Can self-retraction motivate researchers to correct the scientific record?

In the recent years, the number of papers published each year globally is so high that it would not be wrong to say that we are witnessing a publication boom. While one of the primary motives behind publishing a paper is getting the word out on the latest findings, equal emphasis is not given to correcting the scientific record. Issuing corrections is one way of amending published research. Another way of alerting readers to questionable or invalid research is, of course, retractions.

Often times, retractions are issued by journal editors when they detect a serious flaw in papers that overrule the credibility of the findings. Due to the fact that most retractions stem from misconduct on the authors’ part, retractions have earned themselves a bad name. Close to 600 retractions are issued per year, and less than 20% of these can be attributed to honest errors. However, there have been several instances in the past few years wherein authors have themselves stepped up and accepted that their paper has major problems, and have requested for a retraction to the editor; and they have been applauded by the science community for this. Nevertheless, the stigma around retractions remains strong, deterring most authors from openly disclosing errors in their published work. Thus, when nobody but the authors come to know about a problem in their paper, and they are reluctant to retract it, the scientific record remains uncorrected. The self-correcting nature of science, though highly valued, is held back due to the generalized negative opinion about retractions.

Want to avoid article retractions? Join us for an expert-led webinar on April 16, 2024. Find all the details about the event and registration here. 

Some time back, Daniele Fanelli proposed a self-retracting system that will make it clear when authors pull out papers due to honest errors as against misconduct. He suggests that just like self-citations are considered a category in itself, self-retractions should be classified as a type. If all the co-authors of a paper spontaneously request a retraction due to a “documentable flaw,” journal policies should allow them to file a self-retraction. Conversely, if the retraction is based on misconduct, authors should not be allowed to sign the notice, making it evident that the retraction is not initiated by the authors and is not an ‘honest retraction.’ This suggestion has been welcomed by many researchers and even editors.

The one hitch in this proposal is: what if researchers volunteer a self-retraction to cover misconduct? To this, Fanelli says, “In the worst-case scenario, it would be only authors who have falsified one or two papers who might benefit from dishonestly self-retracting. Should that be considered a problem?” This is agreeable, but it only hints that developing policies around this would need a lot of consideration on the journals’ part; such as setting up a way of confirming whether a self-retraction is based on honest mistakes or not, steps to take if one of the co-authors is reluctant to sign the retraction notice, and so on. However, if self-retraction of papers becomes possible, it would be a positive and encouraging step and is certain to benefit science in the long run.  

Retractions are central to academic publishing, but they have been too journal-centric and stereotyped. They are viewed as punishments and can have a far-reaching effect on researchers’ careers, and using them without any distinction has worked against the correction of the scientific record. For long there have been discussions around making retractions more author-centric rather than it being a tool used predominantly by editors to announce misconduct. Thus, a self-retraction system might be the right step forward in this direction.

What are your views on this issue? Please share your thoughts and opinions in the comments section below.  

Recommended reading:

• Retraction notices: Stepping stones to science's self-correcting nature
• Should plagiarism lead to retraction in all circumstances?
• Is the journal retraction rate rising?