Writing
Over the course of your time in the lab we will write a lot together. Manuscripts, fellowship/grant applications, conference abstracts…
Writing is a craft, and scientific writing is a specific kind of writing. Writing effectively takes a lot of time and work. From the get go, I recommend everyone watch this lecture on how to write well. Watch it now, before you have to write anything. And watch it again in a year or two.
Part of your training in the lab is to improve your writing skills, and IMO it’s one of the most important parts. Remember: your three main outputs in the lab are figures, ideas, and code. Writing is a tool for developing your ideas, and it’s the most tangible way you will communicate those ideas. Writing well will get you a job, and it will improve your science, because writing is a thinking tool as much as it’s a communication tool (WARNING! these are not the same, see the lecture 😉).
A heads up: when you get a draft back from me, it will likely have a lot of red ink. More red ink is good. Make sure to give yourself, and me, enough time to work on drafts together (e.g. I want to see abstract drafts a week before conference submission). For any document you submit, expect a lot of comments/suggestions and few rounds of back-and-forth editing. This is a dialogue of iterative, collective development of your document, your project, and your abilities as a writer; not a corrective exercise. I prefer to use google doc because I think it has the best tools for co-writing.
Tips and tricks
Editing is the majority of writing. Editing is a craft you can hone. Here is a collection of tips and tricks for editing I’ve collected over the years.
- Always consider your reader. Who are they? Why are they reading your document? How can you help them out? They will always thank you for making things clearer, more concise, and easier for them to follow.
- For each sentence, interrogate it: "what is your purpose?" It should contribute something specific to the document at hand, and that something should never be “tell the reader you know this thing”.
- Tell your reader only what they need to know in order to understand and care about what you’re telling them, no more. Your job is not to convince your reader that you know stuff.
- Say it in a lot of words, then get rid of most of them. When you don’t know how to say something concisely: don’t worry about concision, write it out in as many words as possible, then edit them down. You’ll often find that the reason you couldn’t express your idea concisely is because it wasn’t fully developed, and the act of writing → editing will help you develop the idea. Note: this is the vomit method on a small scale.
- Write in filler sentences. If you don't know how to say what you want to say, use a filler sentence that serves the same purpose, which you will edit/expand later. Separate out "what is this sentence for" from "what are the words in this sentence".
- Look out for referents. Words like “this”, “that”, et al. What do they refer to? Is it clear to the reader? Can they be replaced with something clearer, or more informative?
The first draft, on the other hand, is dark magic voodoo shit. Where editing is a craft you can hone, writing from scratch is a different dumpster fire every time you put a pen to a blank page. Here are a few methods to get it out and move on with your life.
- The vomit method/shitty first draft. This can be summed up as: You need to write words. It's ok if they are garbage words. You can fix them later. TBH, all first drafts are shitty first drafts whose only purpose is to exist so you can move on to editing, and the vomit method is just one way to get it out of you as quickly as possible.
- The outline-first method. This sounds like elementary school stuff but it can really work. Collect the pieces you want/need to include in your document and arrange them in a bulleted list. Organize the overall structure of the document, identify missing pieces. The idea is to focus on pieces and structure not on language. Once you have the pieces in a structure you like, turn them into complete sentences and delete the bullet points.
- The San Antonio method. This is basically the vomit method with at least one coauthor and a whiteboard. TL;DR one person’s job is to write on the board and everyone else’s is to throw out ideas. The whiteboard author should take an attitude of “yes, and”, note down anything that’s been said, and should try to prompt thinking about how they relate. Afterword, arrange your notes into an outline.
- Generative AI. I would not recommend using generative AI for your first draft, unless you’re recombining words you’ve already written. For example, if your fellowship asks for a community service statement, and you’ve already written three of them for past applications but the prompt was slightly different and you’ve done some new stuff since then. You can throw your old writing samples and updated CV and ask: “I’m writing a statement for a fellowship application. Here is the prompt and some previous statements I’ve written. Please combine them to fit the prompt, preserving my writing voice and, ideally, the actual language I used, as much as possible”. Once you have a collection of past submissions this will save a lot of time, but I would highly recommend against using GenAI to help with writing anything new - it tends to write many fluffy nice-sounding language without any real content, which is the exact opposite of what you want for subsequent editing, and it robs you of the idea-generation phase of writing which is arguably the most important part.
Develop a regular writing practice
Advice on specific pieces of academic writing (Links/advice to come)
- Writing an Abstract
- Structuring your paper
- Specific Aims: by the time your reader gets to your aims, they should be able to guess what they are without reading them. That is, you should have given them exactly and only the information they need to have a burning question, which your aims spell out a way of asking and answering. See Anatomy of a Specific Aims for more.