How to Write a Book in 647 Steps (Part Three)
According to Oxford University Press, my book checks in at a healthy 452 pages. I still can’t quite believe that I wrote something so long — approximately 165,000 words, 130,000 more than anything else I’ve ever written. Writing the book was extremely fun, but the hard work before the writing, the researching and the outlining, was often anything but. It’s that process that I want to describe in this post, which tries to answer a simple question: what do you do once you’ve obtained a contract to write a book?
Planning the Research
I was, of course, incredibly excited when OUP offered me a contract. The excitement quickly wore off, however, when I realized that I had no idea how to actually write a book. My only consolation was that, because I had done so much research for and thinking about the proposal, I had a very good idea of what each chapter would contain and — just as important — how they would fit together. I anticipated writing 12 chapters (I ultimately wrote 16, because I ended up dividing four chapters that were longer than I planned), so I tried to convince myself was that I was writing twelve 10-12,000 word essays instead of one 150,000 word book. That made the project seem somewhat more manageable.
The next step was to figure out how to research the book. I divided the potential sources into four categories: (1) the NMT judgments themselves, along with various orders and motions in the trials; (2) archival material about the trials, particularly documents that explained the prosecution’s decision-making processes and the thought processes of the various judges; (3) secondary literature, both past and present, about the trials; and (4) contemporary jurisprudence that cited the judgments, necessary for the final chapter of the book, which would discuss the impact of the trials on the development of international criminal law. I decided to begin with the judgments, because I did not want my interpretation of them to be affected by either the secondary literature or contemporary jurisprudence. (Auguste Comte’s “cognitive hygiene” in practice.) I would then turn to the secondary literature, the archival material, and the contemporary jurisprudence.
The judgments were, of course, the heart of the project. I had decided early on, when conceptualizing the proposal, that I would write a thematic book, not one organized by trial. I wanted to explore what the tribunals in general said about war crimes, modes of participation, defenses, rules of procedure and evidence, sentencing, etc., so that readers would know when the tribunals agreed with each other and when they disagreed with each other. (As I pointed out in this post concerning aiding-and-abetting, courts have often cherry-picked individual judgments, implying that a minority position was actually the “NMT” position as a whole.) But that meant identifying not only every issue that I wanted to discuss in the jurisprudential chapters of the book, but also what every tribunal said about those issues — no small task, given that the judgments ran approximately 2,500 pages, nearly 2,400 more than the IMT judgment.
At this point, I had not read all of the judgments in their entirety. So that is what I did first. As I read, I highlighted important facts in yellow, highlighted legal issues in green, and added keywords for the legal issues in red pen in the margins. That process yielded a huge amount of highlighting and dozens of keywords. I then hired students to type the highlighted text into Word files, one per judgment, while I organized the legal issues. I created a Word file for each chapter, reduced the chapter descriptions in my proposal to outlines, and assigned the various issues to individual chapters. I then worked with each individual chapter file, arranging and rearranging the issues until I had a workable overall outline. I knew that the outlines would change as my research deepened and as I actually wrote, but it was a good first approximation of the structure of each chapter.
The outlines also provided me with a way to arrange all of the facts and legal considerations in the judgments. Once my typists had finished their job, I cut-and-pasted each individual quote (which I made sure they identified by page number!) into my outlines. That was an incredibly time consuming task, because it meant that I had to constantly move back and forth between all of the outlines. But it was worth it, because at the end of the process I not only had outlined chapters, but outlined chapters with all of the supporting material from the judgments in the right place.
The process was essentially the same with the secondary literature. With the help of two research assistants, I scoured books in libraries and articles in the various on-line databases — particularly Lexis, Westlaw, and Hein Online — for discussion of the twelve judgments, Control Council Law No. 10, or the NMTs as a whole. I then highlighted what I needed and had my typists create Word files for each secondary source. Finally, I cut-and-pasted the relevant text into my chapter files. To avoid confusion with the text from the judgments, which was green, I changed the text of early secondary literature (which I arbitrarily determined was pre-1975) to red and left the text of contemporary secondary literature black. (I wanted to differentiate the two eras when I discussed secondary sources in the book.)
I dealt with legal citations in the same way as the other sources. This was a much less difficult process, because I had decided, as noted above, to discuss the impact of the NMT judgments in the final chapter of the book, limiting the earlier chapters to the judgments themselves. I began by identifying the citations with the help of a research assistant, which was easier for the international tribunals than for domestic courts. The latter required examining as many domestic databases as I could find, as well as OUP’s own excellent Reports on International Law. I then created a list of issues that international tribunals and domestic courts had discussed and arranged them in the same order that I has used in the earlier chapters of the book.
This was, not surprisingly, the most difficult category of source material. Because of the relative dearth of secondary literature on the NMTs, I had little idea what material was out there. Fortunately, due to Hilary Earl’s excellent book on the Einsatzgruppen trial and Valerie Hebert’s very good book on the High Command trial, I knew that nearly all of the material I would need was located in two places: Telford Taylor’s papers at Columbia Law School, and military records in the National Archives in Baltimore. So, with the help of a generous grant from the University of Auckland (where I was teaching at the time), I planned a research trip to the United States. I spent three weeks at Columbia Law School and five in the National Archives, working literally 9-5 to obtain the material I needed. It was a very stressful two months, because I was not sure how long I would need and knew that I almost certainly would not have another chance to work with the archives. I nevertheless had a great deal of fun; as anyone who had conducted archival research knows, the thrill of finding a valuable document — and I found hundreds that no one has ever cited — is better than getting a good Hanukkah or Christmas present. Once I had all the material I needed, I repeated the process I had used for the other sources, inserting archival material into my chapter files in blue text.
All archives, of course, are different. But I have a number of suggestions for scholars who are contemplating working with archival material for the first time. First, plan ahead! Most archives have finding guides available on-line; they’re not always easy to use — I’m looking at you, National Archives — but the investment of time is well worth it. Through hours of internet research I had at least a general idea of the boxes that I wanted to look at in the National Archives and Columbia long before I boarded the plane for New York.
Second, and relatedly, get in touch with the archives before you visit them. Archives have very different hours, restrictions, technology requirements, etc.; you don’t want to find out that the archive you need to use is closed or won’t let you scan documents (see below) when you get there. Equally important, many archives have librarians who will go out of their way to help you in advance of your visit. The Columbia Law Library, for example, was unbelievably helpful when I contacted them from New Zealand. The librarians responsible for the Taylor papers pointed me to boxes that I did not know about and made sure to have the boxes I needed ready for me when I arrived. That alone saved countless precious hours.
Third, I strongly recommend using a scanner for the documents, if the archive allows them. Many people take digital photos of documents instead. Photos are definitely quicker, but I think it is much harder to use a photo than a digital scan. If you are not in a hurry, you can scan documents and convert them into searchable PDFs (a godsend later on) at the same time. If you are in a hurry, you can scan the documents and convert them later. I used an older version of this Canon scanner, which was cheap, quick, and relatively easy to carry around. I originally intended to use a pen scanner, but I learned that most archives only permit flatbed ones.
Fourth and finally — and I cannot stress this enough — keep good track of your documents! If you have photocopied a document, write the archival information on it. If you have scanned a document, give it a title that both identifies it and includes the archival information. I learned the hard way about both kinds of documents. In my haste, I often stapled a bunch of photocopied documents together and wrote the archival information only on the first one. I then separated the documents later and forgot to label all of them clearly, which cost me endless hours of work when I had to figure out where the unlabeled documents came from. Similarly, I did not title my scans accurately enough, forcing me to relabel them all (and there were hundreds) later so I could find them when I needed them.
Once I finished my archival research, I finalized my chapter outlines. I now had four different colors of text in my outlines: green for the NMT judgments, motions, and orders; red for WW II-era secondary literature; black for contemporary secondary literature; and blue for archival material. Because the various sources were easy to distinguish, it was not much problem to arrange the material. I was thus left with one file per chapter, each file (1) identifying all of the issues that I would discuss in that chapter in the order that I would discuss them, and (2) containing all of the material I needed to discuss each issue, that material itself arranged in the order I would use it. I knew that if I had done my job well, I would rarely have to look at the judgments themselves when I wrote each chapter. All I would need was the outline.
It didn’t always turn out that way, of course. The outlines evolved as I wrote the various chapters. But that is a story for the next post.