A safe and supportive forum for philosophers.
Owned & Moderated by Marcus Arvan (University of Tampa)
Contact: marvan@ut.edu
Blog mission & moderation policy
In our newest "how can we help you?" thread, a reader asks:
I struggle with proofreading. I never manage to completely eliminate small errors from my published work, and I am embarrassed by this. Does anyone have any tips for more effective proofreading? Has anyone managed to leverage AI to effectively help with proofreading and editing?
I'm not a great proofreader either and am curious whether AI is helpful too.
Do any readers have any helpful tips to share?
Three independent tips, though you can certainly combine them, too:
1.) Print it out. Working with a different medium–a physical one–changes how your eyes scan.
2.) Read it out loud. This slows you down considerably, and means less scanning/skimming.
3.) You can also try using a different font (either from a serif font to non, or vice versa), colour, and size (slightly larger). New and different things will pop out.
Several years ago a friend suggested listening to my work to proofread it and that’s what I’ve always done since. Listening to text out loud will catch typos, repeated words, and grammatical awkwardness that your eyes miss. With this method I just recently caught a number of remaining errors in a final document that had already been read by 3-4 people. I can even multitask/zone out a bit while listening to my papers since spoken typos tend to be very jarring and catch my attention. MS Word has a “read aloud” function you can use for this under the review tab. I don’t use AI tools myself but I guess some of them should have a similar text-to-speech function in case you don’t have a word processor with this function built in.
If you are not an effective proof reader, you should pay a professional editor to do it. It does not cost that much for one article. But it really looks bad if most of your articles have typos in them. I have edited volumes and such before, and I really do not appreciate working with authors or contributors who do not proofread the proofs of their articles carefully. It just shifts the burden to someone else.
Honestly “Get a professional” does it “really look bad?” Why? Why should it if the argument is correct or well argued? Suggesting that people with this weakness spend extra money to fix some typos seems very silly to me. I am not this person, but the argument in this book strikes me as completely correct. Very few typos mess with comprehension.
https://www.routledge.com/Why-Its-OK-to-Have-Bad-Spelling-and-Grammar/Flanigan/p/book/9780367333232?srsltid=AfmBOooHtCGKhGMQtOZUbz6SjDgxZUwp8rdoKHtRGXOAiqBPR2IU2S1Y
There are two issues here, copyediting and proofreading.
Copyediting involves identifying errors and inconsistencies before the proofs stage. Proofreading involves the resulting proofs (typeset materials).
For copyediting, which involves much more than just eliminating typos, it can be helpful to use a checklist; you can find many online. Keeping a list of frequent kinds of errors can help. Add them to your checklist.
Aso, break your editing into stages. Don’t check for everything all at once.
Then, when you think you are done, set the paper aside once more. Then read it again, later. Most of my errors come from being rushed or distracted and not taking time for that final pass.
I have taken multiple, professional courses in copyediting, even engaged in some academic editing professionally, and I’ve learned that editing involves more than most academics realize. So, I second getting a professional involved or at least a second set of eyes. Still, despite this training, I find it harder to eliminate errors in my own work than in the work of others (something most copyeditors will themselves admit).
As far as proofreading, when submitting work to journals, there should be a stage at which professional copyeditors intervene. (Reviewers might point out typos at submission, but they are not copyeditors, nor should they strive to be.) This is the stage at which you want to remove most of your errors, not the proofs stage. While AI can help, it is notorious for incorrectly flagging correct English and missing errors.
If you have typos that are getting past the proofs stage, find out what they are, and how they got there. Sometimes typesetters introduce errors. Is this happening? If so, can you predict what kinds of issues are likely to recur? Add them to your checklist.
In addition to printing off the essay and also reading out loud, as others have suggested, my own nuclear option is to read the essay in reverse sentence order. In other words, read the last sentence of the essay first. Then read the second-to-lost sentence, etc. This completely kills your ability to read quickly and skip over something. It’s a very slow and awkward way to read, but it ensures that every sentence receives your full attention.
Contra Get a professional, I have never heard of someone paying a professional editor and I wouldn’t do it. Ask friends or family members to do you a favor and read your finished paper; they don’t have to understand it to give proofreading tips.
I’m not a native English speaker, so my perspective may be different. For me, having a native proofreader was a must. In recent years, I’ve been using Grammarly and DeepL a lot – even this post is being corrected while I’m writing it. However, AI systems didn’t eliminate the need for a human proofreader who can still suggest a lot of improvements far beyond the abilities of AI.
Get VoiceDream to read your document to you (also useful for turning any other reading into an audio). Excellent way to catch typos and even stylistic problems.
I have started using AI for proofreading, but I’m not doing anything sophisticated. I just upload my document to Claude and prompt it by saying something like “please proofread this paper for spelling and grammatical errors.”
It usually catches a few things. It usually picks on a few things that I think are fine. And I’m sure it misses some stuff. But it’s useful as a first pass.
Also, because it’s such a people-pleaser, it usually can’t help itself from saying something like “this is a well argued paper defending the thesis blah blah. The writing is clear, and the style is consistent and compelling.” So that’s a nice ego boost as a bonus–Claude likes my paper! (I’ll know it’s bad when even Claude can’t muster any praise.)
Leave a Reply to listen aloudCancel reply