For conferences that "peer review" and "publish" full papers, when, if ever, can you resubmit essentially unchanged versions for publication in a peer reviewed journal?

In my field, when conference papers appear as book chapters, people often "republish" them as journal articles.

I have just had a talk accepted at a conference which is now (decided post submission) planning on publishing the proceedings as a special issue of a journal. It appears the journal is complete crap with an extremely light peer review process in general, no impact factor, and essentially not indexed. I think this "article" will be worthless, but I am concerned that I will not be able to republish the results in a respectable journal.

  • 5
    The "special issue of a journal" designation sounds worrisome to me. Even if it is just the usual conference proceedings repackaged, I think you are right that it will cause problems for later publication. (Even if the second journal agrees not to count the first as a "real" journal publication, it may still appear to onlookers like you are publishing the same paper in two journals.) Commented Jul 30, 2012 at 11:54
  • 7
    You always have the option not to submit to the post-proceedings, or the associated special-issue in this case. If you are worried about crappiness of the venue which is effectively imposed on you, simply do not submit your work to the journal and inform the editors about the decision. After all, nobody can publish your paper without your consent, such as a formal copyright transfer in the case of "traditional" journals. Just save your good stuff for a better venue.
    – walkmanyi
    Commented Jul 30, 2012 at 15:42

2 Answers 2


In math, FPSAC (Formal Power Series and Algebraic Combinatorics) accepts and referees "extended abstracts" for presentations. Presenters are selected based (in part) on the quality of these abstracts, and the abstracts appear in a special issue of DMTCS. The submission guidelines for FPSAC state

The authors will retain the right of publishing a full version of their work in another journal. Authors who do intend to publish a full version elsewhere should however make sure that their conference contribution is clearly an extended abstract of this full version.

Several conference centers publish proceedings of their workshops, for which presenters are asked to submit extended abstracts of their talks after the fact; for example, both AIM (the American Institute of Mathematics) and Oberwolfach (formally known as Mathematisches Forschungsinstitut Oberwolfach) do this. Again, the norm is that the abstract should be a summary of your presentation, with less detail than the published version.

I think that the main answer here is going to be "see if the conference organizers have made a statement, and check with some more senior people as to what the unwritten norms are".


This varies by field. In (most) fields that have journal-based publications and no tradition of selective refereed conferences, it might be quite hard to publish a paper in a journal after it has appeared in some proceedings, without substantial changes.

In computer science, in which conference publication is the norm, it is expected that journal papers have first appeared in conference proceedings. This is especially true if the full paper does not fit in the page-limit of the conference proceedings, and proofs or other material had to be left out of the first publication. Typically you have to make some small changes between the conference and the journal version (i.e. include full proofs, include a fuller discussion of related and subsequent work, etc), but in CS, the delta between conference and journal version can be relatively minor.

  • What happens if a CS conference publishes the page-limited papers in crappy journal?
    – StrongBad
    Commented Jul 30, 2012 at 11:52
  • It probably depends on the specifics. If it actually appears as a paper in a journal, that might be problematic. On the other hand, plenty of conferences publish their proceedings using journal publishers like Springer, and its fine. Hard to know more without knowing the specific situation.
    – Aaron
    Commented Jul 30, 2012 at 12:07
  • 2
    In some fields of CS, the delta can be relatively minor. In theoretical computer science, the delta can be arbitrarily close to zero (although this is starting to change). Other subfields of CS impose a lower bound on the amount of truly unpublished content (typically 25%-33%) in any journal paper.
    – JeffE
    Commented Jul 30, 2012 at 13:26
  • 6
    Related: The question about the delta on CStheory
    – Gopi
    Commented Jul 30, 2012 at 13:57

You must log in to answer this question.

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