I peer-reviewed a manuscript and am about to send my conclusions to the editor. It occurs to me, though, that because there are many details to be checked/changed in various places of the paper (mathematical notations, in particular), it will take quite some time for me to type it into a proper review (probably using TeX, because it is math-heavy).

So, I am wondering: since my hand-writing is decent and people don't usually find it too hard to decipher, could I simply scan my annotated copy of the paper, along with one page of notes, to the editor? Or will that be unacceptable to them?

I think it raises a few questions, going from practicalities:

  • will it be as useful as a thoroughly written-down review?
  • does it give more work to the editor?
  • should I also include a summary of my comments, in computerized form?

to ethical questions:

  • does handwritten notes breach rules of anonymity?

I suppose people used to do that in older times, but it have never received a hand-written review so far, so it is (at least in my field) unusual.

So: should I do it? if I do it, what precautions should I take?

  • 4
    Is your handwriting so distinctive that you suspect the authors will be able to identify it? Nov 12, 2013 at 13:04
  • 1
    I have received hand-written reviews twice, on a sample of ca. 20 submitted papers in applied mathematics. Once it was comments from a (not-anonymous) editor; the other time it was supposed to be anonymous. It ended up quite easy to identify, but it probably would have been identifiable even from a pdf. The two were from the same workgroup, so they are not really "independent samples". Nov 12, 2013 at 18:26
  • 4
    Many reviews, at least in math, are of mediocre quality -- based on a skimming rather than a thorough reading of the paper. In contrast, it sounds like you are writing a review that both the editors and the authors will be extremely grateful for.
    – Anonymous
    Aug 20, 2014 at 13:23
  • 2
    Re your last point: you could also worry about anonymity being breached not because of your handwriting, but because your scanner will quite probably add metadata to the scanned image. Aug 21, 2014 at 11:14
  • @BSteinhurst The style of handwriting can give away a lot about the writer, including: where they learned to write (country/region), how old they are, their gender. If the field they work in is small enough, it will surely help in guessing their identity. But then there are plenty of other things that can give away the review anyway, compared to which handwriting might be a small thing.
    – Szabolcs
    Aug 20, 2015 at 11:29

6 Answers 6


I do not think that there is anything wrong with a scanned handwritten report. As for anonymity: you may even sign your referee report as it is your decision to stay anonymous or not. As for usefulness: I once received an annotated scanned manuscript as a referee report at is was tremendously helpful (as there were several suggestions for formulations which greatly increased the readability). A computerized summary would also be helpful for the editor to form the decision and also for the authors to find out what you main points are (so that they can distinguish between just typo corrections and serious remarks).

  • 4
    In a blind peer-review process, I think revealing your identity without the editor's consent is not allowed
    – F'x
    Nov 12, 2013 at 12:13
  • Not sure - I've heard somebody say that he always writes reports non-anonymously (see the answers to the questions here mathoverflow.net/questions/98308/…).
    – Dirk
    Nov 12, 2013 at 12:26
  • 2
    then that would be a nice separate question (/me winks)
    – F'x
    Nov 12, 2013 at 12:29

A hand written scanned review is acceptable, if readable. But, the time spent on typing it into LaTeX (or any other suitable format) to produce a pdf is strongly advised. The reason is that even the most neat hand-writing, can include letters, symbols, writing details, that can be ambiguous to others (particular of other nationalities). So although, such reviews are acceptable, providing it in digital format reduces the risk of misreading and misinterpretation. In addition, a scanned hand-written review may be conceived as the result of someone not caring too much (however wrong that conclusion may be). So in the interest of clarity, I would suggest spending the time typing the review in.

  • 1
    agreed. I once received a handwritten referee report that was basically illegible, more due to the actual scanning process than the referee's handwriting. but it basically took me twice as long to respond to the report as it would have done otherwise because I spent a ton of time trying to decipher the handwriting.
    – user10636
    Aug 19, 2015 at 15:05

If your copy of the paper is in pdf format, then you can attach typed notes to it fairly easily. A program called Skim does that, and I think modern pdf viewers also have that capability. (It would be good for me if more people used that option, so that authors who get a referee report of that sort can't easily infer that I'm the referee. My own handwriting has become so lousy that handwritten notes are no longer an option.)

  • 2
    Are you sure that people can't identify you from a Skim'ed PDF? I'd wonder about Skim silently adding metadata. Aug 21, 2014 at 11:11

If your handwriting is clear and you make a lot of small corrections, I would personally prefer to get your notes in the paper itself, scanned. The reason is that it's much easier to understand what's going on, and much easier to see where the mistakes have appeared. I would say that adding a separate report that contains long remarks (on structural things, mistakes in math proofs etc.) is a good idea and should be done, since long text is hard to squeeze into the page margins1.

Just a remark to finish: please, use a red pen and make a color scan.

I don't agree with people saying that doing a lot of corrections is necessarily wrong (well, it's another question). Many reviewers do a poor job, and many reviewers are too sensitive. But it's very likely that a good paper needs a lot of corrections.

1 Ask Fermat, he knows something about this.


I have received hand-marked manuscript review previously, and it did not raise an issue to neither the editor nor myself and my co-authors. However there is a point you should consider:

Is it really a substantial and reliable paper if it needs such an extensive correction even in mathematical parts? I have been rejected on the basis of too many typos or too poor grammar, which I don't think should necessarily be a deciding factor. I will only reject people based on bad grammar only if it significantly hurts the understanding of the paper's material.

Extensively erroneous mathematics sounds something like a substantial problem with the quality of the research and not only the quality of presentation.

  • 1
    I don't agree with the last paragraph. "Mathematical typos" are very common and are not necessarily cause for rejection. As a reviewer, I've submitted reports with multiple pages of mathematical corrections and a recommendation of "strongly accept". And if I were an editor, I'd be ticked off if a reviewer suggested rejection of an otherwise strong paper because of easily corrected minor errors. Aug 20, 2015 at 8:02

One point that no-one has mentioned yet: after receiving your review, the authors will (probably) take your comments into account, make changes, and then write a letter explaining the changes which they submit along with the revised version.

In their letter explaining changes, it is common (at least in my field) to answer reviewer's comments point-by-point and to include verbatim the reviewer's questions, e.g.

  • Reviewer 1, Comment 2: "Jones et al. (2001) needed more froobiz to reach the wiznish point. Why do you find a different value?"
    • In this study we used an improved floopnosh technique which inherently requires less froobiz; see also the review by Smith (2004).

So, if the authors decide to go this route and you submit a hand-written review, they will have to TeX-ify your review document, possibly making errors in the process.

  • Well, I don't expect authors' comment on every small thing I pointed out. usually something like "We thanks for the minor suggestions from the reviewer, we considered them". After all, I don't expect the authors to agree with everything I suggest, but I expect them to want their paper to be good, so I expect them to check all my suggestions.
    – yo'
    Aug 20, 2015 at 9:58
  • No, I definitely agree that authors shouldn't agree with every point a reviewer makes. But when the authors disagree, they should state why. And preferrably with something (like a citation) to back it up. Some times, the paper is improved by adding a sentence like "Here, one might think that <insert reviewer comment>, but as was shown in <recent paper/equation X/section N>, this is not correct." Aug 20, 2015 at 12:52

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .