How appropriate is to make multiple submissions of research findings to same venue/conference?

Question

Assuming the following scenario:

Person A is a recognized senior researcher, and Person B is an enthusiastic junior researcher. Both of them have been working more or less equally on a research project and are interested in submit the results of the project to a particular venue. Based on the report written by Person B, Person A has written a paper (PA) to be submitted to the venue. Person B doesn’t like the style and some contents of PA, so he makes comments to Person A, who them modifies PA. Person B still doesn’t like PA, so writes a separate paper PB. Since PA and PB haven’t been published, there are citations between them.

At this point, Person A submits PA, which list authors as "Person A & Person B", and Person B submits PB, which list authors as "Person B & Person A."

1. Is it appropriate to submit PA and PB?
2. What would be your impression if you were part of the venue’s committee?
3. As reviewer, what would be your impression about reviewing two different papers about the same work for the same venue?
4. Any additional consideration?

Answer 1

Most venues forbid simultaneous submissions, or overlapping submissions. For instance, in the call for papers of ACM CCS 2012:

Submitted papers must not substantially overlap papers that have been published or that are simultaneously submitted to a journal, conference or workshop. Simultaneous submission of the same work is not allowed. Note that submitted papers cannot be withdrawn from the process after the first phase reviews are received by authors.

So, in order to answer your question, the main problem is determining the overlap between PA and PB. If it's substantial (only some elements of style change, minor comments), then it should be forbidden.

On the other hand, if the papers are different enough (for instance, PA presents a more theoretical approach and PB a more practical description, with a case study), then there is no problem. That being said, both papers must be self-contained, and cannot rely on the acceptance of the other, unless they are both also published as technical reports (or equivalent).

To make things explicit, if I were a reviewer of a member of the PC, and if I were to think that the overlap between the two papers is substantial, I would recommend rejecting both.

Answer 2

You cannot submit both papers. As others have pointed out, most venues explicitly forbid simultaneous submission of substantially overlapping papers. The answer to your first three questions is the same: Immediate rejection.

In fact, I would go further: Neither PA nor PB can be submitted alone. It is unethical to submit a paper without the explicit consent of all authors, because authorship implies endorsement of and responsibility for the content of the paper. On the other hand, it is unethical to revoke authorship from someone who has made a substantial intellectual contribution to the work. The two authors must work out their editorial differences, like grown-up professional adults, before they can submit anything.

If they can't agree on what to submit, then tough noogies — no paper.

Answer 3

You cannot submit to papers on the same data and that is what this sounds like. It doesn't sound like paper A is the data and paper B is a model.

I would suggest Person B concede to the wishes of Person A and then never collaborate with Person A again.

Answer 4

By venue, do you mean a conference? Depending on the conference, you could certainly focus on different aspects of the same study, or different (but not contradictory) conclusions from the same data (some datasets generate dozens of papers!). There could be a bit of overlap, i.e. in the introduction, but not too much. I'd say the most important thing is to be clear and honest about what you want to present. But if you have together done a large project and are going to write a peer-reviewed paper about it, I don't see any problem in presenting different aspects of the study at the same conference.

Of course, they should not be exactly the same.