Well the second claim is true even if the first claim is a bit of an exaggeration.
For anybody who has ever turned a PhD dissertation into a book (uh, if you're reading this, odds are you have), you may wonder, just how different was my book from my dissertation. Hopefully a lot, because dissertations almost always suck.
My adviser, Orlando Patterson, was pretty good about letting me write my dissertation with the final book in mind. For one thing, that means my dissertation didn't have any statistical regressions. That's rare in sociology these days, at least at Harvard. But it's still a dissertation. Nobody fails to get a PhD because of passive verbs, or advising that more research is needed, or for having an extra chapter that should probably be cut.
Before my book gets set in type by a man who knows the printing press (I like to imagine Princeton University Press does things the old-fashioned way), I wanted to make sure my book is plagiarism free. Academics stay up at night worrying about things like that. And I'm talking about the ones who don't plagiarize. Given constant reworking, and all the things I've read, and my cut-and-pastes from a file of excerpts from books I've read, well it's easy to see how something could slip through. Or how I could read an idea and then come up with it on my own four days later. My family is famous for that. Half the time my brother says anything half intelligent I protest than I told him that, six months earlier. The other half of the time he read it in the Economist. Sometimes both. He doesn't care. Blood runs thicker than plagiarism.
Professors have the great resource of [lest they benefit from my praise, the name has been redacted by a cease-and-desist order from this company. They say I violated the licensing agreement by submitting my book.]. It's a plagiarism detection system on which I check all my students' papers. I haven't taught a single semester without failing at least one student for plagiarism. Sad but true, but that's for another post.
I submitted my book to [the company that wishes to remain unmentioned, despite my praise], just as I submitted my dissertation 3 years ago, to make sure it comes back clean. It did. If you exclude quotations (and I think it can only exclude some of those, because most of my quotes are long and without quotation marks), 25% of my book matches my dissertation. I think if you removed all the juicy quotes, it would go down to 15-20%. In other words, 75-85% of my book is new material, not in my dissertation in any form.
I don't know what the significance of that is or if the percentage is high or low. But I think it's interesting to know, and undoubtedly a credit to my great editor, Tim Sullivan. It's certainly great for the readability of my book.