“The main problem with empathy is that it works like a spotlight, highlighting certain people in the here and now, making their suffering salient to you. This can sometimes be a good thing. Indeed, one of the best arguments in favour of empathy is that it really does make you kinder to the person you are empathising with. This is backed by laboratory research, by everyday experience and by common sense.
So if the world were a simple place, where the only difficulties one had to deal with involved a single person in some sort of immediate distress, and where helping that person had positive effects, the case for empathy would be solid.
But the world is not a simple place. One problem is that empathy is innumerate, favouring the one over the many. In one classic series of studies, psychologists asked some subjects how much money they would give to help develop a drug that would save the life of one child and asked others how much they would give to save eight children. People would give roughly the same in both cases. But when a third group of subjects was told the child’s name and shown her picture, the donations shot up – now there were greater donations to the one than to the eight. All of these laboratory effects can be seen as manifestations of what has been called “the identifiable victim effect”.”

