Anti-empathism: Jesse Prinz

Abstract

In this chapter, the criticisms advanced to empathy by Jesse Prinz are taken into account and carefully analysed. In particular the author attempts to trace the structure Prinz gave to his criticisms by dividing them into three: those concerning moral judgement, those concerning moral development, and those concerning moral motivation and conduct. This structure will be of fundamental importance to the work, because it is by tracing this structure that the author will respond to the criticism of empathy and simultaneously present his theories.

What I call ‘anti-empathism’, that is, the tendency by scholars dealing with empathy to criticise this phenomenon either at the ontological and epistemological level (by contesting that it is really possible to feel what another feels)¹ or at the ethical/moral level (by questioning the role it is supposed to play in the moral sphere or as the founding element of an ethics),² is very recent and, for this reason, it cannot yet count on many supporters. Nonetheless, as usually happens with young movements advancing a provocative thesis, the number of academics advocating this view is rapidly rising.

In what follows, I will analyse the perspective on empathy of the most representative among the ‘ethical’ anti-empathists, namely, Jesse Prinz and Paul Bloom (a perspective which is very often complementary, when not overlapping), and then criticise it in order to make room for an alternative point of view on the matter. My intention is to reject most of the criticisms made of empathy by Prinz and Bloom whilst at the same time distancing myself from empathy-enthusiasts (whom I shall label ‘pro-empathists’). As I have said more than once, my intention is to present a third alternative way of thinking about the moral dimension of empathy, which, although perhaps less provocative, and for this reason also less glamorous than the two maximalist positions, will on the whole be considered—so I hope—more reasonable and plausible.

Jesse Prinz developed his anti-empathic position in two articles that appeared in 2011 and that carry the very telling names: Is Empathy Necessary for Morality? and, naturally, Against Empathy. In the first of the two articles, Prinz divides morality into three different fields, namely, moral judgement, moral development/education, and moral conduct, and he then questions the role that empathy can play in all these three areas. In the second paper, the American philosopher widens his view by being, if possible, even more critical of the real potentialities of empathy. To the question if empathy can be a precondition of moral judgement (as the theory of Hume and those of contemporary philosophers like Michael Slote seem to imply), Prinz answers that:

empathy is not a component, a necessary cause, a reliable epistemic guide, a foundation for justification, or the motivating force behind our moral judgements. In fact, empathy is prone to biases that render it potentially harmful.³

In the next chapter we are going to examine more closely what arguments he used to justify this strong conclusion.

1 Unnecessity of Empathy for Moral Judgement

Let us start our analysis with the first article. Here Prinz describes empathy as ‘a kind of vicarious emotion: it is feeling what one takes another person to be feeling. And the “taking” here can be a matter of automatic contagion or the result of a complicated exercise of the imagination.’⁴ Two crucial characteristics in this description should be noted: first of all, empathy is defined as feeling what one takes another person to be feeling. This assertion is interesting, since for Prinz it seems (although he is a little unclear on this issue) that we can empathise even when we fail in the attribution of mental states. In fact, it is not required to ‘feel what another feels’, or ‘to feel something that is congruent, similar, or similarly valenced to what the other feels’, but to ‘feel what one takes another to be feeling’. This definition appears already rather problematic to me, since it turns empathy in a process founded only on belief and not (also) on acknowledgement. It seems, in other words, that if you believe that a certain subject ‘A’ is feeling ‘x’ and you feel ‘x’, too, then you are empathising with ‘A’. I think that such an assumption would take empathy to be too arbitrary, but let us set this matter aside momentarily and continue with the analysis of Prinz’s position.

Secondly, emotional contagion is for him a case of empathy, too. In this sense, he does not make the difference that I made between emotional contagion (which is per se not a case of empathy), low-level empathy (which is a kind of direct, unmediated, and conceptually poor empathy), and high-level empathy (which is empathy obtained thanks to what Prinz calls: ‘a complicated exercise of imagination’). On the contrary, his definition seems to encompass all forms of attributions of mental states by giving them all the label of ‘empathy’. This is, of course, a strategy that is totally licit to adopt. I contend, however, that my systematisation is more fruitful and, on the whole, more fitting than his. In fact, it can circumvent some important objections, it is productive in academic discussion on the topic of empathy, and it is even nearer to the usual concept employed by laypeople when discussing this phenomenon.

Once Prinz has chosen the definition of empathy with which he is going to work in the paper, he asks whether empathy can be necessary for moral judgement, moral development, and moral conduct. Finally, he also takes into account the question of whether empathy plays and/or should play an integral role for morality. As we are going to see, the answer to all these questions is negative.

Good sentimentalist as he is, Prinz starts his analysis with a quote from the Treatise of David Hume. As Prinz interprets him, the Scottish philosopher sustained the following theory: that virtuous actions provoke satisfaction in the people who receive them, that vicious actions produce uneasiness instead, and that, thanks to our empathy (what Hume calls ‘sympathy’) for the recipients, we can get a feeling of approbation in the first case and of disapprobation in the second. These two feelings, respectively, of approbation and disapprobation, constitute our judgements that a given action is morally right or wrong. Therefore, empathy seems to be, in Humean ethics, the very precondition for the formation of any of our moral judgements. Now, the question of whether Hume should be read exactly in this way is still open to dispute,⁵ but in any case, this seems to be a legitimate view of the possible role played by empathy in morality and specifically in grounding our moral judgements. Is it, nonetheless, also correct? Prinz contends that it is not, and in order to make this point, he resorts to six different arguments.

The first argument Prinz uses to criticise the role empathy can play within morality is one in which, as he himself asserts, ‘deontological considerations overrule utilitarian principles’.⁶ In particular, he imagines that one might judge that it is bad to kill an innocent person even if their organs could be used to save five others who desperately need transplants, and reaching this judgement is a proof, for him, that empathy does not play a role in these kinds of considerations, as one would feel cumulatively more empathy for the five needy people than for one single healthy person. But this is a hasty (and wrong) conclusion, for two different reasons: the first is that Prinz gives the impression of arguing that empathy has a natural inclination to favour utilitarian/consequential kinds of considerations as opposed to Kantian/deontological ones; the second reason is that, contrary to what Prinz believes and to what prima facie might seem, we actually feel more empathy for the one single person than for the other sick people. I will start with the analysis of the second reason.

The dilemma mentioned by Prinz is very similar to the famous trolley problem, particularly in its ‘fat man’ variation: would you be ready to push one fat man onto the track, in order to stop the trolley that would otherwise kill five people? The story is well-known⁷: there is one trolley going downhill. The brakes have failed and in front of it there are two tracks: the one which it is on leads to five people who would all be killed by the impact, whereas on the other track there is only one person. In the original version of the dilemma, it is asked whether one would be ready to pull the lever and switch the track, so that the trolley will hit one person instead of five. In this scenario, the majority of people would be ready to sacrifice one life in order to save five others. But, interestingly enough, when the only way to stop the train is to push a fat man onto the track, the vast majority of people disapprove.⁸ In other words, the majority of people have a strong tendency to act utilitaristically/consequentialistically in the first case, but then seem to apply a deontological principle in the second case, thereby showing a surprising (ethical) inconsistency. Various possible explanations have been fashioned to illuminate and justify this change of behaviour. After all, from a logical point of view, there is no apparent difference: one man should be sacrificed in order to save five others: why is almost every person ready to do it in the first case, but not in the second case?

One possible explanation appeals to a difference in the (moral) intention: in the first situation, the person who pulls the switch does not have the intention to harm anyone, therefore, the killing of one man is seen as a kind of side effect. In the second situation, on the contrary, killing one individual is an integral part of the more general intention to save five lives. In other words, the killing of one person in this case cannot be seen as a side effect, but as an exemplification of the principle that the end justifies the means, a principle that not everyone would be prepared to apply.⁹

Another possible explanation could be founded on the justification principle connected with the more passive or more active role one plays as agent in the situation. In the first case, the scenario is so construed, that someone is going to die regardless, so my choice as ‘lever-puller’ is ‘just’ to decide who, and my moral reasoning would take me, normally,¹⁰ to prefer the ‘lesser evil’. To give an analogy with an example that Philippa Foot makes in her article of 1967, the circumstance which I am in as ‘lever-puller’ is the same as that of a pilot whose airplane is about to crash: if there is no likelihood of directing the plane to a totally unpopulated area, they will at least choose the least populated one they are able to find (e.g. a park with a few people strolling in it rather than a school).

The ‘fat man scenario’, on the contrary, introduces a variable that is very difficult to deal with (morally). At the end of the day, when I watch the trolley heading towards the five workers on the track, the fat man is just a bystander as I am. What authorises me to kill him (because this is a case of killing and not just of letting someone die) in order to save the workers? It seems that I may have a justification good enough to let one person die (as opposed to five people), but that I do not have one good enough to kill one person.

Both the possible solutions presented are unsatisfactory in several respects. The supposed inconsistency between an utilitaristic and a deontological kind of moral reasoning can hardly stem from a change in intention. After all, the intention of a person who chooses to pull the lever in the first case, but to abstain from pushing a fat man onto the track in the second, remains the same: hurting as few people as possible and saving as many people as possible. If we can talk about ‘side effects’ in the first case, then why do not do it in the second? In other words, if my thoughts sound like this in the first instance: ‘I just want to save the five people and avoid a slaughter by switching the track. That there is a person on the other track who will die is merely a side effect/collateral damage’, then why cannot they sound like this in the second: ‘I just want to save the five people and avoid a slaughter. If there were any other solution to stop that trolley than pushing the fat man standing next to me on the track I would try it, but there isn’t. So, I push him.’? As you see, my intention continues to be the same, the number of lives lost would be the same, and analogous would also be my interference with the other’s subject agency/free will/right to decide (in both cases I condemn a man to death without his consent). Nevertheless, something stops me from being so consistent with my goal. What is that something?

Even using the distinction between ‘killing’ and ‘letting die’, or, put in another way, between an instantiation of ‘taking a life’ (fat man scenario) and ‘refraining from saving a life’ (man on the track scenario) as the fundamental variable which should save the ethical consistency of a person who chooses to sacrifice one life in the first situation, but refrains to do so in the second, appears to be a weak move if it remains founded in merely agential terms. My suggestion is, on the contrary, that in the fat man case, emotions make their voice heard so loudly that we simply cannot ignore them.

In the first case, I just have to pull a lever. This is a simple, neutral movement. It will have serious repercussions, and I know this when I pull it, but in some sense, I also have a certain physical and emotional distance from what happens. This permits me to treat the situation with a sort of ‘cold logic’ made by pros and cons. The lives at stake become countable, and I can adopt a more consequential perspective. In the second case the distance I had in the first situation totally disappears. My claim is that there are degrees of involvement (and of related responsibility and sense of guilt) in the things we do to other people, and these degrees reflect the levels of our emotional proximity. That is why we find easier to let someone die by pulling a lever, than to kill them ourselves. And that is why, if forced to kill the fat man before pushing him on the track (I am imagining another possible variation to this mental experiment) I am sure everyone would prefer to shoot him in the head with a gun (maybe while he is not looking at us, so that we can avoid watching his face and so that he will not be terrorised at the thought of his imminent death) than to stab him vis à vis with a knife. In the second case, in fact, we would have an inevitably strong empathic involvement: we would feel his shock, his terror, his pain when stabbed. We might also project ourselves in his situation and thinking how dreadful such an experience would be. On the other hand, in the first case, our empathy would be less elicited. Hence, in this experiment, empathy is never out of the picture: it is simply partially¹¹ dormant in the ‘lever’ case and then it becomes more and more active the more my emotional involvement with the context increases.

Now that we found this directly proportional relationship between emotional involvement/personal proximity on the one side and empathy on the other, we can understand why I have asserted that it would be wrong to say, as Prinz does, that we feel cumulatively more empathy for the five people in need than for the one healthy person: this would be so only in the case in which the one single person was also dying. In that instance, all things being equal (i.e. being the one and the five in the same conditions), we would probably feel more empathy for the five. But the case so as it is described by Prinz is more similar to the fat man scenario, where we have to kill an innocent person in order to save five others. The thought of killing this man in order to harvest his organs and transplant them in other patients would horrify us and hold us back. Of course, we would be empathically sad for the five people, but this sadness would not be great enough to overcome the empathic emotions of horror, disgust, and guilt we would feel in contemplating the idea of killing an innocent person just for their organs.¹² By empathising with this individual, we would soon and easily discover that we could not possibly want something so terrible to happen to anyone: we would not at all wish to suffer something similar, so why do it to others?¹³

I think I have argued enough over what concerns the conclusion by Prinz in believing that we feel more cumulative empathy for the five people in need than for the one healthy person. Now it is time to briefly move to the question of whether empathy always favours consequential principles as opposed to deontological ones. As it can be seen quite clearly in the arguments hitherto made, it does not. Depending on the situation, on the moral convictions of the empathiser, and on the level of emotional involvement, empathy will foster deontological or consequential principles alternatively. Hence, in the country of morals, empathy is, so to say, an essentially stateless citizen: it does not belong to and it does not favour any specific kind of ethics. Or, to put it another way, empathy is morally neutral. As some readers may think that by stating the fundamental moral neutrality of empathy I wish to imply that empathy has no influence in moral matters¹⁴ (which would run against what both anti-empathists and pro-empathists, with diametrical opposite conclusions, sustain), a brief clarification at this stage is in order.

With the expression ‘morally neutral’ I do not wish to intend that empathy is completely ineffective in the moral sphere, but only that its effect may vary, from positive to negative, depending on when, how, and with whom one chooses to empathise. Of course, this will also depend on how the empathising person acts on the basis of the empathised emotion. Therefore, feeling what another feels at a certain moment is not, per se, an intrinsic good or bad act. It is just an act. And, as it happens with any kind of act, in order to judge it as ‘moral’ or ‘immoral’, we need to analyse the action in the light of the moral principles we choose to follow. Thus, typically, a utilitarian will deem empathy as morally good if it leads to the increase of the general happiness/well-being of the greatest number of people. This also means that the moral weight of empathy varies much in the same way in which the moral value of an action can be deemed as morally good or morally bad, depending on a series of conditions which include the intention of the agent, the context in which the action occurred, and the way in which it was enacted. For example, saving the life of a person drowning in a river is usually judged as a moral thing to do, but what if that person is or will be responsible for the deaths of many more people? From a utilitarian point of view, letting that person drown appears the moral thing to do.

In the same way it seems at a first glance that empathy is subjected to the same variations within the moral spectrum. Hence, empathising with the pacifist feelings of altruism, inclusion, and solidarity of a Gandhi-like persona seems prima facie a moral thing to do, whereas empathising with the lust for blood of a Jack the Ripper seems not. However, the reality is more complex: further in the book we will see that what appears as a simple truism (it is good to empathise with good people and bad to empathise with bad ones) is a misleading oversimplification. We will also observe that empathy’s effects for morality in general are, on the whole, positive.

I hope that all of these considerations have made clear why the first objection of Prinz to empathy ought to be rejected. The second (very brief) argument he makes about empathy is that, from a Rawlsian veil of ignorance, we could consider the distribution of resources to the needy as moral, because we might be needy in the future. In this case, our judgement would drive from a concern for the self and not for the needy. Here, I think that the American philosopher is probably conflating empathy with sympathy. What he demonstrated with this example is not that we do not need empathy to reach a Rawlsian kind of moral judgement, but that we do not need sympathy for it. Insofar as empathy means feeling what another feels, it can be compatible with a certain degree of egoism. In fact, it can even be argued that at least a basic form of empathy is necessary for the Rawlsian mechanism to work well: by walking in the shoes of the needy I understand that I would not want to live in those conditions, therefore, my personal interest for my future well-being would lead me to distribute my resources more equitably. Of course, it is not the aim of this book to argue for the necessity of empathy within the Rawlsian theory of justice, but it is imperative to point out that the Rawlsian idea of a veil of ignorance does not ipso facto divest the role empathy can play in moral judgement.

Now that we have examined two of the minor (at least in terms of length of the argumentation) objections to empathy made by Prinz, it is time to move on to the analysis of the most significant ones. As it would be quite laborious to proceed by presenting a criticism and then immediately trying to reply to it, I will instead summarise the critiques of Prinz and then answer all of them at the one time. This will help develop a more homogeneous argument in favour not of empathy per se, but of another way of looking at this phenomenon.¹⁵

As mentioned previously, the aim of Prinz is to show that empathy is unnecessary for all three of the main moral dimensions of morality, which are moral judgement, moral development, and moral conduct. If empathy were necessary for moral judgement, then we could not be able to express a moral verdict without making use of empathy. However, it seems that we are capable of judging something as moral or immoral without resorting to empathy. For instance, we can judge that we have been wronged, or, to put it another way, that someone treated us immorally, without the need to empathise with ourselves.¹⁶ Furthermore, there are cases, such as bootlegging CDs or DVDs or evading taxes, that are commonly seen as morally wrong, even though there is no salient victim to empathise with. There is also a series of transgressions which are generally judged as immoral without thereby having grounded this judgement in empathy or compassion. Jesse Prinz offers the following list: ‘necrophilia, consensual sibling incest, destruction of (unpopulated) places in the environment, or desecration of a grave of someone who has no surviving relative.’¹⁷ In all these cases, empathy can hardly be the cause of our moral disapprobation, for we have no one to empathise with.

If we consider all these critical points, empathy turns out to be contingent upon moral judgement, since we can express moral judgements without having to rely on empathy. However, one might object, empathy turns out not to be necessary in these cases, only because they are all cases in which others are not really involved. Specifically, if we take empathy to be fundamental only for the regulation of moral behaviour between two or more individuals, then we may discover a necessary moral role for it. In other words, empathy might be necessary for a special class of moral judgements: those made in a ‘social’ situation between ‘real people’, as opposed to ‘juridical people’, like the government, or material goods, like the environment, are examples of these. However, even in this context there are some issues. Jesse Prinz has an interesting way of illustrating this. He imagines the following state of affairs: suppose that I come to eat the last delicious cookie from a packet I have been sharing with a friend of mine. After doing it, I feel a pang of guilt. Is this feeling of guilt coming from empathy for my friend? It does not seem the case. In order to feel guilty, I just have to construe my action as greedy. Quoting Prinz on this issue:

Morally significant actions can be recognized without empathy, even if those actions are ones that involve harm. We need not reflect on the harm to see that the action is bad. Perhaps you are delighted that I ate the last cookie. I recognize that, empathetically, and I still feel guilty; I still think that I should have offered the cookie to you.¹⁸

In other words, Prinz is persuaded that on any given occasion, our judgement that something is morally good or bad derives from a sentiment which leads to feeling the appropriate emotional response. Sentiments are in fact for Prinz ‘dispositions to have emotions’.¹⁹ Hence, if I have a sentiment of disapprobation towards greed, I will feel anger or scorn when I see someone acting greedily, and guilt and/or shame when I myself have performed a greedy action.

After having made these points regarding the unnecessity of empathy for moral judgement, Prinz asks himself whether empathy could be diachronically necessary for morality. In other words, granted that we do not need to resort to empathy every time we express a moral judgement, is empathy essential for the acquisition of the ability to judge morally? In short, is empathy necessary for moral development/education?

2 Unnecessity of Empathy for Moral Development

This view seems prima facie appealing. After all, when we think of what morality essentially involves, we think of the principled regulation of our behaviour in relation to others. Therefore, empathy, which seems to many psychologists to have a strict connection with moral behaviour,²⁰ might turn out to be central. The well-known, insightful, and very influential book of 2000 by Martin Hoffman²¹ is committed exactly to proving this intuition. He claims that if a child were not empathic, they might be indifferent as to how their actions affect other people, and as a consequence, they might not come to understand and to fully appreciate when and why their actions are morally wrong. But is this really the case? Is empathy really a precondition to developing a capacity for moral judgement and, in general, for a sense of morality? There are many studies showing how children engage in empathic reasoning when making moral judgements,²² but unfortunately none of these studies was able to prove that there is a relation of causation between empathy and moral development and not only a mere correlation. But this is exactly what is needed if we want to demonstrate that empathy is at the base of our sense of morality. In order to prove this intuition, more and more researchers have shifted their focus to the study of pathological populations taken to be completely devoid of empathy: psychopaths. These seem, in fact, to lack both empathy and compassion and, if the result were that their well-known amoral or even immoral behaviour was due to this deficit, then we would have good reason to conclude that compassion and empathy are necessary prerequisites for morality. What is needed is indeed evidence of the fact that empathy can produce moral behaviour and not that empathy and moral behaviour are merely correlated. However, Prinz believes that psychopathy does not seem (at least at the present stage) to offer this kind of evidence. In fact, even if a plethora of scholars see in a lacking or defective empathy the central characteristic for explaining all the amoral features typical of psychopathy, this view can be challenged.

Now, what does it mean to be a psychopath? Robert Hare famously takes note of the following characteristics: glibness and/or superficial charm, grandiose sense of self-worth, deceitfulness, manipulativeness, lack of remorse or guilt, shallow affect, callousness, irresponsibility, poor behavioural control, lack of realistic, long-term goals, and impulsivity.²³ Cleckley offers another very similar, classical description of the typical psychopath:

Vexation, spite, quick and labile flashes of quasi-affection, peevish resentment, shallow moods of self-pity, puerile attitudes of vanity, and absurd and showy poses of indignation are all within his emotional scale and freely sounded as the circumstances of life play upon him. But mature, wholehearted anger, true or consistent indignation, honest, solid grief, sustaining pride, deep joy, and genuine despair are reactions not likely to be found within this scale.²⁴

Furthermore, psychophysiological studies have revealed that psychopaths generally show a very pronounced lack of responsiveness to the distress of others as well as a lack of fear, shock, or sadness in disturbing situations involving physical or psychological harm to other people.²⁵ Psychopaths also have difficulty in distinguishing between different types of violations and to correctly evaluate their different degrees of severity. For instance, they cannot distinguish a difference between violations of moral and those of merely conventional rules.²⁶ All these various features are often explained by a common denominator, that is, the lack of what psychologists call ‘Violence Inhibition Mechanism’ or ‘VIM’.²⁷ The explanation of the term is approximately as follows: normally developing children have an innate proclivity to empathise with observed distress, so if one child causes another child to cry, the child responsible for the harm will catch the observed emotion and feel badly. These unpleasant feelings will serve as an inhibition signal which will lead the child to cease the actions causing the distress and even drive them to associate bad feelings with that sort of action in the future.²⁸ So, following the theory, violence inhibition is mediated by empathic distress that is then associated by children with moral rules, as opposed to conventional rules, the violation of which does not involve empathic distress. Hence, normally developing children can distinguish moral rules from conventional ones because the former are the only ones to be empathically grounded. Empathy constitutes therefore the emotional basis of moral rules. Psychopathic subjects lack this system of inhibition because they lack empathy for others, that is, they don’t feel any negative feelings when hurting someone else. This is—for many psychologists, including James Blair—the cause of their amorality.

At first glance, this model seems to be very attractive. However, Jesse Prinz is of a different opinion. It is not the lack of empathy that causes an individual to be a psychopath, but rather a severe deficit in moral emotions altogether. Prinz does not want to deny that empathy is absent in psychopathic individuals, nevertheless, the point is to show that this absence is not the primary cause for psychopathic amorality, but the consequence of another, more primordial, more fundamental lack. If we go back to the previously cited description of typical psychopathic individuals, we find the following features: ‘lack of remorse or guilt, shallow affect, callousness, irresponsibility, poor behavioral control, […] and impulsivity’. Among all these various features, there are some characteristics which can be taken as primordial and which lead psychopaths to other kinds of deficits, including the lack of empathy. These characteristics are poor behavioural control, impulsivity, and, above all, shallow affect. Prinz’s suggestion is that the constitutive impossibility for psychopaths to experience mature, wholehearted emotions brings them to be callous, irresponsible, unempathetic, and, at the end of the day, amoral. After all, empathy as we have defined it (and as Prinz describes it, too) is the capacity to experience the emotions of others, and if one is unable to be deeply moved even by their own emotions, they will remain all the more indifferent to the emotions of others. Put in another way, psychopaths do not feel empathy because they cannot feel any kind of emotion in a wholehearted manner. Being emotionally almost dead, they are also not concerned about others.

Thus, considering the status quo of the research on empathy and psychopathy so far, Prinz concludes that we cannot assert with certainty that empathy is necessary for moral development. On the contrary, it seems safe to affirm that in normally developing children with a normal emotionality, methods of moral education founded in punishments, love withdrawal, positive feedback, as well as the offering of positive role models, among others, appear to be both necessary and sufficient for the formation of a mature morality. However, there is another dimension of morality that could still be strictly connected to empathy, and it is that concerning moral conduct or, in other words, moral motivation.²⁹

3 Unnecessity of Empathy for Moral Motivation and Conduct

Prima facie, the connection between empathy and moral motivation appears quite natural: if I empathise with someone suffering, I might be motivated by vicarious sadness for this person to do something in order to help them. However, Prinz challenges this view: for him, empathy has a contingent connection with moral behaviour and when it effectively motivates, it does it as a motivational force among others. These other forces are simply emotions, which are very often not based on empathy. Using the words of Jesse Prinz:

[…] moral judgments have an emotional basis. Token moral judgments contain emotions such anger, disgust, guilt, and shame. Emotions are motivating states, and each of these moral emotions has a behavioral profile. Anger promotes aggression, disgust promotes withdrawal, guilt promotes reparation, and shame promotes self-concealment.³⁰

For Prinz, ‘normal’ emotions do all the job empathy is taken to be doing, they provide us with all the necessary motivation for carrying out certain actions and avoiding others and offer us explanations for our behaviour: ‘I ran away, because I was scared’; ‘She yelled at him, because she was angry’; ‘He bought him a beer, because he was happy to see him’, for example. However, there is more to be considered. Asserting that moral judgements ‘contain emotions’ means that they are intrinsically motivating: indeed, we have a natural proclivity to avoid negative, valenced emotions and to pursuit positive ones. Thus, for Prinz, if we anticipate that an action will make us feel guilty, we will try to avoid that, whereas if we believe that by doing another kind of action we will feel pride or gratification, we will be motivated to carry out that action. Empathy appears to be, in this regard, a useless complication to a sentimentalist framework that works perfectly well the way it can be conceived.

Prinz then goes on with the citing of several psychological studies that demonstrate, in his words: ‘[…] that empathy is not a major player when it comes to moral motivation. Its contribution is negligible in children, modest in adults, and non-existent when costs are significant.’³¹ However, I will not analyse these examples, for many good reasons: the first is that they barely constitute an argument against empathy, for Prinz himself does not discuss the results of these different studies at length, instead, he just mentions them very briefly—so it seems—to drive the point home.³² Presenting good counter-argumentations for every single one of these results would require a vast amount of space and shift the focus of my present book and would also overstep my task as a philosopher. Furthermore, I think that the mere citing of psychological studies without an appropriate problematisation would easily lead to a biased activity of cherry-picking. There are a lot of psychological studies about empathy, and the results among them are far from being uniform: no matter how many pieces of (psychological) research someone can quote to show that empathy is unnecessary for morality: there will be many others proving exactly the opposite.³³ Hence, I prefer to concentrate my attention on other weightier arguments.

These particular arguments can be found in a rather summarised way at the end of Prinz’s article. There, besides repeating some claims for which he has presented in the course of the paper, Prinz makes the following contentions: empathy is prone to biases and parochialism (e.g. cuteness effect and preferential treatment), it can be easily manipulated, it can motivate harmful actions, and it interferes negatively with the ends of morality. As we shall see later, these objections will be further developed by Paul Bloom.

To get the full picture of Prinz’s hostility to the supposed moral dimension of empathy, some other notable criticisms contained in another article should be mentioned. In Against Empathy, the American philosopher contends, contra Hume and contra Michael Slote,³⁴ that empathy is not a precondition for (moral) approbation or disapprobation, or in other words, for moral judgement. This time, however, his arguments are even starker. Since my argumentation will sustain a thesis that is in many aspects opposed to that of Prinz, I cannot forgo mentioning—for the sake of the exposition—his critiques; however, I will offer a summarised version of those in order not to weigh down my own presentation.

4 Other Intrinsic Shortcomings of Empathy

The first polemical targets of Jesse Prinz are, as already said, Hume and Slote. Hume can be taken to support what Prinz calls a ‘patient empathy constitution thesis’, whereas Slote represents an ‘agent empathy constitution thesis’.³⁵ The former represents the empathy I feel for the recipients of a good action that constitutes my approbation for it. For example, if I see an old woman asking for help and then I see a young man helping her out (say, crossing the street), my empathy for the positive feelings of the old woman (happiness, relief, pleasure, or others) constitutes my moral approval of the action. Conversely, if I see this same man ignoring this old lady, thanks to empathy for the feelings of sadness and discomfort of the woman, I will disapprove of the action.

Prinz criticises this thesis by asserting that it is quite odd to talk about empathy in these cases. In fact, the emotion I feel when observing another subject carrying out an action that I approve (or disapprove) of is hardly the same one the patient of the action feels. If A gives money to B, B may feel gratitude towards A, but I will feel admiration and not gratitude for A. In much the same way, if A robs B, B might feel a sense of fear, despair, and vulnerability, but I will feel anger or outrage. There are even cases in which the victim of an action I disapprove of does not feel anything at all. For instance, if A does not notice that she was robbed, she will not feel any negative emotion, but I, as an observer, will feel anger or outrage towards B regardless.

The ‘agent empathy constitution thesis’ seems prima facie more promising. Here, in fact, I empathise with the doer of a certain action, particularly, with their intentions. Therefore, irrespective of what the receiver of an action may feel, I will approve of an action if I positively empathise with its agent and disapprove of it in the case where I negatively empathise with them. However, Prinz is not persuaded. In fact, there can be cases in which I am perfectly able to empathise with the agent, but I still disapprove of their action (Prinz makes the rather bizarre example of a recovering paedophile, who, while able to empathise with another paedophile, would condemn an eventual action of paedophilia on the part of the other). Furthermore, moral disapprobation appears to be constituted by feelings of blame, or something similar, not by a lack of empathy: if lack of empathy were directly connected to moral disapprobation, we would disapprove of any person towards whom we did not feel empathy, and it is easy to see how this would be nonsensical.

Now, before continuing with the criticisms to empathy, I cannot refrain from noticing and stressing the fact that there seems to be an error in Prinz’s reading of Slote. In fact, Michael Slote in his book On Moral Sentimentalism never claims that moral approbation is constituted by empathy and that moral disapprobation is originated by a lack of it; his vision—if I am right in my interpretation—is actually subtler. The central claim of the book is that, thanks to empathy, one becomes receptive towards the emotions, the intentions, and in general the mental states which lay behind an action and drive an agent to carry this out. Therefore, if, after having empathised with the agent, I discover myself to be ‘warmed’³⁶ by their action towards another subject, then I approve of it; on the contrary, if, after having empathised with this agent, I find myself to be ‘chilled’ by their action, then I disapprove of it. Reducing Slote’s position to the kind of simplistic approach Prinz is describing appears, therefore, a totally mistaken interpretation. Of course, Slote’s theory can still be criticised for other aspects, but it should, at least, be properly conceived, otherwise one runs the risk of attacking a mere strawman.

Prinz then goes on with the question whether empathy may be a causal precondition for moral approbation/disapprobation, or, in other words, whether empathy is at the basis of all our moral judgements. As we have already seen in the other article, Prinz’s answer is negative: our moral responses are grounded on action-types. If I classify an action as an instance of stealing, murdering, terrorism, tyranny, for example, then this provides me with enough motivation to feel moral outrage or ire. In the words of Prinz:

The very possibility of thick concepts depends on a direct link between a form of behavior (taking property, taking life, etc.) and a negative response. We are conditioned to immediately despise these action-types without having to contemplate the suffering they cause.³⁷

Prinz’s position is very simple: we were raised to disapprove of some actions and to approve of others. This could happen as a result of a special kind of associative learning. We were, in other words, taught to associate negative emotional responses (like disgust, outrage, contempt) with actions deserving moral blame and, conversely, positive emotional responses (such as admiration or appreciation) with actions deserving moral praise. Since this kind of teaching does not require empathy in order to work, nor is empathy needed after the inculcation of these emotional associations (because then they become a type of automatic mechanism), then we cannot but consider it totally worthy of dismissal.

After this critique, Prinz turns his attention, once again, to the analysis of empathy and moral development. This time, however, the starting point is different: if in the first article the question about the necessity of empathy for moral development was investigated using solely the literature on psychopaths, here Prinz focuses his attention—along with psychopaths—on the issue of moral education. Is empathy necessary to raise our children? Is it the fundamental tool to acquire a sense of morality? Unsurprisingly, Prinz is inclined to think that it is not. If we look at the techniques by which we educate our children (such as offering of role models, punishments, caregiving, and, conversely, love withdrawal) we will soon notice—so maintains Prinz—that empathy does not play a role. Thus, for instance, punishment instils fear, love withdrawal instils anguish, and ostracism instils shame. Ultimately, all these emotions will drive children to avoid certain actions and to carry out others. What is more, given how skilled children are with imitation, they will also imitate the outrage of their parents towards a bad action in the case where another child is caught doing the same. Hence, traditional ways of educating our children are all, in a sense, ‘empathy-free’.³⁸

Even faced with all these criticisms, a supporter of empathy might still contend that it is undeniable that empathy has an essential role to play at least at the epistemic level. Without empathy we would be blind to the emotions of other people, and emotional impact is a feature we have to consider when dealing with other subjects. Suppose A says something that hurts B. A’s capacity to recognise B’s distress is what can bring A to judge that there was something bad in her choice of words and eventually to regret them. Nevertheless, Prinz stands firm: this supposed epistemic role is not only merely contingent, but even epistemically unreliable, for many good reasons:

• 1. We do not always need to feel the emotions of others to correctly attribute emotions to them.

• 2. Affective empathy can lead to (vicarious) personal distress, that is, if the suffering of another is very intense, we might just withdraw and become avoidant, instead of helping out.

• 3. Empathy inherently suffers from what Hoffman (2000) called ‘similarity bias’ and ‘here and now bias’, that is, it works at best with people who are similar and near to us. Empathising with people of another ethnic group and religion, who live in a distant, foreign land is much harder than empathising with, say, a member of our family.

• 4. The fact that someone is suffering because of what we have done or said is per se not a sufficient element to conclude that what we have done or said is morally wrong. For instance, someone might be offended by our words because deep inside they know we are right or because they are a manipulative kind of person who tries to make us feel guilty.

On the whole, Jesse Prinz is adamant that: ‘If we measure the moral merit of an action by quantifying harm, rather than empathy, we may allocate blame in a way that better tracks our considered standards of wrongness.’³⁹

Empathy seems also ill-suited as a normative precondition for moral judgement, that is, as fundamental for moral justification. It is true that we can appeal to empathy to justify a choice we have made, or that we can empathically imagine the consequence of our actions before morally approving or disapproving of them, but even in those cases, Prinz is of the opinion that principles of justice should be preferred. The philosopher states very clearly (thereby revealing his consequentialist convictions about ethics) that it is the harm caused by a given act that makes that act bad and, conversely, that it is the pleasure brought by an action that makes that action good. Therefore, empathy cannot (and ought not) rise to the status of a normative guiding principle.

Finally, Prinz poses the question of whether empathy should be deemed necessary for moral motivation. Much like in the first article, even here he regards the hypothesis as problematic: as we have seen above, other emotions, for instance, guilt, anger, and shame, are better suited to motivating us. Moreover, we often have moral values that lead us to avoid certain actions and to carry out other ones without further need for empathy.

As if all these criticisms were not enough, Prinz even decided to dedicate another entire chapter (carrying the very telling title of ‘The dark side of empathy’) of his article to what he believes are serious shortcomings of empathy. This time the various flaws are not neatly structured, but presented as supplementary theses to the inadequacy of empathy as a moral principle or skill. Empathy—asserts Prinz—makes us sensitive to secondary qualities and blurs our vision when it comes to the central moral elements. Thus, for instance, it has been observed that jurors are inclined to hand down harsher sentences when the victims are manifestly emotional and lighter ones when defendants show regret.⁴⁰ This is troublesome for Prinz, since what should really matter in such cases is whether the defendants are truly responsible and whether victims were really harmed.

Nevertheless, we may wonder whether empathy must really be this partial, biased, unfair ability Prinz is describing. Is it not possible to adopt a more general, neutral perspective thanks to empathy? Does the capacity to step into the other’s shoes and to see the world from their perspective not allow for that? This was famously the position of David Hume: using empathy to acquire what he called ‘the general point of view’.⁴¹ However, the worst enemy of this general viewpoint is, for Prinz, empathy itself. Trying to reach this perspective by making use of empathy is like hoping to extinguish a fire by pouring gasoline on it instead of water. Empathy, in fact, tends to focus on the individual rather than on the multitude, and on persons, rather than on systematic problems.⁴² Prinz is clear: no objective principle can stem from empathy. In his words: ‘With empathy, we ignore the forest fire, while watering a smoldering tree’, and even more bluntly: ‘The general point of view is not a bad idea, but its greatest hope may lie in the extirpation of empathy.’⁴³

In these two quotes is contained in nuce the whole Prinzian conception about empathy: whereas ‘normal’ emotions can be biased, empathy⁴⁴ is intrinsically, by its proper nature, biased. In fact, empathy is fundamentally a dyadic emotion, regulating the responses between two individuals involved in some kind of personal relationship. On the contrary, emotions, such as anger and guilt, are wider and more inclusive in their scope. Moreover, Prinz goes so far as to say that empathy is often the primary cause of the biases affecting other emotions.⁴⁵ For instance, if I fail to get angry at the injustices committed in another part of the world, it may be for the reason that I have chosen to rely too much on empathy or, in other words, because I have focused on the victims (that are not part of my in-group, that come from another culture and from a far-off foreign land),⁴⁶ and not on the crime itself or on the harm it has provoked.

In the very last part of his paper, Prinz examines another kind of fellow-feeling: concern, with the intention to see whether this could be a better alternative to empathy. He describes concern⁴⁷ as a negative feeling which arises from the contemplation of somebody’s plight or, put in another way, from the recognition that someone is in need.⁴⁸ Whilst this particular kind of feeling has surely some advantages compared to empathy, it is—according to Prinz—neither necessary nor sufficient for moral judgement. It is not necessary, because we normally become concerned for someone (and we feel for them) when we believe something bad has happened to that someone and not vice versa. It is not sufficient, because we can feel concern in circumstances that have nothing to do with morality, as when we are concerned about the health of a friend or about the damages made by a natural calamity. Hence, at the end of the day, even concern’s contribution to morals is negligible.

Jesse Prinz concludes his very critical article with the observation that research on fellow-feelings in general diverts the attention of scholars (and lay people, too) from more profitable fields of investigation, such as the study of moral emotions: anger, disgust, contempt, guilt, for example. Inquiry on empathy, as a consequence, has to stop.