This post is a summary of the responses I have received from journals after they were provided with the evidence for five cases of the re-use of data (self-plagiarism) in seven of Francesco Stellacci’s papers on “stripy nanoparticles”. First, it is important to establish what are the rules of the game.
I discussed the rules for students at MIT in an earlier post, “Well over the line: an update”. To recapitulate, it is stipulated that if data are secondary (so derived from a source other than one’s own laboratory work), then the source must be cited. This means alongside the data, not in the text, so in a legend as in “data from reference #30”. EPFL, where Francesco Stellacci is currently employed, has very clear and explicit rules for its students and staff, discussed in an earlier post. These very correctly state that plagiarism from whatever source, so including self-plagiarism, is contrary to academic practice and punishable.
What about journals? It seems pretty simple there too. First, is the stipulation in the Instruction to Authors regarding the fact that a research article or communication is not a review and the data have to be novel. Re-used data cannot meet that criterion and the source has to be cited. Second, many journals sign up to the guidelines of the Committee on Publication Ethics (COPE), which quite reasonably state that plagiarism is not acceptable practice. This includes five (Chem Commun, JACS, J Phys chem., Nat. Mat. PNAS) of the six journals where data has ben re-used by Francesco Stellacci
All journals were contacted between 17 December 2012 and 31 December 2012. The order of the responses simply reflects the date I contacted them.
Journal of Scanning Probe Microscopy (JSPM) 4, 24–35, 2009 has two figures taken from two previous ACS publications, J. Am. Chem. Soc. 2006, 128, 11135-11149 and J. Phys. Chem. C 2008, 112, 6279-6284.
JSPM responded same day, but very disappointingly, to quote:
“Unfortunately, publishing redundant results is too common nowadays, and there is virtually no way to control it.”
JSPM one should note has not signed up to COPE guidelines. There is, of course, quite a lot one can do about “it”, as an editor, from a correction to a retraction. No action is disappointing.
J. Am. Chem. Soc. 2006, 128, 11135-11149 not only suffers from a figure being used in the above JSPM article, but of part of a figure being re-used within the same article. The editorial office replied in due course:
“Thank you for alerting us to your concerns on this matter. We’ll look into it and take appropriate action if needed.”
J. Phys. Chem. C 2008, 112, 6279-6284 has a figure re-used by the above JSPM article. Their response was:
“Thank you for your concern, and I will take it from here.”
Fair enough for both of these ACS publications. In two instances they are the source of plagiarised material, so they are blameless. The issue of re-use of data within the same paper may be more difficult for them to deal with, since it highlights a failure of the review process. All journals rely on reviewers doing their job, but this is not always the case and journals cannot shoulder all the blame for such failures.
Nature Materials 2004 3:330-33 has a figure re-used four years later in PNAS 2008 105: 9886–9891. As the source of plagiarised material they cannot take any blame here.
Nature Materials, 2008 7: 588-595 uses a figure published six months earlier in Chem. Commun., 2008, 196–198. This was the slowest response, due to my original e-mails going into a different sub-folder (see here):
“Thank you for the note, and also for that of Dec 16th (which unfortunately got automatically filtered into a subfolder, which we look at less often). Indeed, it seems that proper credit to the original images that you refer to would be needed in the captions of the figures where they have been re-used, not only in the main text of the paper. We will look into this issue.”
At one level this is fair enough. In any event we are assured of some sort of action here, the appropriateness of which will be for the community to judge.
PNAS 2008 105: 9886–9891 has a figure from Nature Materials 2004 3:330-33. That this is data re-use has been disguised by changes in contrast, colour balance and scale, but is pretty obvious if you actually look at the figures carefully. Raphael has a post on his blog that makes this fact crystal clear. The initial response from a managing editor at PNAS was rather poor.
“Thank you for bringing this to our attention. We have taken a look at the figures, and while they are similar, they are not identical. Also, the authors have referenced the earlier Nature publication in their PNAS article.”
After pointing out to PNAS that the images are indeed identical and directing them to check out the images side by side, with contrast, colour balance and scale adjusted, on Raphael’s blog, they came back on a much more positive note:
“Thank you for providing a link to Raphael Levy’s blog posting. Based on this additional information and clear explanation of the issue, we will follow up with the authors to discuss the matter.”
One assumes that they also agree that referencing a paper in the text is not the same as stating in the figure legend that the data are from another paper (true at MIT, EPFL and in journal guidelines) and, moreover, that one also has to obtain copyright clearance from the primary publication. The alterations made to the figure published in PNAS really do look like an attempt to disguise the self-plagiarism of the data and I hope they examine this aspect too. It was certainly sufficient to take in a managing editor at PNAS and I also had to extract the published figures and inspect them carefully to convince myself. I look forward to seeing the outcome of PNAS’ discussions with the authors, as I am sure do all the readers of this blog.
Chem. Commun., 2008, 196–198 has a figure that is re-used in Nature Materials, 2008 7: 588-595. The response was:
“I have reviewed these two papers and I can confirm the Chem Comm paper was published first, in October 2007, and the Nature Materials paper in May 2008. Therefore, if the figures are deemed similar enough, it should be the Nature Materials paper which should have an appropriate reference to the first figure. I advise you contact Nature with the information you have provided; which hopefully they can then resolve.”
One might view this as passing the buck, but it is in fact, correct. As I noted above, the source of plagiarised material is entirely blameless and it is the plagiarised material that is the problem. However, one would hope that in this instance and in those of the two ACS publications and the NPG publications whose data have been subsequently re-used that the publishers of the original data would request full acknowledgment of the re-used data and that the publishers of the re-used data will also obtain copyright clearance.
It is rather disappointing that we have clear rules on copyright, plagiarism and self-plagiarism, yet we apply these selectively. Selectively, because rules on copyright are seem to be enforced only when there is a perceived loss of revenue and rules on plagiarism are only enforced universally for students. On a positive note, the new rules at EPFL that I described at the start of this post and those soon to enter the ordinances at the University of Liverpool are excellent responses to the problems of plagiarism and misconduct and bring staff and students under the same regime. One hopes that these represent the start of a trend that will see such rules widely adopted and enforced amongst both research institutions and journals.