“Writing a book is an adventure: to begin with it is a toy and amusement; then it becomes a master, and than it becomes a tyrant; and the last phase is just as you are about to be reconciled to your servitude – you kill the monster and fling him to the public.”
Winston Churchill

Last year I took a vacation for a month to write my dissertation thesis. And it took me that one month to come up with the first draft, which made it into the final version with only minor alterations (but a lot of error checking). While the lack of major alterations might be in part due to my academic advisers (and my) wish to finish the work as soon as possible, I think the major part of this is due to the way it was written, or rather structured.

Doing a dissertation thesis is a major project, the writing itself is a different but not less complicated animal. I think it is a mistake to start writing in sentences unless you know the structure and the content. Once you write sentences, they stick together and are hard to change. And I think it is nearly impossible to write a 200+ pages work if you do not structure it beforehand, and there is a great way to do so: Outlines.

Most people know outlines from school. Many teachers try to give this valuable hint for exams. Plan what you write before you start writing. An outline for a dissertation is similar, but not quite the same. For one thing, it is much more detailed.

How detailed? Well, everything you want to write later should be included in it, without the actual sentences. Metaphorically it should contain the bones of the text, the whole skeleton, and hints for everything else. This means

• the order you want to write the different pieces of information that make your theory
• the notes you made about your studies, the design, the participants, the instruments, the procedure
• the results of any statistical analysis you made
• the ideas for and the issues you want to raise in the discussion

It also includes any notes you do not want to forget and any ideas, e.g., for further studies even if you cannot realize them (a valuable hint from my informal academic adviser: you will have ideas of things you want to realize but you cannot realize everything, so make notes and raise these points in “future work”).

Given that the outline only contains the information, but not the sentences, it is easily changeable. And once you get in the flow of adding flesh to the bones, you can write really fast. An additional benefit of using outlines: I used the same outline as a basis for the articles I wrote about my dissertation. The outline also allows you to focus only on the relevant part by using the hierarchical structure: You can arrange the information similar to the structure you use for your PhD thesis and simply fold in the parts you do not need at the moment. This way, thousands of lines of text become easily manageable. For example, you can fold the parts between the introduction and the discussion to write parts of the discussion while simulateneously seeing parts of the introduction. Sure, you could do something similar with Word’s “split view”, but not as easy and with this focus on the parts you want to see.

Personally, my outline for my dissertation was a 66.5 MB Circus Ponies Notebook file, containing 333,215 words (> 2.2 million characters, equivalent of about 1305 pages). I made sure to write down everything I did, the results of any analysis, etc. It was more or less structured in the way I wanted to write my dissertation. With this outline next to my writing program (Scrivener), it was possible to come up with a good first draft within a month. Why? Because I first read the whole outline, taking care to move the information that did not fit where it was to the correct place, then sorted each sub-point (e.g,, theory, results of Study 1) in the correct order, and then used this sorted outline that contained all the information I needed to write it as a guideline to write that chapter. Given that the sources were marked in the outline (see Academic Workflow) I did not have to check other sources for the actual writing. I didn’t even have to re-check statistical printouts — it was all in that one huge outline (and then in a smaller one that dealt only with the chapter).

I created the outline before I started to write, during the last year of my PhD. But thinking back, it would have been much easier to create the outline during the whole PhD thesis time, as soon as the topic and the first experiments were decided. Noting the decisions (and the reasons for doing so), the results, etc. while planning and doing the studies would have made it much easier in the end, but it also worked this way.

So, I can only highly recommend creating a detailed outline prior to writing and using it for the writing process. It makes an insanely complex work manageable.