Retraction Watch

Tracking retractions as a window into the scientific process

What happens when a correction is retracted?

with 2 comments

Seyed Rasoul Mousavi, assistant professor in the department of electrical and computer engineering at the Isfahan University of Technology in Khomeynīshahr, Iran, has been working on a way to help biologists assemble genomes with as little information as possible.

Last year, Mousavi submitted a manuscript to the Journal of Theoretical Biology, received acceptance after peer review, and got proofs back to edit.

For technical reasons, Mousavi says his corrections didn’t make it into the final article published on Jan. 12, so the publishers issued a separate corrigendum with the editorial changes. Then, on Feb. 8, the publishers retracted the corrigendum that accompanied the original article. The retraction of the corrigendum reads:

This Corrigendum has been withdrawn at the request of the authors. The Publisher apologizes for any inconvenience this may cause. The full Elsevier policy on Article Withdrawal can be found at

In an email, Mousavi gave us more details of what happened:

The story is that when the paper was accepted and I was asked to do the paper proof. I was supposed to do it online using Adobe Acrobat. For some technical yet unknown reasons, the changes I made were not visible to the typesetters and they missed many of the required changes. Then, when I noticed the paper was online without my requested changes being applied, I asked why they did not make my requested changes and then we noticed that they had not seen my requested changes online. They also said that, because the paper was just online, I had to make a Corrigendum for the missing changes! However, they said that, after a while, the corrigendum would be retracted and the missing changes in the paper proof would be applied to the paper. So, that’s what happened as far as I know. In summary, the content of the Corrigendum was only the missing amendments requested at the time of paper proof.

He added that the original correction included mostly fixing details.

Nature of changes: sort of changes I usually make at the proof stage. For example, several of the changes were simply because of a wrong publication year in two of the references (2005 instead of 2010 and 2008 instead of 2010, as far as I remember), which in turn required amending these years in the text. Few typos in the text and algorithm; there were not too many changes and the corrigendum was too short. Some closing parentheses were also missing. Such changes never affect information/conclusion of course.

Elsevier’s Janet Stein, journal manager at the Journal of Theoretical Biology, took care of the case and further explained in an email:

As I recall, the author sent me an email stating that his corrections had not been implemented in the online version of the manuscript; when I checked, I discovered that he was correct, and I processed a Corrigendum.  A few days later a version of the manuscript appeared online which contained the author’s changes. I am still unclear about exactly what occurred; however, as the online version of the manuscript was correct, the Corrigendum was retracted.

To my knowledge, no other situation like this has occurred. The author is satisfied, and the online version is correct.


I was concerned why I need to make a corrigendum while I have already specified the changes in the paper proof. However, I knew that the changes were so minor and should not be so important. (I did not know much about detailed publication rules such as corrigendum and/or retraction.)

We haven’t seen a retracted correction before, although we’ve seen a retracted retraction. That one turned ugly.

But this one seems pretty straightforward, and Mousavi is used to editing challenges, namely with his first name, which came in three different versions during our communications: Seyed (on the university website), Sayyad (in the journal) and Sayed:

Years ago, I thought ‘seyed’ was the right one, but I now understand that it should sound like Sayyed or at least Sayed. I have decided to use Sayyed. However, Seyed has been previously used, for example in my emails; Sayed is a solution in between these two.

Hat tip: Clare Francis

  • omnologos July 23, 2012 at 10:02 am

    You’ve spoken with somebody from Iran? Does Apple know about it already?

  • Anonymous July 25, 2012 at 11:27 am

    This is extremely common at Elsevier. They hire typesetters that are unfamiliar with English, with scientific writing, and with even elementary mathematics. One of my recent publications in an Elsevier journal involved 5 proofs and over 300 corrections. Something like 290 of these corrections were not errors in the submitted source files that they simply had to compile. I recommend that anyone publishing in Elsevier journals request a 2nd, 3rd, 4th proof until there is a version in which no typesetter-induced errors can be found. Otherwise, one’s left in the situation of the author above.

  • Post a comment

    Threaded commenting powered by interconnect/it code.