Editing half-baked drafts into publishable work?

In our newest "how can we help you?" thread, James asks:

People say most of the work of writing a publishable paper comes out in the editing process. Let's say you have a first draft. Can you advise on the way to edit a half-baked piece of writing into a polished work of philosophy? Sometimes it's easy to get "stuck in your head" or overly perfectionistic to the point that you give up on the draft altogether. Thanks.

This is a great question, and a couple of other readers submitted replies to weigh in. An Easter Coaster writes:

James, I would do the Mumford method, https://sites.google.com/site/stephendmumford/the-mumford-method with intellectual peers you respect.

I find this an interesting suggestion, because I basically do everything Mumford says not to do! I mostly just draft papers from beginning to end, working out my ideas on paper but using the drafting of the paper's introductory paragraph as more or less the paper's "outline" so that I at least know where I am headed. Then, after a half-baked version is done, I basically go through the paper from the very beginning again and again day after day, progressively refining the ideas and writing from beginning to end. Because I always start from the beginning each day, the early parts of the paper get polished more quickly, but over a period of time the whole thing ends up polished. This seems to work well for me at least, but if you haven't tried Mumford's method, it seems probably worth a shot!

Another reader writes:

I found that making it into a talk helps. Talks are different from papers, but the process of making a paper into a talk really helps myself rethink each argumentative step in the paper. If I can't comfortably say something to an imaginary audience, or find myself being unsure about the transition from one point to another, then I know I need to revise something or rethink the order of presentation.

I've found this to be helpful too, though I normally only do it if I have a talk to give! One nice thing about it is that if you use PowerPoint for talks (like I do), then the size of a PowerPoint slide constrains how much stuff you can cram into a slide. I've found that this is helpful, in that it can encourage you to break each step in the argument down to a clear, manageable number of points (which can also be helpful in deciding where to place paragraph breaks when writing). I'll also just ad that actually giving talks (at conferences) can be really helpful in terms of turning a half-baked draft into a polished piece, as audience feedback can be really helpful in determining what you need to fix/polish.

But these are just a few quick thoughts. What are yours? How do you turn half-baked drafts into polished, publishable work? Any specific tips for the OP?

Comments

[Louis deRosset]

Developing talks and handouts for talks is very good for some things, and less good for others.

Talks are good for boiling things down to essentials and honing formulations of key claims, examples, and arguments.

Talks are not good for developing the best version of the position of a target of criticism. They are not good for giving a sense of the landscape of alternative views, or carefully explaining why you isolate the question at issue in the way that you do, as opposed, say, to the way it is standardly understood in the literature.

You can do a few of these things in a single talk. But a good paper will do all of these things, and will then keep doing them for the entire length of the paper. So, I think the advice to develop a talk and associated handout is a good start, but only a start.

Perhaps an adaptation of the "develop a talk and associated handout" method might serve: go ahead and develop the talk for the paper; but then do something similar for each of your sections. This will force you to confront the question of what the point of each section is, and will allow you to hone formulations, etc., at the requisite level of detail.

After you've got all that nailed down, you still will need to draft and then edit (ruthlessly) for style and punchiness.

[Anon Postdoc]

I will second Marcus's method. I don't know if this is the optimal method, but it is the only method I have found to work. Although I don't necessarily start each day at the beginning: Say I got up to section 4 one day, I might start from section 4 the next day. If it has been a while then I'm more likely to start from the start.

I'd also add to the advice to give a talk: I often find that the actual process of trying to say everything out-loud to an audience helps me. It tells me which parts are find because they are easy to cover, and it tells me which sections aren't so good because I struggle with them. I think you lose this benefit if you give a talk where you have a completely prepared script that you have memorised, so it is best used when giving talks in venues where a slightly rough and ready presentation is more acceptable (a WIP group or similar, rather than the departments big weekly seminar). And I also don't think you get this benefit by just trying to talk to yourself or by writing the talk (although that might have other benefits)—the key ingredient is being forced to keep talking without doubling back to fix what you just said.

[Marcus Arvan]

On Twitter, Malcolm Keating writes: "On the question about half-baked ideas. @Going_Loopy has a blog post on exactly this issue

https://handlingideas.blog/2018/09/02/how-do-you-turn-a-half-baked-idea-into-a-paper/