The Blank Slate has also been undermined by brain science. The brain obviously has a great deal of what neuroscientists call plasticity—that's what allows us to learn. But the newest research is showing that many properties of the brain are genetically organized, and don't depend on information coming in from the senses.

The doctrine of the noble savage has been undermined by a revolution in our understanding of non-state societies. Many intellectuals believe that violence and war among hunter-gatherers is rare or ritualistic, and that the battle is called to a halt as soon as the first man falls. But studies that actually count the dead bodies have shown that the homicide rates among prehistoric peoples are orders of magnitude higher than the ones in modern societies—even taking into account the statistics from two world wars! We also have evidence that nasty traits such as psychopathy, violent tendencies, a lack of conscientiousness, and an antagonistic personality, are to a large extent heritable. And there are mechanisms in the brain, probably shared across primates, that underlie violence. All these suggest that what we don't like about ourselves can't just be blamed on the institutions of a particular society.

And the ghost in the machine has been undermined by cognitive science and neuroscience. The foundation of cognitive science is the computational theory of mind—the idea that intelligence can be explained as a kind of information-processing, and that motivation and emotion can be explained as feedback system. Feats and phenomena that were formerly thought to rely on mental stuff alone, such as beliefs, desires, intelligence, and goal-directed behavior can be explained in physical terms. And neuroscience has most decisively exorcised the ghost in the machine by showing that our thoughts, feelings, urges, and consciousness depend completely on the physiological activity of the brain.

EDGE: What's the influence of evolutionary psychology in all of this?

PINKER: Evolutionary psychology is one of four sciences that are bringing human nature back into the picture. (The others are cognitive science, cognitive neuroscience and behavioral genetics.) There's a sense in which all psychology is evolutionary. When it comes to understanding a complex psychological faculty such as thirst or shape perception or memory, psychologists have always appealed to their evolutionary functions, and it's never been particularly controversial. It's no coincidence that the effects of thirst are to keep the amount of water and the electrolyte balance in the body within certain limits required for survival—without such a mechanism, organisms would plump up and split like a hot dog on a grill or shrivel up like a prune. Likewise, it can't be a coincidence that the brain compares the images from the two eyeballs and uses that information to compute depth. Without such an ability we'd be more likely to bump into trees and fall off cliffs. The only explanation, other than creationism, is that those systems evolved because they allowed our ancestors to survive and reproduce better than the alternatives.

Evolutionary psychology is taking that mindset and applying it to more emotionally charged aspects of behavior, such as sexuality, violence, beauty, and family feelings. One reason that evolution is more controversial in these areas than it is in the study of thirst is that the implications of evolution are less intuitive in the case of emotions and social relations. You don't need to know much evolutionary biology to say that it's useful to have stereo vision or thirst. But when it comes to how organisms deal with one another, common sense is no substitute for good evolutionary theory. We have no good intuitions about whether it's adaptive, in the narrow biologist's sense, to be monogamous or polygamous, to treat all your children equally or to play favorites, to be attracted to one kind of facial geometry or another. There you have to learn what the best evolutionary biology predicts. So evolutionary thinking in those fields is more surprising than in the rest of psychology.

EDGE: How are your ideas informed by the debate between Frank Sulloway and Judith Rich Harris which has been featured on Edge?

PINKER: Both of them, to their great credit, have addressed what may be the most important puzzle in the history of psychology. It's one that most psychologists themselves don't appreciate, and that most intellectuals don't understand even when it's explained in Newsweek or the daily papers. Here's the puzzle. We know that genes matter in the formation of personalities. Probably about half of the variation in personality can be attributed to differences in genes. People then conclude, well the other half must come from the way your parents brought you up: half heredity, half environment, a nice compromise. Right? Wrong. The other 50% of the variation turns out not to be explained by which family you've been brought up in. Concretely, here's what the behavioral geneticists have found. Everyone knows about the identical twins separated at birth that have all of these remarkable similarities: they score similarly on personality tests, they have similar tests in music, similar political opinions, and so on. But the other discovery, which is just as important, though less well appreciated, is that the twins separated at birth are no more different than the twins who are brought up together in the same house with the same parents, the same number of TV sets in the house, same number of books, same number of guns, and so on. Growing up together doesn't make you more similar in intelligence or in personality over the long run. A corroborating finding is that adopted siblings, who grow up in the same house but don't share genes, are not correlated at all. They are no more similar than two people plucked off the street at random. So no, it's not all in the genes, but what isn't in the genes isn't in the family environment either. It can't be explained in terms of the overall personalities or the child-rearing practices of parents.

Both Harris and Sulloway, and a handful of other psychologists like David Rowe, Robert Plomin, and Sandra Scarr, have called attention to this puzzle: what are the non-genetic determinants of personality and intelligence, given that they almost certainly are not the family environment. Many people, still groping for a way to put parents back into the picture, assume that differences among siblings must come from differences in the way parents treat their different children. Forget it. The best studies have shown that when parents treat their kids differently, it's because the kids are different to begin with, just as anyone reacts differently to different people depending on their personalities. Any parent of more than one child knows that children are little people, born with personalities.

Where these two differ is that Sulloway argues that the unexplained variation comes from the way that children differentiate themselves from their siblings in the family. They take these strategies for competing for parental attention and resources outside the family and react to nonrelatives using the same strategies that worked for them inside the family. Harris argues that the missing variance comes from how children survive within peer groups—how they find a niche in their own society and develop strategies to prosper in it.

I think that Sulloway has captured something about the dynamics among siblings within the family. But I'm not convinced that these strategies shape their personalities outside the family. What works with your little brother is not necessarily going to work with strangers and friends and colleagues. And indeed most of the data that support Sulloway come from studies in which siblings rate their siblings or parents rate their children, or in which siblings rate themselves with respect to their siblings. The theory is not well-supported by studies that look at the personality of people outside the home. Indeed, it's a major tenet of evolutionary psychology that one's relationships with kin are very different from ones relationships with non-relatives.