Duplicate submission of papers to academic journals is actively discouraged by journals and publishers. Many publishers have strict policies about duplicate submission, and the reuse of research data. For example, see the following article from Nature Neuroscience:
The journal itself states in its Guide to Authors that submission implies:
‘[…] that there is no significant overlap between the submitted manuscript and any other papers from the same authors under consideration or in press elsewhere.’
Duplicate submission (and subsequent publication) can cause confusion, and also wastes time during the peer review process. Papers which are published and then found to be published elsewhere risk being retracted, and may potentially cause reputational damage to the authors involved.
However, are there any circumstances under which duplicate submission is acceptable?
A useful resource is the ICMJE website, which lays out the main contexts in which duplicate submission may be encountered, and in which it could be acceptable (please see: http://www.icmje.org/recommendations/browse/publishing-and-editorial-issues/overlapping-publications.html, [accessed 24 April 2016]).
This resource is also important as the majority of academic journals follow the same or similar principles. These relate to:
Let’s summarize each of these:
1. Duplicate Submission
‘Authors should not submit the same manuscript, in the same or different languages, simultaneously to more than one journal.’
The reason for this is to avoid a situation in which both journals publish the same or similar paper (and putting the work into peer review and preparation for publication), and avoids disagreement between journals claiming the right to publish a paper that was received by each journal at the same time.
2. Duplicate Publication
‘Duplicate publication is publication of a paper that overlaps substantially with one already published, without clear, visible reference to the previous publication. Prior publication may include release of information in the public domain.’
If authors submit a paper that does overlap in the way described above, they must declare this to the journal they are submitting to (failure to do so may result in immediate rejection on discovery of the overlap). Duplicate publication can be problematic due to the risk of double-counting data.
Exceptions to this can include: publication of a full article where previously a preliminary paper or report was published, or papers from scientific presentations which were not published in full, or that are being considered for publication in a proceedings publication. Authors should declare these circumstances in their covering letter to their target journal.
Note also that there may be circumstances in which journals will consider joint publication where this is in the public interest, particularly on matters of public health.
3. Acceptable Secondary Publication
There may be circumstances where secondary publication is both useful and important, and generally this relates to public health contexts in which governmental or other institutions may wish to disseminate information to as wide an audience as possible, and in multiple languages.
It is important to follow journal guidelines, but the following circumstances may enable duplicate publication to go ahead:
- If the authors have received the formal approval from both journals to proceed.
- If both journals’ editors agree to a publication timeline, enabling the designated primary journal to publish the article first.
- If the paper for secondary publication is targeted towards a different audience. In such circumstances an abbreviated version could be sufficient. (The authors should discuss this with the editors.)
- If the secondary version reflects the data and interpretations of the primary version. (Again, authors should check with the editors of both journals on the acceptability of the dual publication.)
- If the secondary version of the paper declares that the paper has been published in whole or in part elsewhere—for example, with a note such as, ‘This article is based on a study first reported in the [journal title, with full reference]’ — and the secondary version cites the primary reference.
- If the title of the secondary publication clearly indicates that it is a secondary publication (complete or abridged republication or translation) of a primary publication.
Note that the National Library of Medicine does not consider translations to be republications and does not cite or index them when the original article was published in a journal that is indexed in MEDLINE. 
4. Manuscripts Based on the Same Database
It is not uncommon for multiple research groups to access and analyse the same databases. Journals will generally still consider such papers individually on merit since the study hypothesis or analytics may differ between the papers, and indeed the conclusions. If very similar papers are received by coincidence, it may be the case that the journal will give preference to the paper submitted first, but this could depend on other factors such as the quality of the analysis and write-up. Authors should also declare if they are aware of similar data analyses from other research groups.
Secondary analyses should always cite the primary source.
Our advice is always to consult the guidelines of the journal you are submitting to, and to declare in your covering letter if your paper, in any form, has been or is being considered for publication elsewhere and the circumstances for that. The editors can then advise if this contravenes their policies or conforms to valid exceptions which allow for multiple publication.