I recently submitted a paper for publication and received feedback that the paper was too complex. It contained two relatively independent concepts. The material would best be served by splitting it into two papers and submitting them independently.

I have to say I agree with this editorial analysis.

However, I am now stuck as to what to do. The first concept can stand on its own, but it is largely useless without the second one. The second one has to assume the first one exists so it can build upon it.

But, as I understand it, for a journal (or a good one, anyway) to publish an article, its research must be not only sound but properly justified as well. Only, I can't justify the research in the first article without referring to the second one, and I can't present the research in the second one without using the first one as a basis.

To further complicate the matters, I'm working at an intersection, and the two articles would probably work best if submitted to different journals in somewhat different fields.

How do I format my abstracts, introductions and conclusions to pass this message across and improve my chances of acceptance?

  • 8
    Arxiv, arxiv and arxiv. Upload both papers in arxiv, with each citing the other. Then submit the two papers in a journal, each citing the arxiv version of the other one. Of course you must find a journal that allows preprints on arxiv (e.g. Springer, ACM)
    – Alexandros
    Oct 8, 2015 at 11:49
  • Interesting. How do I go about finding if the journals I am aiming for allow such preprints? Oct 8, 2015 at 12:38
  • If published in the same journal, such double submission is known as companion papers.
    – Alex
    Oct 8, 2015 at 14:47
  • I am not sure of your use of the term "properly justified". Wouldn't it be enough to say that the results have applications in (your second paper) and are the subject of ongoing investigation (or independent investigation if you want to submit both papers at the same time). You are then free to deal with the reference to the second paper in any way you wish. You could, for example, state "submitted to ..." and update the status in your first paper as the second paper makes its way through the revision process. Oct 8, 2015 at 18:34
  • +1 for Alexandros's suggestion. (I have done the same with a combinatorics paper that had to be split in two, except that I put the whole long paper on arXiv instead of the two separate parts; this arguably led to the complication that I had to do referee-induced corrections twice, both on the papers-to-be-published and on the arXiv version. But it was worth it.) Oct 9, 2015 at 3:08

2 Answers 2


This is one of the things pre-print services can address quite readily, even for fields that don't normally use preprints. Putting both papers on arXiv (or another field-equivelent) allows them to exist in a citable form, and as one or both of them make it through the publication cycle, the linking citation can be updated to reflect this. FigShare will also let you upload documents and provide you with a DOI, though I've never actually seen this done in practice, and it's likely arXiv or an equivalent is more appropriate for a full-sized manuscript.

If you're in a field that has conference presentations and the like that aren't equivalent to publication (for example, much of biomedicine) you may also be able to cite a poster or presentation given at a conference.

  • Since Alexandros did not turn his comment into an answer, I'm accepting this one. Oct 13, 2015 at 15:18

If both paper would fit into the same journal, you could ask for a back-to-back publication of the papers. If that is not an option, you can independently submit the papers and cite the first papers as submitted in the second (editors usually do not like it). You would have to provide the first paper as part of the second submission to the reviewers.

  • "ask for a back-to-back publication of the papers" - could you elaborate what is meant by that? As I understand it, the primary issue here is that at the time of submission, it is not known yet whether, and in what form, the other paper will be published. Oct 8, 2015 at 11:49
  • 1
    As for "cite the first papers as submitted in the second (editors usually do not like it)", I'm not sure why editors wouldn't like it, given that this is a standard procedure even when submitting without circular references, if only to cope with the delay between submission and publication. This might be field-specific, though, as in my CS-subfield, there is no tradition of providing access to preprints of any sorts or using something like arxiv. Oct 8, 2015 at 11:51
  • I was under the impression that citing submitted work was only valid for papers already accepted for publication but not yet actually physically published. Oct 8, 2015 at 12:39
  • @JoãoMendes you can cite manuscripts "in preparation", if you wish so. But I find it useless (the cite cannot be checked) and annoying
    – Davidmh
    Oct 8, 2015 at 14:06
  • 6
    Perhaps I'm just old (fashioned), but journals used to happily publish 'series' of papers with Roman numbers in their titles to indicate where in the series they were (On Fizzbang I: Analysis of Fizzbang; On Fizzbang II: Application to SNAFU; On Fizzbang III: Universal Applicability). These could be back-to-back if all the reviewers were on time, or came out whenever they were ready. I'd contact the editor and see if this is a possibility for that journal, given the feedback.
    – Jon Custer
    Oct 8, 2015 at 14:21

You must log in to answer this question.

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