Against Empathy is a book about the negative effects of trying to feel what other people feel by Canadian American psychology professor Paul Bloom. The book makes the case that concern and compassion function better in the absence of empathy. It also makes the case that empathy is a driving force behind much of the cruelty and irrationality in the world. The book is divided into six chapters and two shorter interludes, each of which explores a different aspect of empathy.

Bloom begins by defining his terms and laying out the case he intends to make over the whole book and in each chapter, as any good academic would. Adherence to definitions for the purpose of avoiding confusion is done well throughout the book, and is especially necessary when a word as widely defined and misused as empathy is in play. Rather than arguing in favor of psychopathy, Bloom advocates thinking with our heads rather than our hearts so as to reach a more consistent and helpful morality. Nor does he argue that empathy is completely bad; only that it does more harm than good.

The first chapter makes the distinction between cognitive empathy (recognizing another person’s feelings without feeling them oneself) and emotional empathy (experiencing the world as one thinks that someone else does). The shortcomings of the latter are the primary focus of the book, namely that empathy can lead to ignoring unidentifiable victims, denigrating logical choices that have superior results, letting our biases lead us astray, overrating present costs versus future costs, and sending unnecessary aid. The chapter ends with responses to objections raised by Bloom’s colleagues during the writing of the book.

In the second chapter, Bloom explores the neuroscientific aspects of empathy, including mirror neurons, the role of preconceptions of other people, and the difference between understanding and feeling. The difference between cognitive empathy and emotional empathy is important here, and it can be detected in fMRI scans. Bloom then discusses how empathy is currently measured, as well as the shortfalls of such methods.

Read the entire article at