Resubmitting a Paper to a Mathematics Journal After Questionable Rejection

Question

Summary

I have recently submitted a paper to a fairly prestigious journal in my subfield of mathematics, along with a co-author. I am a junior author, but my co-author is well-established. It's not the top journal in the subfield, but it is well-considered -- one wouldn't think "wow, amazing job!" [getting a paper there], but more "that's a solid, reputable journal, congrats".

About a week ago we received a rejection letter, stating as the primary reason that it was not sufficiently novel -- there have been two other papers on a similar topic (although these are far from complete answers).

My co-author was pretty angry by this -- I have very little experience, and so don't really know how these things happen. They even emailed the editor questioning it (s/he says s/he has never done this before, but felt this was extreme enough to warrant doing so); the editor said that s/he did not discuss such things, and would not go against referees. (The editor has now finished his/her term.)

The journal uses an electronic-style submission, via an online management system. As such, I assume that the new editor would be able to see that I had submitted before; as such, it would seem the done thing for me to add a cover letter, explaining that this is a resubmission.

Regarding why I might want to stay with the same journal, it stands slightly alone in terms of reputation. In the field, there are a few top journals, say with ratings 9 or 10 out of 10; then there's this one, at say 7 out of 10; then the remainder are 5 out of 10 or below. I'm not sure my paper warrants the top journals, so I would like this one if possible.

Further Context; can be skipped if desired, with explicit question below

There were two full reviewers; one was very positive, but the other negative -- albeit I'm not convinced this person read it so carefully, as s/he makes a logical error in arguing that one proof is 'immediate', and questioned one theorem, but stated it incorrectly (the correct statement answers her/his question). Additionally, there were two 'informal reviews', "without scrutiny of the technical details".

Unfortunately, the result is rather subtle. (I presented it to someone, well-established, recently who at one point said, "It looks like it would be fairly straightforward, but it is actually quite subtle.") It appears that at least one of the 'informal reviewers' (actually, the more positive of the two) didn't get the subtly, suggesting that there was only a specific difference between two other papers, but actually they are extremely different. (I'm not claiming my paper is better than those -- in fact, I submitted it to a less prestigious journal -- just different.)

The main point of rejection by the editor, though, was "originality of problem posing, ideas and methods -- in light of the existing results" (citing the two aforementioned papers). As stated above, the paper is actually really quite different from the ones in the field -- it's not just me that feels this -- but in order to really appreciate this, one needs to have a reasonable understanding of the paper, or at least the other papers.

I feel that perhaps we did not do a good enough job of explaining the subtleties in the introduction. It is one of my first papers, and I did the writing, with comments from the other author. Since I am very familiar with the paper and the subtleties, I perhaps took it for granted that a reviewer would read it fully and carefully.

---

Explicit Question

Does rewriting the paper somewhat and resubmitting to the same journal sound like a reasonable idea, or should I just chalk it up to "the randomisation of the review process"? The editor has now changed, and the person that handled my paper is not an associate editor.

The positive reviewer gave lots of helpful comments -- the negative reviewer only gave a few, and (other than the one where s/he hadn't read the statement carefully enough) were typographical. I would implement these changes (including the typographical ones), where appropriate. But the major change would be in explaining subtleties in the model better, and comparing with other models, explaining key difference (that are perhaps obvious to me, who has spent 2yrs on this project, but not to someone giving an 'informal review').

Bear in mind that my co-author has already contacted the editor, who said "I have no reason whatsoever to question the professional authority of the associate editor and reviewers", and would not comment on any of our concerns. (I do understand her/his point of view here, to some extent.)

As I said, I originally assume we'd just resubmit elsewhere, but since my co-author thought it worth contacting the editor, who would not consider it, I wonder whether it's worth doing this. I would probably make the changes anyway, and if we got rejected we could just submit elsewhere after...

---

As you can see from this question, my writing can sometimes be rather long with unnecessary details, and the key details slightly hidden -- I am trying to improve this!

Incidentally, none of the people that looked at it noticed a fairly obvious mistake, where there was supposed to be a 3/2 factor... in the one display of the main theorem!

Answer 1

Resubmitting to same journal is a bad idea. I would move on.

A. You haven't been given any encouragement by the journal to do so. They are going to see same paper coming at them. Even if you can clean up the darned subtlety.

B. Just psychologically aren't you going to kick yourself if you stay at this journal and get bounced again? Take a new shot.

ALSO, take a look at the issue of previous work, how you cover it, how you differ, and rewrite the paper to better explain the compare/contrast.

EDIT, to expand versus several comments throughout whole thread:

1. MITM: It's OK whatever you do. After all, if it works for you "woot!" Of course if it doesn't, then "on your head, be it". Sounds like you will be fine to handle it either way. You are a sharp guy and producing.

2. My impression is that the pick a 7 journal was a reasonable stance when you first submitted. However, now, getting a negative result, you need to reevaluate. Think of this more like a real option decision tree in a probabilistic Bayesian universe than a Euclidean proof.

   This will sound flippant, but is a classic example from economics and math: When I wildcat a well, I make some judgement to take the risk. Fine. However, having returned a dry hole, sinking a second well in the SAME BLOCK is a VERY different calculus. I've learned something to change my prior. Of course, if the block is incredibly promising or oil prices are very high, I might sink another well before moving on. And I can (in parallel with your case) tell myself that I didn't have an optimal completion, etc. etc. (this time will be different, better, blabla). But bottom line is I got some negative feedback and need to adjust my prior.

   I liked the approach of aiming for a "7". But if it was in any way borderline at the time...well the "market" just gave you feedback. At this point, I would move on to a "surer block" and sink my next well in a "5" or "4" block. (Lower risk.) Don't go to a predatory journal. But I would be fine to going to more of a "datapoint collector" type journal. I don't know your field so well, but in chemistry, there are reputable journals that will still collect a lot of less interesting results.

   Perhaps you can go back in the future and argue about the subtlety or draw out some of the differences. In future work, reviews, talks, etc. But at this point, think it is more important to just get the finding out there.

   The good thing is nobody is kvetching about the math so at least the equations are right. (Well fix the 3/2. And, uh, take a good look since what else did you miss? But still, nobody is fighting the proofs.) But on notability. This is a harder thing to argue/fix even if you are "right". Sure, people could just be not following all the arcane subtleties. Or conversely you all could be doing the old math trick of trying to take credit for "generalizing/improving" other people's basic work. At a certain point, it really doesn't matter which (and I actually guess it is more the former). But just get the result out there, so it is in play.

   Move to a less prominent journal; get the result out there; collect your notch; and don't bang your head against a wall legalistically. And then move on to next piece of work. If you keep producing, good things will come. If you get diverted by reviewer debates and editor debates and the like, it may derail the production.

Answer 2

Fix the paper anyway

You got feedback, so implement it and (re)submit to Arxiv. Do think of a referee missing some points as feedback, and, to the extent convenient, try to address that, too. If they misinterpreted a theorem, should you give or recall some definition right before it, add an example after it, or emphasize how it is different from other results wherever you discuss prior art, or maybe even after stating the theorem?

Why not submit to a different journal?

If there exist other journals of roughly equal prestige and roughly similar readership, there seems to be little lost if you were to submit there, rather than in the original journal. If, on the other hand, no equivalent journals exist, I would still prefer going for a different journal if reasonable choices are to be found. Reasonable journal might be a good enough generalist journal or a slightly less reputable specialist journal - one would have the same weight in terms of prestige, while the other would have the same amount of exposure to the interested people, if your field is at all like mine.

Resubmitting to the same journal, and to a different editor (as seems necessary) without telling that that is indeed what you are doing seems somewhat impolite; not clearly wrong, but not quite something I would like to be caught doing, either. Submitting to the same journal with an explicit cover letter or similar would be ethically fine, but there might come a (hopefully quick) rejection.

Answer 3

If you can re-write the paper in such a way that you can address the reviewers concerns (novelty), then it is fine to resubmit it. But the paper shouldn't try to "argue" the case. Either the results are new(ish) or not. But I expect that novelty is a harder thing to address than some other things. Perhaps the reviewer just knows a lot about the field and thinks your results are minor. Rewriting to make things less subtle might help or not with this reviewer, but it probably would in general.

But you can, of course, resubmit elsewhere. But I would only suggest doing so if you can, in some way, address all reviewer suggestions.

But as second author to a more senior person, you may need to take their lead on this.