10

I have seen the questions here and here, but my reasons for wishing to withdraw are utterly different than theirs, so I don't think they're especially relevant.

I work in mathematics, say in field A. In recent years, field A has found a number of interactions with another part of mathematics; call it field B. About 7 or 8 months ago I finished a project that was an application of some results from field A to a question in field B. Despite using some results from field A, the resulting paper (both in techniques and scope) was essentially purely a paper of field B, and so I submitted it for publication in a journal J from field B.

Last week, while working on a completely unrelated project in field A, I realized that the techniques I had used in the abovementioned paper can be translated, essentially word-for-word, to get some results in field A. Unfortunately these results are no longer at all in the scope of journal J, as they are really applications to field A rather than field B. Thus, I thought the right thing to do would be to withdraw the submission from journal J, merge the two results into a single paper, and then publish in a journal that lies more in the intersection of field A and field B. So a few days ago I emailed journal J to inform them of the situation and request withdrawal of my submission.

Unfortunately, I think that email has not yet arrived to the editor handling my submissions, because I have now received an email informing me that my paper has been accepted to J modulo minor revisions and enclosing two referee reports. The timing is really unfortunate; I wish I had realized the application to field A a few months ago rather than just last week.

I still think the right thing to do is to withdraw the paper. However, I feel terrible for wasting the time of the journal, and especially of the two referees, who have put a great deal of time into reviewing the paper! Is it unethical to withdraw? More generally, is it insulting to the referees? If so, what is the best way to handle this situation?

Paper Withdrawer
  • 109
  • 1
  • 4
  • 3
    What do you mean "modulo revisions"? – Azor Ahai -him- Apr 17 '23 at 01:42
  • 9
    Just out of curiosity, if you'd published the paper first and then realised it also had applications in field A, would you feel these new results deserved a new paper, separate from the one you already published? If so they might deserve a new paper anyway. – N. Virgo Apr 17 '23 at 10:08
  • 2
    Your current paper was obviously good enough to publish on its own, why do you think adding new observations is necessary? Are the new observations not sufficient in scope for a paper of their own? What wold you have done with these observations if you had made them after the first paper was published? – Cris Luengo Apr 17 '23 at 15:45
  • 3
    @AzorAhai this is mathematics slang, see here. – Dan Romik Apr 17 '23 at 16:03
  • @DanRomik To be honest, I have no idea how "In mathematics, the term modulo ("with respect to a modulus of", the Latin ablative of modulus which itself means "a small measure") is often used to assert that two distinct mathematical objects can be regarded as equivalent—if their difference is accounted for by an additional factor." really fits in here – Azor Ahai -him- Apr 18 '23 at 01:13
  • @AzorAhai-him- https://english.stackexchange.com/q/70018/193515 – GoodDeeds Apr 18 '23 at 14:21
  • 1
    @AzorAhai-him- The submitted and accepted versions of the paper can be regarded as equivalent, their difference being accounted for by revisions. – Servaes Apr 18 '23 at 14:44
  • 2
    @Servaes Thanks for actually explaining. However, I don't really understand how that makes sense to state - either they are the same or not. – Azor Ahai -him- Apr 18 '23 at 15:19
  • 4
    I agree with @AzorAhai-him- that the math jargon is a little opaque to people outside mathematics. I would recommend editing it out of the title because this Q and A might be valuable to people outside math that will be confused by the use of the term modulo – qdread Apr 18 '23 at 16:01
  • 1
    @AzorAhai-him- Two things can be the same on some level, but different on a more refined level. Two cars can be the same model, even if their colour is different. They are essentially the same thing, modulo paint. The same goes for the submitted and the accepted papers, in case of very minor corrections. They are essentially the same, the differences being (superficial?) revisions. I think that in this informal context, the use of 'modulo' suggests that the difference is small or negligable in some sense. – Servaes Apr 18 '23 at 22:44
  • @Servaes It seems like it could just be replaced in both instances with "with minor". Am I wrong? – Azor Ahai -him- Apr 19 '23 at 00:19
  • @Servaes I was asking about the question, not your example. – Azor Ahai -him- Apr 19 '23 at 03:07
  • @AzorAhai-him- "My paper has been accepted with minor revisions" could be taken to mean that the revisions have been made and the paper has been accepted. "My paper has been accepted modulo minor revisions" means that the revisions have not yet been made. – Servaes Apr 19 '23 at 18:40
  • @Servaes Then I invite you to edit the post to be clear to someone whose training isn't in math :) (although I don't think I'd ever read "accepted with minor revisions" in that way but it doesn't matter) – Azor Ahai -him- Apr 20 '23 at 00:27

2 Answers2

34

The paper is yours until you release it to a publisher. But, I suggest that you consider what is best for the advancement of understanding, rather than the wishes of a particular publisher.

However, it might be best on all accounts if you let the current paper be published by this publisher and write a follow up paper, referencing the current one as needed.

The reason behind this suggestion is that the current paper is relevant, I hope, to those in field B, and they might benefit from seeing it and might not if it is published elsewhere.

You might have additional work to do, however, to achieve sufficient distinction between two papers to have them both published.

But, think of how best to advance scholarship, not the economics of a particular outlet.

Buffy
  • 363,966
  • 84
  • 956
  • 1,406
  • thank you Buffy. this is a good point about field B, and in fact in many respects maybe I should prioritize the connection there. I think it is however definitely mutually exclusive to publish the current version in journal J and publish the new observations somewhere in a journal of field A; indeed this would essentially feel like "salami publishing" to me and I definitely wish to avoid that. what about the following compromise: publish the current version in journal J, and then just put the new version with the content on field A on arXiv? does that seem like a sensible idea? – Paper Withdrawer Apr 16 '23 at 14:29
  • 1
    That might be a good short term solution so that you don't get scooped on the ideas. But longer term, you might think about extending to a full paper for field A. – Buffy Apr 16 '23 at 14:49
  • 3
    exactly my gut feeling after reading the post. – Walter Apr 16 '23 at 17:37
  • 10
    On a side note, for OP personally it might also be better to have two papers instead of just one. There are issues with salami-slicing results but this seems to be a case where having two distinct papers is justified. – quarague Apr 17 '23 at 00:03
  • 3
    @PaperWithdrawer If field A journals and field B journals don't overlap much, having separate publications saying "you can use technique X for A/B" seems justified, as the intended audience wouldn't be otherwise aware -- especially if it's not immediately obvious that the results generalize. (Which is likely the case if you, the expert, didn't realize it until after submission.) If it feels too much like salami slicing, you can always go more "in depth" in the field A publication, looking at corollaries and special cases which might be A-specific. – R.M. Apr 17 '23 at 17:00
  • 3
    @PaperWithdrawer There's a huge difference between intentional "salami publishing" of several papers in a row and actually realizing there was more to this one paper after submission. Plus, you might want to check if there's a journal in field A that accepts short publications (sometimes called "letters" or "communications"), maybe that's an option? – Sabine Apr 17 '23 at 18:11
2

"I used research already known/published in field B to show that it works really well/is useful in field A" is, on its own, a very important (albeit arguably not innovative) paper to publish, separate to "Look at this method/theorem in field B".

Ander Biguri
  • 5,571
  • 2
  • 28
  • 37