In today’s New York Times book review, Robin Marantz Henig reviews a new book on evolutionary psychology by Paul Bloom, How Pleasure Works. I haven’t yet read this book, but the review indicates that it may be a more reasonable specimen of the genre, seeing the human penchant for pleasure-inducing activities as a nonadaptive byproduct, or spandrel, of other evolved tendencies. But there’s a slight problem with the review. Can you spot it?
Pornography is another example of pleasure via essentialism. Why do some men spend more time looking at Internet porn than interacting with flesh-and-blood lovers? There may be “no reproductive advantage” to liking pornography, Bloom writes, but there is an advantage to its source: an urge to look at real-world “attractive naked people,” which makes us want sex, which in turn is good for continuation of the species. Pornography uses the same pleasure mechanism as actual sex, which is handy since “there aren’t always attractive naked people around when you need them.”
The paragraph contains one of the most common misconceptions about evolution: that traits evolve because they’re “good for continuation of the species.” (This is probably Henig rather than Bloom’s misconception.)
This isn’t the way that most traits evolve under natural selection. They evolve not because they help the species to persist, but because they help the genes to persist. Or, more accurately, selection favors those traits that enhance the survival and reproduction of the genes that produce those traits. Genes that make us want to have sex (and, as a byproduct, make us look at pornography that can gratify the sex drive) are favored because individuals that carry them have more offspring than those with sexual lassitude. This says nothing about the good of the species as a whole, but about the good of individuals and the genes they contain.
Indeed, sex may be a trait that reduces the persistence of some species or groups. This is because most sex requires two individuals to reproduce, and it may be hard in some circumstances for two individuals to find each other. If population density is sparse, perhaps because individuals live in marginal habitats, the need to find somebody else to mate with may not be as adaptive as those forms of reproduction (say, parthenogenesis) that enable you to reproduce without a partner. This may explain the observation that asexual species often occur in marginal or disturbed habitats.
Natural selection may occasionally favor the evolution of traits that favor the persistence of groups (“group selection”), though I don’t know of any such traits. And selection can operate directly on the genes themselves, as in the phenomenon of “meiotic drive”, in which one form of a gene outcompetes another because it’s better at getting into the sperm during cell division. There’s also “kin selection,” in which behaviors like parental care can evolve that may be deleterious for an individual but useful for its offspring or kin.
But must evolutionists think that the main level on which selection operates is that of individuals, not groups. It is the differential success of individuals, not groups, that causes the spread of the genes they contain.
Remember that the next time you see a nature program on television that explains some behavior as having evolved “for the good of the species.” They’re invariably wrong.

