because lawyers are extensively trained in dealing with case precedents, resources are available to locate such cases, and all agree that even a 5 to 4 verdict counts as a precedent. Adversarial deliberation helps to ensure that hopefully independent and competent judges will fully hear both sides, and appeals are possible. This is unlike casuistry in bioethics in which one or several individuals offer cases, probably by making them up, and then a decision is made about which is closer to the case at hand.

Learning to do bioethics using either specification or casuistry is unrealistic. We doubt that a sufficient number of specifications or cases can be developed, learned, and applied. We can question the legitimacy of decisions encapsulated in specifications and in paradigmatic cases. We can question whether a particular specification or paradigmatic case applies. Specification relies on formulaic rules and casuistry relies on similar, yet different, case paradigms. Using these techniques tends to hide actual conflict and impedes moral debate. Both obscure the fact that in actual moral decision-making, conflicting moral considerations need to be carefully considered in relation to each other.

IV BALANCING

Balancing is a metaphor for the attempt to determine the relative importance of conflicting values in particular cases or classes of cases in order to come to a conclusion mainly about moral obligations. Balancing may be intuitive or deliberative, or a hybrid of the two. In intuitive balancing, reasons are not offered to support the decision that one value is of greater importance than another involved in a particular conflict. Deliberative balancing provides reasons for believing that one value has greater importance than another. For example, suppose parents object, on religious grounds, to possibly life saving treatment for their minor child. Balancing would involve an articulation of the possibly competing values. In this case the values may be parental rights, risk, uncertainty, avoidance of harm or death, and the burdens of treatment. In an easy to resolve case, a child will surely die without a relatively burden and risk free treatment. It can be argued, from a variety of perspectives, including the parent's responsibility to protect a child, that parental decision-making is of lesser importance under such circumstances. The harm is extremely serious, and the risks and burdens of treatment are minor. Reference might be made to the risks and burdens people normally assume in everyday life. This balancing might be encapsulated, more or less as argued, in a rule of thumb, but the key point, that the potential net benefit is great enough to override parental rights based on religious motivation, would be stated in order to make at least implicitly clear that with lesser benefits, or greater burdens, the conclusion might be different. The result of such a balancing, which may be expected to have wide applicability in many cases, may thus resemble a specification although it would not conform, as we believe it should be stated, to the requirements that Richardson places on specification.

Balancing is helpful because it is attentive to the fact that the issues involved in many actual cases form a potential continuum. For example, risk can involve any percent, harm avoided can vary from the almost inconsequential to death, and a parent's reasons may vary from simple convenience to deeply held religious conviction supported by extensive involvement in a religion. Although both casuistry and specification implicitly depend on offering reasons for considering one value to be more important than another, balancing is explicitly designed to make such comparisons, and should make them explicit or implicit in the resulting judgments. It is the only method of the three that explicitly concentrates on the true complexity of actual circumstances, which tend to form a full range of concerns.

Richardson distinguishes two types of balancing: contextual and global. Contextual or piecemeal balancing is considered unobjectionable. In particular, Richardson objects to “global balancing,” but finds acceptable “contextual balancing.” However, the distinction is not clearly drawn, except to say that contextual balancing is “dictated by the content of some principle …” (Richardson, 2000, p. 286). For example, the principle of nonmaleficence requires consideration of net harm in terms of a balancing between harms and benefits. In criticizing Gert, Culver, and Clouser (1997), Richardson points out that “harm” underscores all of their rules, and so in balancing the demands of diverse conflicting rules, they engage in “global” balancing. Richardson points out that the problem with the approach of Gert, Culver, and Clouser (1997) is that the harms involved are of different types and so they cannot be balanced on a single scale. Thus, Gert, Culver, and Clouser resort is to intuitive balancing. The main difficulty with the approach, according to Richardson, is that balancing “tends to mask the real reasons at work. At the very least, it fails to encourage the articulation of the real reasons….” It provides “an excuse for laziness” (Richardson, 2000, p. 297). This all suggests that contextual balancing involves good reasons for weighing commensurable values.

Richardson's distinction creates problems because it suggests that, appropriate to its name, “global” balancing improperly involves the reduction of all values to a single scale. This, however, is not the intention of Gert, Culver, and Clouser who, as Richardson states, recognize distinct types of harm. Despite the suggestion that “global” balancing involves the reduction of all values to one scale, the problem with it is that it compares incomparable values intuitively. Contextual balancing seems, by contrast, to properly stick to an interpretation of a single principle. So it would seem that any theory that involves comparing values among diverse principles or norms is guilty of a mislabeled “global” balancing. Ironically, the contextual balancing that Richardson finds unobjectionable involves the same problems Richardson finds with global balancing. The risks, benefits, and harms of surgery are not comparable on a single scale. Consider the differences among death, behavioral changes, loss of physical function, incapacitation, monetary cost, and risk and uncertainties. Balancing harms and benefits in case of certain types of surgery, say involving deep brain stimulation, may involve more complex items and more incommensurable values than cases of balancing involving the rules of Gert, Culver, and Clouser. Yet such balancing is essential to even simple applications of the principle of nonmaleficence.

We conclude that the distinction between global and contextual balancing is unhelpful. Each involves similar considerations. Also, many if not most cases of troubling moral conflict involve different norms. If it is the case that the involvement of different norms is “global” balancing, then we support it as well as “contextual” balancing, which we view as operationally similar, and perhaps the same. ¹²

We do not reject intuitive balancing. As long as this is clear, it can be rebutted with good reasons or accepted or rejected by those with similar or different intuitions. When values are nearly evenly balanced, then intuitive balancing may make good sense. However, in our opinion, intuitive balancing does not advance moral debate and inquiry to the degree of deliberative balancing.

The term “intuitive” in the philosophical literature typically means an unreasoned judgment, one that is immediately made without reference to apparent thought processes. This is unfortunate because it makes too many judgments appear to be “intuitive.” Much of what humans do in making “decisions,” moral and otherwise, is habitual. Habits are built over time, and can be deliberately fostered or taught based on good reasons. Furthermore, good reasons may be provided in support of habitual practices. Most people would see little need to offer reasons in favor of the claim that life is typically more important than temporary and minor suffering. A health care professional might act out of habit to save a life under such conditions and might be nonplussed if asked to justify his or her apparent “weighing” of life over minor suffering. Balancing in many cases may seem to be intuitive, but if called upon, supporting reasons may be offered.

In difficult cases, balancing involves a careful consideration of the moral import of various conflicting concerns. Often this is quite complex, involving separable issues that must be given consideration. For example, whether a patient has decisional capacity may not be clear, so the import of this uncertainty must be weighed against the patient's right to make the decision. Risk versus benefit is often a separate issue. In a decision about, for example, whether surgery is morally appropriate, the risk/benefit balancing may support a decision or it may need to be balanced against the patient's autonomous desire. In each, good reasons may be offered.

Balancing involves a consideration of the relative importance of all issues involved. Sometimes important issues may be closely balanced. Suppose for example that the risk/benefit ratio is thought to be of highest importance, yet risk and benefit may be close. In this case, the decision of the patient, even if of questioned capacity, may be the determining factor. Again, balancing recognizes that these types of close determinations involve a loss in spite of a morally appropriate choice. At times these are Pyrrhic victories where the choice has no winners because of the unavoidable loss no matter which way a decision is made.

The point is not that balancing always produces the “right” decision. Instead the point is that balancing and the reasoning behind it are crucially important in reaching a decision and that this should be made clear partly so that those who disagree will understand the basis of the decision. Balancing, and not casuistry or specification, fosters both good decision-making and respectful disagreement. Effective moral debate is facilitated when the closeness of the decision is honestly admitted.

Our emphasis on balancing does not mean that case presentations are unhelpful. They are indispensable in bioethics. First of all, careful examination of cases indicates the concerns that must be balanced. Secondly, cases help people to appreciate that multiple issues must often be taken into account and that there are various kinds of reasons that support the different degrees of importance assigned to different elements. Also, balancing does not negate the use of rules-of-thumb that may result from the balancing process. These ought to make explicitly or implicitly clear that the values have been balanced to determine which, in a particular case of class of cases, is considered most important.

V TEACHING

As indicated previously, balancing provides important elements for teaching ethical decision-making. In order for a framework or method to be useful, it must be able to be applied well by those for whom it is most relevant. The advantages of balancing are that it emphasizes constant attention to details of the case at hand, prompts individuals to understand their limits, and provides a method on which other methods implicitly rely. Even though there may be a loss of value during balancing due to sacrificing one value for another, there can be gains in moral reasoning in terms of good process. In the end, the teaching or doing of ethical resolution using balancing brings bioethics back to the basics of moral decision-making by stripping away unnecessary and unreliable mechanisms imposed by casuistry and specification.

Most medical encounters do not need explicit ethical balancing since no value conflicts occur or because good habits resolve the value conflicts that exist. The benefits of teaching balancing reside in developing good habits, attentiveness to shifts in cases that necessitate new balancing, and methods for addressing ethically complex cases. Although straightforward to teach, the practice of balancing takes considerable care and practice.

In teaching balancing, case examples can develop an appreciation for the range and diversity of ethical concerns and provide practical examples of good process. In order to be useful, cases do not need to be considered paradigms, as in casuistry. Rather, the teaching should be done in the context of cases and not simply using cases as justification of a decision. In the former, cases are used to exercise the mind and give practice. In the latter, cases are used to solidify and narrow a point of view. A simple example of a teaching seminar elucidates these points.

For consistency we use the “imperiled” newborn scenario. This is the topic both Strong and Richardson use in their debate. For the sake of teaching balancing, we need to make the case discussion both more general and specific, but in different ways. Assume our audience is a group of physicians training in a high-risk obstetrics fellowship. In clinical practice, these fellows play a role in perinatal planning for pregnant women for whom prenatal fetal abnormalities have been identified. A variety of ethical challenges face fellows in these situations, but for the sake of this seminar we focus on perinatal ethical decision-making concerning resuscitative measures. In order to help this audience prepare for encountering actual cases, the general topic needs to be addressed with specific examples while still providing education about broad areas to be balanced. The case examples need to be given depth of texture and variation of circumstances. Strong's trisomy 18 disease poses a good example, but would need to be given greater texture about family desires and sub-specialist recommendations as well as about the current co-morbidities. In Strong's use of this example, he does not account for a family who has discovered “Internet success stories” of children with Trisomy 18 who have reached their teen years. The knowledge of these success stories often changes what information needs to be given to place it in the correct context. Failing to grapple with the reasons a family may not agree with a professional judgment hinders productive ethical deliberation. Strong does not account for a fetal ultrasonographer saying that that the usual cardiac malformations found in Trisomy 18 are not present in the case. In this way, our case does not act as a paradigm, but rather an opportunity to explore why a family may request aggressive measures and the variation of elements that might be encountered.

A single example is not sufficient in this seminar as it would not give the physicians a balanced understanding of the complex variations. The perinatal planning for care of a fetus with Klinefelter's syndrome might also be instructive to raise similar issues but with a changed history of the disease. One might also discuss amniotic band syndrome and Trisomy 21 as raising different ethical challenges, and present instances in which the family or physician take differing views of resuscitation in order to show the variation of stances within the profession and among families. These cases provide instances for the physicians to reflect on how morally relevant elements may be balanced, prior to having to face an actual patient. When faced with the patient, they can then search for the salient ethical challenges. They also can understand that many moral decisions involve a sacrifice of one value for another. In difficult cases all participants need to be cognizant of what price in values must be paid for any given outcome.

In the above example, the physicians are not taught to dissolve ethical dilemmas by simply applying a specification. Also they are not given the “standard” cases to which they measure or judge all of their tough cases. They are given examples and practice for identifying issues, resolving dilemmas through balancing, and acknowledging the cost of morally troubling cases.

VI CONCLUSION

Balancing provides a key to resolving many application problems in training medical professionals and ethics consultants and underlies deliberation in casuistry and specification. Casuistry and specification add unnecessary and unhelpful overhead to the process. Further, the danger of these systems lies in that they can give the impression of providing definitive algorithms such that there may be a loss in recognition of the underlying uncertainty of situations and of the incommensurability of conflict between some values. Balancing allows for a simple process that continues to pay attention to the inevitable losses occurring during the resolution of tragic situations where one value is selected against another value. There are cases where no good solutions exist, yet decisions concerning valued ends must be made. This goes to the point of teaching attentiveness to the challenges of particular cases rather than simply categorizing or typifying them. Overall, balancing avoids the problem of encapsulating moral decision-making, whether in a rule or in a case. It displays the reasons behind moral decision-making, and thus enhances good more decision-making and furthers appropriate moral debate.

Notes

1. For example, DeGrazia (1992, p. 512) claims: “I will … argue that a ‘specified principlism’ is the most promising model — though it requires development.”

2. A conflict is resolved when one or another norm is considered to dominate in a particular case. A conflict is dissolved when it is determined that norms do not conflict. Specifications are often intended to encapsulate the resolution of conflicts, and thus, as results of inquiry, tend to dissolve moral conflicts by offering a new set of norms that do not conflict. In this sense, the conflict under consideration does not arise in future deliberation.

3. Although this specification seems to violate Richardson's qualification that specifications do not merely add exceptions by adding disjunctions, it does seem to provide substantive qualifications to the norm supporting parental authority. The case was first presented in Richardson (1990).

4. Technically, the specified principle should be stated: “Respect patient autonomy by following a patient's advance directive whenever it is clear and relevant.”

5. Although cooperative development of a set of norms, without the inclusion of supporting reasons, seems to be the goal, restatement of this goal is difficult to find in the body of Richardson's work. By and large, he talks about individual specifications by individual deliberators and the argumentation that produced the specifications. However, having a specification does introduce a new action-guiding norm. If they do not count in future deliberation, then having them stated and supported makes little sense because the decision could stand without the specified norm. As a result, we take his considered view to be that the set of cooperatively developed norms stand as the main guidance to moral decision-making in bioethics.

6. We support Richardson's statement that in the face of conflicts, efforts might be made to change the world in a way that avoids conflict. We are claiming that balancing is a better way to underscore the importance of the values overridden in particular cases, and that, despite his statement, specification is often hostile to that attempt. See Richardson (1990, p. 301).

7. The case was first presented in Richardson (1990). Carson Strong's account of the case is in Strong (2000).

8. To answer this objection by calling for a further specification not only further overpopulates the moral universe but also simply puts off squarely facing the fact that balancing an indefinite range of circumstances is required. That, after all, is the way specifications are typically formulated.

9. These are factors that would be carefully considered in balancing.

10. The reader may immediately suspect that with additional cases, another case, with a different outcome, might be closer.

11. Finding a large number of helpful agreed-upon cases is not on the horizon. For example, consider the moral debate over whether a psychiatrist or psychologist should breach confidentiality in order to prevent harm to third parties. Even the well-known California court case, Tarasoff v. Regents of University of California, (1976), involving breach of confidentiality due to a credible death threat, was decided by a split verdict. In that case, some justices decided that the balancing tipped in favor of disclosure while others thought it did not. To use Tarasoff as a paradigmatic case would take as morally settled a decision that has not secured the agreement among bioethicists, even though it has the force of law.

12. Richardson's acceptance and use of at least some balancing should not obscure the difference between specification and balancing. Balancing is a way to resolve moral problems involving conflicting values; typically it does not result in a new moral rule. Specification necessarily involves the production of new moral rules as the result of moral deliberation regardless of whether balancing is employed.

Tarasoff v. Regents of University of California. (1976). 17 Cal.3d 425.