Edit to polish the writing, or edit to polish the writer?

Image: Polishing the chimney of a Burrell Traction Engine.  © Oast House Archives, CC BY-SA 2.0.  What? You think this is only tenuously connected to the post?  My friend, tenuous connections are my thing.

One of the most exciting parts of being a mid-to-late-career researcher is seeing the scientific writing produced by the early-career researchers (ECRs) I’m mentoring: Honours undergrads, grad students, postdoctoral fellows.  It’s a treat to see a new manuscript (or more often, a new piece of a manuscript*) ping its way into my inbox.  A treat, but of course also a new obligation, because I put a lot of effort into editing ECR manuscripts.  The question, though, is how much effort?  And what kind of “editing”?

Once upon a time, I would simply take an ECR manuscript and make “track changes” edits until I was happy with the results.  In other words: I would polish the writing (albeit with the use of “track changes” so the ECR could see and learn from the edits I made).  I don’t do that any more.  I still make liberal use of “track changes” (as my students will attest!), but now I make fewer edits and more (and longer) comments.  I now try to explain what writing problem I see and suggest fixes that the ECR might choose to pursue – that is, my intent is to edit to polish the writer, rather than to polish the writing.  I realize that this sounds trivial.  But I don’t think it is: it’s hard for me not to simply polish a piece of writing (and sometimes I still fail).  I think it’s hard for others too.

I can think of three reasons why you might decide you should polish the writer, not the writing:

The first is naïve: you might think (as I once did) that making fewer edits would take less time.  Not so! I can fix fifty instances of a subject/verb disagreement in number in less time that it takes me to highlight one and write a marginal comment along the lines of

“The verb in a sentence needs to agree in number (singular vs. plural) with its subject, not with the noun that happens to be closest to it (which is often part of a modifying phrase – one that could be removed from the sentence without disrupting its basic meaning).  Check for this error and fix throughout.”

And that’s not even counting the non-zero probability that the writer will not, in fact, actually “check and fix throughout”, meaning that I have to go through the process a second time with the next draft.  And so help me, maybe a third.  (We’ve all experienced that writer.)  So we can drop “less time” as a reason; it just doesn’t pan out.

The second reason to back off from polishing the writing has to do with voice.  We’ve done a marvellously effective job, as a scientific community, of scrubbing nearly every hint of individuality from our writing.  I think that’s something we should regret, and I’ve realized that I’ve been part of the problem: I’ve edited my ECRs manuscripts to make them sound like me, rather than like the best versions of themselves.  If I flag a writing problem rather than just fixing it, I let the ECR I’m editing find a solution of their own. If it’s not the solution I would have found, it may be just as good.  It may even be better.  I’ve become more deliberate about this recently: if a bit of writing stops me, I ask whether I’m objecting because I would have written it better, or merely because I would have written it differently.**  If the latter, I try to keep my editorial big mouth shut.  (I sometimes fail.)

The third and best reason to edit the writer, not the writing, is of course that part of my mentorship lies in teaching my ECRs to write better and more easily.  An enormous fraction of a career in science is writing, so teaching better and easier writing could have equally enormous value – if only I could do it!***  Simply correcting an error maybe efficient editing, but it’s poor teaching.  That’s true even if I use “track changes” so that, in theory, the ECR in question could look at and think about each correction.  Word simply makes it too easy to accept a change – even to accept all changes in a single uneducational mouse click.  I owe it to my mentees to explain what seems wrong with a piece of text, and to suggest more than one way it could be better, because that’s one of the ways that ECRs can learn better writing.

I’ve held some weird beliefs over my career about how my ECRs could learn better writing.  I once thought that ECRs could simply read some papers in the discipline, and they’d automatically be able to produce similar work.  I once thought that ECRs would naturally look at each “track changes” edit and generalize from the specific case to a general rule, then adopt that general rule correctly for all time.  I once thought that ECRs would, all on their own, discover and read writing books, and that doing so would immediately make them brilliant and prolific writers.  These beliefs, and other equally weird ones I’ve held, have in common the notion that even if we aren’t born brilliant writers, at least improving is easy and natural.  But here’s the thing: not only is writing hard work for me, so is getting better at writing.  So how can I expect anything different from my ECRs?  My students and postdocs are likely going to learn better writing just as I did: slowly and laboriously, by deliberate attention to practicing the craft, and by integrating lessons from many sources.  One of those sources is me, and my editorial comments on their manuscripts.  Comments, not corrections, as much as I can manage that (although I’ll admit that for reasons of efficiency, I usually still edit hard on a last draft before submission to a journal).

I know I’m not the only person who’s thought about this.  If you’re an ECR, how do you want your mentors to edit your manuscripts?  If you’re an old fogey mid-to-late-career-researcher like me, how to you handle your mentees’ manuscripts?  Please share your tips and tricks in the Replies.

© Stephen Heard  September 24, 2019

---

*^Typically, I ask students to send me sections as they draft them – usually, Methods first, then Results, then Discussion, and then finally Introduction.  (More experienced writers, such as postdocs, are more likely to send me a full draft manuscript all at once.)  This lets us catch some issues before they’ve shaped a whole paper.  It also helps disabuse new writers of the misguided notion (perhaps we can blame The Sound of Music) that they should start at the beginning and write in a straight line until they reach the end.  That may work for some writers, but I think it doesn’t work well for very many.

**^Answering that question, though, takes a surprisingly large amount of time and thought.  There are an awful lot of things about writing that I do by instinct – by familiarity with language and structure garnered from extensive reading, rather than from knowledge of a set of rules.  Often, then, I don’t actually know why I object to a particular bit of text until I think about it, or turn to a usage book, or to Google.

***^I do teach a formal course in scientific writing, and of course I’ve written The Scientist’s Guide to Writing, a guidebook on the subject.  You’d think I’d have this down, wouldn’t you?