The January 2012 issue of Biosystems Engineering has a commendably thorough retraction notice regarding a case of plagiarism in its pages.
The notice, regarding the article “Advanced techniques for Weed and crop identification for site specific Weed management,” by Karan Singh, K.N. Agrawal, and Ganesh C. Bora, of North Dakota State University, speaks for itself:
This article has been retracted: please see Elsevier Policy on Article Withdrawal (http://www.elsevier.com/locate/withdrawalpolicy)
This article has been retracted at the request of the journal editors.
It has been brought to the attention of the editors that this review paper has been constructed, in substantial part, by verbatim copying of paragraphs etc. from papers that had already been published in scientific journals. The copied papers include:
Gee, C., Bossu, J., Jones, G., and Truchetet, F. (2008). Crop/weed discrimination in perspective agronomic images. Computers and Electronics in Agriculture, 60(1), 49–59. doi:10.1016/j.compag.2007.06.003
Longchamps, L., Panneton, B., Samson, G., Leroux, G. D., and Thériault, R. (2009). Discrimination of corn, grasses and dicot weeds by their UV-induced fluorescence spectral signature. Precision Agriculture, 11(2), 181–197. doi:10.1007/s11119-009-9126-0
Manh, A.-G., Rabatel, G., Assemat, L. and Aldon, M. -J. (2001). Weed leaf image segmentation by deformable templates. Journal of Agricultural Engineering Research, 80(2), 139–146. doi:10.1006/jaer.2001.0725
Okamoto, H., Murata, T., Kataoka, T., and Hata, S. (2006). Plant classification for weed detection using hyperspectral imaging with wavelet analysis. Weed Biology and Management, 7, 31–37. doi:10.1111/j.1445-6664.2006.00234.x
Piron, A., Leemans, V., Kleynen, O., Lebeau, F., and Destain, M.-F. (2008). Selection of the most efficient wavelength bands for discriminating weeds from crop. Computers and Electronics in Agriculture, 62, 141–148. doi:10.1016/j.compag.2007.12.007
Tellaeche, A., Burgos Artizzu, X. P., Pajares, G., Ribeiro, A. and Fernández-Quintanilla, C. (2008). A new vision-based approach to differential spraying in precision agriculture. Computers and Electronics in Agriculture, 60, 144–155 (authors incorrectly listed by Singh et al by forenames not family names). doi:10.1016/j.compag.2007.07.008
Wang, N., Zhang, N., Wei, J., Stoll, Q., and Peterson, D. E. (2007). A real-time, embedded, weed-detection system for use in wheat fields. Biosystems Engineering, 98(3), 276–285. doi:10.1016/j.biosystemseng.2007.08.007
The editors apologise to the authors and publishers of all papers copied and to the journal’s readers for the inappropriate publication of this paper. One of the conditions of submission of a paper for publication is that authors declare explicitly that their work is original and has not appeared in a publication elsewhere. Re-use of any data should be appropriately cited. As such this article represents a severe abuse of the scientific publishing system. The scientific community takes a very strong view on this matter and apologies are offered to readers of the journal that this was not detected during the submission process.
Perhaps not coincidentally, the same issue of the journal also has an editorial decrying the “on-going problem” of plagiarism.
The editorialists, including editor in chief Bill Day, state:
Plagiarism, the practice of taking someone else’s work or ideas and passing them off as one’s own, and its ancillary self-plagiarism, in which individuals republish work that they have already published, represent significant challenges to scientific journals. Authors have a right to be acknowledged as the source of their own work, and new authors must present their work in their own words. For more details and examples, readers are encouraged to look at the website of the Committee on Publication Ethics at www.publicationethics.org.
The answers to two questions that are regularly raised with the editors of Biosystems Engineering may help explain some of the problems: firstly, “Can conference papers be published in the journal?” and secondly, “Surely the words of the original author are the best way to express the science I need to quote in my paper?”
This journal has traditionally been prepared to consider publishing papers that are developed from work published at scientific conferences – with the proviso that the journal paper must be of greater depth and breadth in its science, and repetition must be minimised. Conference papers are publications – though they may be less accessible, they have been made available to the scientific community, and increasingly are distributed electronically and placed on the Internet. However, this is apparently not always understood. Papers have been submitted that are nearly identical to a conference paper, sometimes openly acknowledging the conference paper but sometimes not even referring to it. Such extreme examples of self-plagiarism are a form of redundant publication. It is not acceptable practice, and such papers will be rejected when detected.
In principle, journals expect papers to be written in the author’s own words. It is their interpretation of the science that is important, and using their own words demonstrates understanding, so significant strings of words should not appear from other published works. Of course some repetition will arise by chance and some because standard phrases or descriptions of equipment or methods need to be reused. However this is not a justification for extracting text from the introduction, review, results or discussion of other papers. If it is important to use the actual words of another author, they should be put in quotation marks and be clearly referenced.
The availability of software tools like iThenticate has made it possible to quantify the extent to which text from previously published papers has been copied. Where the copying of significant amounts of text is detected during the review process, authors may be contacted and asked to amend their text where appropriate. If the extent of copying compromises the integrity of the submission as new work, the paper will be rejected. Occasionally it is discovered that a paper that has been published in the journal includes extensive plagiarised text. In such cases the editors will challenge the authors and will, where appropriate, issue a correction or retract the paper. Retracting papers is not undertaken lightly, but it is important for the journal and its authors that we maintain the highest standards for this publication.
The swift retraction — the now-withdrawn paper was published in May of this year — seems to have prevented the plagiarists from stealing any credit for their “work.” The paper hasn’t been yet been cited, while the oldest of the papers it plagiarized has been cited 41 times, according to Thomson Scientific’s Web of Knowledge.
A basic task nowadays in reviewing any paper is to look for very ‘well-written’ sentences, and stick them in a Google search to look for exact or close matches. I have caught several articles this way. Other tools are available, but copying-and-pasting out of the .pdf for review and into Google is quick and trivial. A writing style is like a fingerprint, and if reviewers see changes in writing styles within a document, Google the suspect sentences.
I agree.
I think programs like this should be utilized by editors before they send the paper out to reviewers. I remember back in my undergraduate physics labs: There was some sort of program we had to paste our lab reports into before we submitted them online. This program would presumably compare your text to the existing database of other reports, and a score would be returned to you. If you were below a certain threshold, then you could submit it.
Sure, this would probably involve some extra time for the editors, but they get paid to organize studies for the journal anyway.
You really have to question the intelligence of a person who plagarizes a paper in such a poor manner that it can be discovered in minutes.
How stupid are they?
The notice indicates that “It has been brought to the attention of the editors that this review paper has been constructed, in substantial part, by verbatim copying of paragraphs etc. from papers that had already been published in scientific journals.” This suggests that this is not the detective work of the editors, but of someone else. I wonder who?
Yes, Joe, “you really have to question the intelligence of a person”…but think of all the work and stress that went into stealing from not one or two papers, but at least seven different papers…that really took some time and effort…stupid, yes, but really, really hard working.
And Sierra, “stick them in a Google search to look for exact or close matches”…excellent idea. I’ve been in the habit of using very short strings on Google; what is the maximum # of characters you can “stick” in a search?
Apparently Google restricts queries to 32 words, and will often remove connecting words (e.g., and, or, etc.) to get to the limit. I’ve had success picking 1-2 sentence long strings that look suspicious (i.e., flawlessly constructed sentences that come after reading a few sentences that look like they were written by someone whose first language is not english – and are clearly not the original work of the same writer) and searching these.
What is most bizarre is why Elsevier does not have a formal system to pick up this stuff upon submission. All manuscripts submitted to the Elsevier system must be in .doc format, and since all the journals listed in the retraction notice above are also Elsevier journals with full-text online, one wonders why the submission system doesn’t auto-flag any manuscripts where strings of text in the paper match strings of text in already published papers. Over time, I’m sure such a system would pay for itself.
Thanks. “There oughta be a law”: Elsevier could/should have a system for detecting plagiarism. Doesn’t look good to have a private individual running plagiarism software on their own time and finding what the journal overlooked.
Also, I’m under the impression that if you put a query into quotes, like “Four score and seven years ago”, the whole string is taken as a literal to match rather than using the individual words–thus, if I search for myself, I quote “Conrad Seitz” and I get better matching. Might be a way to keep bridge words like “and” in the search.
I’ll speak from my experience. I suspect plagiarism is not rigorously taught in schools in Korea, unless you are in university level humanities, sociology, and science courses that specifically teach how to write research papers. The very concept of giving credit where credit is due seems to be something entirely foreign to much of the Korean population.
Knowledge and appropriateness of paraphrasing, citing, primary sources, secondary sources, differences between original research versus reviews, bibliographies, footnotes, writing critically just isn’t there in many students I have come across. Is it just me or does someone have a similar experience in parts of the world outside of Western culture?
I would say that Western culture also has a problem in these areas, so perhaps the concepts are not being as rigorously taught here as one might hope. Leaving out the folks who are deliberately cheating, there are many Western students who are hazy on the concept of plagiarism. They will say, “It’s okay. I got it from the web,” is if words on the web are not written by someone who deserves credit (or by a person paid by an organization that deserves credit). They will ask, “How many words do I have to change?” and they will change “before” to “prior to” in an attempt to reduce the number of words copied verbatim from a source, as if substituting a couple of synonyms constitutes original writing. The difference between original research and a review paper, well, that cuts awfully close to the bone for many in the academic grove. Those jokes about moving bones from one graveyard to another have a basis in fact. Not to knock reviews, because they certainly have an important place, but do you know people who built their careers on writing reviews rather than producing original research? I’d say Western culture has a ways to go.
JKR asks if anyone else has experience regarding researchers’ lack of knowledge of appropriate acknowledgement of sources and citation.
The researchers I work with do not have English as their first language. Inaccurate citation remains a frequent issue in all settings I’ve worked in, and occurs in manuscripts written by researchers at any point in their career–from grad students through very senior tenured people with administrative responsibilities.
But despite its frequency, I have never in any case seen evidence that the motive for re-using text copied from somewhere else was to deceive readers or steal credit for the ideas–although this could be a motivation in highly competitive areas.
Since the early 80s I’ve worked on manuscripts with researchers in Spain, and more recently on manuscripts by researchers from Eastern Mediterranean countries. It is not unusual to find sentences or even paragraphs copied verbatim from other published articles. Usually a reference will be given, but often there will be no quotation marks. Recently a very senior author (in the West) whose manuscript had this problem justified the lack of quotation marks by saying they were “not needed because the reference was given”.
The google check Sierra Rayne describes is something I do frequently while editing manuscripts, so that I can document to the authors where their copied-and-pasted text came from. Occassionally, the reference given in the manuscript is to an article different from the one I find on the net as the source. In these cases I ask the author to recheck the reference and correct it as necessary. I also wonder whether the copied part has been previously re-used (or plagiarized) by the authors cited, who were then in turn “re-plagiarized” by the authors I happen to be working with–who where honestly unaware that the authors they cite had themselves “borrowed” the text without providing appropriate attribution.
Researchers do need better education and training in acknowledgment and citing (as part of training in good writing, reporting and publication practices), but they also need better role models. Sloppy citation is very common in published articles, and researchers often find cases of missing or incorrect references, missing quotation marks and the like when they do their own literature search and review. Perhaps they assume that if the Big Names can get away with it, why can’t they get away with it too?
“Authors have a right to be acknowledged as the source of their own work”
And how does “self-plagiarism” violate this?
Jon, self plagiarism is a tough one because the argument can be made that “I wrote it”. But the true ethical reasoning lies with the original publisher. If I write something that gets published in Journal A and then copy and paste my words to Journal B, it takes away from the original journal. The journals compete with the content and originality of their words, so I must cite myself. Then the reader can go to that original journal for further information.
Right, but just because publishers lobbied for certain copyright rules does not mean that they are right or that we should accept them uncritically as absolute ethical rules. They are not and they should be changed. Authors should have ample latitude in reusing their own work. I would not agree with copying a review article verbatim, but using paragraphs, especially if rearranged in a new flow should not be a problem.
The issue of “self-plagiarism” is made up by publishers trying to protect an obsolete and unfair business model. Authors should hold the copyright to their work, end of story. Luckily, that’s the direction in which things are moving with open access journals.
Sue, I do not believe the copyright is the issue. I can copyright anything I write, but as soon as I turn it over to a publisher I have given them the right to produce it. If I give that same right for the exact text to someone else (especially if they don’t know about the first publisher) I have violated a copyright law. Copyright is for original text. It is no longer original if it has already been copyrighted/published.
That is not a very defensible position. I am aware of a person who published the same article in very remote and different locations (Australian journal, Swedish journal), and did it to get more on the CV. After a while, the CV was good, and a better job came around. In other cases, the same stuff is published in marketing and statistics areas. It’s resume padding.
First, we are not talking about empirical papers. Republishing the same empirical paper is never acceptable as it is misleading and it may create problems with meta-analyses and all that. Second, republishing substantial parts of one’s own review articles should be allowed, especially if they are placed in a different context. The reason for that is not just to pad one’s CV, it’s to have a larger audience read your ideas. All journals have a limited audience, especially when people have to pay to read the articles. Publishing similar reviews in journals that are read by different audiences, especially if they are fine-tuned for the different audiences, is perfectly reasonable in my opinion. I want scientists to spend their time doing good research, not waste it on High School English paraphrasing projects to make the “self-plagiarism” police happy. 🙂
Paul, when I have published I have always had to sign something that says what I am submitting has not been published somewhere else. Publishers do not mind if you write about what you have already written about and you can cite yourself all day long, but re-submitting an exhisting paper is plagiarism whether it is your own work or not. I know someone that “padded” their resume and they were fired based on the ethics code of the institution.
Self-plagiarism seems like a silly concept to me. Logically one cannot take one’s own words and falsely pass them off as one’s own, nor can one re-use one’s own words without permission from the original author. Copying one’s own output is copyright violation if someone else (say, a journal) owns the copyright. Redundant publication, even if one owns the copyright, is justly frowned upon as a sneaky attempt to inflate one’s resumé.
It is unfortunate that ethical proscriptions against lying and stealing are enforced these days only for economic reasons like wanting to sell copies of a journal or a book. I’m not suggesting a return to the bad old days when an absence of copyright laws meant that authors couldn’t legally protect themselves against unauthorized publications (see Miguel de Cervantes as a wonderful example). But I do suggest that we should make a distinction between the ethical sins of plagiarism and resumé enhancement and the legal/economic sin of copyright violation. The concept of self-plagiarism seems like a word-game intended to smear a mere legal violation with the connotations justly attached to ethical violations. The two categories overlap at times, but we should be clear about what scientists oppose on ethical grounds and what publishers oppose on economic grounds.
I see self plagiarism as an ethical problem aside from the legalistic issues of copyright holding. If you write something and publish it once, you are creating an original work. If you then publish it again in a different journal without attribution, it is no longer original. Even if you’re not doing it to pad your resume, you are misleading the readers of the second work. It’s a subtle thing.
Contrast that with a fiction writer, who can publish a short story in a magazine and then republish the same story in a book as part of a collection of stories. In that case, the frontispiece will contain a listing of previous publication dates and locations. This is accepted. The critical thing is that you are not misleading your readers as to the originality of the publication.
“you are misleading the readers of the second work”
How so? If the reader of your second work ALREADY have seen your first work they will think WTF and so they will not be misled. If the reader of your second work has not seen your first work, then that proves that publishing a second version in a different journal was worthy because some people had not read it! So, please stop your little witch hunt, since there is no justifiable rationale for it.
Such is not a “witch hunt”, and in many cases, the witches actually exist and they need to be hunted. Self-plagiarism is just lazy slop – nothing more, nothing less. And self-plagiarism is economically inefficient. No reader should have to pay to read effectively the same article twice, and the only way you find out if such is the case is by having to purchase the second article. Can I (or my institution) get my money back for such duplicate publication? I thought not.
In the 21st century with the comprehensiveness of the e-search tools available, trying to claim that some journal review article isn’t appropriately ‘covered’, and needs duplicate publication in multiple venues, is nonsense.
If the only way you can get people to notice your work is by plagiarizing yourself and effectively reprinting your work in various different journals, then bad luck.
One notices that these ‘tactics’ are more common in the biomedical and agricultural applications sectors than they are in other traditional fields of research.
I agree Conrad, plagiarism is an ethical issue. If it were jailable we would probably have about 25% (or more) of students populating our jails. LOL The key to your post is “original work”. No one is saying you can’t cite yourself, but to copy and paste for further publication is wrong. I believe it was Seirra that stated it was lazy slop; I have to agree. If a writer/researcher can’t find anything additional to write about their origianal work, they need to move to a different subject.
Publishing the same paper twice is similar to inventing a widget and giving the rights to distribute it to Company A. Then contacting Company B to distribute your widget without Company A or B knowing of each other. This is also not a jailable offense, but could get you quite a law suit. Witch hunt, Jon B? Hardly. It is ethically wrong and shows the lack of moral character of the individuals that partake in this practice.
Agreed. And another issue with self-plagiarization is who actually wrote the original words being plagiarized. In many cases, review authors plagiarize text from prior articles (both primary research and review manuscripts) that they may have been co-authors on, but which they did not write.
I fully agree that so-called self-plagiarism is an ethical problem aside from the legal issue of copyright violation. But I think it should be called what it is: redundant publication or duplicate publication or multiple publication of the same stuff. I don’t see how self-plagiarism is logically possible.
And I completely agree that misleading readers is wrong. My argument is with what people call it when an author misleads readers in a certain way versus in another way.
JudyH, I totally agree with you!
With which part? That the word “self-plagiarism” is silly or that the activity itself is unethical no matter what it is called? Both, I hope.
Are you saying, JudyH, that “self-plagiarism” is derogatory in an excessive way? We could use any word, I suppose, to avoid even the appearance of bias; I think that is appropriate. There are so many kinds of redundant publication, from the lifting of paragraphs to using the same experimental results over again. Just lifting a few paragraphs from yourself is a venal offense; reporting the same experiment twice (especially if it is disguised as something new) is more of a capital crime.
In this case, it appears that the responsible parties have elevated plagiarism to an industry, if not an art form. Seven, count ’em, seven separate papers.
Dr. Seitz, my objection is to what seems to be an internal contradiction in the word and to the bleeding-over of justifiable negative connotations from a heineous sin to a lapse of proper procedure. I warn readers other than Dr. Seitz that the following comment may be longer than they care to read.
Plagiarism is the theft of a felicitous arrangement of words (or a brilliant idea) with the intent of pretending that the thief produced the sentence or paragraph or treatise (or idea) him/herself. As expressed in most definitions and in the publisher’s definition associated with the retraction of the Singh et al. weed paper, plagiarism is theft from another person. There is no need to steal from oneself. If I can think up a graceful, insightful, and original sentence, I can think it up again. And in any case it is my sentence, the product of my brain. If I think it up only once but I use it a second time, I am not pretending that it is my own sentence. It ismy own sentence. Based on this traditional definition of plagiarism, I cannot see how self-plagiarism is a logical concept.
The practice of re-using previously thought-up sentences becomes objectionable in two instances. (Or maybe three. As the Monty Python troupe says, nobody expects the Spanish Inquisition.) One is when the copyright to such sentences has been sold (or extorted from the author by some coercive method such as refusal to publish a scientific paper unless the author transfers copyright to the journal) to another party. This party now owns the right to copy the sentences, to redistribute them for money, and to sell this right to yet another party. If I sell my graceful, insightful, and original sentence to one publisher and then sell it again to another publisher, the first buyer complains. But this is copyright violation, not plagiarism. It is a commercial violation. In selling the sentence a second time, I have not stolen it from the original author nor have I used it without the original author’s permission. I have given myself permission to sell it again. The fact that this permission is no longer mine to give throws the act into the realm of copyright violation. It is an ethical violation, to be sure, and one that you, Dr. Seitz, term a venial sin. In the gift economy of science, the capital crime would be theft of another scientist’s gift to the community, the theft of another scientist’s original sentence or brilliant discovery. In this latter case I would be hoping to capitalize on the value of someone else’s sentence or discovery to falsely enhance my own reputation in the community. In the case of re-using my own sentence, I am not expecting to profit duplicitously from another’s creation. Any reputational enhancement that I gain is what I have earned because the sentence or idea in question is my own creation. The publisher is annoyed, to be sure, because s/he has lost sales that s/he was expecting to make on the basis of having a unique product. But the scientific community is not harmed.
A second instance in which the re-use of one’s own material is objectionable is, as you point out, when the author passes off the re-use as a new thing under the sun. Reporting the same experiment twice with the intention of misleading readers as to the number of experiments that were performed does harm to the scientific community by falsely increasing the weight of one set of results within the total pool of reported results. As you say, this is a capital crime. But it is not plagiarism. Nothing has been stolen from the original author nor reported a second time without the original author’s permission. As in the example of selling the same sentence to two publishers, the publisher is annoyed because s/he has lost sales of what was advertised as a unique product. But the publisher is not truly concerned about the damage to science. Scientific publishers are in business because scientific publishing is a profitable enterprise, not because they are selflessly advancing the cause of science.
A third instance (aha, Monty Python was right) in which the re-use of one’s own material is objectionable is the partial or complete re-publication of a journal article in order to inflate one’s resumé. I see this as different from the previous instance because there is not really an intent to mislead the scientific community about discoveries. The intent is to live up to the standard of output established by some authority for purposes other than science. I specifically exempt from this category the translation of previously published articles into another language. In agreement with Jon Beckmann in previous comments, I see no problem with this as long as the re-publication is clearly marked, both for the readers of the translated article and for the institutional inspectors of the author’s resumé, who are counting beans rather than evaluating scientific worth. Also in agreement with Jon Beckmann, I do not object to re-use of large chunks of one’s own prose if they are incorporated into articles for a significantly different audience, as long as these chunks are flagged with a phrase such as “as I have explained elsewhere (Me, 1982)” or something similar. This is not plagiarism, because the re-used chunks of prose are my own and I have given myself permission to re-use them. A publisher would regard this as copyright violation because s/he holds the copyright. But as in the example of selling the same sentence to two publishers, the scientific community is not harmed. In contrast to these exceptions, re-use of material is objectionable when it inflates a publication record and thereby allows some scientists to defeat others in the competition for industrial jobs, academic positions, and research funding. The bean-counting aspect of today’s culture is deplorable and I deplore it as much as anybody can. There are more routes to resumé inflation than simply re-publishing one’s own output, and these should be discountenanced along with re-publication. But again, re-publishing is not plagiarism. It is an ethical violation, yes, but we need another word for it. If “duplication publication” is not sufficiently explicit, I’m open to other words. Let’s think on it. We should be able to come up with something from the Greco-Roman lexicon.
Now, six victims. Only six? The pikers! I could show you some master’s theses and doctoral dissertations that put Singh et al. to shame. I have just recently explained to a graduate student that he cannot construct an entire thesis introduction out of paragraphs copied nearly verbatim from other people, even if he does carefully cite each source. Original writing is necessary. Our Western system is not very good at explaining and enforcing this when I can point out both foreign and domestic students who have not understood the criterion for acceptable performance.
RE: “Plagiarism kills weed paper”: http://www.retractionwatch.com/2011/12/28/plagiarism-kills-weed-paper
Ganesh Bora lists a similar article by the same three authors under his Publications and Reports webpage at North Dakota State University: http://www.ndsu.edu/aben/publications/ganesh_c_bora
“K. Agrawal, K. Singh, G. C. Bora and Don Lin. 2012. Weed Recognition Using Image-Processing Tecnique Based on Leaf Parameters. Journal of Agricultural Science and Technology, Vol.2 (8B): 899-908.”
Here’s the Google Scholar cache of the HTML: http://scholar.googleusercontent.com/scholar?q=cache:_P0IKI6xgHgJ:scholar.google.com/+Weed+Recognition+Using+Image-Processing+Tecnique+Based+on+Leaf+Parameters&hl=en&as_sdt=0,19
I wonder why Ganesh Bora continues to credit himself with this work on his official university website? I hope someone from NDSU investigates this.