How to Handle Plagiarism Without Destroying the Author

Dear Colleagues,
Cindy Mulrow and I are struggling with how to handle an episode of plagiarism. After publishing an article, we heard from the author of a related paper. This person noted a striking similarity between the structure and some of the language of several paragraphs in the Discussion of his paper and several paragraphs in the Discussion of the paper that we had published. Our correspondent did not question the accuracy of the observations in the paper that we had published. We verified his observation and then asked the dean of our author's medical school to investigate. The investigation disclosed that our author had inserted language from the earlier paper and used the structure of several of its paragraphs. He and his supervisor apologized formally to the authors of the earlier paper, who accepted the apology. The dean of the medical school formally disciplined our author. All parties felt that the act of plagiarism had been the result of inexperience and bad judgment rather than dishonesty.

Cindy and I don't want to do more to punish our author, but we also sense that we need to notify readers that this article contains material lifted word for word from someone else's work without acknowledgement. The plagiarized material was accurate, so it's not a matter of correcting an error. The article itself was valid, so it's not a matter of scientific fraud and withdrawing the paper.

We appreciate any advice that helps us think this problem through to a satisfactory resolution.

Thanks in advance,
Hal Sox and Cindy Mulrow

___________________________

First let me applaud you for pursuing this to the level of reporting the episode to a medical school dean, and their having a formal investigation. That is an unusual level of thoroughness and due process, and one which I would like to see emulated more often.

WAME policy suggests that responses to scientific misconduct be titrated to "fit the crime". In this case, there has been a formal apology and formal discipline, and all feel that this was more of a mistake than a deliberate transgression.

In that case, it might appear there is no benefit to publicly chastising the transgressor further. If you think there is a benefit, then you should publish a notice to that effect. However, it seems to me that it might be equally appropriate to publish a notice in your journal, simply reporting the fact that in a recent article such and such occurred, had been reported to the medical school, and there had been a formal apology and discipline—ie, pretty much what you report in this WAME message. You could then add that since this was an error of inexperience rather than malice, you have decided not to specifically identify the author or article, as there is no error involved, but that you would reserve the right to do so in less benign circumstances.

No doubt that will seem too little to some, but it goes a LOT further than most such cases and serves public notice to others what the consequences of such an action might be. It will also help educate the uneducated—you might even wish to expand this a bit further to explain to readers what the definition of plagiarism is. In this era of Web searches and cut and paste, many perfectly well-meaning authors and scientists actually perceive no problem in using others' words without attribution.

Michael Callaham

___________________________

The case sounds like all-too-common laziness rather than heinous dishonesty. The latter should result in public disclosure and humiliation, but with what we have here, publicly naming the individuals could be almost career-ruining, which I think would be very disproportionate to the crime. It's interesting how this parallels the (depressingly) most common form of student plagiarism. Often students take screeds of notes from their readings and then under pressure of deadlines, pour their notes into their assignments, "forgetting" that what they are adding was not their words. These habits seem to persevere in some.

Simon Chapman
Tobacco Control

___________________________

I agree with the 3 comments posted earlier. It would not be appropriate to name the transgressing author. I would like to add that, for authors who are not native speakers of English, defining plagiarism can be extremely difficult. Sentences may be plagiarized not for their contents but for their syntax. Because appropriation of proper syntax cannot be considered plagiarism, the author can honestly feel that he/she didn't do anything wrong at all. I feel that this type of problem is inevitable when much more than half of the world's scientists must write in a language that is not their own. A theoretical solution would be to screen all submitted manuscripts for potential plagiarism, but I cannot think of any practical way to accomplish this.

Roberto Refinetti
Editor-in-Chief, Journal of Circadian Rhythms

___________________________

Seems to me that the editors should publish a note under a Corrections and Additions section (almost every journal must need one) or a statement from the original unreferenced author in the letters section. Either should identify or reproduce the involved paragraphs and state that they were from an unattributed source and then give the original reference.

I doubt that any more is called for after the action already taken.

Wallace Sampson
The Scientific Review of Alternative Medicine

___________________________

I once wrote the following: On detecting plagiarism in students' assignments: We have the following JISC service:

Plagiarism Advisory Service
http://www.jiscpas.ac.uk/, http://www.jiscpas.ac.uk/apppage.cgi?USERPAGE=6238, and http://www.jiscpas.ac.uk/apppage.cgi?USERPAGE=6588

It is based on the famous turnitin service: http://www.turnitin.com/

The effectiveness of these services in detecting plagiarism depends on the coverage of the underlying database(s) they are checking the submitted manuscripts against. One can also use Google/Google Scholar: by selecting from the document in question a sequence/phrase made of 10 words (Google's maximum) that looks suspicious, and searching for it as a quoted phrase in Google. Again, you are limited to what Google is covering in its index (though Google now covers PubMed and PubMed Central and a huge number of PDF documents on servers of UK and US academic institutions). Also, the selected sequence should be "unique," ie, not a common phrase.

Other signs of plagiarism include:

1. Unexplained switching between UK and US spelling in the same document (some segments written in British English and others written in American English);
2. Uneven writing style (alternating between a poor English style and that of a fluent native speaker, with no explicit attribution) suggests borrowed paragraphs;
3. No or very few references cited when you are expecting more;
4. Segment pasted from the Internet into Microsoft Word sometimes retain their unique HTML formatting, e.g., table and other markup; and
5. The level (language and research matters) of the manuscript is much higher than the known student level (from previous work submitted by the same person). It should be noted that students sometimes rely on the services of professional copyeditors to "polish" their manuscripts (this is usually allowed if the copyeditor is only working to improve the style and flow of the document).
6. See also: http://www.millikin.edu/staley/plagiarism.html, http://plagiarism.dal.ca/, and http://plagiarism.phys.virginia.edu/Wsoftware.html

Maged N Kamel Boulos
Lecturer in Healthcare Informatics
School for Health, University of Bath

___________________________

I am unable to give useful advice on the current issue but this experience is yet another thing that tells us all that people need to be trained and reminded all the time about the dos and donts of medical writing. Let us train the upcoming researcher on the art, rules and regulations of scientific writing.

Adamson Muula
Malawi Medical Journal

___________________________

I agree with the several other commentators that, so far, this case has been admirably handled. I also agree that it would be disproportionate to 'name and shame' the miscreant.

But I'd like to focus on a perspective that, until now, seems to have been missing from this discussion, namely that of the reader. You do not mention whether the earlier paper was cited (as well as copied) in the later piece (although you imply that it was not). If it was cited, then I suggest no further action is called for, but if it was not, then one must ask whether readers (as well as the author of the original piece) would be helped by a citation.

Would it be disingenuous to offer a correction stating that the reference [+details] was omitted from the paper?

I appreciate that this is not the whole truth, and a few diligent readers might even spot the plagiarism, but this solution at least provides a link to the original work and recognition of the original author without a public dressing down.

If you did choose this course of action, you could also report the case as you suggest, or simply publish a note reminding potential authors about the problems of plagiarism without mentioning what prompted it.

Liz Wager
Medical Writer and Editor
Sideview

___________________________

I regularly share with collegues on my staff comments made within this Group. I find the problems raised and the suggested solutions most enriching. Here is a reaction from Eva Ndavu:

"Dr Morfaw,
It is difficult for me to understand how someone publishing medical information or research could plagiarize due to "inexperience and bad judgment". There are ethics in publishing just as there are ethics in medicine. An unpublished author submitting a document for publication for the first time should take special care in using only original material and giving due credit for any words and ideas which are not original. Whereas the editor would want to encourage new authors and researchers, certain standards must always be followed. It is surprising in the example given that none of the reviewers discovered the plagiarism. The editor in this case has obviously taken a humane approach. The author is lucky and hopefully has learned from the 'mistake.'
Eva R. Ndavu"

Fidelis

___________________________

I'm glad you've mentioned the link between plagiarism and non-native speakers of English who are just starting to write in this language. I've helped so many such writers of English who were novice writers to avoid plagiarism that I know what you say is true. Many novice authors simply don't know the citing protocols, as they're fresh out of academic experiences in which just tacking on a list of references is enough to satisfy their teachers. And I've found a few such cases of "translated plagiarism" for a journal whose cover-to-cover translation I supervise.

In all the cases I've found, we've been able to save the novice author and the journal embarrassment by explaining how citing must work to the author. I've never had a case in which the author was trying to put something over on anyone. As you say.

You say it's unreasonable to screen for plagiarism, but I think it might be easier than you think. At least, I have a sense of when it's happening and I think it would be even easier for proper peer reviewers/editors to develop because they are familiar with their literature in a way I'm not. (I'm an author's editor, translator, teacher of English for Specific Purposes, and researcher in writing—I work with several medical specialties so I can't be highly familiar with one.)

I know I'm looking at a candidate paragraph to be checked for plagiarism 1) when a whole paragraph reads too smoothly—most novice manuscript authors aren't that good; 2) when premises, principles and facts are being interwoven beautifully in succession, with only a single reference after a series of sentences, or 3) when the flow of information (from the ends of sentences to the beginnings of others) changes very suddenly—this means that chunks are being copied without paraphrasing, although citations may be properly in place.

I know this last type is less serious in scientific writing, judging by studying published articles, but it's simultaneously an editing/readability problem that novice writers must learn to handle and editors need to notice. In these cases, editing to improve the flow of information has the virtue of obliging the author to paraphrase from the source s/he's citing.

M. E. Kerans
Barcelona, Spain

___________________________

We had a similar case at our journal, in which we invited the authors of the plagiarizing article to come to our office and explain. They proved that the text of the article was used, that their own data were right and they confessed not to have cited the original article. We published a correction, signed by all authors, stating that they had used the first article extensively and omitted to refer to the article. Would that help you?

John Overbeke

___________________________

I agree that this looks like all-too-common laziness and that public humiliation is very severe. Yet, if there is no way to notify the community that the something has been done about it, the plagiarism involved will be discovered and rediscovered by others, and give the impression that the journals involved do not care. Perhaps there could be a category of scholarly transgression, similar to "erratum", that allows public notification without the overt humiliation. A yellow card instead of a red card. An embarrassment, but not a career-stopper. Perhaps if the community had a standard mechanism for dealing with non-heinous disclosures, they would NOT be career-stoppers, but help diffuse community standards. Moreover, without such a mechanism, the failure to cite properly is perpetuated in print.

For example, the offending writers could publish a "erratum" that says "Portions of document X were inadvertently cited unskillfully, and should be attributed to references Y and Z.

It seems like the error described is at least as serious as the mislabeling of a gel, and deserves equivalent correction.

John Rodgers

___________________________

I think John Rodger's suggestion has great merit... this is indeed a 'yellow card' issue. Readers need to know that journal editors care about such transgressions...

I B Pless
Editor, Injury Prevention

___________________________