.....and here's why.
Let's say you are a space alien on the moon busily engaged in observing human beings and their various aircraft. You write a book on aerodynamics wherein you provide a mathematical model of computation that describes the outward behavior of what you see. But in the introduction of the book you claim that you can model the flight of planes, rockets, and balloons without any need to consider air. Of course aerodynamics without air is as stupefyingly dumb as claiming to understand the mind without understanding the neuronal basis of the brain. But this is what Pinker actually does in his book, and even trumpets the fact!!
Thus to quote Pinker: "This book is about the brain, but I will not say much about neurons, hormones, and neurotransmitters. That is because the mind is not the brain but what the brain does... That special thing is information processing, or computation." (p7)
In other words, by saying that "the mind is what the brain does", Pinker neglects to define the brain! By reverse engineering the mind, and attributing behavioral functions to wholly inferred computational modules somehow selected by evolution, Pinker neglects the massive corpus of findings in neuro-psychology that have detailed in painstaking detail the motivational systems in human and mammalian brains. The sub-cortical systems that are critical for the generation of human emotions and human motivation are not 'computational' by any stretch of the imagination, and must be incorporated in any model of how the brain actually works. Without this, understanding the mind is impossible. Nonetheless, Pinker wears this ignorance like a badge, a badge that discredits his own argument even before its substantive products are considered.
Pinker of course adheres to commonly held viewpoints in evolutionary psychology that postulate behavioral mechanics through an appeal to historical selectionist pressures spanning eons. but it does not logically follow that neurobiological perspectives must be so utterly ignored. Indeed, affective neuroscience, a branch of neuro-psychology that embraces these perspectives, also is wholly informed by evolutionary principles, and assigns much more of our behavioral repertoire to a complex interaction between general purpose neocortical structures and basic emotional systems arising from mid brain systems. In other words, behavioral tendencies are not ingrained in our brain like the bee line a honey bee makes to a flower, but come from complex interactions between brain systems that cannot be 'reverse engineered' as Pinker would have it. Rather, you have to 'go into' brains and have a look.
In common with nearly all practitioners of evolutionary psychology, Pinker accepts the metaphor of the mind as a modular computational device. But the more credible view arising from neuro-psychology is that this is false. Evolutionary psychologists have rightly dismissed creation 'science' because the latter blatantly ignores the overwhelming facts of evolution. The great emerging irony is that neuro-psychologists are rightly dismissing much of evolutionary 'science' (although not evolution itself) because it ignores the facts of the brain.
And this is why Pinker's book is the worst book ever written on the human mind. Not because of a lack of intelligence, style, creativity, or wit, but because of a hubris that psychology cannot afford. And indeed, if there's anything that antagonizes the gods, as well as this writer, it's a prideful arrogance that thumbs its nose on the facts, and this book, by building its case in large measure on conjecture, builds its case on sand.
As a final note, the position of modern neuro-psychology, which counters much of the intellectual baggage of evolutionary psychology, has been most cogently stated in the book 'Affective Neuroscience' by the distinguised neuropsychologist Jaak Panksepp. His book, far more than Pinker's should rightfully be titled 'How the Mind Works'. A second book that makes a similar case, but which is less technical, is Gerald Edelman's 'Bright Air, Brilliant Fire'. Both are highly recommended.