Volume: 130(11) - November, 2008
Recycling used resources for reuse is always good, except when it comes to our own research work. Republishing our own research should be done with extreme caution.
Research is the generation of new knowledge, and as such it is always based on previously generated knowledge, including our own. It is natural when we write a new paper to include references, results, or background material from other work we have done ourselves as well as work of others. The advent of the wonderful electronic publishing and particularly cut-and-paste technology has increased our ability to reuse material from various sources, including our own intellectual product.
However, there are some fundamental tenets of scholarship: We cannot publish the same work in more than one publication venue—implying each time that it is new, even by omission. Also we cannot publish other scholars’ work—implying it is ours, even by omission. Outright use of other people’s work almost verbatim and without attribution is very rare, although it happens. These cases are easy to assess, and the follow-up editorial actions may include banning authors from publishing for several years, and contacting the authors’ employer. However, the boundaries in other cases can be occasionally fuzzy, so I will give you some examples.
Multiple submission of the same research to different journals is an absolute no-no. The community of scholars, as you dig into specific topics, is surprisingly small, and the chances are high that a reviewer will become aware of a multiple submission and alert the editors. When I face such an allegation, my immediate action is to reject the paper and alert the authors giving them the opportunity to clarify things. Since I may have no access to the other submission and since I cannot share the JMD submission with a third party (other than JMD associate editors and reviewers), I have no objective way to test the allegation. Rejecting the paper is a drastic action that gets everyone’s attention, reviewers and authors. The manuscripts do not need to be identical; substantial overlap is sufficient to trigger this action. The authors have the opportunity to argue their case and give me access to the non-JMD submission and permission to share it with all the reviewers assigned to their JMD submission. Then I can ask the reviewers and the associate editor to determine whether a multiple submission has occurred. Based on their input, I can re-instate the paper or maintain its rejection. In a repeat offense, I may ban an author from publishing in JMD for a period of time, as it is customary with some other journals. I will also alert the other journal editors.
ASME has a policy to allow papers published in the ASME conferences to be submitted to ASME journals, which is the idea of transactions. However, an ASME conference is an archival publication and cannot be submitted to a non-ASME journal without: (i) asking ASME for copyright release; (ii) stating to the non-ASME journal editor that the paper has appeared in an ASME conference. Such submissions may make sense if an ASME transactions journal editor has rejected the paper or found the topic unsuitable. Still, we must communicate this information to the other journal.
A more ambiguous situation is when a paper is reviewed and rejected by one ASME journal and then submitted to another—without alerting the new editor that this has happened. Formally, there is no requirement for such disclosure. Journal scopes often overlap and the editors often make paper re-assignments prior to initiating reviews. Occasionally, this may be appropriate after a review is completed, even if negative. However, the appearance of “shopping around” for a journal that will accept a paper is never a good one. Alerting the editor of the paper’s history is always a good idea, and it does not prejudice the next review, particularly if the paper has been improved given the previous input.
Another ambiguous example is the amount of cut-and-paste you can do from one paper to the other, particularly if they are related and some information must be repeated to make the papers reasonably self-sufficient, as reviewers occasionally request. We can use a couple of criteria here: (i) Is there a clearly new idea or information in each paper that warrants a separate publication? (ii) Is the cut-and-pasted material a minor portion of the overall paper’s length? If yes, we should be OK. Mind you that changing the wording, rather than straight cut-and-paste, does not change how we use these criteria.
As these examples demonstrate, there is considerable judgment that we must exercise each time, and two people may have differing interpretations. In papers with multiple authors, some less experienced, all authors need to be engaged in these judgments. We are not perfect and neither is our judgment. Paying conscious attention to these issues is always important for all authors. Reviewers are also authors and they need to be paying attention to another set of issues, as I will comment in a future editorial.