Introduction

Empathy is a thing constantly asked for and stressed as a central skill and character trait of the good physician and nurse. To be a good doctor or a good nurse one needs to be empathic—one needs to be able to feel and understand the needs and wishes of patients in order to help them in the best possible way, in a medical, as well as in an ethical sense. The problem with most studies of empathy in medicine is that empathy is poorly defined and tends to overlap with other related things, such as emotional contagion, sympathy, or a caring personality in general (Pedersen 2010). It is far from clear how empathy fits into the general picture of medical ethics and the framework of norms that are most often stressed there, such as respect for autonomy, nonmaleficence, beneficence and justice (Beauchamp and Childress 1979, 2009). How are we to look upon the role and importance of empathy in medical ethics? Is empathy an affective and/or cognitive phenomenon only, or does it carry moral significance in itself as a skill and/or virtue? How does empathy attain moral importance for medicine? In this paper I will attempt to show that a comparison with the Aristotelian concept of phronesis makes it easier to see what empathy is and how it fits into the general picture of medical ethics.

The Aristotelian concept of phronesis, most often translated as “practical wisdom” or “prudence”, appears to come close to capacities we associate with empathy, at least in some understandings of the latter phenomenon. Is being empathic and being wise in Aristotle’s understanding of phronesis in many ways really the same thing? And, if so, could Aristotle’s understanding of what it means to be wise in practical matters help us to better understand the phenomenon of empathy? This could be the case even if empathy and phronesis were not found to be identical in essence, namely if the two phenomena were found to be importantly related to each other in some way. Perhaps the phronetic person must be skilled in empathic matters or perhaps the empathic person must be wise in at least some of the ways that Aristotle holds to be characteristic of a person having and exercising phronesis?

Different understandings of empathy

The questions are made harder to answer by the fact that we do not have a general consensus among researchers on what to mean by empathy (for an overview see Coplan and Goldie 2011). There are lower-level definitions of empathy, making it essentially a kind of automatic mirroring process of a bodily-feeling type, and there are higher-level definitions of empathy, expanding the emotional component to include cognitive and imaginative processes of mind. Providing this disclaimer of failing consensus, I think most empathy researchers would agree that a low level definition of empathy transforming the phenomenon to a sort of reflex is not sufficient to get hold of what we are to mean by empathy in any full blown sense. At least not if the understanding of the lower level bodily processes in question are not tied to some version of what the phenomenologist calls intentionality (Zahavi and Overgaard 2012). The discovery of mirror neurons in the early 1990s was an important step in the research on empathy (Rizzolatti et al. 1996), but the fact that we are unconsciously affected by the emotional expressions of other people by mirroring them does not by itself make us more or less empathic. In order for an emotion to qualify as empathic it must be a feeling about the other person, not only a feeling that has been caused by the other person’s emotional expressions.

What is being discussed in the empathy literature is most often not if the higher levels of empathy exist or not—in the sense that they should be included in what we are to mean by empathy—what is being discussed is what the higher levels look like and if and how they are dependent upon lower, bodily levels of more or less automatic mirroring processes. I think the concept of phronesis could be fruitful to these discussions and also contribute to the discussion of another issue brought up in scrutinizing the processes and essence of empathy, namely the question if empathy in its full blown meaning must not in some way have a moral content. Empathy seems to include a caring for the other person in the sense that an impulse to relieve her suffering, and perhaps also a reflected judgment that one ought to help her, is built into the process in question (Slote 2007). In an everyday understanding of the concept, being empathic most often means to be morally good, whereas lacking empathy is a moral defect (Battaly 2011). Professional researchers of empathy, in contrast to this, typically want to keep the empathic and the moral realm separate (Prinz 2011). Empathy is not the same thing as sympathy they point out. And getting to know the predicament of the other is not the same thing as coming to the conclusion that one ought to help her, or, even less, automatically taking action in order to help her.

It is true, of course, that being empathic in not the same thing as being morally good. Many other things influence what moral conclusions we form in situations being faced with the misery of other people, and if we will transform these conclusions into morally righteous actions. But in performing conceptual moves to restrict the meaning of empathy from expanding into the moral sphere, I nevertheless think empathy researchers often pay the price of losing some of the true experiential content of the phenomenon in question. It is possible to demonstrate that empathy has a central moral significance and yet explain why having empathy is not enough to be morally wise. I will come back precisely to this issue, but I first I need to say a little bit more about what Aristotle means by practical wisdom, since this is the concept I will now bring into the analysis of empathy.

Aristotelian ethics

Phronesis is thematized in the sixth chapter (or book) of the Nicomachean Ethics as one of the excellences, arête, found in the five different forms of human activities Aristotle associates with seeking and having knowledge: episteme (scientific knowledge), techne (technical expertise), phronesis (practical wisdom), sophia (philosophical wisdom) and nous (intellectual insight) (Aristotle 2002, p. 1139b15).Footnote 1 Practical wisdom is characterized by Aristotle as a kind of knowledge of how to act in situations that cannot be judged by applying algorithms (rules of action), but only by thoroughly understanding the concrete situation at hand and judging what to aim for in this particular case. Phronesis is therefore not identical to scientific knowledge, in which general truths are found which can be applied, or to technical expertise, in which case the goal of the activity is given beforehand, since the technician aims to produce a certain thing—good wine, shoes or a house, etc. Practical wisdom is nor yet the same thing as philosophical wisdom, which is not directly focused on taking action in human matters, and neither the same thing as nous, intellectual insight, the exact meaning of which is notoriously hard to explicate from Aristotle’s writings.

Phronesis is, according to Aristotle, an intellectual ability which is perfected by experience—actually he claims that only old men can have it—but this does not mean that practical wisdom is only concerned with thinking in contrast to feeling or acting. In the passage preceding the definition of the five different forms of knowledge-excellences that humans can have Aristotle discusses how good actions (eupraxia) are dependent not only on intellect but a drive and desire to do the right thing (orexis) (Aristotle 2002, p. 1139a31). Practical thinking is therefore rooted in feelings that guide the deliberation in question, and, as I mentioned, the territory of phronesis is exactly the realm of human interaction. Aristotle uses the expression “intellectual excellence” (aretai dianoetikai) to distinguish practical wisdom from that which he calls the moral excellences: temperance, courage, generosity, friendliness, righteousness etc. They are all arête (“excellences”, sometimes also translated as “virtues”), but practical wisdom involves a kind of reflected deliberation that one does not find in case of the moral excellences, which guide one’s action in a more direct and un-reflected way. The morally virtuous (excellent) person not only needs to embody and cultivate the different moral excellences, he also needs practical wisdom to understand and judge the situation in which he is to take action.Footnote 2 Without practical wisdom he will not be able to act in a good way even though he is courageous, friendly, generous, moderate, etc.

I will not go into the question of how, exactly, practical wisdom and the different moral excellences relate to each other, but I think it is clear that they are mutually reinforcing and necessary for each other in Aristotle’s understanding. Not only does the person who has the particular excellences Aristotle names moral (arête ethike) need phronesis to act well. The phronetic person (having arête of the intellect, dianoia) must necessarily embody moral excellences, since without them it would not be possible for him to see, understand and judge the situation at hand in the appropriate way. If I am ungenerous, unjust, a coward, or intemperate, I will not see what is at stake in a precarious situation—what we usually refer to as a moral dilemma. I will perhaps even not understand why and that it is a moral dilemma, because I am unable to understand the conflict at hand in for instance a situation in which I am tempted to lie rather than causing suffering. Perhaps I will not even understand that telling the truth will cause the other person to suffer because I do not understand what things are like from her perspective in the first place.

Phronesis and empathy

My idea of introducing phronesis to the investigation of empathy is not to further complicate the issue by bringing all the questions and distinctions of Aristotelian practical philosophy to the scene. Ideally, the concept of phronesis should make us able to see clearer what empathy is, not hiding it behind clouds of further distinctions and problems regarding the essence of human nature, knowledge and the good. How so? A minimal notion of empathy is that it consists in feeling and knowing the state and predicament of another person. Empathy in this way is a kind of discernment, a way of seeing what is going on in a world that we share with other human beings. This, I think, is in a way exactly what Aristotle means by phronesis. Martha Nussbaum has shown how the Aristotelian notion of practical wisdom rests on an understanding of emotions as containing knowledge about the world we share with other human beings (Nussbaum 1990a). Phronesis is not devoid of feelings, it is rather based in feelings that help the wise person to see and judge what is at stake in the situation. In Aristotle’s famous, but also notoriously vacuous, formulation it is about feeling the right things “at the right times, about the right things, towards the right people, for the right end, and in the right way” (2002, p. 1106b21-3). Phronesis must therefore be rooted in empathy; it must take its starting point in being able to feel and know the state and predicament of the other person in the situation in which we strive to seek the best solution for the people involved. This discernment is aiming to map out what I earlier referred to as a moral dilemma, or, perhaps better, a situation which calls for action, but in which it is hard to know what the best thing to do is.

Some philosophers will deny that empathy essentially is a feeling. If one believes that establishing knowledge about other people’s state of mind is essentially about having a theory about them, one will rather see empathy as a cognitive phenomenon. And if one believes it to be a matter of imagining to be in the other person’s situation, or even imagining being him, one will stress that this may be dependent upon feelings, but not restricted to feelings only.Footnote 3 I think it is important to stress, as the simulation theorist does, that empathy is not only a feeling in the sense of a bodily sensation. Empathy is an emotion containing knowledge about the other person, as Nussbaum underscores in her understanding of Aristotle’s practical philosophy (1990a). This emotion containing knowledge about, for instance, that the other person is sad, and about why she is possibly sad, can be further cultivated by imagining possible reasons for sadness, or trying to remember how things have worked in similar situations one have experienced in life meeting this person or sad people before. However, these further empathic acts do not transform empathy into a cognitive phenomenon only; it rather underlines that the feeling in question is rich and sophisticated in its emotional content.

Getting beyond the feeling or thinking dispute of empathy we still have to answer the important question of exactly what sort of feeling empathy is. Empathy is potentially rich in cognitive content, but as an emotion it does not appear to be of one certain type, such as love, hate, jealousy, pride, etc. Taking our starting point in the experiential characteristic of empathy—how it feels—it seems to be hard to pin point what sort of feeling it truly is. Empathy is not like pain or pleasure in the sense that it always feels bad or good. Typically, but not exclusively, empathy is about suffering, but it is not my pain I am feeling, it is the pain of the other person in the sense that I feel that the other person suffers. Even if we agree to limit empathy to cases in which I feel the suffering of the other, and not, for example, her joy about winning an Olympic gold medal, this feeling of the other’s suffering can be pretty different depending upon how she suffers. To see and feel someone wriggling in bodily pain does not produce the same empathic feeling as seeing her losing the Olympic bronze medal by two hundredths of a second and feeling sorry for her. This is the point at which I would like to enter the empathy debate with a more specific suggestion, namely that empathy could be viewed as the feeling component of phronesis. This is what makes empathy a specific type of feeling and this is what makes phronesis special in comparison with the other forms of human knowledge listed and analyzed by Aristotle in the Nicomachean Ethics (2002, p. 1139b15).

Empathy and the virtues

Just like empathy researchers have pointed out that empathy does not appear to be a specific feeling, they have also pointed out that empathy is not a specific virtue (e.g. Battaly 2011). The latter argument has been part of the strivings to empty empathy of moral content and sometimes also of moral significance, as I touched upon above (e.g. Prinz 2011). It is probably right that empathy cannot be put among other moral virtues as being of the same type as, for instance, courage, temperance, friendliness, trustworthiness, righteousness, etc. A proof at hand for this that is often brought up is the example of the psychopath. The psychopath understands exactly how the other person is suffering, but he does not feel any urge to help her, or, at least, he does not transform any such urge into helpful actions, rather he acts in ways that utilizes the suffering of the other for his own gain, or perhaps even to feel raw pleasure, if he is a sadist as well. Typical for the moral excellences, according to Aristotle, is that one acts upon them directly and in an un-reflected way. If one is a courageous person one will act in a courageous way when faced with a situation demanding courage—leaving aside for a second the question of weak will: akrasia. This does not seem to be the case with empathy, as the psychopath example shows, and therefore empathy is not a moral virtue, so this argument goes.

Perhaps the psychopath case could be shot down by the counter argument of akrasia—the psychopath has the knowledge but lacks the will to help, or, rather, he has other competing wills that makes him ignore that he ought to help the suffering person. But there appears to be something wrong with this counter argument, since empathy, at least to my mind, partly consists in an urge to do good by relieving suffering. This urge is visible already when we watch somebody in pain and feel not only his pain, but also a kind of un-reflected urge to relieve it in some way. This caring urge is not felt by the psychopath, it appears to be blocked in some way, but this does not stop the psychopath from cultivating his feeling of the other person’s predicament in a sophisticated way as concerns how and why the person in question is suffering. Maybe one could claim that the different moral virtues always work in combination with each other and that it is the total lack of some other essential moral virtue that the psychopath lacks and which makes his feelings and actions morally defect even though he embodies the virtue (excellence) of empathy. If we include being a caring or compassionate person among the moral virtues these would be good candidates for this essential moral virtue that is missing in the psychopath case.

A related but more elaborate way of putting empathy into the context of Aristotelian moral philosophy is to view it, not as one of the moral virtues, but as an integrated part of phronesis. This is my suggestion. Phronesis is, as I have pointed out, not a moral, but an intellectual virtue in Aristotle’s theory, although, of course, importantly related to the moral virtues. Empathy would then be, as I phrased it above: the feeling component of phronesis. Let us now attempt this interpretation more in detail.

The reason why the psychopath does not feel the urge to help or comes to the moral conclusion that he ought to help is, indeed, that he lacks in other moral virtues, such as friendliness or righteousness, if we are to stay with Aristotle’s list of virtues, or compassion and carefulness, if we expand the list of moral excellences to accommodate other considerations than the ones at play in Aristotle’s practical philosophy. What remains for the psychopath is the feeling-understanding component of empathy unguided by the disposition to act in a morally good way provided by the other moral virtues which provide the means for phronetic discernment together with empathy. In this way it is possible to be empathic without exercising phronesis even though empathy is the starting point of phronesis without which it cannot be performed. The wise deliberation of phronesis must be guided by an emotional discernment of the ways other people feel and think and other moral virtues. My guess would be that the psychopath still feels some kind of un-reflected urge to do something when watching somebody in pain, since this feeling is close to an automatic reflex in such situations, but that this urge is not transformed in any caring way. Instead the urge provides energy for other cognitive moves and actions that take advantage of the suffering person or at least neglects her suffering.

This pattern explains why empathy is most often looked upon as a moral virtue even though closer philosophical exploration makes us skeptic about it having a moral content in itself. Phronesis cannot be exercised without having the moral virtues, but it can also not be exercised, indeed, even be initiated, without having empathy. Phronesis partly consists in empathic capacities. Typically persons embody the basic moral virtues to some extent and they also have empathy. The psychopath has empathy but lacks in basic moral virtues. Other people may embody most moral virtues to a large extent, but lack in empathic skills, which makes them come to unwise decisions about what to do in ethically precarious situations even though they are virtuous in the sense of being moderate, generous, brave, friendly, righteous, etc.

In what sense is this solution different from claiming empathy to be a moral virtue in itself? In addition to dealing better with the psychopath counter argument, I think it explains why empathy is utterly morally significant without being a moral excellence in itself. It is so significant because lacking empathy will mean lacking phronesis, without which the moral virtues cannot come to a wise expression. Lacking in the different moral virtues will also have bad effects on phronesis, but lacking in empathy means to lack phronesis itself. Because of this, empathy will be highly significant and necessary for being a morally virtuous (excellent) person despite not being a moral virtue in itself.

Are empathy and phronesis then, after all, not only related but also identical things? No, they are not. The psychopath example makes this obvious, but I think it is possible, and perhaps even common, to have at least robust forms of all moral capacities (virtues), be skilled in empathic matters, and yet not be a phronetic person. What would lack in these cases are the kind of life experiences in moral matters that makes Aristotle say that only old men can have phronesis. What Aristotle is wrong about when he says this, is the restriction in kinds of experiences that will count as cultivating phronesis. Not only the political life of the polis, but also experiences from other professions than that of the politician and from everyday, private life will make persons wiser in ethical matters. The example of the experienced doctor having met and helped a large number of patients, and the example of the doctor who have grown wiser by falling seriously ill himself and temporarily occupied the position of the patient, make this point obvious. Other sources of phronesis are the imaginative exercises of art and literature, as Martha Nussbaum suggests in her writings (e.g. Nussbaum 1990b). These matters bring us to a final question I want to sort out in this paper, namely how the relationship between empathy and phronesis I have suggested could be of significance for the way we look upon the role of empathy in medical ethics.

Empathy and medical ethics

As Jodi Halpern has pointed out, in her important study From Detached Concernto Empathy, to exercise empathy in the clinical encounter is not only a kind of ethical icing of the cake making health care professionals nicer and kinder to patients in addition to being skilled in medical matters (Halpern 2001, see also Halpern 2014). Empathy is actually a capacity making the doctor more able to make a correct diagnosis, and also a quality of the medical meeting that contributes to empowering patients and improving upon their recovery (Halpern 2001, p. 94). The reason for this is that empathy is not only about being influenced by the feelings that patients display—emotional contagion—or feeling sorry for them—sympathy. Empathy is one of the basic capacities that make the doctor able to understand what the reasons for complaints and suffering are about and what can and ought to be done to help the patient in the best possible way. In this process of clinical understanding the very fact that the patient feels that the doctor is interested in her problems and wants to help her will contribute to making the patient more able to deal with her health problems, and it will also improve upon recovery (what is often referred to as placebo effects). Doctors should allow themselves to be moved by patients and the feelings they display—they should not be detached—but in this being moved by the patient it is crucial for the doctor to not conflate the feelings of the patient with her own feelings, or forget that she is actually there to help the patient and not to feel sorry for her. The reason why empathy is sometimes looked upon as a faulty or risky strategy for the doctor, leading to non-objective judgments or burn out on the side of the doctor, is that it is confused with an emotional merging with the experiences of the patient or a feeling sorry for her. Writes Halpern:

“Writers on empathy either base empathy in detached reason or sympathetic immersion. Against these models I describe empathy in terms of a listener using her emotional associations to provide a context for imagining the distinct experiences of another person. Therefore, empathy is a form of emotional reasoning, with the risks of error that such reasoning involves. To empathize more accurately physicians need to strive to be self-aware, thus avoiding projecting their own unacknowledged emotions onto patients.” (Halpern 2001, p. xxiii)

Emotional reasoning in the form of imagining what the patient is experiencing and what these experiences are caused by and are about is the best way of forming a good clinical judgment that at the same time displays a caring for the patient. This way of phrasing the role of empathy in the clinical encounter comes very close to the Aristotelian concept of phronesis that I have been investigating. Maybe Halpern’s model of empathy is essentially a model of phronesis? I think this is the case, but I would be quick to point out that Halpern in her model of empathy points towards how empathy is a necessary part of good clinical judgment, rather than covering the whole ground of clinical understanding. This is exactly the point I have been trying to make myself about empathy in relation to phronesis. Empathy does not guarantee that the doctor will develop an adequate understanding of the patient’s problems and find the best way to help her, but without empathy the doctor is in many cases bound to fail, because she will not even see—be perceptive of—what the problem really consists in. As soon as life-world questions have some kind of bearing upon what and how a patient presents symptoms to the doctor, the capacity for and attitude of empathy will be necessary to assure the doctor of forming a perspective that goes deep and wide enough to address the reasons for illness.

If we acknowledge this we can see how empathy forms an important part of medical ethics as a capacity and attitude of health care professionals making good care for patients possible. Empathy does not enter the scene after the medical problems have been understood and addressed—in order to guarantee that patients are treated in a humane way in some kind of addition to being helped with their medical problems. Empathy is a core part of clinical understanding and this understanding is itself morally significant because of the duty to understand and help on the side of the professional.

In earlier studies I have tried to show how the concept of phronesis can be considered a kind of hermeneutic virtue in medicine because the handling of ethical dilemmas is exactly about being able to see things from the patient’s point of view without losing touch of one’s own horizon of understanding as a doctor or nurse (Svenaeus 2003). This philosophy of medical practice as a form of hermeneutical understanding is inspired by the work of Hans-Georg Gadamer (Gadamer 1990; Svenaeus 2000). The explication of phronesis as a hermeneutic virtue in medicine is actually developed by Gadamer himself in his late work The Enigma of Health, a collection of papers in which the author also addresses the phenomena of health and illness as different forms of being-in-the-world, a concept dating back to his teacher Martin Heidegger (Gadamer 1993). Lou Agosta, in his recent study Empathy in the Context of Philosophy, has attempted to show how empathy is not only a way of developing thoughts about the state of the other by way of simulation processes, but actually a way of being-together-in-the-world (the German concept is “Mitsein”) in an authentic way (Agosta 2010, see also Agosta 2014). Agosta’s attempts to provide empathy with a founding role in a theory about intersubjectivity and the good life is in consonant with phenomenological analyses of empathy, such as we find them in Edith Stein or Max Scheler (see Bornemark 2014). As is the case with Gadamer’s interpretation of human interaction, Agosta’s interpretation of empathy take its starting point in the philosophy of Heidegger developed in Being and Time (Heidegger 1986). These phenomenological-hermeneutic suggestions to develop empathy as an essential ethical concept squares well with my own attempt in this article to show how empathy is a core part of phronesis.

In combination with, and addition to, addressing phenomenological-hermeneutic theories about empathy, the easiest way to show how empathy has an important role to play in medical ethics would be to argue for the importance of virtue ethics in medicine more generally (Pellegrino and Thomasma 1993). By bringing in virtue ethics, phronesis will be provided a role, and then, by way of my interpretation, empathy will also enter the scene as an important concept and phenomenon for medical ethics. This way of showing empathy to be an essential part of medical ethics is by no means wrong, but it is not the only way of providing empathy with moral significance. Medical ethics is often viewed as the project to balance between different prima facie principles, such as respect for autonomy, nonmaleficence, beneficence and justice in ethical dilemmas (Beauchamp and Childress 1979). This set up brings in different central principles from different moral theories—utilitarianism, Kantian ethics—and the difficulty is to strike a proper balance between the principles through an interpretation of how they attain significance in a particular situation. Phronesis as a hermeneutical skill will surely be central to this interpretative endeavor and empathy in forming a necessary part of this practical wisdom will have to be cultivated if the ethicist should come to sound conclusions in balancing the ethical principles in question.

In this way empathy is not any kind of substitute for respect for autonomy, or some other of the central principles of contemporary medical ethics, it is rather a necessary condition for being able to see how the principles in question can be respected and acted in favor of in the real world of patients and their suffering.Footnote 4 Empathy is a basic condition and source of moral knowledge by being a central component of phronesis, and, by the same power, it is also a motivation for acting in a caring and helping way. The comparison between empathy and phronesis in this way makes it easier to discern the place and function of empathy in medical ethics.