"Babes-Bolyai" University, Cluj-Napoca, Romania
In which self-plagiarism is concerned, considering the current state in the field, there are only two ways to go. The first way to go is to agree upon three minimal criteria for ethical writing (1. a new publication based on an old one is intended to target a new audience; 2. copyright laws are respected; and 3. it is made clear to the reader and in the author's CVs that the new paper reproduces old ones or parts of them) and to follow them in order to allow for the full expression of the humanistic spirit of science (i.e., disseminating knowledge produced to solve various problems). The second way to go is to elaborate clear rules and guidelines to avoid self-plagiarism, endorsed by all the major actors in the field; from that point on self-plagiarism can be considered misconduct. However, these rules cannot be applied retrospectively, to a time when they did not exist and/or were not lawful. All things considered, the current state of the field is unfair for scientists! As there are no clear lawful regulations regarding self-plagiarism, most scientists are like Schrodinger's cats, neither guilty nor not-guilty! It depends on who, on how, and on if someone is looking...!
I have read Dr. Roig's article carefully. I must say that I still have the feeling that the author is more focused on the form rather than the content of a scientific publication. In my view, the major and ultimate goal of science is to generate knowledge used to solve various theoretical or practical problems (see David, 2007). Thus, the dissemination of scientific knowledge through publications is the ultimate and one of the fundamental components of any scientific endeavor (David, 2007; David, 2008).
In my previous publications on this topic (David, 2007; 2008) I have pointed out that if we implement three minimal criteria, which Dr. Roig seems to agree with in his article, there is no need for other inquisitional, excessively detailed procedures (sic!), to avoid the so called self-plagiarism (i.e., stealing ideas from oneself). These three minimal criteria are as follows (David, 2007; 2008): (1) the new publication based on an old one is intended to target a new audience; (2) copyright laws are respected: (3) it is made clear to the reader and in the author's CVs that the new paper reproduces old ones or parts of old ones (and to what extent).
In his article, Dr. Roig says that he basically agrees with these criteria. However, when discussed in his paper, it becomes clear that they are not enough, and that we need detailed specifications on how to re-write our own work and ideas, reminding of inquisitional science (Chapman, 2007; David, 2007; 2008) etc. We analyze Dr. Roig's arguments as follows.
Concerning the first criterion, Dr. Roig says that in the age of the internet there is no need to target new audiences by publishing the same material in more than one Journal. We strongly disagree, as least from the perspective of someone coming from an Eastern European country, part of the former communist block. First of all, I would like to remind the readers that in less developed countries, and even in the newly emerging Eastern European democratic countries, there is no full access to the internet, or if there is access, people still cannot afford to pay the fees required for the full version of articles. Thus, some Journals are still on a market of their own, without yet being players on the global market of science. Second, at least in less developed countries, many Journals are not available online, and therefore articles they publish are not accessible via the internet! Therefore, the re-publication of an article in a different Journal, to target a new audience, is still legitimate. Moreover, during the communist period, this was a straightforward state policy, scientists being encouraged to publish some of their best papers both in the country and abroad! Obviously, this is not an argument, but just an interesting example.
Regarding the second criterion, Dr. Roig maintains that being allowed to re-use about 500 words (he exemplifies this using APA rules) in a new article, makes it impossible to use your own old text in a new one and still avoid copyright violation. Again, things are more nuanced. For example, Elseviere (www.elsevier.com), and many other Publishers, allows the author to prepare derivative works, to extend the journal article into a book, or to re-use portions or excerpts in other works, without asking for permission, but only with the acknowledgement of the original work.
Concerning to the third criterion, Dr. Roig rightly notices that the connection between the old and the new work is sometime not obvious, and that oftentimes, authors do not make it clear in their CVs that a paper duplicates another paper. We agree with him that this criterion must be fully satisfied, and we suggest the following ways to go about this issue: (1) provide full disclosure by mentioning if the new and/or derivative scientific work incorporates texts previously published; and (2) cite the old scientific work in the new one (David, 2007; 2008).
To make a long story short (see also David, 2008), we believe that ethical behavior in publication is to promote the humanistic spirit of science (i.e., to disseminate knowledge to the largest possible audience) following existing laws (e.g., copyright) and ethical writing, as circumscribed by the above mentioned three minimal criteria. Any effort to redefine ethical writing in an inquisitional manner (e.g., coming up with rules on how to re-write our own text to avoid text recycling; condemning and forbidding ab initio "salami slicing" and redundant/duplicated papers from one language to another or from a local Journal to an international one), leading to the limitation of the humanistic spirit of science, should be strongly resisted (David, 2008)!
Our view is that, considering the current state of the field, there are only two ways to go about this issue. The first way to go is to agree upon the three minimal criteria for ethical writing (as suggested by David and others) and to follow them in order to allow for the full expression of the humanistic spirit of science (i.e., disseminating knowledge produced to solve various problems). The second way to go is to elaborate clear rules and guidelines to avoid selfplagiarism (as Dr. Roig and others suggest), endorsed by all the major actors in the field; from that point on self-plagiarism can be considered misconduct. However, these rules cannot be applied retrospectively, to a time when they did not exist and/or were not lawful. All things considered, the current state of the field is unfair for scientists! There are no clear lawful regulations regarding selfplagiarism, and therefore, most scientists are like Schrodinger's cats, neither guilty nor not-guilty! It depends on who, on how, and on if someone is looking...!
Chapman, S. (2007). What's so precious about originality. British Medical Journal, 334, 1251.
David, D. (2007). Free science versus inquisitional science: Where to go? Revista de Politica Stiintei si Scientometrie (Journal for the Policiy of Science and Scientometrics), 5, 137-141.
David, D. (2008). Duplications spread the word to a wider audience. Nature, 452, (6), 29.
* Correspondence concerning this article should be addressed to:
REPLY TO DAVID'S "SCIENTISTS AS SCHRÖDINGER'S CAT"
St. John's University, New-York, USA
The rules regarding the various forms of self-plagiarism, as well as those that apply to other areas of responsible research conduct could always benefit from further clarification. However, it may not be possible to formulate guidance that covers every possible scenario. An ethically mindful attitude toward full disclosure and transparency in scientific research and publication may be more useful than the formulation of additional guidance.
In his commentary on my paper (this issue), Dr. Daniel David proposes two options with respect to the problem of self-plagiarism. He suggests that scientists be allowed to republish their work according to the 3 criteria he had previously outlined (David, 2007; 2008) or that clearer rules be formulated and endorsed by major scientific regulatory agencies and professional organizations, but that these should no be applied retroactively. I admit to being very sympathetic to Dr. David's concerns, particularly with respect to scientists working in nations where issues regarding responsible research conduct have not received their due attention. I also appreciated his clever use of the physics analogy that, because of a lack of consensus and/or clear guidance, scientists who recycle text or commit other forms of self-plagiarism may find themselves in the same position as Schrödinger's cat; not knowing whether they are guilty or not. The reader should note, however, that although clear guidelines are always desirable, they are unlikely to cover every possible scenario that can arise during the conduct of scientific research. As such, the Schrödinger's cat analogy can be equally applicable to a variety of other situations in which scientists must make decisions based on limited guidance.
Let's consider, for example, the International Committee of Medical Journal Editors' (ICMJE), (2007) rule of authorship in journal articles that states that only those individuals who have "made substantive intellectual contributions to a study" merit authorship. As with other forms of proper research conduct, most of us can identify instances that represent substantive contributions to a study (e.g., designing and executing a study; specifying and carrying out data analyses and interpretation, conducting a literature review and writing up a manuscript). Likewise, most scientists can point out a wide variety of nonessential contributions that would not merit authorship (e.g., entering or coding data for subsequent statistical analyses, administering study materials, and scoring questionnaires). There is little or no ambiguity in these examples and the decision to assign authorship in each of these cases would be relatively straightforward. However, in the day-to-day world of big, collaborative science there are a myriad of situations in which deciding whether a study contributor merits authorship in a paper can become a difficult dilemma. The fact is that the existing guidance on authorship, like that for text reuse as a form of self-plagiarism, raises a number of questions. For example, how many contributions deemed to be substantive merit a listing as an author of a study? Would one suffice? Do several nonessential contributions equal one substantive contribution? There have been a number of instances of authorship assignment disputes in the sciences and the problem of "gift authorship", though apparently common in the biomedical sciences, has long been recognized as problematic (Flanagin, Carey, Fontanarosa, Phillips, Pace, Lundberg, & Rennie, 1998, Smith, 1994). All of these concerns are relevant to the social sciences. For example, I recently reported some preliminary evidence suggesting that in the United States, students in psychology are sometimes included as authors of conference presentations even though some of them may not have earned it (Roig & Marks, 2006). One might dismiss this situation not only as harmless, but also as a great benefit for students who have an interest in embarking on a career in psychological science. But, others have long recognized the deceptive nature of this practice and its negative consequences (see Fine & Kurdek, 1993). At the professional level instances of inappropriate authorship can result in more serious negative consequences. A case in point occurred recently with a paper jointly authored by a team of researchers one of whom later admitted to have fabricated some of the data (Wade, 2005). Although a second researcher played no role in the fabrication, a thorough investigation by a panel of experts from his university revealed that his contributions to the paper did not merit him senior authorship and he was charged with "research misbehavior' for the ethical lapse (Wade, 2006). Thus, in the eyes of some, issues of authorship are not trivial matters. Neither is the problem of self-plagiarism in all of its forms.
Indeed, ambiguous situations can arise in many areas of research in which scientists are forced to exercise clear ethical judgment in the context of limited available guidance (e.g., the treatment of data outliers, decisions regarding statistical analyses). Let's again reconsider the issue of text reuse as a form of self-plagiarism. As noted earlier in my article, the existing guidance may vary somewhat (e.g., up to 10% reuse is acceptable in some journals whereas some authors suggest less than 30% reuse may be acceptable). However, in spite of any discrepancies, there should be a general understanding that certain practices are invariably wrong. Thus, copyright issues aside, it would be highly inappropriate for an author who successfully replicates one of his/her previously published studies to simply reuse the old publication verbatim, substitute the old data with the new data, and resubmit the paper to the same or different journal. I cannot imagine an editor who would knowingly accept such a submission even if the new paper makes casual reference to the earlier published study. But then, what about less blatant scenarios? How about if the author decides to reuse about three quarters of the prior publication or perhaps one half? Should we deem these as acceptable practices? Given the growing concern over unethical scientific conduct, it seems to me that most journal editors would not be in favor of accepting these types of manuscripts, even for authors with limited English proficiency. Of course, there should be no question in anyone's mind that translations of manuscripts or republication of a paper for the purpose of reaching a different audience is a perfectly acceptable strategy assuming that there is full disclosure as per the guidelines instituted by official bodies, such as the ICMJE.
I strongly believe that issues surrounding the question of self-plagiarism and other questionable research and publication practices are best approached within the context of mindful, ethical scholarship. Scholarship refers to that set of principles that establish the validity of our claims. These principles include the proper use of citations to enable the reader to consult the previously published works on which the current paper draws from, correct authorship assignment, appropriate reporting of results, as well as other elements some of which I discuss in my on-line instructional resource on ethical writing (Roig, 2006). Exercising great ethical care over our research and writing can go a long way in preventing us from finding ourselves in the same position as Schrödinger's cat.
I do agree with Dr. David that dissemination of our research findings should be the principal goal that guides publication ethics. However, as a form of scholarship, science has a long-standing set of rules and procedures that guide publication and these are continually evolving. Recent concerns over professional misconduct have led to an increased emphasis on ethical transparency in research and publication practices by, for example, the mandatory inclusion of statements in articles regarding the approval of animal and human subjects protection committees, full disclosure regarding potential conflicts of interest, and the listing of authors' contributions. Thus, given the current ethical climate, certain practices, such the verbatim recycling of large portions of our previously published text are not consistent with newly evolving notions of scholarly excellence and can be deemed problematic (again, Schrödinger's cat). Of course, as Dr. David clearly points out, there are situations in which republication and/or translations of works play an important role in informing non-English speaking scientists from emerging economies. Again, assuming full disclosure as per established rules discussed in my article, these instances of duplication should not in any way be considered cases of self-plagiarism.
We should also not overlook the fact that much of the concern and the development of guidance on matters of research integrity originate in the socioeconomic context of abundance, accessibility (e.g., internet, print journals), and linguistic dominance of Western nations, particularly the United Kingdom and the United States. One has to wonder whether our ideal of a responsible research climate is sufficiently sensitive to the often insurmountable challenges faced by scientists from less affluent nations. Be that as it may, the fact remains that most cases of self-plagiarism that are detailed in journals are covert in nature, are perpetrated by scientists from a wide spectrum of nations, including the United States, and suggest either ignorance on the part of the authors or a simple disregard for ethical standards. In more than a few cases, repeat offenders can be characterized by an attitude that is more concerned with the cranking out of publications to enhance the authors' vita for tenure and promotion purposes and/or to maximize their chances of obtaining funding, than with genuine efforts to advance science. I refer to this type of approach as assembly-line science: To produce as many knowledge products (i.e., publications) as economically as possible, regardless of the product's contributions to science. Such an approach is antithetical to the careful, thoughtful, approach expected of dedicated scholars and scientists and should be strongly discouraged.
David, D. (2007). Free science versus inquisitional science: Where to go? Revista de Politica Stiintei si Scientometrie (Journal for the Policy of Science and Scientometrics), 5, 137-141.
David, D. (2008). Duplication spreads the word to a wider audience [Letter to the editor]. Nature, 452(6), 29.
Fine, M. A., & Kurdek, L. A. (1993). Reflections of determining authorship credit and authorship order on faculty-student collaborations. American Psychologist, 48, 1141-1147.
International Committee of Medical Journal Editors (2007, October). Uniform requirements for manuscripts submitted to biomedical journals. Available at http://www.icmje.org/.
Roig, M. (2006). Avoiding plagiarism, self-plagiarism, and other questionable writing practices: A guide to ethical writing. Available at http://facpub.stjohns.edu/~roigm/plagiarism/.
Roig, M. (2007). Some reflections on plagiarism: the problem of paraphrasing in the sciences, European Science Editing, 33(2), 39-42.
Roig, M., & Marks, A. (March, 2006). An exploration of students' and professors' authorship contributions to EPA conference presentations. Poster paper presented at the 77th Annual Meeting of Eastern Psychological Association, Baltimore, MD.
Smith, J. (1994). Gift authorship: a poisoned chalice? British Medical Journal, 309, 1456-1457.
Flanagin, A., Carey, L. A., Fontanarosa, P. B., Phillips, S. G., Pace, B. P., Lundberg, G. D., & Rennie, D. (1998). Prevalence of articles with honorary authors and ghost authors in peer-reviewed medical journals. Journal of the American Medical Association, 15(3), 222-224.
Wade, N. (2005). Korean Scientist Said to Admit Fabrication in a Cloning Study, The New York Times, December 16th, p. A1.
Wade, N. (2006). University panel faults cloning co-author. The New York Times, February 11, p. A12.
* Correspondence concerning this article should be addressed to: