As Michael has pointed out in his latest blog-length contribution to our conversation, every writer is different. How Michael approached writing Sand was very different from how I went about composing Written in Stone, so rather than try to lay out a precise set of writing rules I will attempt to summarize my unique writing experience.
By the time I signed with Bellevue Literary Press in the autumn of 2009 Written in Stone was represented by three complete chapter drafts, a smattering of notes and partial chapters, and an outline in the form of a book proposal. Unlike a novel, however, I had not started with chapter one. In order to grab the attention of agents and editors I had written what I thought would be the most immediately interesting chapters first (the ones on the origins of birds, whales, and humans), so when it came time to write the whole thing I went back to the proposal to remind myself of the story I wanted to tell.
The basic layout of the book ended up looking something like this. The introduction, through the discussion of a famous "missing link", would be a hook which would explain the importance of paleobiology in pondering the origin of our species. In order to carry readers along through the rest of the book, though, I knew I needed to provide them with the appropriate tools, and rather than lay out all the background information in a textbook-like chapter (something I have always hated in pop-sci books) I decided to use history as a way to introduce basic concepts. Chapter 2 would cover some of the basics of paleontology and geology by tracing the work of Steno, Cuvier, Hutton, Lyell, and others, while Chapter 3 would tie these discoveries to evolution by recounting how Darwin attempted to reconcile his evolutionary theory with paleontology.
From there I decided to arrange the chapters about evolutionary transitions such that the earlier chapters would provide some context for the later ones. The chapter on the origins of tetrapods, for example, comes before the chapters on birds and early mammals, and the chapter on early mammals precedes those on whales, elephants, horses, and humans. Following convention, the chapter on human evolution comes last, but not because I intend to cast our species as the pinnacle of evolution. Instead it seemed like the best place to put the chapter for dramatic effect, especially since our changing understanding of human evolution goes to the heart of the ongoing public controversy about our origins. This is capped off by a conclusion which considers what the fossil record can tell us about how the world we live in came to be as it is, especially the much-debated question of whether or not the origin of creatures like us was inevitable.
Hence I had a pretty good idea of where the larger story in Written in Stone was going, but there were also some subplots that I had to keep in mind. I knew that I was going to talk about using carbon and oxygen isotopes to gain insight into the diets and habits and extinct animals, for example, but the question was whether to introduce it in the chapter about whales or the chapter about elephants. I also wanted to make sure that I mentioned punctuated equilibrium, but in order to do so I needed a solid example of stasis and rapid change from the vertebrate fossil record. In some cases I lucked out and I was able to perfectly embed these discussions within the narrative, and other times I had to place something in a footnote (or toss it altogether) because there was no place to put it without derailing the story.
All of this planning had to be done before I jumped back into the active writing process. (I had taken a break while the proposal was making the rounds in case an editor wanted to substantially change the content of the book.) It was not unlike preparing to write a fictional story. Not only did I have to come up with a list of scientific "scenes" in each chapter, but I had to make sure that each one kept moving the larger story forward.
I did not expect to get everything right the first time, though, especially since my early drafts of each chapter were intentionally rough. I had a general format in mind (open with some historical quirk, build that into a story about changing views of the fossil transition in question, and then shift into a discussion of how new discoveries and methods have further altered the picture), but in order to fill out this template I needed to first educate myself about the subject at hand. I spent hours reading papers, riffling through historical documents, perusing books, and catching up with new research so that I could highlight the most interesting parts of the story. Once I felt like I had enough background information I would write a rough draft of the chapter from what I could remember. This established the story arc of the chapter, and with that in place I started to go back to the technical literature to flesh out the abstract with particular examples. After that I would start editing the chapter, usually three or four times before I was satisfied with what I had put together.
Chapter by chapter I repeated this process, but I saved the introduction and conclusion for last. I knew they would be the most difficult sections to write. I wanted to make sure the introduction provided a solid hook and that the conclusion contained a solid payoff for the reader (especially since no one likes a crummy ending), and as expected I wrote and re-wrote several versions of each.
Once I had completed the entire book I went back to the beginning and started to edit the work as a whole. This shift in focus allowed me to see where I had repeated the same arguments in multiple chapters and helped me to identify which chapters needed the most work. I ended up sending the first version off to my editor just before Christmas, and I anxiously awaited her thoughts on what I had put together.
When the first batch of edits arrived I was very nervous, but, as it turned out, the cuts were not as painful as I had feared. Nearly all the revisions made sense to me; my reaction was not "How could she have done this to my work?" but "How stupid of me to have missed that!"
Perhaps the most important effect of the edits was making me recognize some of my little tics. I often included original quotes in the manuscript only to rephrase them for clarity in the next line, and this hindered the narrative more than it helped it. Likewise, I often took such pains to make sure readers understood my examples that I risked becoming repetitive. I was so concerned with being clear and accurate that I did not realize when I started to get in my own way, and I am certainly grateful that my editor helped me learn what to watch out for.
All of this work has been carried out during nights, weekends, and the occasional snow day. I do not earn enough from my writing to make freelancing my sole occupation (at least not yet), and working another job nine hours a day, five days a week greatly restricts the time I have to write. During the few free hours I have each day I try to juggle the two blogs, academic papers, popular articles, and the book, and at the end of almost every day I feel like I still have not done enough. Even so, I did not start writing this book because I expected a huge advance or because I thought it would be a financial success. I have spent so much time on it because I simply could not resist writing it. I had to do it, and I am grateful that I have a chance to share my enthusiasm for paleontology with the public (though, should the book place me in a better position to transition into science writing full-time, I won't complain).
As of this date I am still working with my editor to trim down and tune up my manuscript. It has gotten better with each successive iteration. With any luck Written in Stone will soon be completed and go into production, and once it is sent off I will jump back into the proposal process as I try to find a home for my second book. So much to write, so little time...
Tomorrow: Figuring out illustrations, especially when you are on a tight budget!
For more on writing a popular science book, see the latest posts by David and Michael. The title of this entry was unceremoniously stolen from Tom Levenson, who wrote a similar series about his book Newton and the Counterfeiter last year.
Do you think it you'd offered your book to publishers as "Brian Sweetext" rather than your actual name, you might have had a psychological advantage?!
I always write the introduction and conclusion last. It amuses the Spousal Unit to no end, but you get a better frame that way. :)
I intentionally wrote the introduction to my thesis late in the process to make sure I knew what I was introducing.
I tend to write intros and conclusions last too to make sure that I don't say I'm going to do one thing and then end up doing something else.
My series on writing Newton...and by extension other such works ain't dead; it's just resting. (More precisely, it's pining for the fjords.)
BTW -- concur w. everyone else: body of the text first, conclusion next, introduction last.