This is a great thread — but it’s important to emphasise that there is no “correct” method of academic writing. Perhaps “write shit first and quickly” and then edit is a process that works for some people. It does not work for me. My first drafts are much closer to my final ones. https://twitter.com/Dr_Joe_McIntyre/status/1291210221763747842

2. That does not mean I do not edit on the back-end. I do. But I do my “editing” at the front end — in terms of developing my outline. I outline obsessively and with increasing detail. At the beginning, I have nothing more than my section headings.

3. As I read, I add more headings and move them around to where I think they belong. I also cut and paste quotes from readings into the relevant sections and make notes within the sections about what I think I will argue there.

4. This becomes a recursive process: (1) read something new, add quotes and notes into the outline sections; (2) rearrange and refine the outline sections themselves; (3) read something new...

5. By the time I am ready to write, I have an incredibly detailed outline. It not only has all the headings, it has all the arguments I intend to make within each heading, arranged in logical order. And I have all the quotes I need underneath each argument.

6. If I have done my “editing” well — which is rare — I can write the entire article or chapter without looking at anything other than my outline. And I can write any section I want, because by the time I start typing they are (or should be) in the right place.

7. Personally, I like to write from start to finish, from introduction to conclusion. But again, if I’ve done my outlining well, I could write from conclusion to introduction. (Just thinking about that, however, makes me a bit nauseated.)

8. When I’m done with the first draft, I follow @Dr_Joe_McIntyre’s approach and kill my babies -- stripping out my hyperbole, prioritising better arguments over weaker ones, etc. But I rarely have to make major structural changes, because the global logic was considered ex ante.

9. Again, there is no “right” way, The key, as Joe says, is to think not only about your overall argument — the “big idea” — but also about how every piece of the logical puzzle fits together. Every section, every heading, every argument, every paragraph must need to be there.

10. In other words, I should be able to turn each and every paragraph of your article or chapter into a bullet point and reverse engineer the kind of outline I start with. If I can’t do that, there is something wrong with the work’s internal logic.

11. And yes, my outlines are kind of absurd. A 15,000 word article will usually have a 4,000 word outline. (More if you count the quotes I will use, which are already slotted in.) But hey — I’ve basically written a big chunk of the article before I ever put fingers to keys!

12. The incredibly laborious work of outlining is precisely what enables me to write actual text very quickly. Writing for me is basically little more than turning each bullet point in my outline into a paragraph of text. Call it Kantian writing: freedom through planning.