For the second time in a week, we’ve seen a journal retract a paper because it duplicated something in its own archive. Yesterday, it was a case of plagiarism in a plant journal. Today, we find that the Journal of Anatomy has retracted an article it published earlier this year by a group of Slovenian authors who took a page (or several) out of their 1995 article in the same periodical.
The article, “Muscles within muscles: a tensiomyographic and histochemical analysis of the normal human vastus medialis longus and vastus medialis obliquus muscles,” came from researchers in the department of orthopedic surgery at the University Medical Centre in Ljubjana. According to the abstract:
The aim of this study was to show the connection between structure (anatomical and histochemical) and function (muscle contraction properties) of vastus medialis obliquus (VMO) and vastus medialis longus (VML). The non-invasive tensiomyography (TMG) method was used to determine the contractile properties (contraction time; T(c)) of VML and VMO muscle, as a reflection of the ratio between the slow and fast fibers in two groups of nine young men. VML and VMO significantly (P < 0.01) differ in the proportion of type 1 (59.6: 44%) and type 2b (6.3: 15%) fibers. The VML muscle is almost entirely composed of type 1 and type 2a fibers. In many samples of this muscle no type 2b fibers were found. The proportion of slow-twitch type 1 fibers is nearly twice as high as the proportion of fast-twitch type 2a fibers. These observations indicate that VML is a slower and more fatigue-resistant muscle than VMO muscle. These characteristics correspond to the different functions of the VML, which is an extensor of the knee, and to the VMO, which maintains the stable position of the patella in the femoral groove. Our results obtained by TMG provided additional evidence that muscle fibers within the segments of VM muscle were not homogenous with regard to their contractile properties, thereby confirming the histochemical results. T(c) can be attributed to the higher percentage of slow-twitch fibers – type 1. The statistically shorter T(c) (P ≤ 0.001) of VMO (22.8 ± 4.0 ms) compared with VML (26.7 ± 4.0 ms) in our study is consistent with previously found differences in histochemical, morphological and electrophysiological data. In conclusion, the results of this study provide evidence that the VML and VMO muscles are not only anatomically and histochemically different muscles, but also functionally different biological structures.
But as the retraction notice explains, the article got some performance enhancement, in the form of a pair of tables:
Regarding: Travnik L, Djordjevič S, Rozman S, Hribernik M, Dahmane R (2013) Muscles within muscles: a tensiomyographic and histochemical analysis of the normal human vastus medialis longus and vastus medialis obliquus muscles. J Anat222, 580–587. doi: 10.1111/joa.12045
The following article from Journal of Anatomy, “Muscles within muscles: a tensiomyographic and histochemical analysis of the normal human vastus medialis longus and vastus medialis obliquus muscles” by L. Travnik, S. Djordjevič, S. Rozman, M. Hribernik and R. Dahmane, published in Volume 222, Issue 6, pages 580–587 (available through http://www.onlinelibrary.wiley.com) has been retracted by agreement between the authors, the Journal’s co-Editors-in-Chief Julia Clarke, Thomas Gillingwater, Anthony Graham and Stefan Milz, and John Wiley & Sons Ltd. The action has been agreed due to the fact that Tables 1 and 2 and associated data have already been published in “Histochemical and morphometric characteristics of the normal human vastus medialis longus and vastus medialis obliquus muscles” by L. Travnik, F. Pernus and I. Erzen, published in Journal of Anatomy, Volume 187, Part 2, 1995, pages 403–411, and their use in the current study was unacknowledged.
We regret any inconvenience or harm that this error may have caused.
Raja Gosnak Dahmane, the last author of the paper, tells us:
We agreed with the journal’s decision to retract the article.
The truth is that we were not aware that we have made a mistake since
we published in the same Journal …so it was an improper understanding of copyright!
If our intention was to deceive anybody we should send our work to another
journal. So, in the name of my co-authors, I ask you to save our reputation and be soft when writing your post.We are also planning to re-write the paper (which is really interesting), but properly this time.
Soft, we were.
A snafu at PubMed Central led to the temporary retraction the 1995 paper as well. Julia Clarke, who edits the Journal of Anatomy, tells us the earlier article will be restored. Meanwhile, she says, the retraction marks the first for the publication in at least 22 years.
The retraction notice rather strongly hints at self-plagiarism rather than copyright infringement. While I can readily be ‘soft’ on an authors for being unaware of the latter, the same is not so easy for the former.
Copyright is not infringed; actually you have a choice of about 1/6 of all existing journals where you can republish. No such thing as self-plagiarism does exist, so, missing reference here does not indicate any misconduct. Retraction was totally inappropriate.
‘No such thing as self-plagiarism does exist’
That sounds to me like a controversial statement. Can you elaborate?
The popular, and – official, definition of plagiarism is unclear, usually it’s a form of stealing. Actually, it’s falsification, a FALSIFICATION OF THE FACT OF AUTHORSHIP. One says that he did some original work while the fact is that someone else did it. I many times in many places tried to make this point. Now, obviously, if the fact of authorship is stated correctly, there is no plagiarism in any form or shape.
Answering the next comment by ferniglab: very true, scientific report is about original work. And I agree that republication without reference is absolutely wrong. But it’s very important that it has nothing to do with plagiarism in any shape or form. I believe that in this case the proof of intent to deceive is weak, this can be a missing reference in its pure form; the remedy should have been a correction. A missing reference to one’s own work doesn’t look so bad in the landscape – that is the reason to spare an otherwise good paper.
Is this not a moot point here? Only one of the most recent reports author’s was on the original… If I pass off the work of myself and Dr. A as the work of myself and a Dr. B, I have, by any definition, plagiarized the work of Dr. A.
I didn’t look at the names. The question as you put it, has that your easy answer. However, the official rule says that each author can “claim” the whole work, but this rule is clearly what is called a “legal fiction”. In a similar situation somewhere else, you never know what the 7 authors may tell to their 7 attorneys.
I still believe the editor here made a reckless decision, with no respect for the scientists or science and feeling no responsibility for the consequences. The paper has already been published; retraction is the extreme punitive measure. There are proper ways to deal with a car lacking, for whatever reason, its licence plates too, you cannot set it on fire.
A research article has to be original work. if it uses previously published work without clear and direct attribution, then a reasonable conclusion is that there was either intent to deceive (misconduct) or a mix up of data (so not so good record keeping). In both cases the work is clearly no longer original and so fails the first editorial test at the journal. Retraction is a reasonable outcome.
Interestingly, you are using the word “attribution”, usually – it’s in reference to plagiarism.
Does anyone have a recommendation for an on-line plagiarism checker, preferably one that’s free and good for science articles?
The issue here seems simple, but is complex. First, it is not really possible to plagiarize yourself. You can, however, violate copyright which you have transferred to someone else. But that can’t be the case here since it is the same journal that published both articles.
Instead the issue is that the data (or some of it) was not novel. Most journals (but not all!) require the data in a paper to be new and unpublished. The question then becomes, if they did not acknowledge the reuse of data (and what possible reason could there be for failing to cite themselves?) is there sufficient new data in the article to warrant publication?