Paul Bloom and I live in different worlds, his consisting of labs, mine of texts. He has managed, however, to bridge the divide between us. Though a scientific illiterate, I followed his argument in Against Empathy without difficulty, and it held my interest throughout. It helped that on page 96 Bloom cited my very favorite Jewish joke, and that on page 154 he mentioned something else common to us, “a cold and repressed Canadian heart.” (True, mine is naturalized while his is presumably native, inasmuch as he currently teaches psychology and cognitive science at Yale.)

Most of all, however, Bloom disarmed me by his respect and obvious affection for Adam Smith. The Theory of Moral Sentiments pops up in his argument again and again. This is a good sign, because of modern moral philosophers Smith is the most sensible. Just as we can understand Smith’s project as an updating of Aristotle’s Nicomachean Ethics in the light of modern philosophical psychology, so we can understand Bloom’s as an updating of Smith in the light of current experimental psychology. Indeed Bloom goes so far as to admit that “after many years and many millions of dollars, [of] the three major findings from the neuroscience of empathy research, none…are [sic] exactly new—they reinforce ideas from philosophers hundreds of years ago—but they add to our knowledge in valuable ways.” If Bloom publishes a second edition, he might usefully include an appendix specifying those valuable additions.

In his prologue Bloom suggests that if pressed for time readers might content themselves with his first chapter only, and in fact it effectively presents the main lines of his argument as a whole. That argument depends on taking empathy in a strict sense, as “mirroring” or “the act of feeling what you think other people feel—experiencing what they experience.” He distinguishes this “emotional” empathy from “cognitive” empathy, which consists of grasping the experience of the other without necessarily feeling it. Although both forms of empathy may play a role in supporting salutary actions or deterring harmful ones, neither is necessary or sufficient for either task. While endorsing compassion and kindness, he sharply distinguishes them from empathy.

* * *

Bloom’s critique of empathy begins with the commonsense observation that there are many other sources of moral conduct. His next stage is to show that these may moreover be in tension with empathy. An enlightened consequentialism (Bloom’s provisional moral prescription) will require more reliable guidance than empathy provides. Empathy is “like a spotlight,” he writes, “directing attention and aid to where it’s needed.” “But spotlights have a narrow focus, and this is one problem with empathy.” In fact one might conclude that this is the core problem with empathy, of which the others that Bloom urges are variations. Ultimately our capacity for empathy is both too narrow and too idiosyncratic, too biased on behalf of ourselves and our own to provide a true moral compass.

The remaining chapters develop different aspects of this argument. In “The Anatomy of Empathy,” Bloom enlists modern studies in brain science to confirm both those possibilities and those limits of empathy that he has already discerned on the basis of everyday experience. In “Doing Good,” he reports on some perverse effects of empathy as recorded in the laboratory, including its “innumeracy” (i.e., its lack of proportionality). In an “interlude,” “The Politics of Empathy,” he refutes the notion that empathy always votes Democratic. “Political debates typically involve a disagreement not over whether we should empathize, [but with whom].”

* * *

In “Intimacy,” Bloom considers the role of empathy in various private relationships. In this context he explores the distinction between empathy and compassion, drawing from Buddhism. Compassion need not involve empathy in the strict sense of sharing the other’s suffering, although it may benefit from something quite different, namely, our own past experience of similar suffering. Where empathy does figure in even the closest relationships, its effects are at best mixed. “Empathy makes us too-permissive parents and too-clingy friends…. We can often do much better.”

In another “interlude,” “Empathy as the Foundation of Morality,” Bloom finds little evidence for the claim that empathy, being fundamental to all morality, explains the emerging morality of children. In “Violence and Cruelty,” he opposes the popular notion that the world’s violence is due to a want of empathy and could be abolished through its cultivation. He argues, among other things, that violence is “a part of human nature, shared with other animals, evolved for punishment, defense, and predation—and unless we are transformed into angels, violence and the threat of violence are needed to rein in our worst instincts.” He further argues that contrary to much belief both popular and scholarly, morality in general plays a major role in violence, as does empathy in particular. “[Although] some people think about kindness [when they think about empathy,] I think about war.” Here again he cites Smith on sympathetic resentment as a spur to action against perceived offenders.

In the final chapter, “Age of Reason,” Bloom sets out to vindicate Aristotle’s claim that man is a rational animal. Though conceding much evidence to the contrary, he invokes both experimental and common sense evidence that we are beings capable of reaching rational conclusions in moral matters, even when these conclusions run contrary to other powerful inclinations. Of the claim that rationality is merely “a white male Western pursuit,” he remarks that here “the extremes of postmodern ideology circle around to meet with the most retrograde views of a barroom bigot.” He considers at length the claim that the modern scientific materialism to which he subscribes (the presumption that mind is nothing more than brain) refutes the possibility of human rationality. He concludes, however, that “there is nothing in the claim that we are rational animals that clashes with the findings of neuroscience.” That claim remains to be examined on the basis of the empirical evidence.

Although we could then still be rational, many psychologists argue that we are not. Bloom concedes the force of the many experiments in which human subjects have acted irrationally and even ludicrously, but insists that they don’t dispose of the question. That we often behave unreasonably simply can’t establish our inability to behave reasonably.

* * *

Against Empathy concludes with positive arguments for human rationality: in morality as in other matters, there is a strong positive correlation between intelligence and appropriate action. In this final stage of the argument he cites Smith not just once but twice, concluding with Smith’s paean to reason as conscience, which is to say the impartial spectator.

Bloom has written a book rich alike in arguments for sensible positions and in refutations of fashionable but unsound ones. Among the questions that remain is whether we can successfully analyze the relation of empathy to morality without an account of what morality requires of us. As a psychologist Bloom rejects this question as above his pay grade, choosing to make his case against empathy in the name of morality as such.

Yet suppose that morality does depend, as he thinks, on some contribution of empathy and other passions guided by reason asserting its powers of independent judgment. In that case we could not determine whether these resources are equal to the task unless we knew what morality demanded of us. Nor would we know to what education to submit them. In order to do justice to this range of questions, we would have to follow Smith’s example and write a Theory of Moral Sentiments or, if we aimed at radical improvement, Jean-Jacques Rousseau’s Emile.

Bloom doesn’t pretend to meddle with philosophy, and is not primarily a scholar of its history. It might be instructive, however, to construct an encounter between him and Rousseau, whose critique of compassion so remarkably anticipates Bloom’s critique of empathy. In fact the comprehensive fault that he finds with empathy is the same that Rousseau finds with compassion: its unreliability, whether as a motive for moral behavior or a guide for it. Justice requires generality, but compassion and empathy are incurably particular and therefore unavoidably arbitrary. The two thinkers agree that we cannot but enter more readily into some types of sufferings rather than others, into those of some sufferers rather than others, and in a manner that (limited as it is by our capacities of imagination) is unavoidably disproportionate. Bloom cites with approval Stalin’s notorious statement that one victim is tragic, while a million are merely a statistic. He might have cited Immanuel Kant’s earlier remark that although justice would require the generalization of compassion, compassion resists such generalization, being itself bound to the imagination and therefore to particularity. In this Kant simply follows Rousseau, for whom the tension between compassion and justice is the fundamental problem of universal morality (as opposed to that of the citizen, which is a different matter).

* * *

Given the shortcomings of empathy or compassion, then, both thinkers see morality as depending decisively on reason. Yet Rousseau differs from Bloom in stressing the moral ambivalence of reason: although a condition of morality (and even or precisely of compassionate morality), it is not simply friendly to it. Though it is the sine qua non of our decent treatment of others, reason also poses the foremost obstacle to it. This is because its development is inseparable from that of amour-propre or consciousness of oneself as a self among others. Amour-propre fuels an irrational preference for ourselves over others and a Hobbesian treadmill of competition with them, ending only in death.

Because all reason is social, however, developing only in and with society, it inevitably reflects all the flaws of society. While claiming to perfect its members, society in fact deforms them. Rousseau would thus have challenged Smith’s voice of rational conscience: could it be anything but the internalization (and therefore the sum) of society’s defective teachings? (Smith doesn’t resolve this difficulty.) Rousseau points toward (and in my view even beyond) more recent critics of enlightenment such as Michel Foucault.

Rousseau then sees the relationship of compassion to reason, and of both to morality, as extremely complex. He would reject Bloom’s position, as he did the Enlightenment progressivism of his own day, as naïvely optimistic. Still, he would join me in applauding Paul Bloom’s excellent new book.