I would recommend that Angela’s care team try force-feeding one more time, with the goal of stabilizing her health at a level at which doubts about her capacity can be quelled, and then allow her to decide what kind of treatment she is to receive thereafter. Even so, I would want more information from the people closest to her case, and I can imagine learning things that might lead to a different conclusion.
In recommending force-feeding, I do not mean to suggest that Angela’s parents’ wishes trump other considerations. The medical treatment offered to Angela must be determined by Angela’s interests and not those of her family. Moreover, Angela is an adult, and law requires that she not be treated without her consent. On the other hand, some of the care team members closest to her “doubt her ability to consent, because of psychiatric illness, the side effect of drugs, and the disturbing effect of the eating disorder on her freedom of choice.” This is why I think it is reasonable to try one more time to get Angela to a state in which she can make a competent decision about her treatment, even if this involves force-feeding and its attendant risks.
My reasoning here can be organized in terms of the following decision tree. Suppose we begin a regimen of sedation and force-feeding, but it proves too great a strain for Angela, and she dies before she has a chance to gain much weight. This would be sad, but it would not mean that our decision was wrong. She probably would have died soon anyway, and she seems already to have accepted this fact. Also, her quality of life was already very poor. At most, we have inadvertently hastened what was already imminent.
On the other hand, suppose that Angela gains enough weight that her survival is no longer precarious. By itself, this does not show that our decision was right, because our goal was not simply to make her gain weight or to keep her alive in some state or other. Our goal was to stabilize her weight at a level at which she would be more clearly mentally competent, that is, have the capacity to give or withhold valid informed consent to medical care.
Suppose that Angela does not, despite her weight gain, become more clearly decisionally capable. Does that mean that our decision to force-feed was wrong? I do not think so, but to see this we need to consider further possible outcomes. Suppose she continues to lack capacity, but the weight gain changes her attitude so that she retroactively accepts the force-feeding. In that case, she was initially treated against her will, and this may have been traumatic, but because she lacked capacity then, it was not a serious violation of her autonomy. More importantly, she now accepts the treatment and is getting better.
On the other hand, suppose that, after the weight gain, she remains both decisionally incapacitated and opposed to anything more than palliative care. In this case, we should allow her to resume her slow self-starvation—not because it results from the sort of reasonable, autonomous choice that is protected by law but because it is now clear that she is unlikely to get better, and there is little to be gained by continuing to force treatment on someone in this state. Yet it was still right to have made the attempt.
Finally, suppose that Angela gains enough weight that her care team agrees that she has, by the appropriate criteria, capacity to give or withhold consent to treatment. Suppose, that is, that there has been a marked reduction in the factors that previously compromised competence or indicated lack of capacity: she becomes more lucid, more communicative, less obsessive, less duplicitous and manipulative, and less dependent on psychoactive drugs. This would, I think, tend to vindicate our initial decision to engage in force-feeding: we gambled that she could achieve decision-making capacity and our gamble has apparently paid off. But we would not be out of the woods yet, because, again, our goal was not simply for her to gain capacity. Our goal was to get her to have sufficient capacity to give or withhold meaningful consent. Of course, if she becomes decisionally capable and then consents to future treatment, that would be the happiest outcome. It doesn’t mean that treatment will be effective, but our initial ethical problem has been discharged. On the other hand, if Angela becomes decisionally capable but subsequently refuses anything more than palliative care, then, again, we should allow her to resume her self-starvation: not because we have fetishized the value of isolated autonomous decisions, and certainly not because illness is as “valid” a goal as health, but because it is futile and miserable to do anything else.
In the preceding account, I have tried to minimize my appeals to the abstract theoretical considerations beloved of philosophers and to stick to the sorts of concrete reasons that make sense in a clinical setting. Even so, I want my clinical reasoning about Angela to fit with the philosophical view of moral reasoning that seems right to me: the one developed by W. D. Ross in chapter 2 of The Right and the Good.Footnote 1 Ross’s view in this text is backward looking, pluralist, and intuitionist. On it, we are subject to an irreducible plurality of types of duty, by virtue of various features of our circumstances. For example, I am subject to a duty of beneficence whenever I am able to benefit someone, and I am subject to a duty of fidelity whenever I have made a promise. Keeping promises may tend to benefit persons, but according to Ross that is not the only reason I am subject to a duty of fidelity. Ross allows that these duties may conflict and therefore regards all duties as prima facie, that is, conditional on not being overridden. A duty may override or be overridden, but unless it is overridden, we ought to fulfill it. When beneficence, for example, trumps fidelity, we do not say that I have a duty to break promises. Instead, we say that I have a moral reason to do A (keep an appointment with my spouse) but a stronger moral reason to do B (play host to some old friends who have arrived from out of town unexpectedly), and doing B makes doing A impossible. In the absence of a monistic consequentialist decision procedure, how do I decide which of these competing prima facie duties is my duty proper? According to Ross, there is no alternative to a cultivated practical wisdom: “I must study the situation as fully as I can until I form the considered opinion (it is never more) that in the circumstances one of them is more incumbent than any other.”Footnote 2
Ross’s work may not be well known in clinical ethics circles, but it lies behind an approach that is well known: the “four principles” approach popularized by Tom L. Beauchamp, James Childress, and others.Footnote 3 According to this approach, questions of medical ethics are considered within a framework defined by four principles: beneficence, nonmaleficence, respect for autonomy, and justice.
In terms of these four principles, the case of Angela presents us with conflicts between beneficence (helping her to become well and competent), nonmaleficence (avoiding causing trauma and death from force-feeding), and respect for autonomy. Simply recognizing these conflicts, however, does not tell us how to settle them, and precisely this has been a persistent criticism of the four principles approach.Footnote 4 Defenders of the four principles have sometimes replied that their approach is meant not as a decision procedure but as a map of the main values at stake in medical dilemmas. Of course, we still need to make decisions somehow, so proponents of the four principles have often argued that their theory can be supplemented with an account of how, through a process of increasing specification, we can move from these general principles to particular decisions on practical problems.Footnote 5
Though this helps a little, it does not say much more than what is already implied in Ross’s account, namely, that we must reflect on the circumstances and form a judgment about the totality of the morally relevant features of the situation and the reasons that these generate. Ross pessimistically insists that these “judgments about our actual duty in concrete situations have none of the certainty that attaches to our recognition of the general principles of duty” and later suggests that “there is therefore much truth in the description of the right act as a fortunate act. If we cannot be certain that it is the right, it is our good fortune if the act we do is the right act.”Footnote 6 Whether or not we share his pessimism, it is not essential to his overall account. What is essential is that we can arrive at judgments about what to do in concrete circumstances and that these judgments can be guided by practical wisdom in view of the facts of the case, even if those facts do not logically force a unique conclusion on us in the way that a closely theory-guided approach might. Indeed, this points to one of the advantages of an approach to clinical ethics that gives pride of place to practical wisdom over the mechanical application of ethical algorithms: it allows us to take into consideration a wider range of facts and circumstances as relevant to deciding what to do. This is relevant to the case of Angela in several ways, but I will mention only two.
First, the morally relevant circumstances of our decision include not just medical facts about her condition and the likely outcome of particular treatments, or legal facts about the duties and obligations of her care team. They also include relational facts about the parties involved and also epistemological facts, for example, about the grounds of some of her care team for doubting her decision-making capacity. What is the nature of these grounds, and how serious are they? Here I want more information. If these doubts have more to do with the turf wars between some healthcare professionals over expertise—or if they are only slight and hesitant anyway—then they may not constitute enough reason to rebut the psychiatrists’ judgment of capacity. But if these doubts reflect the considered judgment of the people who, day in and day out, actually spend the most time with Angela, then this may well constitute enough grounds to question her capacity (and neutralize the demand for her consent).
Second, in contrast with the timeless and ahistorical feel of some highly theoretical approaches, the application of practical wisdom to this case requires us to consider the chronological (or even narratological) development of this case. What has already happened? How have things unfolded in this particular case? Such considerations are not just relevant to deciding what to do; they may also help to dispel a certain perplexity about my recommendation. Recall: my recommendation was that we give force-feeding one more try, to try to achieve capacity, and then if that fails, allow Angela to opt for palliative care. If we fixate on that conclusion and on the sanctity of consent, we may wonder how force-feeding could have been permissible in the first place. If Angela has a right to self-determination in the end, didn’t she have it in the beginning, too? Conversely, if force-feeding was justified the first time, why wouldn’t it be justified the second time? (And the third and the fourth?) Rossian approaches to ethical reasoning allow—indeed require—us to treat cases not just as individual decision moments frozen in time, in the light of timeless principles, but as courses of action involving real individuals with real relationships and with real histories.
