In the Dark

The topic came up in a recent conversation of the ethical issues surrounding what is sometimes erroneously called self-plagiarism, but is more accurately called duplicate publication (or multiple publication or even redundant publication). This refers to the situation in which an author publishing their own intellectual material (specifically research results) more than once in different journals or other media. This is distinct from plagiarism, which involves an author publishing someone else’s intellectual material without attribution. It is also distinct from copyright violation, which can occur if the author tries to re-use material already published in a journal that has retained the copyright; the solution in that case is simply not to publish in a journal that does that.

Publication practice differs widely in different academic fields so in what follows I’ll concentrate on what applies in Physics & Astronomy. Here there is one type of publication, the Conference Proceedings, in which papers are often near-duplicates of others. That is because speakers tend to give the same or very similar talks at different conferences, and also tend to recycle material when writing up their contributions. I see nothing particularly wrong in that, although one wonders whether a plethora of versions of the same talk is needed. I stopped writing conference papers over a decade ago as they take a lot of time to do and I don’t think they fulfil any useful purpose. In any case such articles should not count as research publications, especially if they are not peer-reviewed (which is generally the case in Astronomy). I know this is different in other fields. In Computer Science, for example, the conference article is one of the main modes of research publication.

The more serious issue is when a researcher publishes (or tries to publish) multiple versions of the same research in different journals in an attempt to pad out their publication list by passing off old material as original research. This is difficult to do nowadays because of plagiarism detection software, but not all journals deploy such tools and some cases do get through the editorial process and make it into the journal as a publication. Sometimes this even happens with high-profile journals.

The question is how one reacts to this kind of multiple publication. I did a totally unscientific social media poll recently and the results were quite interesting. Of my respondents, about 20% said that they thought multiple publication was fine. About 30% thought that multiple publication constituted academic misconduct, and about 50% thought that it wasn’t fine but fell short of academic misconduct.

I suppose the definition of research misconduct varies from one institution to another. For reference here is what it says in Maynooth University’s Research Integrity Policy statement:

Publication of multiplier papers based on the same set(s) or sub-set(s) of data is not acceptable, except where there is full cross-referencing within the papers. An author who submits substantially similar work to more than one publisher must disclose this to the publishers at the time of submission.

The document also specifically refers to “Artificially proliferating publications” as an example of research misconduct.

In the past I would have posted a poll on here but I now have to pay $15 per month for the privilege of hosting a poll so with regret I’ve unblocked my Twitter account to let you vote there:

One reason people might be tempted to indulge in multiple publication stems from the fact that the current system of research assessment depends so much on bibliometric indicators relating to refereed publications. While I regret the emphasis on bibliometrics, I do think that multiple publication of research papers is indeed academic misconduct because artificially boosting the number of such items on one’s CV might be a way of gaming the system. It seems to me that such a strategy is unlikely to work, but I have seen people try it.