A friend of mine, who is working on her PhD in Medieval Literature, recently wondered in my hearing why it is that the dissertation (or, here in Australia, the PhD thesis) is so much more difficult to write than any other paper, ever. I couldn’t help agreeing with her on that observation, although I’m not certain I am able to give a complete answer to the question. One major factor may well be the time scale in which the project is undertaken. A paper which is written over the course of a month or two will be easier to keep coherent—to use a consistent voice throughout, and to make certain that all lose threads are tied up before the conclusion of the discussion. A PhD project on the other hand, takes years to complete. Research which is done early on is either written up straight away, or it is set aside and we have to go back to old notes from (by now largely forgotten) previous works sessions and write it up. While it makes much sense to do the writing as the work is done, for a PhD project with a steep learning curve (aren’t they all?), it is very likely that things learned over the course of the project will mean that what was written for segments done early on will need to be changed, often dramatically, for the final write up. On the other hand, if one chooses *not* to write paragraphs about the work done early on, but to simply leave the data & graphs to stand on their own for writing at a later date then much time is wasted looking over the data again, trying to remember what was done with this sample before the writing can commence.
In my case the writing bits of it here and bits of it there have conspired to add an additional complication. I’ve just received back comments from my advisor on one of my chapters, and he points out that the thesis is meant to be written in a formal style throughout, yet in some places I’ve adopted a casual, chatty tone summarizing what I did to obtain the results. Looking over my work I see that he is correct, the voice used varies greatly from one segment to another, no doubt based upon how I was feeling on any given day, how much sleep I’d had, if I was hungry, and, possibly, even what the weather was like on the day. So not only do I need to finish up the summary of all of my results, I also need to read over everything and convert it all to a consistent writing style. Why didn’t I think about that aspect earlier along in the writing process?