28

I am currently reviewing a computer science conference paper. I like to give constructive feedback both related to the content as well as for language issues, such as typos or grammatical mistakes.

The paper I'm currently reviewing has significant language flaws. Mentioning each and every typo, wrong word order, and grammar issue would take me about an hour. The paper will likely be rejected anyway, so I am not sure if these language suggestions will even be used. It would be different if the paper were likely to be accepted, because improving the language would further improve the quality of the paper.

Does it make sense to give thorough and detailed feedback on a poorly written paper's language quality in a review if the paper will likely be rejected?

  • 3
    You are a reviewer, not a proofreader. Possibly related: academia.stackexchange.com/questions/60576/… – Alexander May 31 '16 at 9:29
  • 13
    For this type of paper, I'll often list every problem in the first paragraph only, and then say that the rest of the paper has similar issues. That makes the point that the writing is beyond redemption, offers some help to the authors in pointing out types of problems, and doesn't take up too much of your time. – iayork May 31 '16 at 12:22
40

From Alan Jay Smith's The Task of the Referee:

Refereeing a paper can require considerable time and effort; don't waste that effort on a detailed critique of a badly flawed paper that can never be made publishable. Finding one or more fatal and uncorrectable flaws excuses the referee from checking all subsequent details.

The trick is to determine what is indeed "fatal". If you believe that your corrections have some non-trivial chance of improving a published paper at some point (possibly in a less-competitive venue), go ahead and make them, as you are contributing something to the body of literature. Otherwise, do not waste your time.

As an addendum, you are not a copyeditor, but a scientific reviewer. Ways to deal with papers with many language/grammar issues include:

  • If grammatical errors are very numerous, point out classes of errors, instead of individual ones.
  • Focus on errors which actually obscure scientific meaning, and let ones which are merely poor or nonstandard style go.
  • 1
    Thank you for the ref, I never saw it before, but it's really great and full of useful insights! – gaborous May 31 '16 at 13:58
  • 2
    For a CS conference paper, I'd go slightly farther than this and say, "don't waste that effort on a detailed critique of a badly flawed paper where you wouldn't trust the authors to fix those flaws without you needing to see it again." Journal papers have scope for multiple rounds of revise and referee; conference papers don't. – David Richerby May 31 '16 at 20:39
  • I don't remember which conference it was, but it was suggested reading for a review at some point - I was also very happy to find it! – Patrick Sanan Jun 1 '16 at 6:53
19

My interpretation of the question is as follows: you are asking if in your report you should spend time correcting all the grammar errors, spelling errors, etc., of which there are many of, especially since the paper will be rejected anyways.

First of all, the answer is completely independent of whether the paper will be rejected or not. The effort you put into a review should not depend on whether you intend to reject it or not - that defeats the entire purpose of a review.

Secondly, you claim that the paper will be rejected anyways, which implies that the errors in writing have no significant influence on the scientific content of said paper. If true, state this explicitly in your review. This ensures that no unwanted misunderstandings occur (since you shouldn't reject a paper due to language errors alone, at worst you should require a major revision prior to accepting it).

Now, should you list all/most of the errors? No. That would come off as quite condescending and probably not useful at all. Instead, you should mention what tendencies you see and use one or two examples to explain what you mean. Since there are so many mistakes that it'd take you an hour to write them all down, chances are that the authors' language skills are so bad that they need to consult a professional, potentially hire an editor to write the paper for them: mention this, again in a polite way, emphasizing that neglecting to do so is likely to harm their future papers as well.

You should also remember that if your feedback on their writing takes up more space than usual, perhaps you should consider giving them even more feedback on their actual content to balance things out, even though you've already clarified that their language errors aren't the reason for their rejection. This is because authors can sometimes be so vain that if you focus too much on their writing, they'd automatically assume that "oh yeah, they must have rejected me for my writing, look how much they focused on it", even if you explicitly state otherwise. You are not responsible for their assumptions, but a good review should be well-balanced and give off the proper signals.

9

I'll answer the question with a different interpretation: Is it worth it to cite every single typo, if there are many of them? IMHO, no.

If there are two or three typos that you saw, then you should list them. Otherwise, it is not only a significant waste of time, but the authors can end up relying on your list and missing stuff that you also missed. I usually say that it needs to be reviewed, because it contains several typos/errors.

9

Your question isn't about the level of language in fact. If I understand well, your real question is "Is it worth it to take 1 hour to correct a paper, since it will certainly be rejected ?"

At this question, I will always answer yes. I am personally a non-native english speaker and I love when people correct my work. I can then improve my English, which is always good.

What do you have to lose by doing it ? Nothing. And the person will even maybe be grateful to you.

  • sure. i have the exact same opinion as you. I always give concise feedback also on language-issues. however, this is more or less an extreme case where i am not sure anymore, if it makes sense to do so, because the author has foundational flaws in his writing style. hence, correcting each error one by one will likely not make him a better writer next time... – beta May 30 '16 at 13:54
  • 1
    Hum, what I would do : correct the non-basics fault and write a short text about what you think (if you have the right to do so) – Gautier C May 30 '16 at 14:02
  • 16
    "What do you have to lose by doing it ?" - Time that could be better spent on other things that are your responsibility, of course. – ff524 May 30 '16 at 14:06
  • 2
    It's his responsability too to give a review, and in a good review, there is always a concise feedback about language-issues. But I agree on the fact that it is maybe not worth the time wasted on it. – Gautier C May 30 '16 at 14:09
  • 3
    Sporadic mistakes can be pointed out in full. However, it is not the reviewer's job to do a language teachers' job of denoting every single language mistake; the latter would probably be paid for it. Point out a dozen errors and then say "there are many more language glitches; please have the paper proofread by a native speaker" or so. – Captain Emacs May 30 '16 at 16:13

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.