I hate checking proofs. Here’s how I make it (slightly) easier.

Image: Proofreading marks, by volkspider via flickr.com, CC BY 2.0

Like many of us, I suspect, I have a love-hate relationship with writing.  I love having written.  And I enjoy certain kinds of writing and certain parts of the writing process (oddly, I really like shortening things; even more oddly, I just added this parenthetical that lengthens this paragraph).  Other kinds of writing (Gantt charts, anyone?) I dislike; and there are a few parts of the writing process that I truly despise.  Checking proofs?  I’d rather remove my own gallbladder with a rusty spoon.

Everyone, sooner or later, is faced with proofs to check.  For those few who have escaped so far: by “checking proof” I mean that very last stage before publication, when you see your article (or, God help you, your entire book) typeset.  You’re asked to check every word, every letter, and every punctuation mark for errors introduced during typesetting.  And it’s hard – so hard – and it’s boring – so boring – and it’s necessary – so necessary.*

What makes it hard is that you’re checking writing that you know very well, and checking it for the kind of small errors humans are good at sliding over as they read.  You know perfectly well, reading your own paper, that the next word is meant to be “shift”; so unless you’re very careful, that’s the word you’ll see – even if actually, the ink on paper has a very important letter missing.

I don’t think there’s any way to make checking proof easy, but there are definitely ways to make it easier.  I don’t mean “faster” – I don’t think it’s possible to check proof both well and fast. Instead, I mean with a high probability of catching almost all errors with fewer than six dozen re-readings.  I have a grab-bag of tricks for this, and all involve confronting the core problem: you’re too familiar with what you’ve written.

An obvious possibility, of course, is to have somebody else check the proofs for you – not a coauthor, but someone who brings completely fresh eyes to the job.  This works a treat; but checking proof isn’t the easiest job to farm out (largely because it’s such an unpleasant one).  Perhaps you’re lucky and have the resources to pay someone to do it.  More plausibly, perhaps you can make a deal with a friend: you’ll check their next proofs if they check yours today.**

What if you’re stuck checking your own proofs (as most of us usually are)?  You need to find ways to break that over-familiarity with what you’ve written.  In The Scientist’s Guide to Writing I call that “reader simulation” – you’re trying to see the text as a new reader, not as the writer.  Here are some techniques:

  • Check proof somewhere else than where you wrote. If you write in coffee shops, check proof at the office; if you write at the office, check proof in your neighbour’s kids’ treehouse.  Human memory is strongly tied to the context in which memories are formed, so consistent visual, olfactory, and other cues can reinforce memory. You want to disrupt it.
  • If you write and edit onscreen, check proof on paper; and vice versa. Same reason.
  • Read backwards – word by word, from the end of the article to the beginning. The idea here is to stop yourself from focusing on the meaning of the text – that’s what you know far too well – so you can focus instead on the individual words. I’ll admit I find this technique a real challenge, and it does take some self-training; but I have colleagues who swear by it.  The next few techniques take less dramatic approaches to the focus on words rather than meaning.
  • Work on paper, and follow as you read with a pencil. Follow along every millimetre of each line – underlining helps. Better still, tap each word – yes, every single one – as you check it.
  • Again on paper, use a cut-out screen – a piece of paper with a slot cut just large enough to show one line of the document you’re checking.
  • Read out loud as you check – deliberately, word by word. This has two advantages: it shifts the way you process the text from silent reading to vocalizing, and it paces your checking no faster than you’re reading.  As an alternative, you can have a friend read out loud and follow along word by word on your own copy.

While we’re on the topic of proofs, two other points:

(1) Back when we submitted manuscripts on paper, everything was retyped by the publisher and errors could creep in anywhere.  With modern electronic submission, most of your manuscript is simply reformatted.  It should therefore be unusual to find typesetting errors in the body of your text.***  However, some journals copyedit and proceed directly to the proof stage without showing the copyedits to the authors.  The fact that this is an awful and disrespectful way to proceed doesn’t seem to stop the offending journals, so if you notice shenanigans of this sort then you’ll need to check everything very, very carefully. Otherwise, you can be strategic: save the bulk of your time for elements that require more intervention by the publisher, and therefore are more likely to have errors introduced.  Tables are notorious for this (they’re often still retyped and are always reformatted), as are equations, lists of authors and affiliations, and special characters such as integral signs, accented letters and diacritics, and the like.  Styled text (bold, italic, etc.) can get messed up too.

(2) You’ll surely be tempted to make a few last edits – not corrections, but new changes to the text – at the proof stage.  At least, I always am: I’ve never written anything that I couldn’t still tinker with.  Publishers will normally instruct you not to do this (or they’ll say that they’ll charge you for such edits).  In practice, you can probably get away with one or two tiny edits – but resist the temptation unless they’re critical to meaning.  Perfection is unattainable anyway.

Now to admit the sneaky reason I wrote this post.  I have more proofs ahead of me (including another whole book’s worth coming in the fall), and I’d love to expand my toolbox.  So: do you hate checking proof too? Do you have a technique that I’ve missed?  Please share your tips in the Replies.

© Stephen Heard  August 13, 2019

*^Some folks seem to use “checking proofs” and “proofreading” interchangeably.  I’m not sure that’s quite right. When I do the very final pre-submission polish of something, I’d call it “proofreading”; I’d reserve “checking proof” for dealing with typeset proofs from a publisher.  Would anyone like to disagree in the Replies?   (Also, I used no fewer than FIVE em dashes in that last sentence, and I’m bizarrely proud of myself.)

**^I have a friend who, rather bizarrely, actually enjoys checking proof and doesn’t ask for any quid pro quo. If you ever find someone like this, treat them like royalty.  My friend likes Calvados, and when she checked the proof for The Scientist’s Guide to Writing, I ordered a couple of bottles – probably the best alcohol-related investment I’ve ever made.

***^Well, except for those you made yourself, and that weren’t caught in copyedit.

11 thoughts on “I hate checking proofs. Here’s how I make it (slightly) easier.

  1. Aaron Dalton

    First of all, proofreading your own material is an exercise in masochism. The urge to tweak is so strong. After a long writing task, I never want to see it again.

    Your list of techniques is great. For me, the only way I can do it with any success is on paper and reading it backwards.

    Really, though, have someone else do this. And I’m not just saying that because I’m a professional editor 🙂 The writer is simply too close and will be distracted by too many other concerns. At least find a conscientious buddy you can pay with a nice bottle of wine or something.

    Liked by 1 person

  2. Emilie Champagne

    There’s something I don’t understand: “Tables are notorious for this (they’re often still retyped and are always reformatted) […]”
    Then why journals insist so much on table formatting?


    1. Aaron Dalton

      I can’t speak for all workflows. When I was in academic book publishing, after editing the manuscript in Word, it would get imported into InDesign. Clean formatting of the Word document will hopefully make rekeying unnecessary, but in any software, proper, clean formatting of tables is a tedious task. It’s never right the first time.

      That said, I’ve seen a number of submission guidelines that go way overboard with their prescriptivism. Any designer worth their salt can get any clean table into InDesign relatively quickly. It’s the formatting that can take time. But the formatting in Word is largely irrelevant, as long as you don’t have weird tables within tables and superfluous cell splits.

      And even that said, if the document is not going to be properly designed (the Word document is essentially what’s going to be published; e.g., poorly done e-books), it’s the author who has to do it, thus prescriptive guidelines.


  3. Chris Mebane

    Another great opportunity for goofs in your proofs are scientific species names. Some copy editors apparently run the manuscript through spelling checkers and, voila! evolution before our eyes, with new species. For instance, the insect “Chironomus dilutes” descended from Chironomus dilutus, and the fish “Cottus confuses“, descended from Cottus confusus. This may confuse us quite a bit, as there are many published articles with these two new species.   My particular irritation with this is toward Wiley who finds it expeditious not mark their changes in proofs, and leaves it to authors to find whether they have published new spell-check-species, for example.

    A second way to expedite proof checking is to publish with PLOS One (and I presume other OA journals operating on a lean margin).  No detailed copyediting, no proofs to check, just read it when it comes out. Or not.


  4. Jeff Houlahan

    Steve, is one option just to care a little less? Does the occasional typo have much affect on the quality (or even readability of a paper)? We know that people often have complete comprehension of a sentence even when the letters in words are dramatically rearranged and that relatively minor rearrangements have little affect on comprehension or even reading speed. I suspect 1 or 2 or even 5-10 single letter typos in an article have something close to zero impact on reabadility.


    1. ScientistSeesSquirrel Post author

      Jeff, I’m amused by my own horrified reaction to this, because (in part) you’re right – readers will skim right past many typos without noticing just as you will as the author. But of course there are two risks. One is that some readers may infer, consciously or not, that sloppiness in typesetting implies sloppiness in science. (How large that group is, is an interesting empirical question). The more important: sometimes the typo does matter. Hence the advice to check equations particular carefully, for example: not only are they more likely to get messed up, errors in them will matter and won’t be easily reader-corrected the way my ‘missing letter in shift’ would be. Numbers in tables, same thing.

      But I just can’t, because I’m a horrible perfectionist… I love that you put your finger on a sensible suggestion that my own warped psychology would NEVER let me see :-).


      1. Jeff Houlahan

        Things you can’t confess…

        As an ecologist – “I don’t like fieldwork.”
        As a scientist – “I’m not a perfectionist.”
        As a dad – “All my kids have moved out and I’m not sad.”
        As a non-native New Brunswicker – “I don’t find Maritimers unusually friendly.”

        Liked by 1 person

  5. Uncorrected

    Copy and paste your proof into a word document. You should probably do this a section at a time to avoid formatting.
    Use Word’s compare function to check it against your submitted manuscript (or section thereof).

    This will find any places your text has been changed by copyeditors. Then read over it once or twice, answer author queries and approve. Most people aren’t going to read every word in the paper and perfection is impossible so don’t waste too much time. I have been doing this for years and never spend more than an hour on proofs.

    Liked by 1 person

  6. Pingback: What can the fields of UI/UX teach scientists? – Brushing Up Science

  7. Pingback: Tricks for reading and correcting proofs | Scientist Sees Squirrel

Comment on this post:

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.