The Way to a Definition

Abstract

‘The Way to a Definition’ offers a clear-cut definition of empathy thanks to a descriptive, genealogical, and conceptual reconstruction of this phenomenon which is based especially on the phenomenological views on empathy by Edith Stein. It is shown how the proposed definition is able to accommodate the perspectives of the old and new exponents of the phenomenological tradition with the Smithian-derived idea widely present in many modern scholars of empathy who see, unlike the phenomenologists, imagination as the basis of this phenomenon.

It is not the aim of this work to offer an in-depth analysis of the philosophy of Lipps, but it was important to start our investigation of empathy with a brief report of his view on the matter, primarily for the reason that he was the first to talk about empathy (not sympathy) and to use this concept in a systematic way central to his own theory and, secondly, because more than hundred years after Lipps’ book, there is still no agreement among the scholars both on the definition of empathy and on the precise role it can play. In other words, although Lipps’ desire was to clarify the concept and use it consistently, his objective is nowadays still far from being achieved. As a matter of fact, for the young academic approaching the subject area of empathy, the landscape can be quite disorienting and distressing: empathy is a ‘hot topic’ nowadays, and it has managed to enter the field of inquiry of disciplines ranging from philosophy to the neurosciences, from psychology to zoology, and from anthropology to economics. Of course, given the incredible amount of work made during the years by scholars with such different backgrounds and opinions on the matter, it cannot be surprising to see how blurred the definition of empathy is. Just to give two concrete examples of the terminological, semantical, and content-affecting chaos which exists in the literature, we can take into consideration the works of Coplan (2011) and Batson (2011). In the first, the philosopher Amy Coplan defines empathy as: ‘[A] complex imaginative process in which an observer simulates another person’s situated psychological states while maintaining clear self-other differentiation’,¹ however, she also notes that the empathic phenomenon has been described as involving one or more of the following features:

• (A) Feeling what someone else feels;

• (B) Caring about someone else;

• (C) Being emotionally affected by someone else’s emotions and experiences, though not necessarily experiencing the same emotions;

• (D) Imagining oneself in another’s situation;

• (E) Imagining being another in that other’s situation;

• (F) Making inferences about another’s mental states;

• (G) Some combination of the processes described in (A)–(F).²

The famous psychologist Daniel Batson refers to empathy using these words: ‘I shall use empathic concern and, as a shorthand, empathy to refer to other-oriented emotion elicited by and congruent with the perceived welfare of someone in need’.³ Nevertheless, he then adds that there are at least seven other states which can be called (and have been called) empathy, which are the following:

• (A) Knowing another person’s internal state, including his or her thoughts and feelings;

• (B) Adopting the posture or matching the neural response of an observed other;

• (C) Coming to feel as another person feels;

• (D) Intuiting or projecting oneself into another’s situation;

• (E) Imagining how another is thinking and feeling;

• (F) Imagining how one would think and feel in the other’s place;

• (G) Feeling distress at witnessing another person’s suffering. ⁴

It is quite easy to understand that this incredible range of different and competing conceptualisations in the literature has created (and still creates) countless difficulties for anyone willing to coherently investigate the phenomenon of empathy. Actually, we have reached the point in which a researcher using the name ‘empathy’ or the corresponding verb ‘to empathise’ is not communicating anything at all if they are unable to provide an unambiguous description of it. Therefore, any good work on empathy should start with a clear-cut definition and the fact that this definition will under some aspects be ‘stipulative’ should not be an issue. As the already cited researcher on empathy Daniel Batson once wrote: ‘In spite of frequent claims that one’s own use of these terms is best, I know no clear basis […] for favoring one labeling scheme over another. In such circumstances, I believe the best one can do is recognize the different phenomena, make clear the labeling scheme one is adopting, and use it consistently.’⁵ This is in fact the strategy used by the vast majority of researchers on empathy and I will make no exception. However, I do not want my definition of empathy to be arbitrary, in fact, I would like the definition offered in this work of mine to be attractive and convincing for other scholars as well as for ‘normal’, laypeople. That is why I want my definition to maintain important insights stemming from traditional researchers on the subject (Adam Smith, Edith Stein, Max Scheler, etc.) and, at the same time, to be in line with both the latest characterisations of empathy and with the normal use people in general make of this concept. This will also help to highlight a coherent evolution of the discussion around this phenomenon that, despite becoming more and more problematised and problematic, revolves nonetheless around the same features.

Now, the greatest and most important difference there is between the interpretation of Smith, the already summarised theory of Lipps, and the phenomenological view on the matter, is the function attributed to the faculty of imagination. In fact, imagination was surely crucial for Smith, who unambiguously wrote:

As we have no immediate experience of what other men feel, we can form no idea of the manner in which they are affected, but by conceiving what we ourselves should feel in the like situation. Though our brother is upon the rack, as long as we ourselves are at our ease, our senses will never inform us of what he suffers. They never did, and never can, carry us beyond our own person, and it is by the imagination only that we can form any conception of what are his sensations. Neither can that faculty help us to this any other way, than by representing to us what would be our own, if we were in his case. […] By the imagination we place ourselves in his situation, we conceive ourselves enduring all the same torments, we enter as it were into his body, and become in some measure the same person with him, and thence form some idea of his sensations, and even feel something which, though weaker in degree, is not altogether unlike them.⁶

Smith is crystal-clear here: senses cannot tell us how other human beings are feeling. The only means we have available to experience what others are experiencing at a given moment is by placing ourselves in their situation thanks to imagination and, through this process, imagine what we would feel. Notice that this process, though imaginative, has the practical effect to make us feel the emotion we imagine. Indeed, Smith writes: ‘For as to be in pain or distress of any kind excites the most excessive sorrow, so to conceive or to imagine that we are in it, excites some degree of the same emotion, in proportion to the vivacity or dullness of the conception.’⁷

The central role attributed to imagination (that is itself part of the bigger mechanism of empathy) brings Smith close to Hume’s theory on sympathy, but marks a difference with regard to the position held by Lipps, who, as we have seen, thought that empathy was something spontaneous and based on an automatic mechanism of imitation. For Lipps, empathy was fundamentally a conceptually poor process. However, the two of them shared the belief that only our own mental states are transparent to us and that those of others have, on the contrary, to be in some way inferred.

Nonetheless, the real problem with Lipps’ conception of empathy was what he called Trieb der Nachahmung.⁸ For Lipps, all human beings have the tendency to internally imitate the observed expressions of other people (gestures, mimicry, facial expressions, etc.), but these imitations are not always visible from the outside, for the very good reason that this kind of proclivity activates impulses in us which are nevertheless not carried out at any time. In fact, human beings have learnt to suppress these impulses. Thus, for instance, every time I see a person yawning, even though I can decide to not yawn in reply, I will still internally imitate this expression.⁹ For Lipps, this mechanism is in place from the most basic forms of emotional contagion and imitation to the highest and most complex forms of empathy.

I think that Lipps’ theory suffers from many weaknesses. Firstly, it seems rather odd to affirm that we always resort to a mechanism of imitation when empathising. There are in fact a great number of cases in which imitation does not occur and to say that it occurs internally and unconsciously certainly appears as a poor explanation. Furthermore, as Edith Stein rightly noticed, if the theory of Theodor Lipps were true, we would ‘not arrive at the phenomenon of foreign experience, but at an experience of my own that arouses in me the foreign gestures witnessed’.¹⁰ This means that by means of imitation we would not grasp the experience of other subjects, but only our own experience, awakened by the observation of the expressions of others.¹¹ For this reason, Stein concludes (and I cannot but agree with her) that Lipps’ theory of imitation cannot serve as a genetic explanation of empathy.¹² We will see now in the next chapter how some of the greatest exponents of the phenomenological tradition define empathy.

1 A Phenomenological View on Empathy

The description Edith Stein gives of empathy has numerous points common with the definition I will offer. This is not surprising, as I think that Stein, together with Max Scheler (whose theory of empathy has influenced me a great deal, but is one which I cannot support in its entirety, in particular as it has too many distinctions between very similar phenomena like Einsfühlen, Mitfühlen, and Nachfühlen that often occur together, thereby risking muddying the waters, instead of clarifying the problem), was capable of grasping the real essence of the empathic phenomenon. To quote Stein:

When it arises before me all at once [the experience of the other person, ed.] it faces me as an object (such as the sadness I ‘read in another's face’). But when I inquire into its implied tendencies (try to bring another's mood to clear givenness to myself), the content, having pulled me into it, is no longer really an object. I am now no longer turned to the content but to the object of it, am at the subject of the content in the original subject's place. And only after successfully executed clarification, does the content again face me as an object.¹³

By analysing this passage, it is possible to identify three different phases: (1) the emergence of the experience, (2) the fulfilling explication, and (3) the comprehensive objectification of the explained experience.¹⁴ In the first phase, we perceive the experience of the other person thanks to their ‘movements of expression’, which are (for Stein as well as for Scheler, Merleau-Ponty and for modern phenomenologists like Zahavi or Overgaard) the direct behavioural expression (and integral part) of their experience. In the second phase, it is the turn of perspective-taking. Notice that for Stein the perspective-taking is different from projection (although in the modern literature ‘projective empathy’ and ‘perspective taking’ are often conflated). In fact, she uses the term hineingezogen, which means something like ‘dragged’ or ‘drawn’. The empathiser is therefore ‘dragged’, as it were, in the other’s experience. While this occurs (and because this occurs) the empathiser tries to ‘presentify’ what the target experiences. This is an intentional and conscious process, in which also imagination plays a crucial role.¹⁵ In order to understand the third and last phase of the empathic process according to Stein, we have to briefly sketch a difference essential in her reflections on empathy: that between primordial and non-primordial.

For Stein, primordial and non-primordial are the qualities of all our experiences, that is, all our acts can be either ‘primordial’ or not. In particular, primordial are all acts which are carried out at a precise moment and at that same moment are experienced by the agent who is doing them, and non-primordial are the contents of cognitive acts at the actual moment in which they are carried out. What does this mean, exactly? A practical example might help to shed some light on the issue.

Imagine the following situation: suppose that I find myself in Lucerne (where I actually am, writing these lines), waiting for the bus. It is winter, and I have the impression that the cold and damp air that pervades the Swiss city is actually penetrating my own bones. Suddenly, I start to think of my summer holidays in Tuscany, where I come from. I remember the sweet hills full of cypresses, olive trees, and vineyards in the Florentine countryside, kissed by the strong Mediterranean sun. I remember that day on the beach south of Livorno and the smell of brackish water. Now, the sensation of coldness I feel is ‘primordial’; primordial is also the perception I have of the bus finally approaching, and primordial is the remembering of my holidays in Tuscany. What is non-primordial is, instead, the content of my remembrance: the hot weather, the hills, the sea, and that joyful mood in which I was and I no longer am, while waiting for that slow bus. Thus, we can state that for Stein, empathy, memory, and fantasy are all primordial acts in which facts are non-primordially represented.

Subsequent to this account, we should be able to understand the third phase of the empathic process. In the second phase, Stein says explicitly that the empathiser is not directed towards the target anymore: they are, as it were, in the target, in their position and they are directed towards the same object the subject is directed to. This also means that the empathiser’s experience of the target here is original. However, and this is crucial, in phase three the target becomes once again an object, the object the empathiser is directed towards. This means to say that the primordial experience the empathiser has at phase two is only a stage in the process: that is, at the end of the process, in what we might want to call the ‘attribution phase’, this experience becomes non-primordial, for the good reason that the empathiser knows that this is in fact the experience of another and consciously understands that what they feel is not primordial. This is the reason why Stein describes empathy as an original act with non-primordial content.

As we have seen, whilst for some authors, like Adam Smith and Edith Stein, imagination is of central importance for carrying out the empathic process, it is not so for Lipps, who instead argues in his texts for an immediate, direct, and conceptually poor process.¹⁶ This creates a tension that remains unsolved even in the works of authors all belonging to a tradition that is far from being unitary like the phenomenological one. In fact, as we are going to observe later in the book, a modern and very influential phenomenologist, such as Dan Zahavi, tends to adhere to a philosophical interpretation of empathy that departs from that of Stein and has much more in common with the reflections of Merleau-Ponty. Following this reading, the mind of others does not have to be considered as something alien to this world and that for this reason must be inferred. The mind, in fact, expresses itself in behaviour, gestures, vocalisations, and facial expressions. Empathy does not have to, in this sense, bridge a gap, thanks to imagination, from the visible to the invisible, from what can be observed to what cannot be observed, because everything can be observed. Mental states are not cut off the world; they exist in it and can be examined by means of the common context of experience. To use the words of Merleau-Ponty:

In so far as I have sensory functions […] I am already in communication with others […] No sooner has my gaze fallen upon a living body in process of acting that the objects surrounding it immediately take on a fresh layer of significance; they are no longer simply what I myself could make of them, they are what this other pattern of behaviour is about to make of them.¹⁷

And, especially:

We must reject this prejudice which makes “inner realities” out of love, hate, or anger, leaving them accessible to one single witness: the person who feels them. Anger, shame, hate and love are not psychic facts hidden at the bottom of another's consciousness: they are types of behaviour or styles of conduct which are visible from the outside. They exist on this face or in those gestures, not hidden behind them.¹⁸

2 Our Working Definition

As you can see, there is a rift between the scholars who tend to see empathy as a conceptually rich and imaginative process and those that conceive it as an immediate and conceptually poor mechanism. Of course, I could choose to give my preference to one of these positions and reject the other, however, my intention is to do something different. My aim is to offer a definition of empathy that can accommodate both positions in its formulation. Later, we will see how this description of empathy is able to describe correctly both the conceptually poor and unmediated kind of empathy theorised by some phenomenologists (what I will call Low-level empathy) and the conceptually rich and imaginative variation typical of approaches valuing the phenomenon of the perspective-taking, namely, High-level empathy.

Empathy, according to my definition, is an other-directed psychological process which allows us to tune into others and thereby understand and feel with a variable degree of approximation mental states and emotions that we deem consonant with those experienced by them, while maintaining a sense of self-other differentiation.

This definition entails a few important consequences that it is imperative to highlight. Let us begin with the first adjective I have used: ‘other-directed’. With this I want to imply that when we empathise we always empathise with another (a human being or an animal), therefore, our focus is directed to the other and not turned to ourselves. Empathy is an other-centred mechanism, and not an egocentric mechanism.

I have also labelled empathy as a ‘psychological capability’ and not as an emotion. In truth, empathy cannot be considered an emotion among others, as, for example, compassion. On the contrary, empathy allows us in principle to feel all kinds of emotions the others can feel; thus, its effects will not follow a fixed characteristic pattern, but will vary, depending on which sort of emotion we have come to feel thanks to it. As a consequence, empathy can be said to possess only one of the three main characteristics of emotions, namely, the intentionality, whereas it lacks a specific phenomenology and it makes no sense to talk about epistemological standards and standards of correctness in relation to its epistemology.¹⁹

Coming back to our definition, empathy, I have also asserted, ‘allows us to understand and feel with a variable degree of approximation the mental states of another subject’. This sentence involves a complex cluster of different elements; hence, it will pay to spend some time explaining it. It is usual among researchers on empathy to make a fundamental (and now classical) distinction: that between cognitive and affective (or emotional) empathy. This distinction aims to clearly identify and correctly describe what they have rightly taken to be two different phenomena: cognitive empathy indicates the experience we undergo when we grasp or understand what another subject thinks or feels in a certain moment,²⁰ whereas affective or emotional empathy denotes the state we are in when we not only understand the emotions of another person, but we also feel them, as it were, ‘on our own skin’.²¹ This clarification involves other related matters and requires further analysis. Indeed, if we agree on the fact that there exists a kind of empathy which implies a mere cognitive understanding of the emotions of others, and another one which makes us feel what other subjects are feeling at an affective level, then we are forced to conclude that the first one will also be compatible with a certain distance and ‘coldness’ on the part of the empathiser. After all, the mere understanding that someone is suffering does not lead us, ipso facto, to suffer with them. On the contrary, the second type of empathy would take us, so to say, ‘nearer’ to the inner emotional world of the target,²² making us undergo feelings that are more difficult to handle ‘at a distance’. However, the question we should ask ourselves is: does it really make sense to speak of ‘cognitive empathy’? In order to find an answer to this question, a few clarifications are needed. Whereas I agree on the fact that it is very useful to distinguish between the sharing of an emotion and the mere understanding of the emotion another person may be feeling, I am not willing to label this second kind of phenomenon as empathy tout court. Doing this would mean running against some characteristics which form an integral part not only of the definition of empathy I propose and examine in the present book, but also of what lay people think of as empathy, of what the common usage of the term indicates. In some sense, speaking of ‘cognitive empathy’ contradicts the very etymology of the term ‘empathy’ insofar as it deprives it both of the ‘pathos’ condition and of the ‘en’ condition. The ‘pathos’ condition requires that the emotions of the other must be felt by the empathiser (the word pathos comes from the Greek verb paschein, which means ‘to feel’, ‘to suffer’, ‘to be emoted’), so that in order to empathise with another, one needs to be in an emotional state caused by the empathy experienced for the other. That means to say the ‘understanding dimension’ is not enough without a corresponding ‘feeling dimension’. The ‘en’ condition entails that the empathiser feels the (vicarious) emotions and generally the other’s mental states, as it were, on their own skin, or, in other words, from the inside. As I have said above, this view of empathy is also shared by lay people and it is unambiguously revealed in our normal, everyday usage of the term.

Furthermore, considering the mere understanding-the-emotions-and-thoughts-of-others,²³ a case of empathy tout court would conflict with the fact that empathy does not only involve feeling some kind of emotion as it were our own, but the fact that this emotion has to be consonant with the emotion the target of our empathy is experiencing.²⁴ For this reason, the famous figures of the ‘empathic torturer’ or of empathy as consistent with the ‘cruelty of sadism’, described by scholars like Stephen Darwall,²⁵ Michael Stocker,²⁶ Martha Nussbaum,²⁷ or Fritz Breithaupt,²⁸ should not be taken seriously: it is no more than a contradiction in terms. But I am going to analyse this issue in more depth in the section dedicated to the critics of empathy. For now, it suffices to say that if a human being feels pleasure at the sight of another person who is suffering, then this is indeed the opposite of empathy. The feeling one experiences in this negative situation is in fact poles apart from the feeling endured by the sufferer.

However, even if cognitive empathy should not be seen as empathy, so to say, par excellence, it nonetheless needs to be considered as the most basic and fundamental part of it. There is no true empathy without cognitive empathy and, inasmuch as it makes us understand what kind of emotions the other is feeling, it suits our definition of empathy (even though at its barest). Having said that, there is one crucial thing that must be highlighted, and it is the following: whereas I find the distinction between cognitive and affective empathy to be a useful one for the sake of clearly distinguishing the two phenomena theoretically (after all, understanding and sharing a feeling are not one and the same thing), these two kinds of empathy very often occur together in the praxis. In other words, the boundaries between these mental phenomena are blurred and it can be quite difficult to tell where cognitive empathy ends and where affective empathy begins.²⁹ Sometimes, in fact, the emotion the other is expressing is so clear and so powerful that we affectively empathise with this person immediately and then, on reflection, we become aware of having comprehended what she or he has felt. On other occasions, the conscious process of trying to understand the perspective of the other by placing ourselves in their position is what triggers consonant feelings in us and shake us from apathy.

Therefore, while it can be fruitful to discuss whether cognitive and affective empathy may diverge in special cases (think of the ongoing debates about psychopathy on the one side and autism on the other),³⁰ I claim that in that part of the population that does not include the autistic spectrum disorder and does not suffer from psychopathy, affective and cognitive empathy are very often intertwined.

In the abovementioned definition of empathy appears the term ‘mental states’. With this I want to designate typical mental contents, such as beliefs, emotions, and desires. More specifically, mental states should be considered in terms of propositional attitudes, namely, in attitudes that one can take towards a given proposition. In other words, mental states have a propositional content, or, to put it differently: propositions constitute the mental content of our mental states. So, for example, in the case of belief, if Paul believes that his wife is cheating on him, his mental content is the proposition ‘my wife is sleeping with another man’ (or something similar); in the case of desire, if Lucy wants an ice cream, her mental state is ‘the ice cream is (for some reason) desirable’, and so on. Hence, when we assert that thanks to empathy it is possible to understand the mental states of another subject, we wish to claim that by means of our empathic faculty we can construe in our mind mental states with a similar or even the same propositional content of those belonging to another person.³¹ This means also that empathy—at least the ‘high-level’³² one—is partly a matter of attributing mental states to other subjects.

Finally, it should be noted that empathy always involves, as already mentioned, ‘a variable degree of approximation’, which means to say that other people’s minds (and related mental states) are never completely transparent to the empathiser and that the empathiser will in any case always differ from the target of their empathy. As a consequence, cognitive and affective contents stemming from an empathic process will at best resemble the original mental contents of the target, but they will never be identical to them. This is an intrinsic shortcoming—or, perhaps better said, an intrinsic limit—of empathy that should not be overlooked. The ‘approximation matter’, moreover, uncovers another crucial issue, namely, what I will call the performative nature of empathy. What I mean by this is that empathy is a capacity that must be actualised through a process that has a certain outcome, and in that sense, it can be compared to a performance. Take the case of playing football, for instance. In order to play this sport, one needs to possess certain capacities (among others, the capacity to walk, jump, and run in a coordinated way)³³ that can be further developed and become skills (one player can be more skilled than others in running quickly, others can have a particular skill in dribbling opponents, for example), which are fundamental in achieving the best possible outcome. Analogously, the natural empathic capacity most people have³⁴ can be developed and become more refined (we can become skilled at empathy) and thus help us to reach better outcomes in our empathic processes. In summary:

• 1. Empathy—like emotions—comes in different degrees (it can be more or less strong and more or less accurate in its understanding and feeling of the others’ mental states).

• 2. Different people have different empathy levels (they are not, in other words, all equally skilled).

Specifying all that is of crucial importance, since it helps us to clearly delineate what is a case of empathy and what is not and to answer some questions regarding the extent of empathy. So, for example, imagine the following situation. Suppose Ryan tries to empathise with his friend Josh, who is disappointed and sad. And suppose that Ryan believes Josh is angry, so he goes to him and says: ‘I know how angry you feel at this moment. I can understand if you want some kind of retaliation and I totally share this sentiment!’. As you can imagine, Josh would look at Ryan with an expression of surprise and confusion on his face and tell him that he does not feel angry at all, but rather sad and disappointed. Now, would you be ready to affirm that Ryan has really empathised with Josh?

In order to answer this question, we should first ask ourselves what, inherently, empathy is. Is empathy constituted and defined by its outcome (i.e. the actual reaching of a feeling consonant with that of the target) or rather by its process (i.e. by the fact that we engage in an empathic process by projecting ourselves in the target’s position, trying to feel what they might be feeling)? Arguably, the majority of people (both scholars and laypeople) would agree on judging empathy by its outcome: it seems rather odd to affirm that Ryan has really empathised with Josh and that, for instance, someone who constantly draws the wrong conclusions about the mental states of others (but who nevertheless always tries) is a good empathiser. However, it seems also unfair to say that someone who has made all the correct moves to empathise with a subject, but who fails to reach the right conclusion, has not empathised at all. This is a complex dilemma, and many academics have decided either to follow one of the two horns of the dilemma by highlighting, respectively, the ‘outcome dimension’ or the ‘process dimension’,³⁵ or else by widening their definition of empathy, thereby simply avoiding the dilemma altogether.³⁶ In light of these considerations, I think that the model I propose can be a valid way, not of bypassing the problem, nor of capitulating to one of the two options, but of providing a solution to the impasse entirely.

Following my model, empathy cannot depend just on its outcome, or we would be forced to label as acts of empathy all those cases in which there is only a mere isomorphism regarding the emotions of both the empathiser and the target (e.g. the cases of emotional contagion, in which, inter alia, there is no self-other distinction and thus—as we will see in the next chapter—cannot count as empathy). But empathy cannot depend solely on the undertaken process either, or we ought to accept that a person who genuinely thinks and feels that another is happy when they are actually sad has really empathised, because he or she undertook all the “right steps” (e.g. imaginative enactment, clear self-other distinction). So, how do we solve this dilemma? Think again about the analogy with football. One can play football masterfully, as Lionel Messi or Cristiano Ronaldo do, or rather mediocrely, as I do, for instance. Are Messi, Ronaldo, and I doing the same thing when we play football? It seems that in some sense we are, and in some sense we are not, but generally, we tend to agree that we are all playing football. Only, we are inclined to make an important difference that we stress using two thin concepts: we say that Messi and Ronaldo play football well (even very well) and that I play football badly (even very badly). This means that although we are carrying out the same performative act (pro forma), I do it at a certain level and with a certain outcome, and they do it at a whole other level with entirely different outcomes. However, imagine that I start to use my hands to take the ball (outside of the box and without being a goalkeeper) and shoot a goal, or imagine that I am simply too incompetent to even control the ball with my feet and defend it with my body, or to make a run. Would it still be possible to agree that I am really playing football? Arguably not.

Why so? The point is that for all performative acts the rule applies that there is a certain boundary within which a performance can still be considered valid, that is, it can still be considered as a valid instance of a given performative act, although it is not a good or a perfect one; on the contrary, all actions falling outside of this range cannot be judged ‘fitting’ with regard to the undertaken performance. My argument is that the same goes for empathy. To a certain degree, instances of empathy in which the empathiser does not perfectly match the target’s mental states can still be considered cases of empathy. Yet, above a certain threshold, they cannot.³⁷ So, for example, the person who says to feel ‘the happiness of the other’ when the other one is actually sad does not meet the minimal requirements of the consonance condition (the incongruence is too great), whereas the empathiser who feels the other is sad when the target of their empathy is depressed has (although not perfectly) met those minimal requirements. Of course, this also entails that there will be ‘borderline cases’ in which judging if a certain act counts as empathy or not will be harder than in others (e.g. academics might debate as to whether Ryan has really empathised with Josh), but since matters of emotions are rarely simple and clear-cut,³⁸ this inherent difficulty cannot be taken as sufficient to reject this view. On the contrary, I think that this view can help to solve complex cases, such as that of Ryan and Josh, without resorting to dogmatisms (like that of isomorphism). A possible solution for the ‘Ryan and Josh case’ could be the following: is there in the mixture of sadness and disappointment experienced by Josh an element of (maybe not entirely conscious) anger? If there is, then we could agree on the fact that Ryan has, to some extent, empathised. Otherwise, we can deem his act as a mere attempt at empathy (like a failed performance).

I think that the choice to be more flexible regarding the ‘isomorphism condition’, making it an issue of ‘consonance’ and not of perfect equality, not only makes our concept of empathy more similar to that of David Hume and Adam Smith, but it can make the discussion about empathy more productive, since requiring perfect equality would be a non-starter, as it is for Peter Goldie’s concept of empathy.³⁹ In one of his latest articles before his death, in fact, Goldie defended an intransigent position for which empathy (‘empathetic perspective-shifting’, to be exact) was unable to operate with the appropriately ‘full-blooded notion of first-personal agency’ that is involved in deliberation.⁴⁰ His argument is complex and well-described and it is not easy to summarise it in a few words. But, to put it as succinctly as possible, since it is impossible for the empathetic person A to take on the full-blooded notion which is typical of the first-personal agency and deliberation of B in conditions of confusion and conflict (when, i.e. decisions are not easy to be made), then either A is forced to usurp B’s agency or they have to conceive of it in an unrepresentative ‘double-minded’ way. It is not my aim to deepen this issue here, because this will take us far away from our primary focus, but it suffices to say that the concept of empathy held by Goldie is uselessly restrictive. No one requires from empathy to be that accurate and to give us the complete full-blooded access to the first-personal agency of the other. To believe it is to ask too much from a notion that is widely used to describe a more modest understanding and sharing of the other’s mental states. Thus, Goldie’s criticisms can be tackled by refusing to expect from empathy outcomes and effects that it simply cannot deliver.