The two types of writing, I’ve found, are very different. For the past 16 months, a large part of my research has been dedicated to so-called change-of-state tokens; in Dutch, as in English, the one we find most frequently is oh. When I started to work out my analysis at first, the writing helped with structuring my ideas and findings. While in that phase it was of course important to discuss these ideas, and thus what had been written, with my supervisors, the quality of the writing definitely did not make it suitable to be read by anyone else.
When at the start of this year I got invited to write a paper based on this analysis, I decided to try writing as it was suggested in books like Writing your Dissertation in Fifteen Minutes a Day. Its author, Joan Bolker, suggests that because writing is hard, you shouldn’t worry too much about it. The best way to get started is simply to get started. Meaning that at the start you shouldn’t be trying to write a good paper, or even an analysis that makes sense to anyone: it is solely about getting words on paper. The result of this of course, was that the first draft – or null draft – was unreadable for anyone but me (and sometimes even for me), and contained way more information than I could ever fit into a journal article.
I ended up with more than 40 pages of analysis, which did not even include structural sections like an introduction or conclusion. Most of this was pretty much useless text, but the trick in writing the actual paper is to figure out which parts of the null draft weren’t useless. It turns out that in the swamp of incoherent analyses, I could detect some sort of pattern on which the actual paper could be based. This didn’t necessarily mean that parts of the null draft could be copy-pasted, but at least it gives an idea of what you should write and which examples are suitable to support the thesis of the paper.
Of course, as it was my first major paper, some things did not go as planned. That is to say, I ignored the handbooks in my approach. Instead of writing up the analysis first, and making sure that I and my supervisors were okay with it, I started with an introduction. Now the problem was not that I wrote a bad introduction, far from it even. It was coherent, well structured, and made a good case for reading the rest of the paper. But because an analysis needs refinement after refinement, the introduction I wrote at the start no longer fitted the analysis I wrote in the end. As a result, the introduction I was pretty proud of, ended up in the trash. Time wasted, but hopefully lesson learned.
The final paper probably took me only a month or two to write. That is I guess the advantage of writing gibberish first: afterwards you have a pretty good idea of what to write and how to write it. Of course, it did go through some revisions based on critique from my supervisors, but nothing I would consider major. The analysis as I had thought it out stayed pretty much the same; the only change was by my own design. Of course, it might turn out that peer reviewers have a completely different opinion, but we’ll just have to wait and see. The paper has been submitted in a form that I’m happy with and proud of, and the analysis is a strong starting point for a lot of extra work (i.e., papers). Now it’s just hoping that it passes review, all you can do is wait.