|Bioethics and Medical
Balancing Principles in Beauchamp and Childress
ABSTRACT: In the latest edition of Principles of Biomedical Ethics, Tom Beauchamp and James Childress provide an expanded discussion of the ethical theory underlying their treatment of issues in medical ethics. Balancing judgements remain central to their method, as does the contention that such judgements are more than intuitive. This theory is developed precisely in response to the common skepticism directed at "principlism" in medical ethics. Such skepticism includes the claim that moral reasoning comes to a dead halt when confronted by competing conflicts between moral norms in a given pluralistic situation. In this paper, I use examples from the text to show that despite the authorss arguments to the contrary, balancing judgements are the product of unreasoned intuitions. Given the necessity of some such judgements in any principle-based system, my argument highlights the degree to which principled ethical reasoning rests upon an arational core.
"Principlism" is the term often used, sometimes derisively, to refer to a method of moral reasoning found in medical ethics and elsewhere. At the core of principlism is the idea that ethical justification rests primarily, if not exclusively, in appeals to more general or "higher level" moral norms under which any more particular ethical claim can be subsumed.
Principles of Biomedical Ethics, by Tom Beauchamp and James F. Childress, has for many critics in medical ethics exemplified the worse sins of "principlism." From its first edition, the authors have argued for the importance and usefulness of general principles for justifying ethical judgments about policies and cases in medical ethics. The organization of their book reflects this conviction, dividing discussion of particular ethical problems under the rubrics of the key ethical principles which the authors believe should govern our moral judgments: principles of autonomy, nonmaleficence, beneficence and justice.
It was always a caricature of their views to label them as straight-arrow deductivists. (1) At the very least, they have from the first insisted on the necessity of making judgments about the proper balance to be struck between competing ethical commitments when they are in conflict. Since Beauchamp and Childress disavow appeal to any overarching framework from which such a balancing judgment could be derived, particular moral judgments could never for them be simple deductions from any single moral principle.
But then one may ask how Beauchamp and Childress handle one of the key criticisms of principlism, which points to the inevitable conflict among principles in the sort of pluralistic system favored by many, (2) and the resulting need to "balance" or prioritize the norms in conflict. The charge is that such balancing will be both ubiquitous and arational in its reliance upon sheer intuition or moral "common sense." And so the vaunted rational order of a principled approach to ethics turns out to be little more than a fašade.
In the latest (now the 4th) edition of their book, the Beauchamp and Childress provide an expanded and more nuanced account of their theory and method, which is intended in large measure to dispel such skepticism about at least their version of "principlism." How well have they succeeded?
As they characterize it, their theory is a "principle-based, common morality theory." (Beauchamp and Childress, 120 ff.) Starting with the common morality shared by members of society, the "theory constructs principles and rules from considered judgments," seeking coherence through a process of reflective equilibrium. Developing this coherence will require both that principles be "specified" in the manner described by Henry Richardson; and that judgments be made how best to prioritize or balance competing moral principles.
Space doesnt permit me to discuss all of the elements of this theory, and so Im going to set aside coherence and specification for the moment to concentrate on the account that Beauchamp and Childress give of "balancing," which even in the new and improved theory retains its central importance. (3) I will illustrate my contention that balancing judgments in Beauchamp and Childress are inarticulate intuitions for which reasons can not be given, and that the defenses which Beauchamp and Childress offer against this charge do not succeed. (4)
Can One Give Reasons for Balancing Judgments?
Beauchamp and Childress admit that in making balancing judgments, "some intuitive and subjective weightings are unavoidable, just as they are everywhere in life when we must balance competing goods." (Beauchamp and Childress, 36) At the same time, they insist that "balancing...is a process of justification only if adequate reasons are presented." (Beauchamp and Childress, 34)
As an illustration of such "adequate reasons", the authors describe the situation of a physician faced with a choice between returning home to take her son on a promised trip to the library, or staying at the hospital to attend to a life-or-death emergency. "A life hangs in the balance, and she alone has the knowledge to deal adequately with the full array of circumstances. Her action of canceling her evening with her son, painful and distressing as it is, can be justified by this good and sufficient reason for doing what she does." (Beauchamp and Childress, 33-34)
Their illustration confuses having provided a good and sufficient reason for breaking the promise to her son, with having provided a reason of any kind for the implicit claim that the obligation to save the patients life is weightier than the obligation to keep the promise. No reason has been offered for this balancing judgment its taken as a given of common-sense morality that has no need for justification. Anyone who demands that further reasons be given risks being rightly labeled a moral nincompoop.
In cases like this one, which arouse no significant disagreement about the proper balance between contending obligations, the failure to give reasons justifying that balance may be unproblematic. But in other sorts of cases, the moral controversy will center around differing judgments about what the proper balance should be. In those cases, simply to assert that one of the obligations constitutes a "good and sufficient reason" for overriding the other begs the question by failing to give any reason on behalf of the balancing judgment which has been made. Balancing is not then a mode or method of justification; it marks the end of reasoned justification as a tool of moral reflection.
In fact, Beauchamp and Childress themselves dont give reasons justifying the balancing judgments which they make in support of their positions on particular issues. They simply assert them. This can be illustrated by the various views they espouse on issues involving potential conflicts between respect for autonomy and duties of nonmaleficence or beneficence. How does one decide what to do with such conflicts? Because they are moral pluralists, Beauchamp and Childress recognize no a priori ranking among these opposing obligations, and so the proper balance will depend on the circumstances, rather than be determined by any general rule. As they acknowledge, "Respect for autonomy...has only prima facie standing and can be overridden by competing moral considerations...The principle of respect for autonomy does not by itself determine what, on balance, a person ought to be free to know or do or what counts as a valid justification for constraining autonomy." (Beauchamp and Childress, 126)
Since respect for autonomy is only a prima facie duty, how do Beauchamp and Childress decide when it is, or is not properly overridden by weightier obligations? Take for example the case they discuss of a patient who does not wish to be told the results of the HIV test hes taken, and which the physician knows has come back positive. On the one hand, the patient has a right of autonomy with respect to the information he wishes to receive; and on the other, his deliberate ignorance may indirectly pose a risk to his sexual partners if it serves to rationalize unsafe sexual practices. It would breach duties of nonmaleficence for the physician to ignore such risks to others. Which duty is the weightier; and why? Beauchamp and Childress answer in one sentence: "In light of the possible consequences, the disclosure was justified despite the fact that the patient did not want the information." (Beauchamp and Childress, 403) No further reasons are offered to justify this claim. It is simply the deliverance of a moral intuition, not backed by any reasons.
In another example, the authors discuss the case of a patient who has had a myelogram which shows some inconclusive evidence of a serious spinal cord condition. Confirmation will require further testing. When the patient asks the physician what the results are, "the physician decides on grounds of beneficence to withhold potentially negative information, knowing that upon disclosure the patient would be distressed and anxious." (5) Is this balancing judgment warranted? Again, Beauchamp and Childress provide a one-sentence answer. "This physicians act of temporary nondisclosure seems to us morally justified, though beneficence is (temporarily) given priority over respect for autonomy." (Beauchamp and Childress, 282)
Sometimes, such summary balancing judgments are implicit rather than explicit. One example is their discussion of the case of Larry McAfee, which is offered as an example of a justified assisted suicide. McAfee was ventilator-dependent and quadriplegic as a result of an automobile accident. At the time described in the example, McAfee wanted to be disconnected from his ventilator, and wanted to have access to the sedatives which would make it possible for him to carry out his intention to die by this means. (6) Beauchamp and Childress strongly endorse the Georgia court finding "that McAfees right to refuse treatment and disconnect himself outweighed the states interest in the preservation of life and in preventing suicide." (Beauchamp and Childress, 239) They offer no further accounting of this balancing judgment. One would seem to be called for, in light of their discussion of "justified paternalism" in the chapter that follows. There they assert that "in general, as the risk to a patients welfare increases or the likelihood of an irreversible harm increases, the likelihood of a justified paternalistic intervention correspondingly increases." They acknowledge, moreover, that a paternalistic decision can be justified even where the interference with autonomy is "more than minimal." (Beauchamp and Childress, 283) Given that whats at stake is McAfees life and the possibilities for future achievements and satisfactions that it may hold, why cant a decision not to honor his demand to discontinue the ventilator or his demand for sedatives be justified on paternalistic grounds?
The recourse to "balancing" does not answer this question with a set of reasons providing a justification. It answers it by the assertion of what is taken to be obvious to moral common sense. As the controversy about a case like McAfee demonstrates, however, this method will gain us nothing when we are confronted with those whose moral "common sense" delivers a different intuition.
Are There Criteria for Balancing?
My discussion of Beauchamp and Childress treatment of balancing has so far left out a prominent element in their account: "As a response to criticisms that the model of balancing is too intuitive and open-ended, we can list a few minimal conditions that reduce the amount of intuition involved. These conditions add content to the requirement of giving good reasons for actions and norms." They then list those conditions.
We should begin by setting aside the first criterion, which cant count as a reason for a balancing judgment. If the "better reasons" it refers to are intended to include the balancing judgments which Beauchamp and Childress believe are essential for good ethical reasoning, then it is hard to see how this first criterion for a good balancing judgment would avoid begging the question.
The remaining criteria are not criteria for evaluating a balancing judgment. Rather, they are criteria for comparing courses of action with respect to selecting the one which entails the least infringement of norms, without regard to which norm is the weightier. If the second condition is not met, for example, it follows that the alternative of not infringing the norm is the course of action which results in less infringement of the norms I value, and so should be preferred. In order to make this evaluation, I dont have to make any judgment about which norm would be the more important if in fact pursuing the one required some sacrifice of the other. Similarly for the 3rd, 4th and 5th conditions. If they are not met, then there is an alternative course of action which offers less of an infringement of norms, because the conflict between the norms in question either disappears or is reduced.
When all of the criteria are satisfied, they establish that there is an irreducible conflict between norms which only then will require a judgment about which norm should be overriding. The criteria are of no further use in warranting the balancing judgment which must follow.
Take as an example Beauchamp and Childress evaluation of policies for mandatory pre-marital screening for HIV infection, required for a while in several states. In their argument against such policies, they point out that screening of this population identifies only a few cases that couldnt be identified by more efficient voluntary means; that theres no evidence that additional HIV infection or illness is prevented; and that spouses and future offspring can be better protected by means of counseling and voluntary testing. (Beauchamp and Childress, 414-415) These are arguments that mandatory pre-marital testing is unjustified because there are morally better alternatives which infringe neither duties of respect for autonomy nor duties of beneficence or nonmaleficence. They are not arguments which decide what we should do in the event there were no such alternatives i.e., in the event we really had to prioritize between competing moral objectives.
So the accounts that Beauchamp and Childress give for the rational nature of balancing judgments do not succeed, and their own use of balancing judgments seems to betray their intuitive and arational character. Beauchamp and Childress might well answer that the rationality of particular balancing judgments will be warranted by their place within an adequately specified set of coherent principles and considered judgments. Evaluation of that claim will have to wait for another day. In the meantime, we should be asking: If balancing judgments are not the product of reason, where do they come from, and how, if at all, can we critically evaluate them?
Notes(1) See B. Andrew Lustigs article for more on this point.
(2) For example, see John Kekes.
(3) So, even though they adopt Richardsons account of "specification,", they do not accept David DeGrazias suggestion that specification and balancing be "merged." (Beauchamp and Childress, 34)
(4) Although I wont argue for it here, I believe that for their part, both specification and coherence rely upon balancing. If so, then there is an arational core at the heart of the tripartite system which Beauchamp and Childress advocate, and perhaps at the heart of any putative system of reasoning based in ethical principles.
(5) Beauchamp and Childress do not explain just how this is accomplished in the face of the patients direct question. E.g., does the physician say something like, "I cant really say for sure what the problem is until we repeat the test. Lets wait til then, when we can go over the results in detail." If the patient acquiesces in this plan, the case presents no ethical conflict with respect for autonomy. Alternatively, does the physician respond with a lie, like "I dont really see anything to worry about; but lets repeat the test just to make sure." Presumably, the balance of obligations could shift significantly if we add the obligation not to lie to the scales.
A similar vagueness about practical details infects a number of their other illustrative cases, including the one that follows this case on the same page.
(6) Beauchamp and Childress do not mention that subsequent to the events they describe, McAfee reversed his demands to discontinue his ventilator support after United Cerebral Palsy of Greater Birmingham and other disability advocacy groups offered him genuine alternatives to the hospitals and institutions in which he had been treated and housed. (Herr, Bostrum and Barton, 34-35)
Beauchamp, Tom L. and James F. Childress. 1994. Principles of Biomedical Ethics, 4th ed. New York: Oxford University Press.
DeGrazia, David. 1992. "Moving Forward in Ethical Theory: Theories, Cases, and Specified Principlism." Journal of Medicine and Philosophy 17: 511-539.
Herr, Stanley S., Barry A. Bostrum and Rebecca S. Barton. 1992. "No Place to Go: Refusal of Life-Sustaining Treatment by Competent Persons with Disabilities." Issues in Law and Medicine 8: 3-36.
Kekes, John. 1993. The Morality of Pluralism, Princeton, N.J. : Princeton University Press.
Lustig, B. Andrew. 1992. "The Method of Principlism: A Critique of the Critique." Journal of Medicine and Philosophy 17: 487-510.
Richardson, Henry S. 1990. "Specifying Norms as a Way to Resolve Concrete Ethical Problems." Philosophy and Public Affairs 19 (Fall): 279-310.