This book is certainly well-written and very stimulating, but readers new to the subject should be aware that it is highly polemical, and not at all a neutral dispassionate introduction to the field. The book is written from a strongly Chomskyan perspective - indeed the constant worshipful references to the Great Man become tedious after a while, and the many shortcomings of Chomsky's Transformational/Generative Grammar theory are not mentioned. It is one thing to argue - as Pinker does, convincingly - that human beings are born with an innate ability to deduce the grammatical rules of any language from a limited input. It is another to claim that there exists a Universal Grammar which applies to any language (this is not proven in the book), and it is another still to claim that Chomsky's grammar (which hardly works for English let alone any other language) is that Universal Grammar. The book contains some basic linguistic mistakes, which make one question the real expertise of the author (who is a cognitive psychologist, not a linguist). Just one example: to claim (p127) that in an agglutinative language eight morphemes can be combined in half a million different ways is ridiculous, supposing as it does that they can be combined in any order (in fact each morpheme has to go into a particular "slot" in the word). Nevertheless, a stimulating read - inspiring on one page, infuriating on the next. But please don't take it as Holy Writ (especially the Chomskyan bits).