I've been buying a lot of books this summer. That's not "a lot of books" by the usual standards, because I've always been a better-than-average bookstore customer. Lately I've been buying a lot of books even for me. But I haven't been buying them to read. I've been buying books to write.

Over the last two years, I have been steadily squirreling away books I need for the book I'm writing. Every scholarly book is built out of earlier books, in pretty visible ways. If you want your book to say something new, rather than just parroting what has gone before, you're actually going to need to build it from more books, not fewer. If you read shallowly, you'll end up saying what everyone else said. If you have a question that someone else hasn't asked, you'll have to dig for hints and clues and small pieces of evidence through dozens-to-hundreds of other books that were after different questions. If people don't already know (or think they know) the answer to your question, there's going to be a lot of research.

(Fiction is also built out of earlier books, but less obviously. A novelist or short story writer can only create something that seems fresh after she's read and digested hundreds and hundreds of other works of fiction. If you've only read a dozen novels in your whole life and then try to write one yourself, what you come up with will sound appallingly close to one of those dozen. You will sound like someone trying to imitate your favorite writer, and it will basically be fanfic. You probably won't even realize how close you are to outright plagiarism. It takes a lot of reading to get a sense of all the possible moves you can make in a story and to digest a bunch of different influences before you stop trying to sound like someone else and start trying to sound like yourself.)

Now, you can't research an academic book without an academic library and inter-library loan. But writing requires keeping a bunch of other books handy for a long stretch of time. You'll have to root back through those books repeatedly to check one thing or another, and won't necessarily foresee exactly what you'll need to look for day to day. So after a certain point, depending on the library for those books becomes either inefficient or anti-social. You need to be able to check facts as you write. You can't always be stopping short and making out a list of things to look up on Monday. On the other hand, the more useful a book is, the less fair it is to hog the library copy while writing something that takes years to finish. You don't want to be the guy who's had the standard biography of Milton out of the college library since 2008.

So, I've spent the last two years buying a book here and a book there, as I happened to find things on used-book shelves, building up a small collection of books I knew I would need for the project I've been researching. Over the years I spent writing my dissertation and then turning it into my first book, I collected a bunch of specialized monographs and reference works I needed for that project. Because my second book takes a different methodological approach, I have to accumulate a new section of my personal working library.

But over the last six weeks, the buying has picked up intensity. Instead of buying a couple of books every month, I've purchased a couple of dozen since May. Part of that was opportunity, the result of visits to used-book stores while traveling with my spouse or of finally seeing things at a good price on abebooks.com. Every individual book was a good or excellent bargain, at least five or ten times cheaper than a new book from a university press. But I bought a lot of them, including multi-volume sets, more than enough to make me reorganize the shelves in my study. I've walked into bookstores and walked out with a box, more than once.

This newly serious book-buying feels like part of a new intensity in the writing process. The work is picking up steam, and so the reading and fact-checking have to keep pace. The preliminary phase of research and drafting is done, and now I'm starting the first really serious push. If I can sustain it, that push will take the next two years or so. During that time, I will have to work on lots of other things beside the book, but it will be the central project and there should be periods of fairly intense and sustained writing. I need a solid core of particular reference material in my writing space in order to keep momentum, need to be able to put my hand on exactly the volume I want at a specific moment in order to keep from slowing down. Assembling the books I need is like stockpiling supplies and equipment for the long push. I'm tooling up.

How this big push actually turns out won't be clear until it's done. At the moment I feel like the next two years will be the central and crucial part of the writing process. But I might be further from the end than I think. It's hard to know exactly where you are in the process of your second academic book, because the experience of writing your first book doesn't give you a clear road map. Writing that original, started-out-as-a-dissertation book is a torturous and often backward process. You start out with a few seminar papers and some ideas, and try to bootstrap them into a dissertation; because you've never written a dissertation before, you have to figure out the rules of that genre as you go, and often piece out the actual, workable structure for that document fairly late in the game. Then you've got a book-like object called a dissertation which you have to reverse-engineer into another, very different piece of writing, an actual publishable book. That process frequently demands more unexpected and sometimes radical restructuring, even very late in the game. The final version of my book was written almost exactly backwards: most of chapter four had been written first, followed in order by chapters five, three, two, one, and the introduction. While I would always recommend writing the intro last, and while I do plan some late tweaks to chapter one so that it sets up the rest of the book smoothly, that's not generally a writing process I want to reproduce.

This time I started with a clearer structure and a clearer plan. Putting a book together the hard way will teach you to appreciate how they're put together. So the sequence of writing might be more straightforward this time around. But writing a book in such a necessarily chaotic and haphazard way doesn't give you a feel for how a more normal version of the process goes, or where you might be in the process. You learn how to do it, and you learn that you never want to do it this way again. Writing the first book is about getting across the finish line alive. Writing the second book is about learning how writing one is actually supposed to work. Or at least that's my story right now; I'm sticking to it as long as I can.