‘Without God’, said Dostoevsky’s Ivan Karamazov, ‘everything is permitted.’ For some psychologists, says Paul Bloom, ‘without God’ should be replaced by ‘without empathy’. What else, in the absence of religious faith, and given our postmodern scepticism about how reasonable ‘reason’ is, could curb our wild, exorbitant egos? Even if reason is impartial, we increasingly agree with the 18th-century philosopher David Hume that it is ‘inert’, too ‘impotent’ to spur moral behaviour. Morality only works, Hume claimed, because, in the same way that plucking one of the tautened strings of a musical instrument makes all the others vibrate, humans are naturally primed to be stirred by one another’s suffering. He called this proclivity ‘sympathy’. The term ‘empathy’ was coined by 19th-century psychologists to distinguish Hume’s nerve-quivering type of sympathy from the more detached, cerebral act of appreciating the predicaments of others. Now, empathy is widely accepted as the sole true source of morality and social progress, each person’s bulwark against wrongdoing and the psychopath’s vital deficit. ‘People often assume that empathy is an absolute good,’ writes Paul Bloom. ‘You can never be too rich or too thin… or too empathic.’

Bloom, a professor of psychology at Yale, admits that he himself once thought along these lines. Recently, however, he has come to the conclusion that empathy is no more successful than reason in puncturing our biased self-concern and bridging the gap between ourselves and others so as to achieve moral action. Uncomfortably sensing other people’s suffering may not provide sufficient incentive for us to try to remove it. We may instead simply remove ourselves and our awareness of it. A woman living near one of the Nazi death camps wrote to the authorities indignantly to complain about being forced to witness atrocities. She demanded that these be either discontinued or else carried on where she could not see them. ‘Empathy’, says Bloom, ‘has to connect to kindness that already exists.’

How useful is it, he wonders, for an aid worker, doctor or psychiatrist to feel her patients’ pain? Her vicarious suffering could well distract her from actually helping them, and, over time, lead to ‘burnout’. It would be better, in fact, for the roles to be reversed – not for the doctor to replicate the patient’s distress but for the patient to catch the doctor’s calm.

Bloom compares empathy to a spotlight. It strongly directs attention to a particular place, leaving the surroundings dark. Precisely because of its visceral immediacy, it is limited with regard to how many people it can encompass, and for how long. ‘One man’s death is a tragedy; a million deaths is a statistic,’ said Stalin, and charities tend to employ his reasoning. Notoriously, rather than mentioning abstract numbers, they focus our attention on photos of particular victims. Stalin may sadly be right about our natural psychology, Bloom concedes, but that means we should make all the more effort to exert calculated reason to help us rise above it. Concern, justice and fairness require maths. The Effective Altruism movement is surely right to research and advise on which are the most financially efficacious charities to donate to.

This book’s publisher declares it a ‘landmark scientific investigation’, but it is, rather, a well-argued polemic. It is generally at its weakest when invoking empirical research. Bloom points to experiments that, he claims, demonstrate by means of brain scans that sympathising and empathising occur in quite different areas of the brain. But are we ever sure whether we’re feeling sympathy rather than empathy, or vice versa? It is speciously circular to claim that neurological procedures can potentially give scientific confirmation as to which of these we are sensing. The experimenters have to rely on information supplied by the subject whose brain they are scanning as to whether he is experiencing either sympathy or empathy, and exactly when. The experiment thus presupposes, and adds nothing to, the merely conceptual, anyway hazy, distinctions we already have. What does such ‘science’ prove?

More persuasive are the experiments, cited by Bloom, that suggest that, contrary to much received wisdom, there is no causal link between low empathy and either psychopathy or aggression. Psychopaths are often highly emotional. And, thanks to the ‘spotlight effect’, the very empathy that inspires kindness and generosity can equally engender anger, prejudice and cruelty. Both the conservative and the liberal, Bloom rightly argues, can be empathic. If one empathises with the robbed, the other with the robber, it is not empathy that determines which of the two is right.

Bloom is not, despite the title, against empathy. He even gestures in Hume’s direction by saying that perhaps empathy is indeed part of the aetiology of morality. He simply wants to redress the balance, pulling us away from Hume and towards Kant. To make others loom as large as you do in the immediate foreground, he says, is a less effective strategy for achieving morality than to shrink and depersonalise yourself until you see yourself as just one in a wider arena of individuals. This rehashes Kant, as does Bloom’s argument that most wrongdoing stems from making special – in other words, empathic – exceptions only for certain individuals: ourselves, those we love, those who happen to touch us at a specific moment of time. Of course, we all thrill to the story of the Nazi who spared the man who looked into his eyes, but is it any consolation? It simply raises the question of why, if he could empathise in that instance, he killed all the others.