Self-Plagiarism: Can You Steal From Yourself?
By Marianna Evola
A few weeks ago, after one of my responsible research presentations, a student attendee approached me as the room cleared. The graduate student was head of his departmental graduate student organization and he requested that I come and speak with his student group about self-plagiarism. I smiled and calmly accepted the invitation while throwing an internal tantrum. Of all of the topics associated with responsible research, self-plagiarism is one of the most difficult issues to address in detail. It is easy to incorporate the issue of self-plagiarism into a broad presentation on responsible research because you can vaguely assert that students must be aware of self-plagiarism as a critical error to avoid while publishing. It is much more difficult to flush out the topic and provide students with solid boundaries and guidelines, because as I discussed in the past, there is much debate and ambivalence about self-plagiarism as an ethical violation. That being said, I took the challenge to talk to the students about this complex ethical issue and thought that it would be efficient to utilize this month's SM column to get my thoughts organized.
When I present on plagiarism, I define plagiarism as theft of words or theft of ideas, basically plagiarism is theft of intellectual property. This is a very straight forward way of thinking about plagiarism. Students understand theft and graduate students can readily internalize the thought of someone stealing their ideas. Graduate students have reached a point in their professional development that they have begun to realize that as scholars, their ideas are a key component of their identity. Thus, it would be a very personal offense if someone were to steal their ideas or creations. Once you get students to internalize the personal nature of the offense, they recognize how repugnant plagiarism is to academics. However, if plagiarism is easily defined as theft, the question remains, "Can you steal from yourself?"
As such, I've realized that I've often left out a key second component to the definition of plagiarism (Merriam-Webster, online dictionary): "To commit literary theft: present as new and original an idea or product derived from an existing source." It is this component of plagiarism that addresses the offense of self-plagiarism because if authors are attempting to deceive their audience into believing that redundant work is original, when they recycle text or submit duplicate publications, they are plagiarizing. As such, a lot of the questions about the ethics of self-plagiarism in the discussion below will revolve around whether the author is attempting to deceive their audience.
The source of self-plagiarism that I frequently address is the need of researchers to recycle the methods section of their publications. In my previous column on self-plagiarism, I described a heated debate amongst academics in the sciences vs. the humanities and the scientists insisted that it was a service to the reader to keep their methodology consistent between publications. There are few changes made between experiments, generally only one variable is changed from one experiment to the next. So between experiments, methods do not markedly change. By keeping your published methods consistent, you enable the reader to cover a long series of experiments/publications very quickly because they only need to assess which variables were changed as the experiments progressed. As a scientist, I know that it is very difficult to write a clear, concise methods section and once it is written, it is efficient for both the author(s) and the reader(s) to keep the methods consistent between publications. This is a historical practice in the sciences. As such, there is no deception on behalf of the author/s when they recycle their methods section between publications. In fact, there is an expectation in the sciences that the publication text of the methods will remain consistent for a research group. We expect text recycling when we read methodology. We do not expect originality. Therefore, the authors are not being deceptive and arguably, the text recycling of methods is not interpreted as self-plagiarism. That being said, if methods are recycled, authors should (and generally do) cite the original source of the text.
Another source of self-plagiarism is duplicate and redundant publication. Once again, there are legitimate reasons for duplicate publications. Often when collaborators publish a unique collaborative project, they are trying to reach two or more distinct audiences, and it can be difficult to select a single publication that will reach the diverse audience. For example if a collaboration was examining the relationship between genetics and behavior, the geneticists would want to reach the audience interested in genetics. However, the behaviorists would want to reach the audience interested in behavior. As such, authors sometimes want to publish in multiple journals so that they can reach both audiences. Or, another example is when international scientists originally publish in English but want to serve their native audience as well, As such, they want to duplicate their publication by submitting the work to a journal that is published in their native language. Arguably, both of these academic agendas are legitimate, the scholars want to reach an audience that could easily miss their work if it is only published in a source or language that is not frequented by a particular audience. The key to legitimate duplicate publication is transparency. The authors should be overtly informing the publishers and editors that the work is published elsewhere and they proposing duplicate publication to reach a different audience. Transparency enables the editors and publishers to decide if duplicate publication is warranted and to obtain the permission to publish from the original publisher without violating copyright. Transparency demonstrates that the author/s are not trying to deceptively self-plagiarize, rather they are attempting to legitimately get their message to a distinct audience.
In contrast, if the author(s) are merely trying to increase their publication count by submitting to as many journals as possible, and they are not disclosing to publishers, editors or their audience that the work has previously been published elsewhere, they are engaging in the unethical behavior of self-plagiarism. Their deception demonstrates that their agenda is unethical and thus they are self-plagiarizing. And contrary to the assertion that self-plagiarism is a victimless offense, it is actually damaging academic misconduct. Self-plagiarism in the form of duplicate publication wastes the time of colleagues by requiring multiple reviewers to review the same work for different publications. Furthermore, duplicate publication wastes the time and resources of editors and publishers, time that should be spent on original work. Duplicate publication consumes critical publication space in journals, and robs other authors of the opportunity to publish original research.
Finally, one of the biggest research concerns of undisclosed duplicate publication is the incorporation of duplicate data into meta-analyses. A meta-analysis is a systematic review of data published in the literature. They are large studies that combine the published data of several smaller studies to increase the sample size and statistical power associated with a large question within a discipline. When authors fail to disclose duplicate publication, researchers that conduct meta-analyses on the literature blindly incorporate duplicate studies into their analysis. Incorporating data from duplicate publication, skews the results of meta-analyses which can invalidate the study. Thus, asserting that self-plagiarism is a victimless crime is false. When authors engage in a deceptive practice of redundant publication, they are harming their colleagues and their discipline. However, if they are fully transparent that the work has been previously published, respectfully argue that there is a need to reach a broader audience and appropriately obtain permission to republish respecting copyright, the action of duplicate publication can be a service to research and discovery.
Another source of self-plagiarism comes from "salami-slicing" or breaking up a large study into a collection of smaller studies so that the author(s) can increase the number of publications. Often in cases such as these, much of the introduction, rationale and interpretation for the research is duplicated because there was a singular rationale that led to a series of experiments. Rather than publish the series of experiments as a larger comprehensive study, the researchers choose to break the studies up into little trivial bites of information. This publishing strategy is wasteful. Similar to duplicate publication, it wastes the time of reviewers, editors and readers and unnecessarily consumes valuable publication space. It also can distort the perceived value of work by over representing a line of research in the literature.
There are right reasons for engaging in text recycling and duplicate publication and there is a right way to go about text recycling and duplicate publication. If the authors are transparent in their practice and agenda and are not attempting to deceive their editors, publishers or audience, they are not self- plagiarizing. Rather, they are serving the academic world. However, if there is an intent to deceive for the selfish purpose of maximizing their publication count, they are engaging in a damaging and unethical practice and as such, deserve the disfavor of their colleagues and the reputation damage that their unethical practices will eventually incur.
As such, the boundaries and practices that I will define for students would be:
- Never try to deceive your audience to increase your publication count, especially in a modern world full of plagiarism detection software. You will be caught.
- Always be transparent if you are attempting publishing work that you know has been previously published. Transparency is critical for both text recycling of methods sections (i.e., citation) and legitimate duplicate publication. Transparency elicits trust. Secrecy elicits suspicion and will cause your actions to be interpreted as deceitful.
Marianne Evola is the director of the Human Research Protection Program in the Office of Research & Innovation. She is a monthly contributor to Scholarly Messenger.