Over at Daily Nous there’s some excerpts of advice on refereeing papers. Much of the quoted advice is things not to do, and one caught my eye:
It is not your job to be copy editor…
…an apropos comment as earlier today I’d been collating my notes on a paper to turn it into a referee report, and the thought crossed my mind “refereeing would be a lot less tedious if I didn’t feel the need to NOT point out grammar and punctuation errors”.
And yet, despite the advice quoted above and my own personal inclination, I believe there are reasons why it is appropriate for referees to provide this sort of feedback (which is why I do it myself).
- When I first graded undergraduate philosophy essays and marked them up for poor grammar, spelling, punctuation, etc., I often got objections along the lines of “Why are you grading my English, this isn’t an English class, you should just grade my arguments, not my language”. My response to this always was: It is not always possible to separate form from content. If your sentences are not grammatically constructed, this will often make your argument unclear, or even incomprehensible if it is not obvious how to rehabilitate the sentence into something grammatical. Misspelled words can introduce ambiguities. Misplaced commas can substantially alter the sense of a sentence. All of these play a role in the strength of the argument.
- Proofreading your own writing is not only tedious, it’s actually very difficult: It is far easier to spot errors in someone else’s writing than it is to spot them in your own. I know that no matter how thorough I try to be, I will overlook duplicated words, missing words, incorrect tenses, subject-predicate match, etc., and thus I appreciate it greatly when someone else catches them for me. So, I proofread as part of my referee reports as a consequence of adopting a sort of Golden Rule of refereeing: I do unto others as I would have done unto me.
- I’m in the enviable position of being a native speaker of English. It is often clear that the person whose paper I’m reading is not. I thus have a leg up on them when it comes to information crucial to producing a publishable paper. They may not have easy access to someone who has this information; it would not be kind of me to withhold it.
- One of the purposes of a referee report is to provide the editor and the author(s) with answers to the questions of “Is this publishable as submitted?” and “Is this publishable after revisions?” For a paper which has grammar, spelling, and punctuation problems, the answer to the first question will be no, even if the arguments are all impeccable. If the answer to the second question is ‘yes’, it would be churlish of me not to identify to both the editor and the author the revisions I see necessary, and that includes the results of proofreading.
All of these reasons are cancellable, such that I wouldn’t go so far as to say that a referee should always copy-edit as they review. But as both an editor and a receiver of referee reports, I am always glad to see, and grateful to receive, detailed comments at the linguistic level.
Which is why as a referee, I should probably end this post and go back to the paper I was working on this morning, and finish typing up all my notes, copy-editing and otherwise!
Letters from the Editors 
I see your point and appreciate your kindness. Yet, I would suggest that there is no need to make an accurate linguistic copy-editing of articles, because:
1) some reviewers (me included) would not be able to do it in a reliable way (and you may not want to exclude them from the already limited pool of reviewers).
2) some reviewers might be able to do so, but would not accept to review articles if they knew they had to also copy-edit the article (and you may not want to exclude them from the already limited pool of reviewers).
3) some reviewers might be able and willing to copy-edit, BUT would then be induced by the number of misspellings, etc. to think that the article is not publishable although it is.
Thus, I would rather suggest to reviewers who are not as kind as you are (and as competent as you are) that they may just add a note saying something like “The English form needs to be checked by a native speaker” (which is, by the way, what I do on top of correcting the minor things I feel confident about).
As I said at the end of the post, all the reasons I gave can be cancelled, and I don’t want to say that all referees should provide this feedback; your comments only support this conclusion. I just don’t think it’s appropriate to say that referees shouldn’t give this sort of commentary.