Navigation
Newsletters Index (06:1)
apaOnline Home Page

APA Newsletters

Fall 2006
Volume 06, Number 1

Newsletter on Philosophy and Medicine

Articles

Compassion, A Double-edged Scalpel

Sandra L. Shapshay
Indiana University–Bloomington and the Indiana University Center for Bioethics–Indianapolis

Compassion is a double-edged scalpel: it is an indispensable tool in forging a trusting relationship between physician and patient, and may help motivate the physician to alleviate the patient’s suffering, but it also poses a problem for maintaining professional boundaries, boundaries that protect the physician from experiencing too much suffering and protect patients by maintaining physician objectivity and impartiality when needed.1 In this paper, I wish to focus attention on these drawbacks of what I will call "unbounded" compassion on the part of the physician, and I will propose a model which will, I hope, illuminate what properly bounded compassion in medical practice should look like.2

I define compassion, following Martha Nussbaum, as "the painful emotion caused by the awareness of another person’s undeserved misfortune."3 Such awareness involves three judgments: that a serious misfortune has befallen someone; that the person did not bring this suffering on herself; and that this person is in some sense valuable, that is, her good is to be promoted.4 Although cognitive faculties are at the core of compassion (more so than in many other emotions), the compassionate person need not articulate these judgments to herself in order to experience the emotion. In addition, the identification of the emotion with evaluative judgments is not meant to limit the phenomenological richness and content of the emotional experience.

John Deigh has identified a weakness with Nussbaum’s definition of compassion, that it cannot account for compassion that is "transformative of the subject’s ends." Consider, for example,

[a] child of affluence…who grew up knowing only suburban prosperity…and without ever encountering urban slums or rural squalor might, when confronted for the first time with such human misery…experience a powerful wave of compassion through which the desire to help the victims of poverty replaces her prior indifference to their plight.5

Since the judgment that others’ ends are important is always already a facet of compassion, for Nussbaum, compassion that awakens a sense of the importance of others’ ends is not accounted for. However, this is not a serious problem for my use of the definition in the context of medical ethics as it is inherent in the physician-patient relationship that the patient’s ends do already matter for the physician—it is part of the role of the physician to promote the patient’s welfare.

In this paper, I will address compassion somewhat separately from empathy—although the two are often interrelated―as the latter is, strictly speaking, neither necessary nor sufficient for having compassion (empathy may be necessary, however, for the most attuned and ideal forms of compassion). I define empathy, following Alisa Carse, as "the ability and disposition to imagine (as best we can) how others feel, what they fear or hope for, and how they understand themselves and their circumstances."6 Thus, while empathy involves a great deal of imagining, one can feel the painful emotion of compassion for a patient suffering from AIDS without attempting to experience vicariously what her first-personal experience of the illness is like. And a torturer can use empathy imaginatively to engage another’s situation as if in her shoes, without experiencing compassion and its corresponding feeling of sorrow because he does not see that the victim’s good is to be promoted.7 Notwithstanding, according to the psychological literature on the topic, it is typically the case that people who engage empathetically with suffering persons will experience compassion for them.8 Additionally, having an appropriate feeling of compassion will often involve empathy. For instance, Gina might feel tremendous compassion for Frank, who is blind, upon meeting him. Gina believes (falsely) that he must suffer a great deal on account of his disability. But let us suppose that Frank has learned to cope very well: his passion in life is classical music, and his blindness has heightened his aural sensibilities. He enjoys an active career as a violin teacher and, from his own perspective, a life that is just as satisfying, if not more, than the average sighted person. In this case, attuned empathy in the course of conversations with Frank would dispel the false belief that Frank is a victim of serious misfortune and, with it, the inappropriate feeling of compassion (though the feeling may linger for a while even after the belief has changed).

Meaning to "suffer with" another (cf. the German "Mitleid"), compassion is an emotion that metaphorically extends a person’s boundaries. This takes an extreme form in the philosophy of Schopenhauer, who saw compassion as the very basis of morality, and affirmed that the compassionate person recognizes that he is, metaphysically, one with another. Schopenhauer writes,

Hence the radical difference of mental habit between the good character [the compassionate one] and the bad [the egoist]. The latter feels everywhere that a thick wall of partition hedges him off from all others…the good character, on the other hand, lives in an external world homogeneous with his own being; the rest of mankind is not in his eyes a non-ego; he thinks of it rather as ‘myself once more’.9

Nussbaum eschews Schopenhauer’s robust metaphysical account of the fusion of self and other and draws attention to the fact that in order to have bona fide compassion for another, one must also attend to the ways in which the other is different from the self. For example, a physician who is not an avid runner needs to attend to the real differences between himself and his patient, which give the patient’s marathon-career ending knee injury supreme importance for him: "these recognitions are crucial to getting the right estimation of the meaning of the suffering for the suffering person. What is wanted, it seems, is a kind of ‘twofold attention’ in which one both imagines what it is like to be in the sufferer’s place and, at the same time, retains securely the awareness that one is not in that place."10 On either construal of compassion—Schopenhauer’s boundary-erasing or Nussbaum’s two-fold attention to another’s as well as one’s own experience―one’s boundaries of concern are extended outward toward another, and we suffer with another.

It is uncontroversial that a certain amount of compassion is essential for good medical practice; accordingly, many codes of medical ethics (such as the AMA’s) stress the importance of compassion, stating that "a physician shall be dedicated to providing competent medical care, with compassion and respect for human dignity and rights."11 Yet, many codes also underscore that the physician "must establish and maintain appropriate relational boundaries…".12 The guidelines thus implicitly speak to the double-edged nature of this boundary-extending emotion. Too much compassion on the part of the physician poses problems. This is captured, in an implicit manner, in the well-known professional directive not to treat oneself or one’s own immediate family. According to the AMA’s practical guidelines for physicians, "[p]rofessional objectivity may be compromised when an immediate family member of the physician is the patient. In emergency or isolated settings, physicians can treat themselves or family members until another physician becomes available."13

Why shouldn’t a physician treat herself or her immediate family? If physicians are exhorted to be compassionate in professional codes, and one typically feels greater compassion for family and friends than for strangers, it would stand to reason that family and friends would be precisely those persons who a physician would be most competent to treat. Yet, the AMA guideline implicitly recognizes that crossing relational boundaries may involve too much emotional engagement with a patient risking physician objectivity and patient welfare. An additional unstated worry is the danger to the physician’s own well being if she crosses the line between work and family; between the professional and the personal—where a bad medical outcome might mean a personal loss and failure that engulfs her entirely.14

Another worry concerning unbounded compassion concerns partiality. Imagine the following case: Ms. Turner, a beloved mother of three and kindergarten teacher, struck down by a drunk driver, lies brain dead in the ICU. Assume that the determination of brain death is not controversial and that her further care is obviously futile according to the hospital’s well-justified medical futility policy. Her family requests, however, that her care be continued for a few days longer; they wish to grieve for Ms. Turner over the weekend at her bedside. Several other patients await a bed in the ICU. The physician is not acquainted with these patients except for the fact that their needs are rather urgent. While it is understandable that a physician’s compassion for this endearing patient and her family might motivate him to keep this patient on life support for a few more days, his compassion may run counter to the demands of distributive justice. It is likely in this scenario that the physician’s professional obligation is to try to convince the family to discontinue life support as soon as possible, and to take steps to do so if the family will not relent. Compassion in this case comes into conflict with the demands of justice and the physician’s professional obligations and needs to be properly bounded.

It may be argued, however, that compassion does not truly conflict here with professional obligations but, rather, compassion for Ms. Turner has obscured the compassion he ought to have for the patients on the waiting list for the ICU. On this reading of the conflict, it is not too much but, rather, too little compassion that is the problem: he lacks sufficient compassion for the patients on the waiting list. But this does not plausibly capture the nature of the conflict. The physician in this case does not know who the other patients are, and without putting a name, face, and narrative to the next patient on the waiting list, it is implausible to say that the physician could or should have real compassion for him or her. Thus, the conflict in question here seems to be more plausibly captured as one between compassion and impartial professional obligations. One can see how compassion is a "double-edged scalpel" in this case, for it may conflict with professional obligations, but it is also indispensable for treating Ms. Turner and her family in a humane fashion. If a physician did not share in the family’s sorrow to any extent, he might act in a very authoritarian manner to discontinue life support. With compassion, the physician is more likely to engage in a sympathetic and respectful dialogue with the patient’s family in order to try to achieve the same end humanely.

In fleshing out these features and drawbacks of compassion in clinical medicine, I wish to propose that a helpful model for the kind of bounded compassion salutary for medical practice is, surprisingly, the model of the emotionally engaged spectator of tragic drama. I am not here giving an argument by analogy to the effect that the guiding normative principles of tragic spectatorship ought, on the strength of the analogy, also to govern the physician’s practice, but I do think that the model of bounded compassion offered by the sophisticated tragic spectator constitutes a heuristic device to illuminate ways in which the physician can and ought to place limits on the compassion she feels in medical practice, in an effort to maintain the proper professional boundaries. But this suggested model might seem, at first glance, rather inappropriate: What has removing a tumor or treating an infection to do with watching a performance of King Lear? Or Death of a Salesman?

Although I acknowledge (and will explore shortly) the significant disanalogies between the virtues and experience of the physician and the tragic spectator, there are two relevant similarities between good spectatorship and good practice of medicine that I would like to explore: first, the good spectator at a (good) tragedy ought to be emotionally engaged, for one does not really "get" the drama if one does not feel compassion (fear and pity, classically speaking) for the tragic hero and in proportion to the horror and undeserved nature of the suffering. Yet, the good spectator does not forget the boundaries of the spectacle and rush the stage to prevent Gloucester’s blinding, for she knows that it is not real, and when the play is over, she can go home and cease to worry about the tragic hero. The compassion is thus bounded by artistic conventions: she experiences the play in a manner that is both emotionally engaged but sufficiently detached so as not to be overwhelmed by suffering, or forget that she is playing the role of the spectator.

Similarly, the physician needs to emotionally engage a patient, and to respond sensitively to the patient’s (and patient’s loved-ones’) suffering. In a study of the clinical role of empathy, Jodi Halpern has made a sustained case that empathetic engagement with patients (which may often lead to compassion) is a superior stance for physicians than detached concern for patients.15 The two main mechanisms Halpern identifies by which empathy proves therapeutic are, first, the case in which the patient recognizes that her physician empathizes with her experience but where the physician shows resiliency in the face of this experience. The second mechanism is "when a suffering person feels that another person is affected by her grief, sense of catastrophe, or fear, and yet remains vital and emotionally present…"16 Clinical empathy involves a good deal of emotional labor, however, and physicians need to do it while remaining resilient, without burning out, without becoming so overwhelmed by the suffering that they can no longer do their jobs competently or leave the painful emotions at work to go home and lead normal lives. Obviously, this is not always an easy thing to do. The effective physician will feel compassion for the suffering patient, just as the spectator of tragedy does, but the compassion experienced in each case is bounded by conventions: conventions of dramatic fiction and conventions of the professional role.

The kind of thinking I’m suggesting on compassion has already been experimented with by the faculty of the Department of Medicine at the VA Los Angeles Health care System and the School of Medicine at UCLA. With compassion-building in mind, researchers developed a program called the Wit Educational Initiative utilizing Margaret Edson’s Pulitzer Prize winning play to educate medical students, residents, and staff providers concerning the "humanistic" elements of end-of-life care. In their report, Dr. Lorenz and his co-authors conclude that the theatrical experience was emotionally powerful for the trainees, and promoted more compassionate care, but did so in a "safe" way:

One reason our program may have been so well received among trainees is that it allowed them to experience the strong emotions that accompany a realistic portrayal of terminal illness, but within the safe environment of a theater seat. …In summary, we found that many learners found the psychologically realistic, emotive experience of the dramatic arts appealing. …Such holistic approaches to medical education should be considered by educators in promoting compassionate end-of-life care.17

The authors of the study were trying to sensitize trainees to the lived experience of dying patients. They do not suggest that they were teaching trainees to adopt the complex attitude involving both emotional engagement and critical distance in the initiative; however, it seems a reasonable suggestion that the "safer" theatrical stance might itself constitute part of the training in good patient care. That, perhaps, in a McLuhan-like fashion, the medium is also part of the message.

Obviously, there are crucial differences between the role of tragic spectator and physician in that the spectator observes and engages emotionally with the drama but is not an active participant in it. Except in cases of experimental theater, the spectator cannot change the outcome of a play. The tragedy unfolds before the spectator’s helpless eyes. The physician, by contrast, can affect the outcome of the tragic situation, either by relieving suffering entirely and helping to restore health (turning what might look like a tragedy into a "Hollywood happy ending"), or by blunting the effects of the misfortune by providing knowledge of the causes of suffering, relief from pain, an additional sense of dignity and control over the situation. Insofar as the physician is in a position to change the outcome of the situation, she is both spectator and co-author of the drama (along with nature, and the patient) simultaneously. Insofar as these are strong differences, there is a danger in completely aestheticizing the situation, and putting the physician in a more passive spectatorial role. No patient wants to hear the following from their physician: "My, my Mr. Johnson, I’ve never witnessed such a fascinating tumor such as yours! It grows at such a sublime rate. You are, without a doubt, my most interesting patient!" Obviously, striking the proper balance between emotional engagement, action, and an aesthetic/critical distance seems to be an art in itself.

Notwithstanding, this important disanalogy between the spectator and the active participant, a second salient analogy is to be found in the mode of attention that the sophisticated spectator and the physician pays to the tragic situation. The sophisticated spectator of King Lear will attend not just to his feelings toward the characters, but also to how the plot unfolds, (e.g., through the use of metaphor and symbol, in specific arrangements of verse, through foreshadowing of certain events). That is, the spectator attends to the aesthetic qualities of the piece in a dialectic of emotional engagement and spectatorial detachment.

Recently, some medical school curricula have been experimenting with courses in art appreciation. Unlike the emotional development of compassion in the Wit program, these are experiments focused on developing students’ observational skills more broadly, to deepen their ability to look with discernment, that is, aesthetically. In a paper called "Learning to Look: Developing Clinical Observational Skills in an Art Museum" the authors describe how Cornell University medical students, with the help of an art educator, developed their observational skills by studying paintings in the Frick Collection in New York City. In addition to learning to pick out markers of health and disease in portraits, for example, the students were also trained to see indications of emotion, character, class, and the subjective life of a person through sophisticated engagement with the paintings.18

For the sake of my arguments here on the proper bounds of compassion, such experiments, while not specifically designed to limit compassion, rely on similar underlying hypotheses about the necessity for a certain amount of critical distance alongside emotional engagement. Similarly, the good physician will maintain enough critical distance from the patient to be able to look and listen to the patient in a discerning manner, to hear what the patient is describing, but also to ferret out what might be going unsaid; to pick up clues as to the patient’s character, emotional state, moments of self-deception, or significant gaps in the narrative.

In a complicated dialectic between empathetic understanding of the patient and critical distance from the patient, the physician can determine the proper medical diagnosis and treatment and, yet, can still emotionally engage a patient to speak with him and to determine the most humane and respectful course of action for the patient. If the physician oversteps those boundaries for compassion, however, the physician starts to play the role of the close friend or family member to the patient. And insofar as she does this, then she shouldn’t be treating the patient because her objectivity is thereby compromised. Thus, the good physician is similarly both emotionally engaged and yet detached enough from the "spectacle" to attend to the medical-scientific features of the situation.

Endnotes

1. Medicine" at the 2006 APA Central Division meeting in Chicago for very lively and helpful discussion. Thanks especially to Lee Brown, who organized the panel and offered constructive criticisms of my paper. Laura Ekstrom, Gordon Greene, Ben Rich, Howard Spiro, and Angelo Volandes offered stimulating and insightful exchanges on this topic. Many thanks to my husband, Steven Wagschal, for helpful comments on this paper.

2. My discussion of compassion is limited to the role of physicians. I believe the role of compassion in nursing is better treated separately.

3. Martha Nussbaum. Upheavals of Thought (Cambridge: Cambridge University Press, 2001), 301.

4. Ibid., 321.

5. John Deigh. "Nussbaum’s Account of Compassion," Philosophy and Phenomenological Research, LXVIII, No. 2 (2004): 465-72 and 469. Deigh borrows this example from Stephen Darwall in Impartial Reason (Ithaca: Cornell University Press, 1983).

6. Alisa L. Carse. "The Moral Contours of Empathy," Ethical Theory and Moral Practice, 8 (2005): 169-95, 170.

7. Martha Nussbaum makes these points in Upheavals, 328-29.

8. See C. Daniel Batson et al., "Is Empathy-Induced Helping Due to Self-Other Merging?" Journal of Personality and Social Psychology, 73 (1997): 495-509.

9. Arthur Schopenhauer. On the Basis of Morality. trans. Arthur Brodrick Bullock (Mineola, NY: Dover, 2005), 141.

10. Nussbaum, Upheavals, 328.

11. American Medical Association, http://www.ama-assn.org/ama/pub/category/2512.html, accessed on April 2, 2006.

12. American Academy of Otolaryngology-Head and Neck Surgery Bulletin, March 2006 (full Code of Ethics found at www.entnet.org/academy/policies/ethics.cfm).

13. AMA’s Code of Medical Ethics: A Practical Guide to Physicians, www.ama-assn.org/ama1/pub/ upload/mm/369/professionalismppt.ppt, accessed on April 2, 2006.

14. My discussion of "bounded compassion" is similar to Alisa Carse’s conception of "properly contoured empathy"; in both notions, role-related obligations are important in delimiting the boundaries of proper empathy and compassion. Carse writes, with respect to empathy, "receptivity to the other must not be confused with self-abnegating absorption into the other or moral subordination to her. A properly empathetic caregiver must, on the one hand, be sufficiently respectful of and open to the one in need, while on the other, sustaining the requisite degree of self-possession, emotional equanimity, and critical distance to avoid self-effacement," p.176

15. Jodi Halpern. From Detached Concern to Empathy: Humanizing Medical Practice (Oxford: Oxford University Press, 2001).

16. Ibid., 142.

17. Lorenz, Steckard, and Rosenfeld. "End-of-Life Education Using the Dramatic Arts: The Wit Educational Initiative," Academic Medicine, 79 (2004): 481-86.

18. Bardes, Gillers, and Herman. "Learning to Look: Developing Clinical Observational Skills at an Art Museum," Medical Education, 35 (2001): 1157-61.

Previous Article | Index | Next Article

APA Newsletters

Fall 2006 Volume 06, Number 1

Fall 2006
Volume 06, Number 1